On the Meaning of the Lithuanian Preposition po: A Cognitive Perspective

Acta Linguistica Lithuanica 2026, vol. 94, pp. 1–19 Contents link / Turinio nuoroda ISSN 2669-218X (online / elektroninis) Copyright © 2026 Ieva Stasiūnaitė. Published by the Institute of the Lithuanian Language. This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. // Išleido Lietuvių kalbos institutas. Šis straipsnis yra atviros prieigos, platinamas pagal „Creative Commons“ priskyrimo licencijos sąlygas, leidžiančias neribotai naudoti, platinti ir atkurti turinį bet kokioje laikmenoje, nurodant autorių ir šaltinį. Received: / Gauta: 2025-11-06. Accepted: / Priimta: 2026-04-13. 1 ON THE MEANING OF THE LITHUANIAN PREPOSITION po: A COGNITIVE PERSPECTIVE Lietuvių kalbos prielinksnio po reikšmė: kognityvinė perspektyva IEVA STASIŪNAITĖ Vilnius University E-mail: [email protected] ORCID ID: https://orcid.org/0000-0001-6787-2602 Fields of research: prepositional semantics, academic writing. https://doi.org/10.35321/all94-10 ------------------------------------------------------------------------------------------------------------------ ABSTRACT Studies of prepositional meaning are rather prolific in cognitive linguistics. However, the Lithuanian po ‘under’ still relies on intuitive explication rather than corpus-based evidence, reflecting the traditional line of investigation. Thus, the present paper focuses on: (1) identifying the primary sense of the preposition; (2) examining its extensions, or major use types, in physical and non-physical contexts, and (3) linking the established senses that differ in status in the form of a network. 1000 concordance lines with po from the fiction segment in the Corpus of Contemporary Lithuanian Language were examined. The Principled Polysemy Model (Tyler, Evans 2003) was used to distinguish between the senses and their contextual variants as well as for primary sense identification. Contextual clues played a significant role as well. The findings show that the primary sense, based on a proto-scene of the Figure beneath the Ground, gives rise to the other senses of the preposition. Some of them are concrete – differentiated by the types of F and G and their geometric and/or functional relationships, while the other extensions are abstract and motivated by metaphor. The paper also considers how the meaning of po interacts with the noun cases it governs. KEYWORDS: Lithuanian preposition po, concrete and abstract senses, principled polysemy. ANOTACIJA Kognityvinėje kalbotyroje nemažai dėmesio skiriama fleksinių kalbų prielinksnių semantikai. Šiame straipsnyje keliami trys tikslai: (1) pagal figūros ir fono tipus bei jų geometrinius ir (arba) funkcinius santykius nustatyti po pirminę reikšmę, (2) išskirti kitas su fizine ir abstrakčia erdve susijusias prielinksnio reikšmes bei (3) ištirti jų motyvacinius ryšius. Išanalizuota 1000 konkordanso eilučių su prielinksniu iš grožinės literatūros registro Dabartinės lietuvių kalbos tekstyne. Tyrime taikytas principinės poliseminės analizės modelis (Tyler, Evans 2003), be to, atsižvelgta į kontekstinius požymius, susijusius su pastovia po vartosenos dažnio raiška. Nustatyta, kad iš pirminės reikšmės, siejamos su protoscena, kurioje figūra yra fono apačioje, galima išvesti kitas penkias prielinksnio reikšmes. Kai kurios iš jų yra konkrečios, o likusios – abstrakčios ir motyvuotos konceptualiųjų metaforų. Straipsnyje taip pat trumpai aptariama, kaip po reikšmė sąveikauja su linksniais, kuriuos prielinksnis valdo. ESMINIAI ŽODŽIAI: lietuvių kalbos prielinksnis po, konkrečios ir abstrakčios reikšmės, principinė polisemija. 2 1. INTRODUCTION This paper aims to contribute to the cognitive line of research into prepositional semantics by focusing on the Lithuanian po ‘under’, which belongs to one of the most highly developed subsets prototypically involving reference to a vertical axis (see also vertical space in Dirven 1993: 74). Its inclusion in the investigation was not random but rather deliberate, motivated by the semantic richness and versatility of the preposition, resulting in multiple lines of scientific inquiry. For instance, Jonas Šukys (1998: 538) treats po in a more traditional manner, focusing on its spatial, temporal, and distributive uses in contemporary Lithuanian, with prescriptive notes concerning case government. Vytautas Ambrazas (2005: 444; 2006: 300), on the other hand, extends the analysis to sequential, gradual, motion, and more fine-grained senses of adjacency, providing historical and dialectal evidence as well. Lithuanian lexicographic sources, though differing in descriptive ambition, likewise play an important role in mapping the semantic range of po. For instance, the Dictionary of the Contemporary Lithuanian Language (DLKŽ8e 2021, hereinafter) adopts a more economical approach, compressing the usage of the item into a set of senses (place, time, manner, distribution, etc.), each illustrated with a few representative examples. The Dictionary of the Lithuanian Language (LKŽe, hereinafter), by contrast, offers extensive data from dialects, older texts, folklore, religious discourse, and