Linguistic Typology

This paper investigates clausal complementation in Mising, an Eastern Tani (TB) language spoken in Assam, India. It examines finite and non-finite complement clauses, analyzing their grammatical structures and underlying conceptualizations. Mising employs at least four complementizers (one finite, three non-finite), while some complement clauses appear without overt complementizers. Each of these constructions exhibits distinct semantic properties.

While primarily descriptive in nature, this study incorporates insights from Cognitive Linguistics. Our analysis suggests that finite complementation aligns with an objective perspective, where the speaker views the complement event as external and distinct. Conversely, non-finite complementation aligns with a subjective perspective, reflecting a closer involvement of the speaker in the complement event. These two perspectives are associated with different conceptual viewing arrangements.

The data for this paper comes from the Pagro dialect of Mising, primarily spoken in the Jonai sub-division of Assam’s Dhemaji district.

The article discusses dialectal and folkloristic features of an oral narrative in a NorthEastern Neo-Aramaic (NENA) variety recently recorded in the Southern Russian village of Urmiya. The dialect of the tale belongs to the NENA varieties originating in the easternmost regions of Turkey, which is corroborated by the tale's speaker, who named its places of origin as Lewən, Albaq, and Gawar. These dialects remain largely undescribed and are highly endangered. The plot of the story is not known from hitherto published Neo-Aramaic field texts. Even though it resembles some of the folktales recorded in Iran and India, the story contains a unique combination of narrative motifs.

Facts fall in place and form a coherent picture once myths have been dispelled. Descriptions of ergativity are fraught with grammatical myths. Strikingly, the puzzlement dissolves into nothingness as soon as one is willing to acknowledge the origin of the mystifications, that is, the confusion of subjects and objects (by functional typologists as well as by Generative syntacticians).

2Институт физико-химической биологии им. А.Н. Белозерского МГУ, г. Москва Интеграза -фермент, необходимый для репликации вируса иммунодефицита человека первого типа (ВИЧ-1). В настоящее время ароматические производные дикетокарбоновых кислот рассматриваются в качестве перспективных ингибиторов интегразы ВИЧ-1. Одно из соединений, относящихся к этому классу, находится на стадии клинических испытаний. Мы создали выборку из 90 производных дикетокарбоновых кислот и провели обучение программы PASS. Точность прогноза антиинтегразной активности при скользящем контроле с исключением по одному, составила 99%. Среди более чем 700 тысяч структур коммерчески доступных химических соединений проведен поиск новых потенциальных ингибиторов интегразы ВИЧ-1. Отобрано и исследовано 8 веществ, для которых эта активность была предсказана с относительно высокой вероятностью; для двух соединений наличие антиинтегразной активности подтверждено в эксперименте. Ключевые слова: интеграза, ВИЧ-1, производные дикетокарбоновых кислот, компьютерное конструирование лекарств.

Gyalrongic languages are known for their complex and conservative morphology, which is valuable for both linguistic typology in general as well as Sino-Tibetan/Trans-Himalayan historical linguistics. Verbal stem alternation is an important component in Gyalrongic morphology, but there is very little work focusing on its origin and evolution to date, even though the history of stem alternation may reveal important clues for the reconstruction of Proto-Sino-Tibetan. Khroskyabs is one of the Gyalrongic languages exhibiting complex stem alternation patterns and can thus open a way for solving said issues. This paper fi rst examines several problems in the reconstruction of Khroskyabs vowels, proposing vowel length contrasts in Proto-Gyalrongic, and then use the reconstruction of long vowels to explain the origin of Khroskyabs verbal stem alternation. I argue that in verbs exhibiting stem alternation, the formation of Stem 2 is based on the Stem 1 form by lengthening the main vowel. I further demonstrate that the vowel length hypothesis proposed for Khroskyabs can explain similar phenomena in verbal stem alternation in other Gyalrongic languages, such as Zbu.

The paper deals with serial verb constructions in Kwa, Benue
Congo and Gur with the verb ‘take’ that is used as an object marker with
the lexical verb. The factors relevant for the choice between the ‘take’
construction and simple transitive construction are discussed. These factors include semantics and pragmatics of the ‘take’-construction itself
(accenting the initial stage of the event, emphasizing the event or the ob
ject), features of the object (pronominality, definiteness), features of the
lexical verbs (presence of more arguments, agentivity, volitivity).

The paper deals with the Enets contextual converb marked by -ʃ(i) in Forest Enets and by -s(e)/-e in Tundra
Enets. A morphological account of this form is provided and followed by a description of its uses in adverbial
clauses, complement clauses and some other contexts. An overview of concurring forms is given. Enets
contextual converb has a wide range of uses that includes both functions of the infinitive and of the contextual
converb proper.

