1.1 Objectives

This paper investigates a selection of seven Finnish verbs of ability (VoAs). They can be seen as belonging to the category of the so-called semilexical or semimodal verbs in that they serve as both (modal) auxiliaries and (lexical) main verbs (see e.g. Dixon Reference Dixon2010, Depraetere & Langford Reference Depraetere and Langford2020). In this paper, we highlight their shared semantic capacity of encoding the modal meaning of dynamic possibility, namely ability (see e.g. Palmer Reference Palmer2001) and conveying additionally a specific lexical layer that determines in terms of what the subject referent is able to do something. We suggest that due to this twofold semantic structure, rather than being halfway between two categories (modal and lexical), these verbs can help us to deepen our insight into the embodied roots of the meaning construction of ability.

The paper aims to provide a grammatical description of each verb and explore their mutual interfaces. Their modal meaning is analysed in terms of conceptual force dynamics (Talmy Reference Talmy2000). To account for the internal organisation of the modal field of ability and the dynamics of the Finnish VoAs within it, we metaphorically employ the term ecology. We also speak of the niche of each verb when describing the processes of semantic–functional specialisation and cohabitation in the realm of ability (for a previous use of the terms ecology and niche within a cognitive linguistic framework, see Taylor Reference Taylor, Radden and Panther2004). The study is strongly corpus-driven. This allows us to track the most frequent patterns for each verb on several different levels (grammatical, lexical semantic, force-dynamic) and observe how these map onto each other. The data include standard and colloquial (mostly dialectal) language use, extracted from both written and oral language datasets.

The selected verbs can all enter a transitive construction with a so-called A-infinitive complement. The A-infinitive can be considered as the basic form of the Finnish verb. The selected seven verbs are presented in Table 1 (examples 1a–g).Footnote ¹ We illustrate their semantics and use with the utterance presented in (1), meaning ‘Liisa cannot eat in Helsinki restaurants’. Each of the studied ability verbs can occupy the position of the modal marker (‘can’). Because of the negative auxiliary ei, they take the connegative form, as seen in column A (VoA.cng). The infinitive form for each verb is presented in column B, associated with a code in subscript that links the infinitive to its meaning, shown in a simplified form in column C.Footnote ² All the verbs encode ability, but they account for the event described in (1) differently. To make manifest the lexical semantic contribution of each verb to the propositional content in (1) an additional explanation is added in column D.

Table 1.

The studied VoAs

These fabricated examples offer a first glimpse of the differences in use and in the lexical meaning displayed by the selected verbs. The first three verbs, sietää _TOLE, suvaita _TOLE and tarjeta _COLD, express different types of tolerance. Tarjeta _COLD encodes specifically tolerance to a cold temperature. The meaning of jaksaa _NRG has to do with energy and effort. The remaining three verbs, tohtia _DARE, malttaa _CTRL and raaskia _SENS, highlight the emotional roots of ability in different ways.

Other types of complements are possible, e.g. an NP complement in (1h), along with intransitive uses (1i). These patterns appear to be verb-specific. Example (1j) illustrates that tarjeta _COLD does not take an NP complement.

The most extensive study focusing specifically on VoAs in Finnish, to our knowledge, is Flint (Reference Flint1980). Flint identifies 45 ‘verbs of possibility and sufficiency’ that display similarities in their meaning construction. For example, they exhibit negative affinity (see Kiuru Reference Kiuru1977). According to Flint (Reference Flint1980:59–60), the verbs form an open-ended semantic system which is internally organised around twelve different gradually overlapping dimensions. One of these is ability. Yet, in Flint’s model, none of the verbs studied here belongs to this dimension. In this paper, we apply a broad definition of ability. We consider that in addition to the prototypical ability verbs (e.g. voida, pystyä, kyetä ‘can, to be able to’), the selected VoAs, too, encode the possibility for the subject to (not to) do something because of participant-internal or -external circumstances. This causative semantics can be accounted for within the theoretical framework of force dynamics (Talmy Reference Talmy2000). It is a model for describing embodied interactions between competing conceptual forces. Each VoA adds to the force-dynamic pattern a specific lexical component specifying the nature of the resisting force. Three types of force-dynamic interaction are distinguished: physical, mental, and interpersonal.

