Quotation is a linguistic device that speakers employ to reflect on language, to attribute utterances to speakers or writers, and to highlight particular expressions within discourse. The functions of quotation therefore range from metalinguistic reference and speaker attribution to indicating non-stereotypical uses of language. At the same time, quotation has posed a challenge for traditional semantic and pragmatic accounts, for instance by disrupting assumptions about compositionality and by blurring the boundary between truth-conditional content and metalinguistic reference across quotation types.
Quotation as a linguistic phenomenon has been granted its own graphic sign in writing: the quotation marks “…”. Yet research has repeatedly shown that this punctuation unit is not a necessary condition for quotation as a pragmatic phenomenon. Still, the correlation between quotations and their explicit marking remains undeniable, and it provides the point of departure for the present study. Based on three corpora (graduation exams, newspaper articles, WhatsApp messages), I demonstrate that quotation marks disclose important aspects of quotation practices. Writers employ them in the corpora with specific nuances. Despite such variation, all uses refer back to a common functional basis. The data reveal quotation marks as flexible resources whose functions shift across communicative settings. They are not mere graphic delimiters but registersensitive indicators, adapting to situational and genre-specific contexts. Describing this variability through the concept of register allows for a systematic account of how quotation marks operate within different communicative environments.
In Catalan and Italian (but not Spanish), the general action verb for making (and doing) derived from Latin FACĔRE can be used as a quotative verb, which introduces direct speech together with a highly mimetic component of the representation of the speech act (Spitzer 1924; Guardamagna 2010). A corpus study of Italian and Catalan shows that the usage of Catalan and Italian fer/fare ‘make/do’ differs from the verb of saying dir/dire ‘say’ in several aspects. Fer/fare in its quotative use is not transitive, but unergative (the quote cannot be pronominalised, the verb cannot be passivised etc.) and it can only introduce direct speech. We discuss the general properties of the verb in the two Romance languages and then the specific properties of quotative fer/fare as described in the literature and as observed in our corpus study. Possible bridging contexts (between sound and speech) will serve as a clue for our analysis, which includes a pragmaticalisation process. Drawing on the phenomenon of mimesis (e.g. Güldemann 2008), we argue that quotative uses of fer/fare stress the manner component of quotation, similarly (but not equal) to colloquial English be like. Thus, fer/fare subcategorises a covert manner demonstrative, i.e. a combination of encoded similarity (cf. be like) and demonstration (cf. Davidson’s 1968/69, 1979 demonstrative theory of quotation). Lastly, we claim that fer/fare is reduced in some contexts of usage, thus possibly at the beginning of a grammaticalisation process, but still to be analysed as a quotative verb.
This study investigates mixed quotation (MQ) and scare quotation (ScQ) in contemporary German-language newspapers, with a particular focus on how to distinguish between them for the purposes of corpus annotation within a broader project on the history of reported discourse. The analysis draws on a corpus of 800 articles from four Austrian newspapers published on 31 January and 1 February 2025. Its aim is to develop a taxonomy of ambiguous cases lying between MQ and ScQ. The results confirm that MQ and ScQ are best conceived as poles of a continuum. At one end are clear cases of MQ, where words are attributed to a definite speaker; at the other are clear cases of ScQ, where quotation marks signal unusual wording, irony, or authorial stance without attribution. Between these extremes lie ambiguous cases, where attribution is uncertain or context-dependent, and gradual shifts from quoting an identifiable speaker to referencing discourse more generally. For annotation purposes, I propose a pragmatic solution: all instances that can be attributed to a speaker, even indefinite or hearsay, should be annotated as MQ; the last steps on the continuum belong to ScQ; truly ambiguous cases should be excluded. Future research, with larger and more balanced corpora, should refine this taxonomy and explore production and reception, including how journalists and readers perceive or exploit the MQ–ScQ continuum.