idiomatic usage, thus prioritising comprehensiveness and empirical documentation. Even though these works provide valuable classificatory and descriptive insights, they treat the senses of po as discrete categories rather than as a radial network. In other words, the lexicographers seem to assume that the meaning of the preposition is too complex and unsystematic to be represented as an interrelated structure. Cognitivists, who view prepositional senses as motivated extensions rather than isolated meanings, also experience challenges, particularly in their early studies. For instance, Inesa Šeškauskienė (2001), in her discussion of po and its Lithuanian near-synonym žemiau, along with their primary translation equivalents under and below in English, argues that general notions (e.g., “lower than”) in dictionaries and linguistic literature are often inaccurate or oversimplified. She, therefore, emphasises not only geometry but also functional relations, or the interplay between the two, in the semantic interpretation of the items; however, that approach applies only to their concrete senses. Other studies on primary translation equivalents of po and their synonyms reveal further difficulties. For instance, Barbara Wege (1991) discusses below, but finds it resistant to analysis in terms of necessary and sufficient conditions. John R. Taylor (1993) and René Dirven (1993), while examining the polysemy of a number of English prepositions, address under though their discussion of its abstract senses remains sketchy. Andrea Tyler and Vyvyan Evans (2003) investigate under and below according to the Principled Polysemy Model; however, their analysis occasionally lacks the depth of argumentation found in their treatment of other prepositions. Ieva Stasiūnaitė (2018; 2025) examines the motivated polysemy of žemiau and below using the same model, showing how their primary meaning gives rise to both concrete and abstract senses. Although the established senses only partially overlap with dictionary entries, a principled network-based approach seems to yield a more systematic account than the previous studies. Other research on inflecting languages, while not involving po, has produced notable findings as well. For instance, Elżbieta Tabakowska (2010) investigates the Polish za (‘behind, beyond, for’), illustrating how the preposition interacts with prefixes and case forms. Its Russian and Lithuanian counterparts are examined by Daria Shakhova and A. Tyler (2010) and I. Šeškauskienė and Eglė Žilinskaitė-Šinkūnienė (2015), respectively, demonstrating how numerous senses form a motivated network around a central spatial sense. Other works focus on the cross-linguistic investigation of prepositions. For example, E. Žilinskaitė-Šinkūnienė et al. (2019a) report that containment and support are rendered differently in Latvian and Lithuanian, while E. Žilinskaitė-Šinkūnienė et al. (2019b) find 3 containment to be similar across Lithuanian, Latvian and Estonian; however, support displays greater affinity between Latvian and Estonian. E. Žilinskaitė-Šinkūnienė et al. (2020) highlight the importance of topological and geometric features in descriptions of spatial relations between two objects, revealing variation in terms for proximity, remoteness and distance in Lithuanian and Latvian (cf. Šķilters et al. 2020). I. Šeškauskienė and Rita Juknevičienė (2020) show that Lithuanian learners struggle with in and on because of noncongruent L1–L2 mappings shaped by semantic and inflectional features, supporting a motivated polysemy approach. E. Žilinskaitė-Šinkūnienė and I. Šeškauskienė (2021) compare the Latvian aiz and the Lithuanian už ‘behind, beyond’, demonstrating how their senses diverge despite a shared domain due to case marking and verbal prefixation, which highlights prepositional meaning as part of broader morphological-semantic networks. However, despite these advances in cognitive-semantic research, no study has addressed the Lithuanian po, therefore, it is important to establish its semantic elements anew to provide an unambiguous description. The present analysis has three main objectives: (1) to identify the primary sense of the preposition, (2) to examine its extensions, or major use types, in physical and non-physical contexts, and (3) to demonstrate how the established senses that differ in status can be systematically linked in a single network. 