The paper proceeds as follows. We first present the theoretical and conceptual foundations of the study (Section 1.2). After this (Section 1.3), we provide an overview of the (very few) previous studies concerning the Finnish VoAs, mainly Flint (Reference Flint1980) and Kiuru (Reference Kiuru1977) (see also Cai Reference Cai2025), account for the selection of the studied set of verbs, describe their syntactical properties (complementation, zero person) and negative affinity, and briefly place the Finnish system in the context of the VoAs in other languages. The following two sections present our research questions and hypotheses (Section 1.4.), as well as the data and methods of analysis (Section 1.5). We then move on to exposing the results. We start by comparing the studied VoAs on three grammatical points, including complementation, negation, and person marking (Section 2.1). Based on this corpus analysis, we then zoom into the particular uses of each verb, detecting the specific types of conceptual forces that are mapped onto the underlying force-dynamic pattern (Section 2.2). After this, we highlight the specific niche each verb occupies within the ecology of Finnish verbs and the modal realm of ability (Section 2.3). The outcome of the study is then discussed in view of its implications to the Finnish system of VoAs and, more generally, to the meaning of ability (Section 3), before the conclusion of the paper (Section 4).

1.2 Theoretical and conceptual framework

As expressions of dynamic modality, the studied VoAs can be paraphrased by ‘it is possible for X to p’, as opposed to epistemic modals conveying the meaning ‘it is possible that p’ (compare examples 2 and 3) (e.g. Perkins Reference Perkins1983, Palmer Reference Palmer2001, Collins Reference Collins2009).

Ability has been defined as ‘the physical, intellectual, and perceptual capacity to do something’ (see Leclercq & Depraetere Reference Leclercq and Depraetere2022). The meaning of ability implies a potential obstacle, the effect of which on p is cancelled. The utterance presented in (2) entails that what could prevent the subject referent from lifting the rock (e.g. insufficient muscular strength with regard to the rock’s weight) is neutralised. Due to their causative semantics, VoAs thus specify that the reason why p is not cancelled is because the subject referent has sufficient resources at their disposal to overcome the challenge (Flint Reference Flint1980:60–62; see also Nadathur Reference Nadathur2016, Johansson & Nordrum Reference Johansson and Nordrum2018).

It should be noted that ability entails not only the capacity to act (see the verb to do in the above definition) but also the capacity of resisting a change (‘not doing’) (see Figure 1(b)) and, when expressed with a nominal object complement or an intransitive construction, the capacity of overcoming unspecified effects that the interlocutor is expected to identify. For example, ‘(not) tolerating Helsinki restaurants’ entails that the subject referent is (not) able to overcome the unnamed negative effects of Helsinki restaurants. On the other hand, Malta! ‘malttaa_CTRL.imp.2sg’ > ‘Hold on!’ calls for the subject referent to resist some implied opposing force that is liable to make them act (see also examples 1h and 1i). Including in the analysis different types of constructions expressing ability allows us to contribute to research endeavouring to link the force-dynamic framework to the analysis of structurally diverse constructions (Hilpert Reference Hilpert2025).

Sensitivity, self-control, daring, tolerance, and regulating one’s energy encoded by the studied VoAs involve embodied properties characteristic of living beings (cf. ‘intellectual and perceptual capacity’ in the above definition of ability). The prototypical subject referent of these verbs is thus expected to be animate (see also Kiuru Reference Kiuru1977, Gregersen Reference Gregersen, Jensen and Schack2024:103). In this paper, VoAs are approached as expressions of different types of embodied interactions between the subject referent and their environment. We analyse the causative meaning construction of VoAs within the framework of Talmy’s (Reference Talmy1988, Reference Talmy2000) theory of force dynamics. It is a model for describing interactions between competing conceptual forces that stem from our perception of the interplay between physical forces in our environment. This framework makes it possible to link the conceptual core of modal meaning structure with the embodied experience of living beings within their environment (Hilpert Reference Hilpert2025:2). The experience of ability emerges from these interactions.