Since completing your PhD thesis on ideophones in Japanese and other languages at Kobe University (Akita 2009), to your postdoctoral fellowship at the University of Tokyo and UC Berkeley, and now as Associate Professor at Nagoya University, where you’ve just hosted one of the largest international symposiums on iconicity (IcoLL2026), you’ve devoted your career to the study of ideophones and iconicity. You’ve written countless articles, authored several books—including a recent Japanese bestseller with the psychologist Mutsumi Imai, Gengo no honshitsu [The nature of language] (Imai & Akita 2023)—as well as editing many more, among them landmark publications like Ideophones, mimetics and expressives (Akita & Pardeshi 2019) and, more recently, the mammoth Oxford handbook of iconicity in language (Fischer et al. 2026). In both English and Japanese, for the general public as well as scholars, you’ve analyzed virtually every aspect of ideophones—their phonology, morphology, syntax, semantics, pragmatics, semiotics, typology, history, and acquisition. Tell me, what is it about ideophones that keeps you coming back for more? What makes them so special to you? I’m curious to know where this passion comes from.
This study investigates the use of camel-related proverbs as speech acts in Jordanian Spoken Arabic (JSA), drawing on Searle's (1969, 1976) Speech Act Theory. The data were collected from 150 native speakers of JSA aged 30-60. The participants were divided into three groups: 20 individuals provided data during the preliminary questionnaire phase, which guided and informed the data collection process; 80 participants completed an extended questionnaire; and 50 individuals participated in an acceptability agreement/judgment task. The analysis revealed that camel-related proverbs in JSA perform 10 distinct illocutionary forces: criticizing, expressing anger, praising, offering condolence, blaming, offering sympathy, threatening, advising, rejecting, and asserting. These illocutionary forces were categorized into four illocutionary acts: expressives, commissives, directives, and representatives. The findings highlight the versatility and contextual sensitivity of camel-related proverbs in JSA, underscoring their role as dynamic communicative tools rather than mere static folklore. Given the study’s focus on JSA, future research could explore the illocutionary acts and forces of camel-related proverbs in other Arabic dialects, offering insights into both shared and dialect-specific pragmatic patterns across the Arab world.
While a number of studies have been conducted about expressing intensification in languages such as English, less attention is given to Bantu languages on the topic. This paper, therefore, explores strategies used to express intensity in Chasu, a language of the Bantu family spoken in Same District of Kilimanjaro, Tanzania. The study is guided by the Functional-typological approach which assumes that linguistic structures such as intensifiers are shaped by their communicative and cognitive functions such as emphasis and expressing degree or speakers' stance. The approach propagates also that the structure of intensifiers is typologically diverse across languages. The study employs a descriptive research approach using data from both spoken and written forms. Spoken data were elicited through sociolinguistic interviews using 30 native speakers of Mamba Myamba Ward of Same District in order to capture conversations of everyday language use. To complement spoken data, written forms were reviewed from a Chasu hymn book to record sentences, phrases, and words with intensifiers. Transcription of recorded data was done using Chasu orthography followed by content analysis. The findings of this study reveal the use of both morphological and lexical strategies. In morphological strategy, intensification in Chasu is expressed by a morpheme -ish-a embedded as a suffix after the verb root and adjectives like in kund-ish-a (love so much) and mbahish-a (very big) respectively. From lexical strategy, intensifiers are expressed through adverbs haiwa and pere modifying adjectives and verbs. In addition, ideophonic adverbs, verbal and adjectival reduplication are strategies for expressing intensification in Chasu. Furthermore, there were few incidences of the use of borrowed Swahili intensifiers sana and kabisa. The study recommends more studies on expression of intensification in Bantu languages to widen up the knowledge scope and theoretical understanding. It also recommends a comparative study examining the structural diversity of intensifiers across languages.
This paper elaborates on the comparison between Harmonic Parallelism (HP) and Harmonic Serialism (HS) as Optimality Theory (OT) models concerning nasal place assimilation in Modern Colloquial Persian. The data of this study were harvested from extant literature on place assimilation, including books, articles, and theses. Also, four native speakers (two males and two females) of Persian from Tehran were consulted to verify data extracted from the literature. The study concludes that place assimilation in this variety of Persian is restrictively regressive due to the impact of the coda condition when dealing with the serial application of place assimilation; hence, regressive place assimilation is executed in two steps where the first one is to delete the place feature of the coda which facilitates the leftward spread of the place feature of the following onset, as the second step; i.e., feeding order. HP encounters difficulty generalizing the place assimilation above since it is inherently derivational. Unlike HP, HS successfully expresses the serial relation beyond nasal place assimilation in Modern Colloquial Persian. Furthermore, HS can explain the directional asymmetry in place assimilation in Modern Colloquial Persian.