The paper deals with the system of future reference forms in both Enets dialects: Future and Debitive
attested both in Forest Enets and Tundra Enets, Hypothetical attested only in Forest Enest and Analytical
Debitive attested only in Tundra Enets. The paper provides brief information on morphological structure
of these forms, numerical data on their frequency in texts and a comparison of the contexts where these
forms are used. Debitive expresses modal meanings of deontic and epistemic obligation and epistemic
possibility. Forest Enets Hypothetical is used for weak intentional future and for weak epistemic possibility.
Tundra Enets Analytical Debitive is limited by contexts of weak epistemic possibility. The most
frequent Future form is used for sure predictive future and sure intentional future, but also is attested in
modal contexts mentioned above. This can be summarized as a privative opposition between the Future
form that has an unrestricted general future meaning and modal forms that are used for more particular
modal meanings. This feature of the Enets verbal system contrasts it with a known cross-linguistic distinction
of future certainty vs. future possibility and therefore Enets data are of interest for linguistic
typology.

Structural similarity between a bilingual's two languages, defined by word order correspondence, reflects the surprisal levels (inverse-frequency effect) associated with sentence processing in the L2. Similarity is predicted to facilitate L2 processing due to the lower levels of surprisal of similar structures and the possibility of relying on existing mental representations and processing patterns from a bilingual's L1. As representations of argument structure constructions map both morphosyntactic structure and underlying meaning, this study investigates the trade-off in L2 processing facilitation by constructions that present equivalent word order in Brazilian Portuguese (BP) and English but differ in their underlying meaning. A self-paced reading task was used to compare processing of causative resultatives and middle voice constructions in English by L1 BP L2 English bilinguals of varying proficiency levels, as these constructions find word order equivalents in BP that fail to convey the intended meaning due to differences in event role assignment of the adjective in resultatives and differences in lexical restrictions on verbs in the middle construction. The analyses of critical words and spillover areas through mixed-effects linear models showed significantly higher RTs for middle voice sentences in comparison to resultative sentences, with L2 proficiency showing interaction with processing of the middle construction only. The results suggest that facilitation comes from overall constructional similarity rather than from word order correspondence alone, since the lexical selection differences in the middle sentences posed higher levels of surprisal to bilinguals than the novel event role assignment of the adjective in the resultative construction.

This article explores the phenomenon of phonaesthemic groups in the English language from a diachronic perspective. I argue that phonaesthemic sound symbolism simultaneously employs the mechanisms of language iconicity and systematicity. The creation of phonaesthemic groups results in semantic groupings of the lexicon based on synaesthesia, that is, on conveying impressions of different sensory modalities by sound. Phonaesthemic sound symbolism, thus, is semantically related to other linguistic ways of referring to different sensory modalities, such as ideophones (visual, tactile, olfactory modalities) and onomatopoeia (audial modality). I argue that acoustic and articulatory parameters of speech sounds used to compose a phonaestheme influence the set of (metonymically transferred) meanings they are paired with. The paper provides examples of the semantic development of the English brand cr-phonaesthemic groups (33 br-words and 53 cr-words), considers their etymology and the pathways of semantic expansion of the respective phonosemantic fields.

Acest articol metodologic prezintă principalele abordări ale comparării datelor în gramatica comparată, cu accent pe sintaxă. Fiecare metodă este ilustrată prin exemple reprezentative – în principal parametri sintactici de variație, așa cum sunt definiți în gramatica generativă și în tipologia lingvistică clasică – evidențiindu-se totodată și limitele fiecăreia. Comparația lingvistică majoră tinde să identifice macroparametri, în timp ce alte metode sunt mai eficiente în surprinderea microvariației.
***
This methodological paper outlines key approaches to data comparison in comparative grammar, with a focus on syntax. It illustrates each method through representative examples—mainly syntactic parameters of variation as defined in generative grammar and classical linguistic typology—while also pointing out their respective limitations. Major comparative approaches tend to identify macroparameters, whereas other methods are more effective in capturing microvariation.

The verb construction patterns of Indonesian and Korean has several differences. Indonesian generally follows the Subject-Verb-Object (SVO) pattern, while Korean tends to use the Subject-Object-Verb (SOV) pattern. However, both have similarities in the use of verbs as the core of the verbal clause and the presence of affixes that change the meaning of the verb. This study aims to; 1) describe the differences in sentence patterns of Indonesian and Korean, and 2) describe the forms of subject pronouns and particles in both Indonesian and Korean. The results of the study show that in the Indonesian Sentence pattern: The common sentence pattern is SVO (Subject-Verb-Object), for example: "I eat rice", while in Korean: The common sentence pattern is SOV (Subject-Object-Verb), for example: "저는 밥을 먹어요" (Jeonun bapeul meogeoyo) which means "I eat rice". As for the BI subject pronouns (I, you, he, etc.) can be omitted if the context is clear, for example: "Eat rice!", and this does not happen in Korean where the subject pronoun is usually not omitted, even in informal language, because the subject marker particle (은/는) helps identify the subject, for example: "밥을 먹어요" (bapeul meogeoyo) still indicates the subject is "I" (저는, jeo-neun) even though it is not stated explicitly. Another pattern found is that Indonesian particles do not have subject, object, or adverbial marker particles as in Korean. Korean uses particles (eun/neun, reul/ul, etc.) to mark the function of words in sentences, for example "은/는" (eun/neun) to mark the subject/topic, "을/를" (eul/reul) to mark the object