In the basic schemas, the central force called Agonist is challenged by an opposing Antagonist force. When the Agonist shows tendency to action, the Antagonist tends towards rest, and the other way around. Figure 1 exposes the modal meaning of ability represented by two force-dynamic schemas underlying the meaning structures in utterances I can lift that rock (Figure 1(a)) and The satellite is able to stay in an orbit around Earth (Figure 1(b)). The Agonist is represented by a circle; the Antagonist is indicated by a concave figure. The arrowhead marks an intrinsic force tendency to action (including movement or change of state), whereas the black dot represents tendency to rest (non-movement or lack of change of state). A plus is placed on the stronger entity, and the line below the Agonist shows the resultant.

Figure 1.

(a) I can lift that rock. (b) The satellite is able to stay in an orbit around Earth.

Figure 1 Long description

Figure 1(a) - The icon features a circle with a greater than symbol (>) and a plus sign (+) inside. The greater than symbol typically indicates a comparison or a positive condition, while the plus sign often signifies addition or a positive attribute. This combination suggests a positive or greater than condition, often used in contexts related to comparisons, evaluations, or positive outcomes. Figure 1(b) - The icon features a circle with a plus sign (+) and a dot (•) inside. The dot indicates an intrinsic force tendency to rest, while the plus sign signifies the stronger entity in a force-dynamic relationship. The icon also features a concave form. It symbolizes the opposing force that resists the tendency to rest. Below the circle and the concave form, there is a horizontal line with a dot in the middle. This signifies that the tendency to rest overcomes.

In Figure 1(a), the Agonist force tends towards action, the Antagonist resists, but the Agonist overcomes, and the resultant is action. The subject referent’s physical forces are sufficient with regard to the weight of the rock. The inverse configuration is also possible. In Figure 1(b), the Agonist tends towards rest, while the Antagonist is inclined towards action. The Agonist overcomes, and the resultant is rest. Earth’s gravity is not strong enough to move the satellite from its orbit. The aspectual semantics of the complement (lift vs. stay) encodes the tendency towards action or rest of the subject referent.

In the negative form, the force relationship is reversed. For utterances presented in Figure 1, this entails that the Antagonist force overcomes and the resultant is contrary to the Agonist’s intrinsic tendency. The altered configurations are shown in Figure 2.

Figure 2.

(a) I cannot lift that rock. (b) The satellite is not able to stay in an orbit around Earth.

Figure 2 Long description

Figure 2(a) - The icon features a circle with a greater-than symbol (>) inside. The greater-than symbol indicates an intrinsic force tendency to action. The icon also features a concave form with a plus sign (+) inside. The plus sign signifies the stronger entity in a force-dynamic relationship. Below the circle and the concave form, there is a horizontal line with a dot in the middle. This signifies that the tendency to rest overcomes. Figure 2(b) - The icon features a circle with a dot (•) inside. The dot indicates an intrinsic force tendency to rest. The icon also features a concave form with a plus sign (+) inside. The plus sign signifies the stronger entity in a force-dynamic relationship. Below the circle and the concave form, there is a horizontal line with a greater-than symbol in the middle. This signifies that the tendency to action overcomes.

Following Talmy’s model, the original schema describing interaction between an entity and an external force can offer a ground for describing forces of different conceptual levels, also those internal to the focal entity (Talmy Reference Talmy2000:413–414). In this vein, each VoA adds a specific lexical component to the force-dynamic pattern specifying the nature of the competing forces and thus the type of the interaction: physical, mental, or interpersonal. The two last mentioned force interactions correspond to what Talmy (Reference Talmy2000) identifies as psychosocial dynamics. They generalise physical notions to mental and interpersonal domains by metaphoric extension (see also Gibbs Reference Gibbs2005:160–161). Physical force dynamics concerns the subject referent’s bodily abilities to act or to resist, with regard to spatio-temporal circumstances. Mental forces are internal to the subject referent: intentions, preferences, emotions. These patterns represent an internalised dialogue between the subject’s desires (the Agonist force with an inherent tendency to either change of state or rest) and social values (the Antagonist force blocking or spurring the Agonist) (Talmy Reference Talmy2000:431–433). Interpersonal force dynamics occurs in situations of interaction between the subject referent and other animate participants. An analogy is thus created between physical force relationships and interactional pressures resulting from social norms and disalignment in expectations and intentions (Talmy Reference Talmy2000:438–440).