The title clearly reflects a comparative focus between Bishnupriya Manipuri (IndoIndo-Aryan and Tibeto-Burman languages, focusing on number, gender, case, and pronoun systems. On the basis of primary field data and secondary sources, the study showed that although the two languages show superficial similarities as a result of prolonged contact, their underlying grammatical organization is remarkably different. The findings show that even in extreme contact zones, core morphosyntactic traits are resistant to contact-induced change. This study contributes to linguistics typology and contact linguistics by highlighting the limits of structural convergence.
Iconicity is a relationship of resemblance or similarity between the two aspects of a sign: its form and its meaning (Meir & Tkachman (2018), and it is a general feature in all sign languages. This work sets out to investigate the manifestation of iconicity in the signing patterns of Igbo home signers, with a view to identifying the forms iconicity take in the language. A wordlist containing some iconically motivated Igbo lexical items is administered to selected deaf Igbo home signers to sign. The pictorial representations of these lexical items are presented and motivations behind them also given. Using the descriptive approach and borrowing a leaf from Edward (2005), the data so collected are analyzed and discussed under the concepts of size and shapes, expression of time, emotive and cognitive signs, and directional verbs. The study observes that the Igbo sign language is highly iconic, and has a great similarity with iconic sings in other African Indigenous Sign Languages such as Adamorobe Sign Language (AdaSL). The study concludes, in line with cross-linguistic findings, that iconicity is an essential of sign languages. The study recommends more scholarly works on iconicity in other African indigenous sign languages.
This article examines the cognitive foundations of strategic manoeuvring (SM) within the pragma-dialectical (PD) framework. Although PD is grounded in the principle of externalisation, which prioritises publicly attributable commitments over private mental states, the strategic character of SM implies goal-directed planning and anticipatory orientation towards interlocutors’ likely responses. Building on Kraus’s (2025) analysis of what the term strategic designates, the paper argues that this orientation presupposes a specific socio-cognitive capacity: perspective-taking (PT), understood here as the ability to model another agent’s epistemic, evaluative-affective, and inferential stance. Three analytically relevant modes of PT — epistemic, evaluative-affective, and inferential — are distinguished and mapped onto the three aspects of the SM triangle: choices from topical potential, audience adaptation, and presentational devices. This mapping shows that SM may be understood as the discourse-level realisation of perspective-sensitive cognitive operations: selecting arguments from topical potential requires managing common ground and tracking the interlocutor’s epistemic landscape; audience adaptation depends on modelling the addressee’s values and affective orientation; and presentational devices involve anticipating inferential uptake and processing effort. The paper further argues that integrating PT into a PD account need not violate the principle of externalisation, because the two operate at different explanatory levels: externalisation governs what may be reconstructed and normatively evaluated, whereas PT provides a psychologically realistic account of how arguers navigate the argumentative predicament. The framework is illustrated through a brief constructed example. The paper concludes that PT supplies the cognitive infrastructure for strategic calibration, while SM constitutes its argumentative implementation in discourse.
This paper examines elision as a syllable structure process in Nzema, a Niger-Congo (Kwa, Bia) language spoken mainly in the Western Region of Ghana and the Ivory Coast. The paper discusses some phonological and morphological factors that necessitate elision in Nzema. The paper discusses the processes involved in elision and the contexts within which the phenomenon occurs in Nzema. It first considers V1 and V2 elision in some phrasal and compound constructions. It also discusses syllable elision that operates within words. The paper shows that segments, syllables, and tones are affected by elision. It is observed that although elision is purely phonological, it is also motivated by morphological factors as vowels interact during compounding, and at word boundary. The analysis shows that in Nzema, morpheme2-initial Vs (i.e., V2) can be targeted for deletion in compounds and across word boundary, while elision of the first vowel, V1, only occurs across word boundary. Again, syllables that are lost word internally are compensated for by lengthening the initial vowel, V1 of the word. Moreover, in a context of V1#V2 where the two vowels are of the same quality, the target for the deletion process is determined by tone. The findings show a link between the phonology and morphology of Nzema in the analysis of elision.