2. DATA AND METHODOLOGY This chapter describes data collection and the methodological framework against which the objectives of the present study have been set. 1000 utterances with po (out of 50,243) were randomly collected from the Corpus of the Contemporary Lithuanian Language (DLKT, hereinafter), excluding its idiomatic or otherwise restricted uses. However, only fiction was selected from a variety of segments as it was believed to represent a wide range of senses of the item studied (such a choice may also be justified with reference to previous research; see, for example, Šeškauskienė, ŽilinskaitėŠinkūnienė 2015). The utterances were glossed according to the Salos Glossing Rules adapted to the specifics of Baltic morphology (Nau, Arkadiev 2015). In the present investigation, a preposition is considered to express a relation between two entities in the extra-linguistic world, e.g., suolas po langu ‘a bench under the window’. Figure (F) is an entity that is conceptually foregrounded, typically smaller and movable (i.e., a bench), while Ground (G) serves as a reference object with respect to which F is located and is therefore usually larger and immovable (i.e., a window) (Talmy 2000). The qualitative method was used to identify and classify the distinct senses, or major use types, of po, with the results presented as relative frequencies, since it allowed a more in-depth examination of meaning. In line with this approach, during the annotation procedure, concordance lines were examined to determine whether F and G represented concrete objects, people, abstractions, or actions, processes. So, space here included, apart from physical, also several other spaces, which follows Gilles Fauconnier’s (2007) Mental Space Theory. In addition to these parameters, the analysis considered the spatial configuration of F and G, their idealised geometric shapes and/or internal structures, and their functional interaction. Thus, during the process of meaning construction, the preposition was assumed to prompt a plethora of conceptual operations grounded in encyclopaedic knowledge. At the same time, certain uses may be motivated by the speaker’s predisposition or communicative intention, allowing the same situation to be construed from various perspectives. Therefore, the degree of specificity and salience assigned to particular aspects of a scene was also considered in the sense-identification procedure, in line with Hans J. Schmid’s (2007) discussion of salience and related notions, such as Ronald Langacker’s (2000) focal adjustments and Leonard Talmy’s (2007) attention phenomena. 4 The qualitative method also implied a principled distinction between distinct senses and their contextual variants. For this purpose, a model of word meaning known as Principled Polysemy was adopted (Tyler, Evans 2003). Despite substantial criticism, it remains one of the few frameworks that offers a systematic and explicit methodology for identifying prepositional senses. First, for a sense to be distinct, or instantiated in memory, it must contain an additional element not apparent in any other senses. In other words, a distinct sense must involve either (A) a successive change in the set of concepts, such as (a) strong interdependencies between the preposition and the dynamic (kinetic) and/or geometric properties of G (e.g., static vs. dynamic, dimensional vs. adimensional), (b) corresponding features defining F, as required by the specific relation expressed, and/or (c) the nature of the relationship between the entities, whether topological (e.g., contact, proximity) or functional, or (B) projection onto a new domain (ibid.: 42–43; cf. Wesche 1986/1987; Lakoff 1987). While (B) spans multiple domains, (A) applies to changes within the same domain, as a locative concept may give rise to another locative concept, a temporal concept to another temporal concept, etc. In other words, transitions between concepts, or, following Ignacio Navarro i Ferrando (2006: 176), shifts, occur gradually. Finally, a distinct sense cannot be inferred from another sense or the context in which it occurs but must instead function as a stable, conventionalised unit (Tyler, Evans 2003: 42–43). The Principled Polysemy Model also provides prompts to guide primary sense identification (ibid.: 47–50); however, in the present paper, they were adapted to reflect the actual semantic tendencies: (1) the physicality of F and G, which is related to such principles of cognitive linguistics as the experiential approach and embodiment, stating that the primary sense is usually more physical, closer to bodily experience (Johnson 1987; Lakoff 1987; Steen et al. 2010); (2) contextual clues associated with a reasonably stable frequency of occurrence (Sinha, Kuteva 1995); (3) the earliest attested sense, which falls in line with other cognitively-oriented works on inflecting languages (Pawelec 2009; Tabakowska 2010); (4) the use of the preposition in composite forms; (5) relations to other prepositions that form compositional, or contrast, sets, dividing various spatial dimensions, e.g., po vs. virš; (6) frequency of use in a network; however, it is not a determining factor due to the overwhelming frequency of occurrence of po when it designates temporal relations (cf. the relevance of the parameter for žemiau and below by Stasiūnaitė 2018; 2025). To specify the primary sense, A. Tyler and V. Evans (2003: 52) also introduce the concept of a protoscene, or “an idealised mental representation across the recurring spatial scenes” associated with a preposition. Finally, a radial category approach (Lakoff 1987), grounded in Rosch’s (1978) Prototype Theory, was used to account for the motivated polysemy of po. Within this framework, word meaning is structured around a central, prototypical sense, from which other less representative senses extend in a