The goal of this paper is to verify whether there is an acoustic difference between a reduced form of the verb poder (po') 'to be able to' and the noun pó 'powder' in Brazilian Portuguese oral discourse. The reduction of the verb occurs in contexts in which it is an auxiliary in periphrases formed with the infinitive. This process results in a form segmentally homonymous with the noun pó ['pɔ]. As the literature states that function words are shorter in duration than content words, homonymy is expected to occur at the segmental, but not at the suprasegmental level. To verify this hypothesis, we analyzed data from Brazilian Portuguese spontaneous speech, selecting the parameters of normalized duration (in smoothed z-score), mean intensity, mean f0 of both ['pɔ] and its contiguous words, and the mean frequency of the formants (F1 and F2) of the vowel of both the verb and the noun. Statistically significant results were found only for duration, confirming that the auxiliary verb (a function word) is shorter than the noun (a content word). It is suggested that the suprasegmental level, namely duration, is an important correlate in differentiation between two segmentally homonymous words but from different categories.

The aim of this paper is to propose a structured argument structure (arg-st) in which the external and the internal argument(s) are defined declaratively and rep-resented separately. Based on the structured argument structure, it proposes a basic distinction between two types of ...

In several Linear A tablets, sign pairs merge consonant-vowel units inside a single cell in ways similar to Prākrit sandhi, where sounds co-articulate fluidly within a syllable. Reexamining material from Hagia Triada and Archanes shows that several signs long catalogued as logograms instead behave as fused phonemic cells. The clearest examples; SE-MO, SE-DI (A608), DATE , and the KA-based A568, record deliberate co-articulation in clay, revealing that the Linear A system could express phonemic combination within a single sign space.

Sinhala is a name of a language belonging to one of the South Asian Languages primarily spoken by the Sinhalese people in Sri Lanka. Some controversies are going on regarding Sinhala language whether it belongs to Dravidian language family or Indo-Aryan language family. Because the Sinhala language seems to contain both Dravidian and Indo-Aryan language features. Most of the discussions, explanations, and illustrations are made regarding Sinhala compared only with Indo-Aryan languages. It is unable to find broad and detailed research activities and linguistic explanations regarding Sinhala compared with the Dravidian languages. Currently only a few numbers of research scholars of the new generation have come forward to do comparative studies between Sinhala and other Dravidian languages to exhibit the Dravidian language features in Sinhala. This research is also one of the attempts to bring out the Dravidian language features in the Sinhala language. According to that, this study talks about the derivational morphology of Tamil and Sinhala and attempts to find the similarities of derived nouns. The goal and objective of this research is to reveal the similar way of generation of derived nouns. Besides, it is going to explore what types of similar nominalizers are used to produce derived nouns in both languages. The suffixes and prefixes are functioned as the nominalizers to form derived nouns in Tamil and Sinhala. Thus, it expects to talk about classmaintaining and class-changing derivational affixes in Tamil and Sinhala. The significance of this study is, most of the Sinhalese scholars say Sinhalese language does not have many Dravidians language features. However, this research is attempting to find the similar derivational morphological features in Tamil and Sinhala and it is going to discuss the way of nominalizing to form derived nouns in both languages.

This paper presents a large scale Syntactic Computational Lexicon, representative of Italian modem language use. The entries were in fact selected on frequency criteria from the nx Corpus and the syntactic patterns encoded were partly inferred from their contexts of occurrence. This lexicon was elaborated in the framework of the EC LE-PAROLE project which developed large, generic and re-usable written language resources in most EU languages. The lexica linguistic specifications, based on EAGLES recommendations and on the GENELEX model, were implemented in the LE-PAROLE model. This paper aims to provide an overview of the Italian instantiation of the PAROLE syntactic lexicon and of the coverage of the syntactic structures handled. It illustrates languagespecific linguistic and lexicographic choices concerning crucial issues to a lexicon building process.

This article examines the relationship between natural language and causal cognition and argues that linguistic structures both reflect and constrain the ways in which humans conceptualize and represent causation. It opens with a survey of foundational philosophical theories of causation, focusing on the tension between metaphysical accounts and judgment-based approaches. The discussion then turns to a range of linguistic phenomena-including causative constructions, conditional sentences, discourse coherence, aspectual interpretation, and argument structure-demonstrating how causal relations are systematically encoded across grammatical domains. Building on insights from linguistics, philosophy, and cognitive science, the article reviews recent developments of a semantic framework for modeling causal knowledge in language. Rather than assuming a uniform mapping between causal relations and their linguistic expressions, the framework accounts for systematic variation in how causality is selected, structured, and communicated. In doing so, it positions natural language as a key source of evidence for understanding the architecture of causal reasoning and its representation in human cognition.

A number of features found in spoken narratives in the Rembarrnga language of Arnhem Land can be seen to serve a cohesive function within the text. These include not only ellipsis of pronominal prefixes and tense marking from the verb complex but also the ...