By observing the interaction between forces of a different nature, we aim to gain an in-depth understanding of the embodied roots of the meaning construction of ability as well as the distinctive features of this modal domain. Johnson (Reference Johnson, Newen, Bruin and Gallagher2018:626–627) describes meaning as emerging from the interaction between an embodied organism and its environment, depending on the specific bodily properties of the living being and the patterns for perception and action provided by the particular situation (in Gibson’s [Reference Gibson1979] terms, affordances). Linguistic meaning is rooted in ‘this much deeper and broader embodied process of meaning-making’ (Johnson Reference Johnson, Newen, Bruin and Gallagher2018:626). Lexically rich expressions of ability, such as the Finnish VoAs, convey a subtle description of the interplay between the bodily properties of a living entity and the situational affordances. By looking at the grammatical and lexical profiles of these verbs, we aim to cast light on the different layers in the meaning construction of ability.

To capture the internal dynamics of ability and the ways in which the Finnish VoAs cover it, we use the metaphor of ecology. In doing so, we aim to describe a specific semantic meaning structure (here, ability) as an ever-moving assemblage constituted by entities (here, different types of abilities) that shape each other and that are shaped by the assemblage and by what is external to it (on the origins of the theory of assemblages, see Deleuze & Guattari Reference Deleuze and Guattari1980), rather than as a category with a predetermined, fixed structure. The multilayered modal domain is metaphorically represented as a landscape on which the linguistic items (here, the Finnish VoAs) cohabit. The aim is to account for the plurality of conceptualised subject–environment interactions constituting the meaning of ability and motivating the relatively high number of ability verbs in Finnish. Approached from another angle, with a focus on the semantic interrelations between the verbs, the assemblage drafted here could certainly be mapped onto an underlying network of force-dynamic scenarios (see Hilpert Reference Hilpert2025; on the usage-based network models of language, see Diessel Reference Diessel2019). We borrow the terms ecology and niche from Taylor (Reference Taylor, Radden and Panther2004), although the focus here is on the semantic layers constituting the modal realm of ability and the location of a set of linguistic items within these, instead of constructional relations.

1.3 Finnish VoAs: State of the art

The lexical meaning of each VoA emerges from the speaker’s perception and experience of the environment and entities within it, but also from the gap between what the speaker thinks the reality is and what it actually turns out to be:

Flint (Reference Flint1980:59–60) enlists 45 verbs of this type in Finnish, forming a system with 12 dimensions, as shown in Table 2.Footnote ³ We have marked the verbs selected for our study in bold letters.

Table 2.

The system of verbs of possibility and sufficiency according to Flint (Reference Flint1980)

In this system, each dimension is internally structured in relationships of adjacency. Sometimes adjacent verbs accumulate into clusters and display varying values with regard to the core attribute of the dimension. For example, within the tolerance dimension, hyväksyä ‘to approve’ and suvaita ‘to tolerate’ display a higher degree of ‘tolerance’ than kestää, kärsiä, and sietää ‘to withstand, to tolerate’. This is liable to stem from the much less specific lexical meaning of these latter three verbs. The different dimensions, in their turn, are related to each other through transitions of varying sharpness: from overlapping (e.g. viitsiä (no. 24 in Table 2) and kehdata (no. 25) which both encode the ability to do something in terms of social norms or personal investment, cf. ‘to dare’, ‘to bother’) to only tangential adjacency (e.g. uskaltaa (no. 31) ‘to dare’ vs. tarjeta (no. 32) ‘to be able to [despite the cold]’).