non-random way, forming a network rather than a list of unrelated senses. Semantic extensions are explained with reference to image-schema transformations and metaphor. Image-schema transformations are systematic modifications of a basic image schema, such as changes in scale, orientation, or perspective, which are grounded in recurring visual and kinaesthetic experiences. Metaphor, in turn, involves mapping structure from a more concrete source domain onto a more abstract target domain (Lakoff, Johnson 1980; Lakoff 1987; Kövecses 2010), allowing spatial experience to shape abstract senses. A preposition thus serves as “an excellent ‘laboratory’ for investigating the way in which spatial experience grounds many other kinds of non-spatial, non-physical concepts” (Tyler, Evans 2003: ix). 5 3. RESULTS This chapter presents the key findings of the present investigation into the semantics of po. First, an overview of its senses, along with their relative frequencies in the DLKT dataset, is provided (see Table 1). Table 1. Senses of po (relative frequencies) As shown above, po has six senses: three relating to physical space and three coding relations in non-physical space. As to the relationship between concrete and abstract senses (31.4%, Total 1 vs. 68.6%, Total 2), the preposition seems to prefer extensions away from physical space. Moreover, po governs three case-forms: GEN, ACC, and INS, the last being the most dominant with concrete senses (21.4%, Total 1). GEN, by contrast, is not used in physical contexts but appears in 62.4% of the data referring to abstract senses (Total 2). ACC is more frequent with po in physical than non-physical contexts (10% vs. 6.2%, Total 1 and Total 2, respectively), yielding a grand total of 16.2%. Moreover, in concrete senses, the preposition combines with a noun in a single case, as shown in Total 3: INS (primary sense: 6.2%, cover: 15.2%) or ACC (pervasion: 10%). In non-physical contexts, similar patterns emerge: GEN (temporal consistency: 3.8%), ACC (distributive quantity: 6.2%), and GEN (temporal sequence: 58.6%). Thus, temporal sequence is the most frequent sense in the compiled corpus. The semantic peculiarities of po, as attested by the DLKT data, are discussed as follows. First, the primary sense is established, and its extensions in physical and nonphysical contexts are examined. Next, motivated relationships between the senses, or major use types, that differ in status are emphasised as a network. 3.1. Defining the Primary Sense 3.1.1. Elements of the Proto-scene Figure 1 offers an approximate image of the conceptual spatial relation between two entities, as coded by po and represented in the proto-scene. Noun cases Senses GEN ACC INS Total 3 1. Primary sense – – 6.2% 6.2% 2. Cover – – 15.2% 15.2% 3. Pervasion – 10% – 10% Total 1: Concrete senses (1+2+3) – 10% 21.4% 31.4% 4. Temporal sequence 58.6% – – 58.6% 5. Temporal consistency 3.8% – – 3.8% 6. Distributive quantity – 6.2% – 6.2% Total 2: Abstract senses (4+5+6) 62.4% 6.2% – 68.6% Grand total: Total 1 + 2 62.4% 16.2% 21.4% 100% 6 Figure 1. Proto-scene for spatial scenes with po The DLKT data attest to several features of the proto-scene associated with the item under study. First, the schematic construal encodes the relation in which F is located lower than G in Euclidean space along a vertical axis: (1) <...>mediniai karveliai po lubų sija <...>. Wooden. NOM.PL.M doves.NOM.PL po ceiling.GEN.PL.F beam.INS.SG ‘<...> wooden doves under the ceiling beam <...>.’ In the example, sija ‘beam’, referred to by means of encyclopaedic knowledge, becomes G (a bold line in Figure 1), while karveliai ‘doves’, designating the boundary of vertical space with its bottom at the lower end, acts as F. The entities are vertically aligned in a sequential order (see the vertical broken line in Figure 1) 1 , so po entails not only geometric relations but also proximity, which is well reflected in its usage with the distance marker tiesiai ‘immediately’. In the following example, akys ‘eyes’ and kakta ‘forehead’ constitute part of a human face, so the absence of distance is more like the consequence of context 2 : (2) <...> po kakta <...> akys <...>. Po forehead.INS.SG eye.NOM.PL ‘<...> eyes <...> under the forehead <...>.’ However, in (3) and (4), F and G are not only vertically aligned but also affect each other due to their adjacency, which is consistent with functional, or extra-geometric, accounts (Vandeloise 1991; 1994; Coventry et al. 2001). In other words, medicine enters the human body through an arm attached to a drip directly over it, while a computer code located beneath the photo gives information about it: (3) <...> [ranka gulėjo] po lašeline su įsmeigta adata venoje <...>. 1 By way of comparison, in the case of žemiau, a seemingly close synonym of the preposition under study, F’s location below G acting as a geometric reference is not clearly specified. It is well illustrated by the following utterance, where the address (F) occupies more space than the stamp (G) but is interpreted as below only in relation to the stamp’s upper or lower boundary (Stasiūnaitė 2018; 2025): Žemiau pašto ženklo <...