This commentary starts by discussing the future of theoretical linguistics in the context the rise of Large Language Models (LLMs). While comparative linguistics will likely persist due to universal interest in linguistic diversity, it is indeed questionable whether Chomskyan generative linguistics will continue as before. Its key contributions are mid-level generalizations, but its claims about innate linguistic knowledge have not been supported. One problem with Chesi's article is his focus on computational efficiency as this diverges from the Chomskyan goals. The commentary lists six core theoretical goals of linguistics and relates Chesi's article to these six goals. In particular, the conflation of theories (language-specific descriptions) and frameworks (general tools) has often created confusion. Ultimately, the text concludes that generative linguistics' decline does not doom the discipline.

When we principally accept the assumption that 'India is a common linguistic area', we have to take the responsibility to prove this hypothesis true with real-life empirical linguistic data and evidence. Since useful evidence is not always easy to come by to frame opinions and hypotheses, in this paper, we have conducted a pilot study with empirical analysis of real-life linguistic data to show how the pronominal systems of some of the Indian languages are similar in characteristics, formation, types, and usages. For this study, we have compared only the personal pronouns from the eight major Indian languages, which are used in different geographical regions of the country. The analysis of empirical data reveals that personal pronominal forms of these eight languages show a large number of similar characteristics, based on which it is possible to reconstruct relational links of commonness between these eight languages. It is also observed that the basic features of the pronominal system of these languages exhibit some notable differences from that of English, even though some of the Indian languages evolved from the same language family. The findings make us radical to form a conjecture of 'India as a Common Pronominal Area', which can either be attested or refuted with results acquired from elaborate analyses of the pronominal systems of all the major Indian languages included in the 8 th Schedule of the Constitution of India, as well as some minor and tribal languages. Some immediate benefits of this kind of study can be visualized in the development of theoretical frameworks for doing comparative studies of pronominal systems of all the Indian languages, as well as developing workable systems for teaching and translating pronouns across the Indian languages.

Nominal sentences and copula in the history of linguistics: the case of Aymara This paper aims to explore the notion of nominal sentence in the linguistic thought of missionary grammarians. This notion, which has been understudied in the linguistic historiography related to the early modern age, provides a better understanding of the interpretation of the concept of sentence in the Western grammatical tradition. In particular, I will attempt to assess the linguistic awareness of European grammarians about the existence of nominal sentences in Aymara, an indigenous South-American language. In fact, both ancient and modern Aymara show no verb 'to be' as it is conceived in Indo-European languages (cf. Benveniste 1960; Kahn 2003 [1973]). Consequently, both the existential function and the copula function are expressed through morphological and syntactic rather than lexical strategies. I will try to show how early grammars of Aymara (Bertonio 1603 [1596], 1612; Torres Rubio 1616), based on the traditional Graeco-Latin model, fail to correctly identify this phenomenon, which is known to be widespread cross-linguistically in current linguistic studies (cf. Stassen 1994, 1997; Bauer 2000). As a result, European missionary grammarians were forced to ascribe to Aymara a full lexical verb 'to be' that did not exist in that language (cf. Adelaar 2004, 2020; Cerrón-Palomino 2000). I will therefore attempt to show how both the existential function and the copulative function of the verb 'to be' in Aymara are the result of a grammaticalization process that gives rise to verbalizing suffixes, either locative or copulative, originating from spatial demonstratives (cf. Diessel 1999; Dixon 2003; Pustet 2003).

The Kukama-Kukamiria language has about thirteen grammaticalized constructions to express different semantic relations between events. These constructions exhibit asymmetric dependency and different degrees of syntactic integration with the main clause. In this study, we focus on four constructions that exhibit restrictions in terms of obligatory coreference between the arguments of the main event and those of the subordinate event. Two of them encode purpose, two manner/simultaneity. A notable fact is that in three of these constructions, the argument that controls the coreference in the matrix clause is the absolutive (S/O). This constitutes a typological novelty given that the most common crosslinguistic pattern is in which the nominative argument (S/A) of the matrix clause controls the coreference of arguments in the subordinate clause.

В работах Дж. Николс и В.Д. Соловьёва, написанных в конце 2000-х гг., был предложен типологический подход к языковой классификации. В тот момент подход был признан несовершенным и отвергнут. Сейчас, с связи с появлением новой  методологии, основанной на использовании подходов компьютерной филогенетики, представляется возможным провести новое исследование и заново подтвердить или опровергнуть ранее выдвинутые гипотезы. Исследование, проведённое с использованием новых методов компьютерной филогенетической лингвистики, подтверждает опасения, высказанные ранее представителями Московской школы компаративистики. Для построения филогенетической классификации, основанной на типологических признаках, необходимо в первую очередь создать соответствующие наборы данных, подготовленные и сбалансированные по генетическому критерию. Помимо этого, необходима тщательная внутренняя реконструкция анализируемых семей.