The dimensions form a conceptually heterogeneous series. The first attributes (I, II) relate to spatio-temporal circumstances, whereas the last ones (IX–XII) stem from the subject referent’s appreciation. In between them, the dimensions III–V are identified with attributes that could be used for describing all the verbs. Indeed, the entire category of 45 verbs is called verbs of possibility and sufficiency in Flint’s study. In our paper, we also consider all these verbs as carrying the modal meaning of ability. One could say that the groups III–V foreground sufficiency, possibility, and ability, respectively, whereas the other verbs combine these with other physical, mental, or intersubjective attributes. The subsequent dimensions VI and VII arise from the subject referent’s strength of different orders. Finally, the meaning of the verbs gathered within the dimension VIII is founded on the presence of an entity exerting some kind of harmful or otherwise unwanted influence on the subject referent (see Godet Reference Godet2023).

In this study, we have selected verbs that are located in the latter half of the series, in other words not highlighting the spatio-temporal circumstances nor the all-encompassing attributes of sufficiency, possibility, and ability. In doing so, we aim to gain insight into VoAs with the most specific lexical meanings involving particular physical conditions (e.g. strength, in jaksaa _NRG, no. 23 in Table 2), emotions (e.g. fear, in tohtia _DARE, no. 28), or environmental settings (e.g. cold, in tarjeta _COLD, no. 32). By choosing to study sietää _TOLE (no. 35) and suvaita _TOLE (no. 36), we can observe the interface between adjacent verbs within a dimension. Looking into the meaning structure and the uses of raaskia _SENS (no. 42) and malttaa _CTRL (no. 43) allows us to investigate a zone that in Flint’s model marks a transition from one dimension to another.

When used with an A-infinitive complement and the past tense, these verbs are implicative, in that they convey an entailment concerning an implied proposition, namely whether or not the event described in the complement took place (Karttunen Reference Karttunen1971, Salminen Reference Salminen2024). This entailment can be inferred from the lexical semantics of the infinitive complement and the modifiers of the finite VoA (TAM, negation, temporal and locative adverbials, causal subordinate clauses) (Karttunen Reference Karttunen1971). For example, the sentence presented in (1k) (modified from example 1) entails that ‘Liisa ate in Helsinki restaurants’. This makes implicative verbs different from non-implicative verbs (1l) and non-factive verbs (1m). The sentences (1l, 1m) do not generate this entailment: it could be that in the end Liisa never ate in Helsinki restaurants, although she had decided or wished to do so. When the finite VoA takes a past tense, as in (1k), the implied proposition is understood as taking place in the past as well (on the role of the past tense in the implicative reading, see Salminen Reference Salminen2024:575–576). When the VoA is in the present tense, it conveys minimal grounding for the process described by the infinitive form, which in itself lacks personal and temporal marking (1n). The implied event is therefore conceptualised as an entailed theoretical possibility (Leech Reference Leech2004:115–118, Visapää Reference Visapää2022). Whether it took place or not remains less determined than in the case of past-time reference (Salminen Reference Salminen2024). When the VoA is under the scope of the negative auxiliary (see 1e in Table 1, ‘Liisa does not dare to eat in Helsinki restaurants’), the underlying proposition is negated too: ‘Liisa does not eat in Helsinki restaurants’. This distinguishes implicative verbs from complement-taking factive verbs. The sentence in (1o) does not imply ‘Liisa did not eat in Helsinki restaurants’.

Given the variation in the complement constructions of the Finnish VoAs, namely the possibility of NP object complements and intransitive uses, which are not analysable in terms of implicativeness (Salminen Reference Salminen2024:585), and what we know of the complexity of the so-called implicative signatures of verbs expressing ability, courage, and effort (see Karttunen Reference Karttunen2012), a closer analysis of this dimension in the selected VoAs goes beyond the scope of the present paper and would deserve a separate treatment.