This study examines the web interface of the Kazakh National Corpus (KNC) to assess its usability for general corpus-linguistic research. Drawing on established literature, we determine whether the corpus interface supports core analytical techniques, including frequency lists, concordances, keyword lists, and collocations and n-grams. We assess the presence, accessibility, functionality, and overall usability of these tools, identifying areas for improvement. To complement this theory-driven evaluation, we compare the KNC interface with those of two established national corpora, the British National Corpus (BNC) and the Russian National Corpus (RNC), to extract practical insights from their interface design and application of these analytical techniques. This two-level approach allows us to highlight both the theoretical requirements and practical considerations for enhancing the usability of the corpus interface for the purposes of international-standards linguistic research.
This paper examines the morphology, syntax and functions of demonstratives in Kiwoso. Similarly to other nominal modifiers in Bantu language, demonstratives are closely tied to the noun class system, exhibiting agreement in number and grammatical gender. Kiwoso distinguishes three-way system of demonstratives: proximal, medial, and distal. Syntactically, demonstratives serve both as pronominal replacing nouns, and adnominal modifying nouns. Canonically, adnominal demonstratives occur post-nominally. However, depending on the functions they serve, they can appear before the nouns, for example in copular constructions. In terms of functions, demonstratives in Kiwoso serve several other functions in addition to spatial deixis, the basic function of demonstratives across languages. Demonstratives are used for emphasis and recognitional purposes. This paper offers a detailed description of demonstratives, situating Kiwoso within the broader linguistic typology of Bantu languages. The findings contribute to a deeper understanding of morpho-syntactic structures, linguistic and communicative uses of demonstratives in Bantu languages.
This study examines how Vietnamese motion events conflate semantic components in verb roots, and serial-verb constructions in light of Talmy's and Slobin's typological predictions. This study employed a qualitative, typologically guided analysis, and it was supported by descriptive frequency counts. The data were compiled from two dictionaries and seventeen online stories, which produced 119 manner-motion verbs, 250 path-motion verbs, 265 cause-motion verbs, and 300 contextual tokens. The results and a limited range of Figure and Ground in verb roots, whereas external components Co-motion are expressed through verb phrases and serial verb constructions. These distributions point to a stable hybrid profile in Vietnamese that combines verb-internal packaging with expansion at the level of constructions. The study provides a more precise refinement of motion-event typology for Vietnamese and offers implications for language teaching, translation practice, and natural language processing applications.
This study aims to analyze the hearer-oriented clitic in Jordanian Arabic from a variational perspective. Even though there has been extensive research to examine this clitic, earlier studies have focused mainly on analyzing this clitic from a grammatical or socio-pragmatic perspective. Recent studies that examine this clitic from a generative perspective show that this clitic is subject to dialectical variation. However, such a claim lacks empirical evidence. This study, therefore, seeks to answer questions related to the sociolinguistic status of this pragmatic marker in Jordanian Arabic. The study examines the impact of sociolinguistic factors, namely age, gender, and region, on realizing this clitic. The data is collected through face-to-face sociolinguistic interviews with sixteen persons from Jordan. To analyze the data quantitatively, the researchers used GoldVarb X. The findings show that Haddad’s (2018) conclusion may lead to hasty generalization. Nevertheless, the study supports Al-Raba’a (2020) observation that hearer-oriented clitics are subject to dialectical variation.