Over the last two decades, an increasing number of scientists have expressed worries as it has been increasingly clear that in a range of fields, a large proportion of scientific results could not be repeated (or “replicated”, or “reproduced”). Is this perhaps also a problem for comparative linguistics? For comparative grammar (or typology), reproducibility was first discussed two decades ago (Plank 2006), and as more and more research projects use advanced quantitative methods, it is of course important to have some discussion of which are the best methods, and whether different statistical approaches give the same results. It seems clear from discussions in other fields that the way scientists use statistical methods is often highly problematic. In 2005, John Ioannidis shocked researchers by suggesting that “for most study designs and settings, it is more likely for a research claim to be false than true” (Ioannidis 2005). ...

and Yokuts. Apart from Miwok-Costanoan, which has found general acceptance through the work of Callaghan (2014), California Penutian is highly controversial and has never been reconstructed to anywhere near the extent of Hokan, for instance. The core of Newman's proposal is a set of 123 sets of possible cognates (in addition to 64 set judged by the author himself to be of lesser quality). Phonological resemblances are typically weak, often semantic stretches are assumed, and onomatopoeia enters many forms. An additional problem is that comparisons are made between Zuni and any number of individual California Penutian languages, with no requirement that possible cognates be found in a substantial set of California Penutian subfamilies. There are as little as 2.7 different California languages per cognate set, and it gets worse when considering the number of different subfamilies: 55 sets provide evidence from one subfamily only, and another 37 sets from two subfamilies only. Applying a criterion that at least two subgroups should be represented would reduce the number of sets to 68. Some of the sets with the best representation of subgroups ('bluejay', 'breast', 'grandfather') are susceptible to sound-meaning correspondences. Using the work of classifies Zuni as Penutian within Amerind. Manaster Ramer (1996), however, is able to muster lookalikes from Indo-European that are just as plausibly related to the Zuni forms as Greenberg's presumed Penutian cognates, using that as an argument against the Zuni-Penutian proposal.

The present article surveys the system of negation in Tocharian, comprising the free syntactic negative, which is movable, and and the bound morphological negative, which is a prefix. For the syntactic negation, the rules of negative placement in Tocharian declaratives, interrogatives and directives are documented. The syntactic negative of Tocharian primarily serves as a verbal negator that can also be moved before nominal phrases to serve as a DP-external constituent negation. When used DP-internally to negate attributive adjectives, the inherited negative particle is incorporated into the adjective as a morphological prefix (DP-internal negation incorporation).

This paper argues for the existence of lexically selected expletives, i.e. semantically vacuous elements subcategorized for by a predicate. It draws primarily on evidence from weather expressions in Basque and Italian, but it also uses independent evidence from Spanish existential haber 'there is/are' and French falloir 'be necessary' structures. These constructions are problematic for an analysis based on either quasiarguments (Chomsky 1981) or 'traditional' (i.e. non-subcategorized) dummies, but they are amenable to an account with lexically selected expletives. Besides offering a unified analysis of seemingly unrelated phenomena, the proposal developed here yields a parsimonious theory of expletives. Moreover, the account provides additional evidence for claim that dummies can appear in strictly subcategorized positions, challenging the traditional assumption that syntactic licensing is necessarily concomitant with semantic role assignment. 1 I am very grateful to Carol Rosen, Nunzio La Fauci, Wayne Harbert, Brian Joseph, and several anonymous reviewers for their insightful comments on this paper or an earlier version thereof. Thanks are also due to all the native speakers who provided their judgments. Obviously, all errors and omissions are my sole responsibility.

Ambifixes are defined as affixes that can occur both as prefixes and suffixes. For example, in Lithuanian the verbal reflexive marker -si occurs either before the stem if the verb contains any other prefix (ne-si-maud-au ‘I do not wash myself’) or after the stem and other suffixes otherwise (maud-au-si ‘I wash myself’). After discussing some methodological issues related to the definition of ambifixes, I present my on-going typological study based on a cross-linguistic convenience sample of ca. 100 languages. My focus will be on the types of factors determining the position of the affix, which range from phonotactic and morphotactic constraints to morphosyntactic, lexical, and semantico-syntactic conditions. In addition to that, I look into the possible historical origins of ambifixes, singling out a number of diachronic pathways to ambifixation, e.g. affixalisation of variable-position clitics or univerbation of constructions containing auxiliary elements bearing an affix in the “right” position but as a whole occurring “on the other side” of the lexical element.

The goal of this paper is to put the semantic functions of the Lithuanian and Bulgarian perfects into a perspective of the grammaticalization cline specific to BE-perfects, i.e., perfects that exclusively use auxiliaries derived from verbs meaning 'to be'. The study is based on synchronic data and on the idea that diachronic conceptual shifts, characteristic of grammaticalization, are reflected in synchronic variation. The analysis follows semantic links between perfect functions from statives and resultative constructions to experientials and evidentials. Quantitative frequency data from genre-parallel collections of Facebook comments in Bulgarian and Lithuanian shows that, in both doculects, auxiliaries are more frequently omitted with less grammaticalized perfect functions, while with cross-linguistically typical functions the auxiliary is more frequently present. In Bulgarian, the auxiliary-drop tendency resurfaces with evidential functions of the perfect. I argue that this auxiliary usage pattern aligns with the grammaticalization cline of BE-perfects.