The studied VoAs belong to those Finnish verbs that have been observed to favour negative contexts (Kiuru Reference Kiuru1977:41–59). In previous studies, this affinity for negation has been accounted for on pragmatic grounds. Kiuru (Reference Kiuru1977:52) notes that verbs of negative affinity are used when explaining and presenting excuses for things not done. Flint (Reference Flint1980:ch. 5) examines a range of types of social situations where these verbs are used in negative answers to invitations and requests, for hedging and explaining politely. Cai (Reference Cai2025) investigates the uses of päästä ‘to be able to go, to access’ (no. 6 in Table 2) with an infinitive complement in spoken everyday conversations showing that the meaning of obstacle is present in the different interactional tasks of the verb. Our study focuses on the embodied semantic roots of the VoA discourse functions, including the motivation behind a possible negative affinity.

Furthermore, according to Flint (Reference Flint1980:16), VoAs are liable to take a zero subject. In these cases, the VoA is in the third-person singular form, but the subject position is left empty. This is a syntactic construction known to be favoured by modal verbs and other types of experience-based verbs (Laitinen Reference Laitinen1995, Reference Laitinen, Helasvuo and Campbell2006). Zero person conveys a non-specific person reference allowing the speaker to present the experience as something that can be shared by anyone in a similar position (ibid.), as in (4) and (5).

Karttunen (Reference Karttunen1971:341) mentions malttaa _CTRL, joutua ‘to have to (go to somewhere)’ and suvaita _TOLE as a part of a language-specific inventory of implicative verbs in Finnish. The existence of a range of verbs of possibility and sufficiency with rather specific lexical semantics appears to be a feature that Finnish shares with Scandinavian languages, at least to a certain extent (Johansson & Nordrum Reference Johansson and Nordrum2018:158). Verbs expressing the sufficiency of time (‘to have time to do sth.’, Flint’s dimension II) have been studied in Swedish (hinna) and Norwegian (rekke) (Johansson & Nordrum Reference Johansson and Nordrum2016, Reference Johansson, Nordrum, Čermakova, Egan, Hasselgard and Rorvik2021). Gregersen (Reference Gregersen, Jensen and Schack2024) analyses uses of the Danish verb gide that are related to the sufficiency of energy and thus come close to some of those displayed by the Finnish jaksaa _NRG. The Swedish orka is part of this group as well (Johansson & Nordrum Reference Johansson and Nordrum2018).

In non-standard Finnish, many of the selected VoAs present phonological and semantic variation.Footnote ⁴ This is also the case for their equivalents in other Finno-Ugric languages. The tolerance reading of sietää _TOLE is widely present in Finnish dialects. Comparable forms are found in the closely related Balto-Finnic languages Ingrian and Karelian. The verb possibly stems from the Germanic lexical base *stē- (cf. e.g. the Scandinavian stå) (SES 2024). Sietää _TOLE can also be used for expressing necessity, e.g. si-tä sietä-ä toista-a ‘dem-part sietää-3sg repeat-inf’ > ‘it ought to be repeated’. In our study, we have included these uses in the tolerance category, considering the necessity reading as secondary (< ‘it bears to be repeated’). In addition to tolerance, suvaita _TOLE and its equivalents are used for describing different types of preference in Finnish dialects and in many Finno-Ugric languages (‘to like, to love’). The form is considered to be of ancient Indo-European origin (*suH-é), as with the permissive verb suoda ‘to allow’ (SES 2024). As for tarjeta _COLD, apart from describing the experience of temperature, it is also reported to express coping with other types of difficulties (related to another person, e.g. at work, in a fight), as well as daring (SES 2024). The Karelian tareta also covers a variety of meanings: ‘to not be cold’, ‘to dare’, ‘to have the heart to do sth.’ (KKS 2024). Variation is observed in the equivalent forms in Estonian and also in the Sami languages (SES 2024). The etymology of the verb is uncertain. One possible origin is shared with the adjective tarkka which in standard Finnish means ‘precise’ but in dialects also ‘sensitive, tender’ (ibid. s.v. tarkka). Jaksaa _NRG is widely present in non-standard Finnish, with a rich array of ability readings (SMS 2024). In a reduced number of dialects, it is used as a communication verb describing fervent, tireless speaking (ibid.). Meanings of the type ‘to take off, to rob’ are also reported in certain regions, as well as in other Finno-Ugric languages. No etymology is provided. Tohtia _DARE displays less semantic variation. The meaning of daring or being brave appears to be present even in its assumed Indo-European origin (*dhers-) (SES 2024). Malttaa _CTRL conveys the meaning ‘to notice, to understand’ in Finnish dialects. This is, of course, related to the meaning of ‘ability to control one’s attention’, presented in example (1f) in Table 1, or that of ‘commitment’, as termed by Flint (Reference Flint1980). An equivalent of malttaa _CTRL is reportedly found in certain Russian dialects due to their contact with Karelian, Ludic, and Veps languages (SES 2024). Inversely, raaskia _SENS is of Slavic origin. It is related to the Russian verb račít´ ‘to want to, to be ready to do sth.’ (ibid.). Raaskia _SENS presents less semantic variation than malttaa _CTRL, although the meaning ‘to dare’ is found in Karelian, alongside with the usual ‘to have the heart to do sth.’.