> adresas. Below post.GEN.SG address.NOM.SG ‘An address below a stamp’ 2 When po and žemiau can be used interchangeably, especially in bodily relations, they evoke different construal of the same spatial scene, cf.: Sukryžiavau rankas žemiau krūtinės / po krūtine. Cross.PST.1SG arm.ACC.PL žemiau chest.GEN.SG / po chest.INS.SG ‘I crossed my arms below/under the chest.’ In the given example, žemiau reflects the spatial configuration of two discontinuous objects, or subspaces, when one subspace is conceptualised as being lower than the other (Šeškauskienė 2001; 2004). Po, on the other hand, is more contact-oriented as arms naturally meet the chest. 7 [arm.NOM.SG lie.PST.3SG] po drip.INS.SG with.PREP stab.PST.PP.INS.SG.F needle.INS.SG vein.LOC.SG ‘<...> [the arm] with a drip with a needle in the vein <...>.’ (4) Po fotografija prilipintas kompiuterinis kodas <...>. Po photograph.INS.SG glue.PST.PP.NOM.SG.M computer-NOM.SG.F code.NOM.SG ‘A computer code was attached under the photograph <...>.’ Thus, in the present analysis, the proto-scene underlying the spatial relation encoded by po, in which F is vertically lower than G, illustrates the least elaborate content of the preposition compared with its other senses. However, the contiguity or spatial proximity of F and G allows them to form a meaningful unit, suggesting a predisposition toward functional relations. This type of proto-scene will be treated as the most representative, or primary, sense of po + INS (6.2% in Table 1). 3.1.2. Types of F and G In this section, types of F and G characteristic of the primary sense of po are discussed. First, various combinations of these entities are presented with relative frequencies, which are subsequently corroborated by the DLKL dataset. Table 2. Types of F and G in the primary sense of po (relative frequencies) G F Human body Concrete object Total 2 Human body 12.9% 27.4% 40.3% Concrete object 33.9% 22.6% 56.5% Action/process – 3.2% 3.2% Total 1 46.8% 53.2% 100% As seen from Table 2, an important criterion for primary sense identification, in addition to its least elaborate content, is the physical aspect of F and G. In 53.2% of the cases (see Total 1), G designates a concrete object, whereas in the remaining utterances (46.8%, Total 1), it is a person or human morphology. Although people functioning as G represent the second most frequent choice, their occurrence in spatial scenes highlights the role of bodily experience in structuring our understanding of the world (Gibbs 2006: 9; Rohrer 2007: 25–47), as attested by studies of conceptual metaphor (Lakoff, Johnson 1980; Lakoff 1987; Kövecses 2010). However, not only concrete objects, including people, feature as G in the primary sense of po; the nature of F is also of utmost importance. Tangible inanimate entities, as its denotata, account for 56.5% of the data (Total 2), while people and human body parts constitute the second most frequent type (40.3%, Total 2). Actions and processes, the location of which is determined with reference to a concrete object, occur in as few as 3.2% of the utterances; however, no human body is selected as G in this case (Total 2). More specific statistical tendencies are further illustrated with concordance lines from the DLKT. For instance, human bodily parts, typically functioning as G with po in the primary sense, are kakta ‘forehead’, akys ‘eyes’, blakstienos ‘eyelashes’, krūtinė ‘chest’, etc. The position of F is then taken by human morphology as well, e.g., akys ‘eyes’, kumščiai ‘fists’, rankos ‘arms’, etc. (12.9%, Table 2), sometimes even forming an integral part of G: (5) <...> po jomis [akimis] atsirado apdribę pilki maišeliai. Po 3.INS.PL.F [eye.INS.PL] appear.PST.3PL sink.PST.PA.NOM.PL.M grey.NOM.PL.M bag.DIM-NOM.PL 8 ‘<...> grey bags appeared under the sunken eyes.’ Inanimate entities, e.g., akmenys ‘stones’, pamatas ‘footing’, laiptai ‘staircase’, kilimas ‘carpet’ or kelias ‘road’, when used with human features, especially kojos ‘legs’, make the most frequent denotatum of G (33.9%, Table 2) 3 . Other physical objects may also function as G, thereby determining the location of people or elements of human anatomy (27.4%, Table 2): (6) Prie šilto tėvo šono prisiglaudęs. Po žvaigždžių lietumi. Near.PREP warm.GEN.SG.M father.GEN.SG side.GEN.SG snuggle.PST.PA.NOM.SG.M po star.GEN.PL rain.INS.SG ‘Snuggling up against the warm father. Under a shower of stars.’ Moreover, when a concrete object, e.g., karnizas ‘a cornice’, atbraila ‘a ledge’, etc., is used as G, a wide range of tangible entities function as F (22.6%, Table 2): (7) Landa dešinėje, po atbraila <...>. Hole.NOM.SG right.LOC po ledge.INS.SG ‘The hole is on the right, under the ledge <...>.’ Concrete entities in the position of G may also appear with a dynamic type of F without affecting the interpretation of the primary sense (3.2%, Table 2): (8) <...> raganius sugebėjo kilstelėti vielas ir spjovė po jomis. Magician.NOM.SG manage.PST.3SG lift.INF wire.ACC.PL and.CONJ spit.PST.3SG po 3PL.INS ‘<...> the magician uplifted the wires and spat under them.’ Thus, various types of F and G, arranged sequentially along a vertical line, appear in the proto-scene underlying the spatial relation encoded by po. However, the typicality of the sense is reinforced by the physical aspect of the entities involved. Further, it will be shown how less typical meanings, or use types, are related to this sense by transforming it in different ways. 3.2. Cover The cover sense of po + INS is the most frequent in physical contexts, occurring in 15.2% of the DLKT data (Table 1). It preserves the spatial configuration of the entities from the primary sense, with F positioned below and directly under G along a vertical axis, emphasising either the endpoint of movement (9) or static relations (10): (9) <...> paėmiau peilį ir sviedžiau po stalu. Take.PST.1SG knife.ACC.SG and.CONJ throw.PST.1SG po table.INS.SG ‘<...> I took a knife from the cupboard and threw it under the table.’ (10) Gulėdamas po antklode, Virginijus įsiklausė. Lie.CVB.SG.M po blanket.INS.SG Virginijus.NOM.SG listen.PST.3SG ‘Virginijus kept his ears open lying under the blanket.’ What distinguishes the semantic structure from the primary sense, however, is functional, or extra-geometric, interaction between the entities. In other words, G, conceptualised as a two-dimensional object, can cover F, which is smaller or at least of the same size and positioned directly beneath it. To do so, G must also be parallel to F, as illustrated in the following examples: (11) Moteris sėdėjo po paveikslu. Woman.NON.SG sit.PRS.3SG po picture.INS.SG 3 Such combinations are impossible with žemiau, which requires a more diffuse, contact-free arrangement of F relative to G, e.g.: <...> *žemiau kojų / po kojomis plikas cementas, o ne purus kilimas. *Žemiau foot.GEN.PL/ po foot.INS.PL bare.NOM.SG.M cement.NOM.SG but.CONJ not.NEG fluffy.NOM.SG.M carpet.NOM.SG ‘<...> bare cement, but not a fluffy carpet, *below/under the feet.’ 9 ‘The woman was sitting under the picture.’ (12) Kažkas rausva gulėjo ant šaligatvio po mano batu. INDEF.who.NOM pink.NOM.SG.F lie.PST.3SG on.PREP pavement.GEN.SG po 1SG.GEN shoe.INS.SG ‘Something pink was lying on the pavement under my shoe.’ (11) illustrates the primary sense of po: the picture, conceptualised as a twodimensional object, hangs perpendicularly on the wall relative to the woman and thus does not cover her but merely indicates her location in close proximity. In (12), on the other hand, the preposition is used in the cover sense, as the shoe occupies a parallel, twodimensional position directly above something pink, presumably smaller, thereby reducing its exposure or visibility. In the following concordance lines, even though earth and water constitute continuous masses, only their two-dimensional horizontal top surfaces are metonymically involved in the process of conceptualisation: (13) <...> gynėjai būstinę įsirengė olose giliai po žeme <...>. Defender.NOM.PL headquarter.ACC.SG RFL.set.PST.3PL cave.LOC.PL deep.ADV po ground.INS.SG ‘<...> defenders built headquarters in caves deep under the ground <...>.’ (14) <...> ji nugrimzdo po okeanu. 3.NOM.SG.F sink.PST.3SG po ocean.INS.SG ‘<...> she sank under the ocean.’ As shown by the adverb giliai ‘deeply’ and the verb (nu)grimzti ‘sink’, contextual clues further support the interpretation of po as expressing a cover relation. Other verbs, e.g., slypėti ‘lurk’, dingti ‘disappear’, (pa)(si)kišti ‘put oneself/something under something’, (pa)slėpti(s) ‘hide somebody/something/oneself’, (pa)laidoti ‘bury’, etc., are likewise significant indicators, aligning with V. Ambrazas (2006: 298) who notes the same semantic contribution of the prefix pa-: (15) <...> fotografija, tūnanti (…) po celofano plėvele mažame albumėlyje <...>. Photograph.NOM.SG lurk.PRS.PA.NOM.SG.F po cellophane.GEN.SG sheet.DIM.INS.SG small.LOC.SG.M album.DIM.LOC.SG ‘<...> a photograph lurking under a sheet of cellophane in a small album <...>.’ (16) <...> šaknys palindusios po pirkios pamatais. Root.NOM.PL bury.PRS.PA.NOM.PL.F po cottage.GEN.SG foundation.INS.PL ‘<...> the roots were buried under the foundation of the cottage.’ Adjectives, such as neįskaitomas ‘undecipherable’, or nouns, e.g., užklotas ‘bedcover’, slėptuvė ‘hiding place’, also frequently occur with po to convey covering: (17) <...> stirti po tuo palaikiu užklotu<...>. Freeze.INF po DEM.INS.SG.M thin.INS.SG bedcover.INS.SG ‘<...> freeze under that thin bedcover <...>.’ (18) <...> šviesios akys neįskaitomos po ilgais kirpčiukais. Light.NOM.PL.F eye.NOM.PL NEG.decipherable.NOM.PL.F po long.INS.PL.M fringe.INS.PL ‘<...> light eyes were undecipherable under a long fringe hair.’ Thus, the successive configuration of the entities along a vertical line is preserved, but changes in the dimensionality and orientation of G cause F to occupy a position of reduced exposure. This modification reinforces an already latent functional interaction, suggesting that the cover sense of po + INS is derived from the primary sense. 