This paper investigates the emergence of phrasal verbs (verb-particle constructions) in colloquial Brazilian Portuguese, a phenomenon typically found in Germanic languages. Drawing on a corpus compiled from the social media platform X (formerly Twitter), the study identifies and analyzes over 90 distinct verbparticle constructions, such as chutar fora ("kick out") and pular dentro ("jump in"). The central argument challenges the typological assumption that Romance languages lack the syntactic and semantic properties required to support phrasal verbs, particularly those that express telicity through particles. Using Talmy's (2000) typology of motion event encoding, the paper shows that informal Brazilian Portuguese satisfies four out of five criteria typical of satellite-framed languages: expression of trajectory through a particle, use of manner or cause verbs, tendency toward telicity, and metaphorical usage. While the final criterion-particles acquiring purely aspectual value-is not fully met, the analysis demonstrates that particles in Brazilian Portuguese contribute to the telic interpretation of motion events, often profiling the endpoint of an action. This constructional pattern diverges significantly from the formal written standard taught in schools, contributing to linguistic stigmatization and sociolinguistic exclusion. The study argues that the verbal action in phrasal verbs is cognitively construed as a means to reach an endpoint, offering communicative nuances unavailable in the formal register. The findings call for a reevaluation of Romance typology regarding motion and aspect, and suggest that Brazilian Portuguese, at least in its colloquial form, is undergoing internal grammatical developments independent of English influence.

The grammars of human languages differ widely and in fascinating ways. Nevertheless, we often come across phenomena in unfamiliar languages which show striking similarities with phenomena in more familiar Ianguages. Very few of such phenomena are universal, in the sense that they occur in every language, but many are found in various Ianguages irrespective of their genetic and areai connections. As such, they must be seen as belonging, in some sense to be further specified, to universal grammar. One such cross-linguistically recurrent category is the CONVERB. Unti] very recently only few people knew the term. lt was first introduced by Gustav John Ramstedt ( ), but it was long ignored. It was picked up again by Vladimir Nedjalkov e Igor Nedjalkov (1987), a fairly inaccessible publication, mentioned by Martin Haspelmath and embedded within the EUROTYP project. The breakthrough is. in Haspelmath and Konig (1995). That the first typological studies of converbs appeared in Russian linguistics comes as no surprise: the languages of Russian colonia] areas of the Caucasus and Northem and Centrai Asia are the richest in converbs. The term itself is adopted from Altaic linguistics. Nedjalkov (1995) defines a converb as "a verb form which depends syntactically on another verb form, but is not its syntactic actant, ie. does not realize its semantic valencies" (Nedjalkov 1995: 25), and rephrases the concept by specifying its nature as "a nonfinite verb form whose main function is to mark adverbial subordination" (Haspelmath 1995: 3). Another way of putting it is that converbs are verbal adverbs, just like participles are verbal adjectives. According to , converbs are thus characterised by the following properties: they are l. verb forms; 2. nonfinite; 3. adverbial; 4. subordinate. Moreover, following Nedjalkov, canonica] converbs can occupy the position of an adjunct, that is of an adverbial, but cannot occupy the positions a) of the only predicate in a simple sentence; b) of nominai attributes like participles, c) of a clausal actant (it cannot be an infinitive or a gerund depending on such verbs as begin or stop), d) of a nominai actant like gerunds in subject or object position.

En este trabajo se analizan dos perspectivas de estudio comparativo de la diversidad lingüística: la tipología gramatical-funcional y la tipología semántica. Ambas se basan en una posición universalista empíricamente anclada por lo que se sitúan en una posición intermedia del continuum cuyos polos están representados por el antiuniversalismo y el universalismo. La principal diferencia entre ellas reside en considerar a las funciones semánticas y pragmáticas o equivalentes funcionales como punto de partida (tipología gramatical-funcional) o como punto de llegada (tipología semántica) en el análisis de la diversidad a través de las lenguas. Comparamos ambas perspectivas de estudio translingüístico y nos centramos en uno de los dominios más importantes de la cognición humana, como es el del espacio. Finalmente, presentamos una breve síntesis del análisis que venimos realizando del subdominio topológico en qomlé’k/toba del oeste de Formosa, lengua de la familia guaycurú, desde el enfoque tipológico semántico.