The model of Derivational Networks, as developed by Kortvelyessy et al. (2020), examines the word-formational potential of lexemes from a unique three-dimensional perspective. However, the original model deliberately excludes all non-affixal word-formation processes, such as conversion, transflexion, and morphological motivation. As noted by Popova (2020), this exclusion results in relatively shallow and sparsely populated networks, particularly in languages where conversion plays a highly productive role in word-formation. This paper aims to provide a more comprehensive perspective on the role of conversion in derivational networks by contrasting the findings of Kortvelyessy et al. (2020) with those of networks where conversion is systematically incorporated. To achieve this, the study reconstructs the original English networks, as compiled by Popova (2020), and the Slovak networks, as compiled byIvanova (2020), integrating conversion, transflexion and morphological motivation. Given that English relies heavily on conversion, while Slovak exhibits limited productivity in this regard, the comparison of these two languages offers valuable insights. The study seeks to determine whether the inclusion of non-affixal processes significantly alters the structure and density of derivational networks, thereby refining our understanding of word-formation patterns across languages.
The Jaton isolect, born from prolonged contact between Javanese and Tondano speakers in North Sulawesi, has a distinct yet understudied morphological structure. This study examines affixation, reduplication, and compounding to assess Javanese influence. Using a qualitative descriptive method, data were collected through observation, interviews, and vocabulary lists from native speakers in Minahasa, then analyzed via morphological classification and element breakdown. Vocabulary was gathered using a basic list method, focusing on core words across specific areas, with Jaton speakers providing equivalents during structured interviews. Results show Jatons' morphology aligns mainly with Tondano, featuring five prefixes, three infixes, nine suffixes, three circumfixes, and two reduplication types involving affixation and vowel change. Jaton's morphological analysis includes affixation with five prefixes, three infixes, nine suffixes, and three circumfixes. Reduplication is divided into wisesa dwilingga and dwipurwa. Compounding forms new meanings, aligning Jaton's structures with Tondano, devoid ofJavanese influence. The study concludes that despite its contact-language origin, Jaton retains a predominantly Tondano framework with minimal Javanese influence, offering insight into isolect development in contact settings.
The study presents a semantic analysis of the English phrasal verb particle up (e.g. pick up, set up, take up) along with the mapping divergence in its equivalent Hindi verb forms. The paper selects multiple senses of the particle up in phrasal verb combinations from Rudzka-Ostyn (2003), collects example sentences and manually translates those constructions into Hindi by looking up relevant dictionaries and corpus resources, including native-speaker intuitions. The study attempts to identify the semantic correspondence in the verbal system of both languages, mainly while mapping English particle semantics in Hindi. The analysis has been extended to the Hindi data to identify the divergence patterns in the English-Hindi sense mapping of the respective particle. The study observes that the Hindi main and light verbs map the literal and extended senses of the particle. Additionally, for the phrasal particle constructions that map onto compound verb constructions in Hindi, the role of the light verb (V2) also maps the semantics of the respective particle.
This paper explores a number of formal similarities and differences between the two samples of intentional lexical blends in English, namely 100 blends in commercial names and 100 newspaper blends. By conducting qualitative, quantitative as well as statistical analyses, it aims to determine how different (or similar) the blends from the two qualitatively different domains-commercial names and newspapers-are as regards a number of their formal characteristics, namely: (a) blending techniques (based on their relative morphotactic transparency), (b) phonemic, graphemic and phonological similarity between the blend's source words, and (c) phonemic, graphemic and phonological similarity between each source word and the blend. Contour blending emerges as the most frequent technique in both domains, but commercial names favor morphotactically less transparent types (e.g., fragment and semi-complete blending), whereas newspaper blends prefer transparent strategies such as complete blends and overlapping structures. Hyphenation is notably more common in newspaper blends, likely to increase source word recognizability. Statistical analyses reveal that in both domains, source words are significantly more similar phonemically than graphemically, both to each other and to the resulting blend. While the difference in phonemic versus graphemic similarity between the two source words is more pronounced in commercial names, this difference between each individual source word and the blend is stronger in newspaper blends. Phonological analysis shows that blends tend to be longer than the first source word and typically match the second source word in syllabic length, with this tendency being more pronounced in newspaper blends. Stress pattern data reveal a strong cross-domain preference for the blend to inherit its primary stress from the second source word, again more prominently in the newspaper sample.