From a historical and cross-linguistic perspective, the semantic background of the selected VoAs is thus rich in overlaps and vagueness. The present study does not aim to track the evolutionary paths of these verbs. It observes the meaning and the use of the VoAs in a heterogeneous dataset, highlighting the functional niche occupied by each VoA unit. A niche is made available by the rest of the units. ‘Each unit has pointers to other units and can therefore be “filed” under a number of different “addresses”’ (Taylor Reference Taylor, Radden and Panther2004:10). These mutual relationships shape the ecology of language units.

By investigating the somewhat heterogeneous constellation of VoAs in Finnish, we can furthermore enquire into the properties of ability as a conceptual unit. The fact that the Finnish lexicon manifests such subtle distinctions between types of abilities and how they stem from different interactions between sentient beings and what surrounds them sheds new light on the internal structure of what has been considered as a semantic category in the linguistic literature on modality (e.g. van der Auwera & Plungian Reference van der Auwera and Plungian1998, Palmer Reference Palmer2001). The Finnish VoAs raise the question of how we should deal with these elements in view of modal classification. We particularly draw attention to the traditional position of this type of items as halfway in the middle of two categories (semi-lexical or semi-modal) (Dixon Reference Dixon2010, Depraetere & Langford Reference Depraetere and Langford2020) and suggest a model allowing them to enter the modal system as fully fledged members.

1.4 Research questions and hypotheses

The paper addresses the following questions.

1. 1. All VoAs have been described as taking an A-infinitive complement and displaying negative affinity (Kiuru Reference Kiuru1977, Flint Reference Flint1980). As expressions of experience, they form a favourable environment for zero subject (see Laitinen Reference Laitinen1995, Reference Laitinen, Helasvuo and Campbell2006). Do the selected VoAs occur in similar patterns, when it comes to complementation, negation, and person marking, or are there deviating tendencies?

2. 2. The meaning of ability can be visualised with two different schematic models, in terms of force dynamics, underlying the ability to do something, on the one hand, and the ability to resist a change, on the other (Talmy Reference Talmy2000). The forces can be of physical, mental, or interpersonal type. Do the selected VoAs display preferences with regard to one of the force-dynamic schemas and force types? Do the results of the grammatical analysis (see Question 1) map onto these preferences?

3. 3. The chain-like linear representation of the category of VoAs (see Flint Reference Flint1980) has the advantage of exposing in clear terms the particularities in lexical meaning proper to different dimensions of ‘possibility’ and ‘sufficiency’. However, the internal structure of these dimensions and the interfaces between them are rather heterogeneous. Furthermore, the linear form may not allow some of the semantic links between verbs to surface. All of this suggests a more dynamic and complex constellation. Is it possible to describe the coalescence among VoAs through the metaphor of ecology, rather than a linear form? What kind of implications does this have on the semantic category of ability?