3.3. Pervasion Po + ACC manifests a pervasion sense which also pertains to physical space (10%, Table 1). What becomes salient in the semantic structure, however, is the distributed movement of F over G: 16 REFERENCES Ambrazas Vytautas, ed., 2005: Dabartinės lietuvių kalbos gramatika [Grammar of the Contemporary Lithuanian Language], 4th ed., Vilnius: Mokslo ir enciklopedijų leidybos institutas. Ambrazas Vytautas 2006: Lietuvių kalbos istorinė sintaksė [Historical Syntax of the Lithuanian Language], Vilnius: Lietuvių kalbos institutas. Athanasopoulos Panos 2017: Language Alters Our Experience of Time. – Conversation, June 13. Internet Access: https://theconversation.com/language-alters-ourexperience-of-time76761?fbclid=IwAR0_clgcDwx53lAnM6xgrj6gVHy1hFXpgmHNutD2eLAMSWoks YDU_YvVtaU [accessed 05.11.2025]. Berg-Olsen Sturla 2004: The Latvian Dative and Genitive: A Cognitive Grammar Account: PhD dissertation, Oslo: Oslo University Press. Boroditsky Lera, Fuhrman Orly, McCormick Kelly 2011: Do English and Chinese Speakers Think About Time Differently? – Cognition 118, 123–129. DOI: doi.org/10.1016/j.cognition.2010.09.010. Coventry Kenny R., Prat-Sala Merce, Richards Lynn 2001: The Interplay Between Geometry and Function in the Comprehension of over, under, above, and below. – Journal of Memory and Language 44, 376–398. DOI: doi.org/10.1006/jmla.2000.2742. Dirven Rene 1993: Dividing up Physical and Mental Space into Conceptual Categories by Means of English Prepositions. – The Semantics of Prepositions: From Mental Processing to Natural Language Processing, ed. C. Zelinsky-Wibbelt, Berlin: Mouton de Gruyter, 73–97. DLKŽ8e 2021: Dabartinės lietuvių kalbos žodynas [Dictionary of the Contemporary Lithuanian Language], vyr. red. S. Keinys, 8-as patais. ir papild. leidimas (elektroninis variantas), Vilnius: Lietuvių kalbos institutas. Internet Access: https://ekalba.lt/dabartines-lietuviu-kalbos-zodynas. Fauconnier Gilles 2007: Mental Spaces. – The Oxford Handbook of Cognitive Linguistics, eds. D. Geeraerts, H. Cuyckens, Oxford: Oxford University Press, 351–376. Gibbs Raymond W. 2006: Embodiment and Cognition, Cambridge: Cambridge University Press. Haspelmath Martin 1997: From Space to Time: Temporal Adverbials in the World’s Languages, Newcastle, UK: Lincom Europa. Hong Tian, He Xianyou, Tillman Richard, Zhao Xueru, Deng Yumel 2017: The Vertical and Horizontal Spatial-Temporal Conceptual Metaphor Representation of Chinese Temporal Words. – Psychology 8, 1679–1692. DOI: doi.org/10.4236/psych.2017.811111. Janda Laura 2002: The Conceptualization of Events and Their Relationship to Time in Russian. – Glossos 2, [1–10]. Johnson Mark 1987: The Body in the Mind. The Bodily Basis of Meaning, Imagination, and Reason, Chicago: Chicago University Press. Kövecses Zoltan 2010: Metaphor. A practical introduction, Oxford: Oxford University Press. Lakoff George 1987: Women, Fire, and Dangerous Things, Chicago, London: Chicago University Press. Lakoff George, Johnson Mark 1980: Metaphors We Live By, Chicago: Chicago University Press. 17 Langacker Ronald W. 2000: Grammar and Conceptualization, Berlin, New York: Mouton de Gruyter. LKŽe: Lietuvių kalbos žodynas [Dictionary of the Lithuanian Language] (1–20, 1941–2002), el. variantas, redaktorių kolegija: G. Naktinienė (vyr. redaktorė), J. Paulauskas, R. Petrokienė, V. Vitkauskas, J. Zabarskaitė; programuotojai: E. Ožeraitis, V. Zinkevičius, Vilnius: Lietuvių kalbos institutas, atnaujinta versija, 2018. Internet Access: https://ekalba.lt. Luodonpää-Manni Milla, Viimaranta Johanna 2010: Metaphoric Expressions on Vertical Axis Revisited: An Empirical Study of Russian and French Material. – Metaphor and Symbol 25(2), 74–92. Luraghi Silvia 2009: Case in Cognitive Grammar. – The Oxford Handbook of Case, eds. A. L. Malchukov, A. Spencer, Oxford: Oxford University Press, 136–150. Markevičius Aurimas 2014: Pastabos apie prielinksnio sampratą, raišką ir vartojimą bendrinėje lietuvių kalboje ir tarmėse [Observations on the Concept, Realization, and Use of Prepositions in Standard Lithuanian and Regional Dialects]. – Žmogus ir žodis 16(1), 70–80. Miliūnaitė Rita 2022: Įvairuojantys ir nauji lietuvių kalbos reiškiniai XXI a. pradžioje: sistematika ir pokyčių kryptys [Variable and New Phenomena of Contemporary Lithuanian at the Start of the 21st Century: Systemization and Trends of Change], Vilnius: Lietuvių kalbos institutas. Nau Nicole, Arkadiev Peter 2015: Towards a Standard of Glossing Baltic Languages: The Salos Glossing Rules. – Baltic Linguistics 6, 195–241. Navarro i Ferrando Ignacio 2006: On the Meaning of Three English Prepositions. – Inroads of Language. Essays in English Studies, eds. I. Navarro i Ferrando, N. A. Crespo, Castelló de la Plana, Universitat Jaume I, 167–180. Oakley Todd 2007: Image Schemas. – The Oxford Handbook of Cognitive Linguistics, eds. D. Geeraerts, H. Cuyckens, Oxford: Oxford University Press, 214–233. Pawelec Andrzej 2009: Prepositional Network Models. A Hermeneutical Case Study, Kraków: Wydawnictwo Uniwersytetu Jagiellońskiego. Rohrer Tim 2007: Embodiment and Experientialism. – The Oxford Handbook of Cognitive Linguistics, eds. D. Geeraerts, H. Cuyckens, Oxford: Oxford University Press, 25–47. Rosch Eleanor 1978: Principles of Categorization. – Cognition and Categorization, eds. E. Rosch, B. B. Loyd, Hillsdale, New York: Lawrence Erlbaum, 27–48. Schmid Hans J. 2007: Entrenchment, Salience and Basic Levels. – The Oxford Handbook of Cognitive Linguistics, eds. D. Geeraerts, H. Cuyckens, Oxford, Oxford University Press, 117–138. Shakhova Daria, Tyler Andrea 2010: ‘Taking the principled polysemy model of spatial particles beyond English: The case of Russian za’. – Language, Cognition and Space: The State of the Art and New Directions, eds. V. Evans, P. Chilton, London, Oakville: Equinox, 267–291. Sinha Christopher, Kuteva Tania 1995: Distributed Spatial Semantics. – Nordic Journal of Linguistics 18, 167–199. doi.org/10.1017/S0332586500000159. Stasiūnaitė Ieva 2018: On the Motivated Polysemy of the Lithuanian žemiau ‘below’. – Kalbų studijos 32, 5–20. DOI: https://doi.org/10.5755/j01.sal.0.32.19290. Stasiūnaitė Ieva 2025: Encoding Vertical Space in Language: A Cognitive-Semantic Study of the English below as Compared to the Lithuanian žemiau. – Taikomoji kalbotyra 22, 156–175. DOI: https://doi.org/10.15388/Taikalbot.2025.22.10. 18 Steen Gerard J., Dorst Aletta G., Hermann Berenike J., Kaal Anna A., Krennmayr Tina, Pasma Trijntje 2010: A Method for Linguistic Metaphor Identification: From MIP to MIPVU, Philadelphia: John Benjamins. Šeškauskienė Inesa 2001: Cross-Linguistic Variation of Space Conceptualization. – Translation and Meaning 5, 85–95. Šeškauskienė Inesa 2004: The Extension of Meaning: Spatial Relations in English and Lithuanian. – Imagery in Language, eds. B. Lewandowska-Tomaszczyk, A. Kwiatkowska, P. Lang, 373–384. Šeškauskienė Inesa, Žilinskaitė-Šinkūnienė Eglė 2015: On the Polysemy of the Lithuanian už. A Cognitive Perspective. – The Baltic International Yearbook of Cognition, Logic and Communication. Perspectives on Spatial Cognition 10, 1–38. DOI: dx.doi.org/10.4148/1944-3676.1101. Šeškauskienė Inesa, Juknevičienė Rita 2020: Prepositions in L2 Written English, or Why on Poses More Difficulties than in. – Nordic Journal of English Studies 19(1), 65–96. DOI: doi.org/10.35360/njes.540. Šķilters Jurģis, Zariņa Līga, Žilinskaitė-Šinkūnienė Eglė, Bērziņa Nora, Apse Linda 2020: Topologic and Geometric Structure of Spatial Relations in Latvian: An Experimental Analysis of RCC. – Baltic Journal of Modern Computing 8(1), 92–125. DOI: https://doi.org/10.22364/bjmc.2020.8.1.05. Šukys Jonas 1998: Lietuvių kalbos linksniai ir prielinksniai: vartosena ir normos [Cases and Prepositions in the Lithuanian Language: Usage and Norms], Kaunas: Šviesa. Tabakowska Elżbieta 2010: The Story of ZA: In Defence of the Radial Category. – Studies in Polish Linguistics 5, 65–77. Talmy Leonard 2000: Towards Cognitive Semantics 1–2, Cambridge, MA: MIT Press. Taylor John Robert 1993: Prepositions: Patterns of Polysemization and Strategies of Disambiguation. – The Semantics of Prepositions: From Mental Processing to Natural Language Processing, ed. C. Zelinsky-Wibbelt, Berlin: Mouton de Gruyter, 151–179. Torralbo Ana, Santiago Julio, Lupiáñez Juan 2006: Flexible Conceptual Projection of Time Onto Spatial Frames of Reference. – Cognitive Science 30, 745–757. DOI: doi.org/10.1207/s15516709cog0000_67. Tyler Andrea, Evans Vyvyan 2003: The Semantics of English Prepositions. Spatial Scenes, Embodied Meaning and Cognition, Cambridge: Cambridge University Press. Vandeloise Claude 1991: Beyond Geometric and Logical Descriptions of Space. A Functional Description. – Spatial Prepositions. A Case Study from French, ed. C. Vandeloise, Chicago, London: Chicago University Press, 3–20. Vandeloise Claude 1994: Methodology and Analyses of the Preposition in. – Cognitive Linguistics 5(2), 157–184. Wege Barbara 1991: On the Lexical Meaning of Prepositions: A Study of above, below and over. – Approaches to Prepositions, ed. G. Rauh, Tubingen: Gunter, 275–296. Wesche Birgit 1986/1987: At Ease with at. – Journal of Semantics 5, 385–398. Zinkevičius Zigmas 1981: Lietuvių kalbos istorinė gramatika [Historical Grammar of the Lithuanian Language] 2, Vilnius: Mokslas. Žilinskaitė-Šinkūnienė Eglė, Šķilters Jurģis, Zariņa Līga 2019a: Containment and Support in Baltic Languages: Overview, Experimental Evidence, and an Extended RCC as Applied to Latvian and Lithuanian. – Baltic Journal of Modern Computing 7(2), 224– 254. DOI: https://doi.org/10.22364/bjmc.2019.7.2.04. 19 Žilinskaitė-Šinkūnienė Eglė, Šķilters Jurģis, Zariņa Līga, Bērziņa Nora 2019b: Containment and Support: Similarities and Variation in Lithuanian, Latvian and Estonian. – Baltistica 54(2), 205–255. DOI: 10.15388/baltistica.54.2.2386. Žilinskaitė-Šinkūnienė Eglė, Šķilters Jurģis, Zariņa Līga 2020: Expressing Distance in the Baltic Languages: Remarks on the Use of an Extended RCC. – Baltic Journal of Modern Computing 8(1), 69–91. DOI: doi.org/10.22364/bjmc.2020.8.1.04. Žilinskaitė-Šinkūnienė Eglė, Šeškauskienė Inesa 2021: When the Search Domain is Back Region in Baltic: The Latvian aiz as Compared to the Lithuanian už. – Studies in Baltic and Other Languages: A Festschrift for Axel Holvoet on the Occasion of His 65th Birthday, eds. P. Arkadiev, J. Pakerys, I. Šeškauskienė, V. Žeimantienė, Vilnius: Vilnius University Press, 433–466. DOI: doi.org/10.15388/SBOL.2021.24.