Building Adjectives: A Typological Overview of Adjectival Constructions
Keywords: Typology; Adjectives; Quality Modifiers; Constructions; Parts of Speech
It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

Following Croft (2001: 66ff.), the “adjective” is defined as the most typical quality modifier construction in a language. A variety sample of 100 languages is, thus, gathered for the present talk and languages are classified on the level of language structure at which the most typical adjectival construction (i.e. quality modifier) is fixed. In the sample, the quality modifier construction is fixed at six main levels. In Latin, Amele, Dyirbal and Lavukaleve the typical adjective is a simple stem; simple stem adjectives are marked by agreement in gender (Latin, Amele) or classifier (Dyirbal), and they can share most of their properties with nouns (Latin, Amele, Dyirbal) or verbs (Lavukaleve). In RigVedic Sanskrit, Classical Arabic and Yimas, the typical quality modifier is a derived stem that is formally different from a relative clause (that is, different markers code the two functions); derived stem adjectives agree in gender (RV Sanskrit, Arabic) or classifiers (Yimas). In Garo and Urarina the typical quality modifier construction is a derived stem that is also a relative clause (that is, a single affix code both functions, as in the relative-participles of Tibetan languages, Genetti 1992, 2005). In Tuscarora and Jamul Tiipay, the typical quality modifier construction is a clause that differs from a derived stem (i.e. a relative clause or a word-sentence marked for switch-reference). In Chukchi and, partly, in Nivkh, the typical quality modifier is a verb stem incorporated into a noun. Finally, in Vietnamese, Lao and Quechua, the quality modifier is a phrasal constituent that is, a quality lexeme settled in the modifier slot of the phrase without overt marking, as in Hengeveld’s “flexible” strategy (see Rijkhoff & Van Lier 2013). Phrasal adjectives can share most of their properties with nouns (Quechua) or with verb (Vietnamese, Lao).

If the preliminary results will be confirmed, the simple stem (i.e. the lexicon), the derived stem, the relative-stem, the phrase and the clause represent the focal layers of the lexico-syntax continuum defined by Croft (2001: 17), and the first sketch of a cross-linguistic theory of the levels of language structure.

Building Adjectives: A Typological Overview of Adjectival Constructions
Keywords: Typology; Adjectives; Quality Modifiers; Constructions; Parts of Speech
It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

Following Croft (2001: 66ff.), the “adjective” is defined as the most typical quality modifier construction in a language. A variety sample of 100 languages is, thus, gathered for the present talk and languages are classified on the level of language structure at which the most typical adjectival construction (i.e. quality modifier) is fixed. In the sample, the quality modifier construction is fixed at six main levels. In Latin, Amele, Dyirbal and Lavukaleve the typical adjective is a simple stem; simple stem adjectives are marked by agreement in gender (Latin, Amele) or classifier (Dyirbal), and they can share most of their properties with nouns (Latin, Amele, Dyirbal) or verbs (Lavukaleve). In RigVedic Sanskrit, Classical Arabic and Yimas, the typical quality modifier is a derived stem that is formally different from a relative clause (that is, different markers code the two functions); derived stem adjectives agree in gender (RV Sanskrit, Arabic) or classifiers (Yimas). In Garo and Urarina the typical quality modifier construction is a derived stem that is also a relative clause (that is, a single affix code both functions, as in the relative-participles of Tibetan languages, Genetti 1992, 2005). In Tuscarora and Jamul Tiipay, the typical quality modifier construction is a clause that differs from a derived stem (i.e. a relative clause or a word-sentence marked for switch-reference). In Chukchi and, partly, in Nivkh, the typical quality modifier is a verb stem incorporated into a noun. Finally, in Vietnamese, Lao and Quechua, the quality modifier is a phrasal constituent that is, a quality lexeme settled in the modifier slot of the phrase without overt marking, as in Hengeveld’s “flexible” strategy (see Rijkhoff & Van Lier 2013). Phrasal adjectives can share most of their properties with nouns (Quechua) or with verb (Vietnamese, Lao).

If the preliminary results will be confirmed, the simple stem (i.e. the lexicon), the derived stem, the relative-stem, the phrase and the clause represent the focal layers of the lexico-syntax continuum defined by Croft (2001: 17), and the first sketch of a cross-linguistic theory of the levels of language structure.

Keywords: Infinitive; Nominalizations; Word-class change.

In ancient Indo-European languages, the infinitive arises from the grammaticalization of a deverbal noun as a part of the verbal paradigm (Kuryłowicz 1964: 158-170; Bauer 2000: 337-349; Ramat 2002; Kapović 2017: 104; Lundquist & Yates 2018: 2110). This process is somehow completed in some languages (Latin, Ancient Greek, Balto-Slavic), where the infinitive shows the accusative government (Lat. mulcere dedit fluctus ‘(he) allowed (you) to rule the sea’ Verg. Aen. 1.66; Gk. líssom’ Akhillêï methémen khólon ‘I beg (you) to let go (your) anger against Achilles’ A, 283), while still ongoing in others (Vedic, Avestan), where it shows both genitive and accusative government (Ved. kó […] gāt práṣṭum etát ‘who will go to ask it? R̥V. 1.164.4d’; dhánānām sātáye ‘for the conquering of wealth’ R̥V. 1.4.9c). As a result, the infinitive has acquired a consistent number of verbal categories in Greek and Latin (namely, tense, aspect and diathesis: Fabrizio 2018a-b-c), unlike in Vedic (García Ramón 1995, 1997, 2009; Lazzeroni 2012): as such, it is conceivable as an action noun (Comrie & Thompson 1985) and cross-linguistically as an Action-Argument (in Croft’s terms: Croft 2001).
This contribution aims at: 1) investigating the morphological encoding of the Action-Argument correlation in three ancient Indo-European languages: Early Latin, Homeric Greek, and R̥gvedic Sanskrit; 2) giving an account of the process of word-class change that the infinitive undergoes.