Based on the literature presented above, we can draw up the following hypotheses.

1. 1. The selected VoAs do not occur in similar patterns, when it comes to complementation, negation, and person marking. Although the VoAs have all been reported to occur in transitive constructions with an A-infinitive complement and showing a special affinity for negative contexts and zero subject, we can expect them to favour different grammatical patterns due to their different lexical components. They encode different kinds of interactions between an entity and its surroundings. This is liable to be reflected, for example, in verb-specific preferences in complementation.

2. 2. The selected VoAs display different preferences with regard to the force-dynamic schemas and force types, because they conceptualise different kinds of interactions between the subject referent and their environment.

3. 3. The metaphor of ecology allows us to highlight the coalescence among VoAs in multiple ways without the limitations of a chain-like configuration. It also helps us to understand the connections between linguistic forms and the embodied experiences and interactions within which they emerge. Furthermore, this alters the overall view on the semantic realm of ability, which now appears as a complex multilayered assemblage based on the conceptualisation of these multiple subject–environment interactions.

1.5 Data and methods

The data sources are presented in Table 3. The written language dataset includes literary texts from the late nineteenth century to the 1950s (LITER), as well as news texts from the turn of the millennium (PRESS). Spoken language data are extracted from dialect interviews conducted from the 1950s to the 1980s (DIAL) and from a smaller contemporary conversational dataset (CONV). By choosing such a large array of data, including non-standard language use and covering a relatively long time span, we aim to gain visibility over the variation and overlaps in the uses of these semantically complex verbs. Modal constructions are known to have undergone recategorisation during the construction of the present-day standard Finnish, in the second half of the nineteenth century, creating a gap between colloquial language use and the norm (see e.g. Laitinen Reference Laitinen, Cheshire and Stein1997).

Table 3.

Data sources

Table 4 shows the total number of occurrences of the selected verbs, per corpus type. We can observe that jaksaa _NRG is by far the most frequent of the selected VoAs, covering more than a half of all the occurrences, whereas tarjeta _COLD and raaskia _SENS display the lowest frequency. The verbs are distributed unevenly when it comes to corpus types. Jaksaa _NRG and sietää _TOLE are particularly well represented in the news text corpora, whereas suvaita _TOLE is mostly used in literary texts. The uses of tarjeta _COLD mainly come from the dialect data. Explaining this variation is beyond the scope of this paper, but we can assume that it results from several intersecting factors, e.g. the situations typically described in each type of oral and written texts, language register, the status of the language users.

Table 4.

The total number of occurrences of the selected VoAs

This overall dataset is dominated by written language use, which is reflected in the total number of occurrences per corpus type. For this reason, and for the needs of the manual annotation, we have reduced the amount of written data for some of the verbs. Table 5 presents the data that we used for the analysis for each lexeme.

Table 5.

The number of VoA occurrences selected for the analysis

The aim of this operation was, first, to respect a balance between written and spoken data. In the reduced dataset, 2008 occurrences were extracted from written data, 2172 occurrences from spoken data. Second, this limitation was necessary in view of the in-depth grammatical analysis, which was partly conducted manually. The omitted occurrences were selected using a random search in the written dataset. We expect that the results obtained from this reduced dataset are representative of tendencies present in the larger corpus.

The two authors performed manual annotation with double verification in the selected dataset. All occurrences were entered into an Excel spreadsheet. For each collected occurrence, the authors recorded the following variables.

• • Syntactic construction: transitive/intransitive, the type of complement in transitive constructions.

• • Negation, identified through the negative auxiliary, a connegative form of the main verb, negative polarity items (e.g. pronouns, particles), and syntactic cues (e.g. partitive case in certain complements).

• • Modality: mood, modal verbs, modal particles and adverbs, interrogative, directive, and exclamative constructions.

• • Person marking, including the zero person and the passive.

• • The type of conceptual force: physical, mental, intersubjective, identified through the lexical semantics of the A-infinitive complement, the degree of animacy of the NP complement, and the meaning of other grammatical and lexical items in the immediate context.