Keywords: Typology; Adjectives; Quality Modifiers; Constructions; Parts of Speech
It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

Keywords: Typology; Adjectives; Quality Modifiers; Constructions; Parts of Speech
It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

Among the studies of possible sources topic markers can develop from, none has mentioned that topic markers can derive from conditionals. Haiman’s (1978) statement that “conditionals are topics” is well accepted, and based on related linguistic facts it has been further stated that “topics are not conditionals” (Shin 1987). The present paper, however, argues that the Rgyalrong topic marker =mənaŋorə has derived from the conditional verb form that has the copula ŋôs as the stem: mə-na-ŋôs=rə [COND-PFV-COP=TOP]. We suggest that three factors have made this development possible. One is that the stem of the related conditional verb form is a copula, which is semantically empty. The second factor is that the conditional clause, like most other clause types in Rgyalorng, is verb-final. The other concerns the fact that in the Rgyalrong dialect examined for the present study, the conditional clause always precedes the main clause. With conditional marking, the copula signals nothing but the givenness of information conveyed by its complement. Since the relative position between the conditional and main clauses is fixed, the path has been paved for [Conditional clause + Main clause] to evolve into [Topic + Comment]. Based on the evidence as observed in Rgyalrong as well as Turkic languages, it could be adequate to state: “The topics were once conditionals.”

Keywords: Typology; Adjectives; Quality Modifiers; Constructions; Parts of Speech
It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

Keywords: Typology; Adjectives; Quality Modifiers; Constructions; Parts of Speech
It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

It is well known that languages do not only differ in the features whereby they define their parts of speech (PoS) and in the number of PoS that they define, but also – and perhaps more importantly – in the levels of language structure at which they do so. As a confirmation, quite a few studies discussed the levels at which the noun can be defined across languages (Hopper & Thompson 1984, Mithun 2000, Lazard 1999). However, barring some notes in Thompson (1988) and Alfieri (2014), a similar approach to the study of the adjective class has never been proposed (exception made for Krasnoukhova 2022), although adjectives are missing more often than nouns across languages (see, e.g., Dixon 1982, Bhat 1994, Hengeveld 1992, Beck 2002; Dixon & Aikhenvald 2004, Haspelmath 2012).

The aim of the talk is to fill this gap and present a typology of the levels of language analysis at which adjectival constructions can be coded across languages.

This chapter gives an overview of the topics in Slavic tense which have generated most interest in the semantic linguistic community. Data from most Slavic languages are considered. The introduction identifies some of the main issues through selected data from Ukrainian, Croatian, and Bulgarian. §2 traces the history of Slavic tense semantics with focus on the formalisation of the temporal domain-the interaction of tense and aspect in a compositional system. The last two sections provide a more detailed account of what we take to be the main theoretical challenges in Slavic tense semantics: the role of temporal auxiliaries, notably the bud-future and the be-perfect (§3) and subordinate tense from complement tense to adjunct tense (§4). This last part centers around the role of so-called "relative tenses" in non-sequence-of-tense-languages.

In this thesis, the influence of culture and social organisation on language and interaction was explored. The areas of interaction that were focused on are the microsociology of questioning and requesting interaction occurring in everyday, informal conversations of ǂĀkhoe Haiǀǀom speakers as well as a language specific communication form called ǂgona: a type of conspicuous standing with the aim of making oneself available to be shared with. The findings were compared to what is known about (American) English interaction, and explanations were given for the commonalities and differences from a cultural perspective.
The overall conclusions of this thesis are threefold. First, I have shown that culture can drive the choice for a grammatical form of an utterance (for example open questions vs. closed questions) as well as the choice of action (for example requesting vs. ǂgona). ǂĀkhoe speakers have a preference for indirect questioning forms in keeping with other aspects of their culture. The form requests take is related to a culturally motivated categorisation of goods and products. The higher the imposition of the request, which is related to the properties of the object requested, the more indirect the request. Moreover, the language specific ǂgona behaviour is inextricably linked and would not exist without ǂĀkhoe specific categorisation of goods, settlement patterns and sharing obligations. The second overall conclusion is that despite these clear influences of the culture and social organisation on language use, the basic structure of interaction is not directly influenced, and this structure of interaction as it was laid out by Sacks, Schegloff and Jefferson (1974) is universal, as long as one includes physical communication along with the verbal communication. And finally, this work shows that these basics of human interaction are present even in very divergent cultures. Not only are they present in large-scale, industrialized, sedentary, hierarchical cultures, but they are also present in small-scale, nomadic, egalitarian cultures.