There is a clear opposition between traditional philology and modern linguistics in the understanding of ancient Chinese modal particles. The former assumes their stylistic and functional heterogeneity, while the latter assumes their stylistic and functional homogeneity. But consequently, the two schools have constructed distinct systems of modal particles and derived different functional classifications. Both perspectives contain conceptual biases that have, to some extent, influenced the study of ancient Chinese modal particle systems and their functions. This paper adopts the traditional philological viewpoint, categorizing ancient Chinese modal particles into two subsystems based on stylistic domains: song modal particles and conversation modal particles. Through this classification, it is revealed that they exhibit significant or even stark opposition in phonetic features, syntactic distribution, function, combinatorial patterns, and diachronic evolution. Additionally, the study employs a hierarchical approach to analyze the functions of song modal particles, while applying the syntactic distribution ternary method and mood structure hypothesis for conversation modal particles, ultimately identifying their subordinate and even more specific subsystems.

Within the framework of Diachronic Construction Grammar, this study investigates the obsolescence of the “jiào” (教) construction. Based on a quantitative analysis of historical texts, three major indicators of obsolescence are identified: declining frequency, semantic diffusion, and increased flexibility in alternation. It is argued that the loss of this construction results from the combined effects of external competition and internal weakening. Externally, the functionally specialized “jiào” (叫) construction exerted competitive pressure; internally, high functional load borne by the “jiào (教) + NP + VP” sequence contributed to its structural weakening . This study highlights the critical role of functional load in diachronic constructional change, demonstrating how the dynamic balance between form and function shapes the persistence and evolution of constructions within a language system.

Drawing on data from naturally occurring Chinese conversations, this study examines two types of collaborative construction in order to explore sentence completing at the discourse level. In abutting collaborative constructions, some syntactically incomplete utterances are interactionally complete, and interlocutors may complete these utterances to display mutual understanding. In overlapping collaborative construction, interlocutors may intervene at an early stage in the production of another speakers’s syntactically incomplete utterance and complete it to demonstrate their own understanding. These findings suggest that the discourse-level sentence completeness is shaped by syntactic projection, interactional action projection, and the use of multimodal resources in conversation. Among these factors, syntactic projection serves as the most fundamental but locally constrained mechanism, whereas interactional projection operates at a more global level. Furthermore, this paper argues for greater attention to complete conversational sequences and to the joint effect of syntax, interactional action, and multimodal resources in analyzing sentence completeness at the discourse level.

The semantics of the Mandarin sentence-final particle “eryi” （而已） has been debated in the literature, with two main approaches: scalar and exclusive. The scalar analysis (e.g., Lü Shuxiang 1999) treats “eryi” as a modal particle whose primary function is to downplay the significance of a proposition. In contrast, the exclusive analysis (e.g., Yip 2024) regards “eryi” as an exclusive element akin to “zhi” （只）, expressing “nothing other than this”. Departing from previous accounts, this study argues that the core semantics of “eryi” is scalar rather than exclusive. Although “eryi” can license non-exclusive interpretations in specific contexts, it becomes infelicitous in the absence of scalar meaning. Experimental evidence from a covered box paradigm (a sentence–picture verification task with partially obscured images) further corroborate this claim. The results show that native Mandarin speakers exhibit significantly higher acceptability of non-exclusive uses of “eryi” than those involving “zhi”  and “zhi...eryi”（只……而已）, while exhibiting no significant difference from baseline sentences without focus markers. These findings suggest that “eryi” is semantically distinct from the canonical exclusive marker “zhi”, and that “zhi...eryi” does not instantiate exclusive concord. Rather, its interpretation arises from the compositional interaction between the exclusivity of “zhi” and the scalar nature of “eryi”.

From the perspective of engagement category, the functional difference between the structure “V-shang yi-verbal classifier (V上一量_动量)”  and the structure “V-shang zheme/name yi-verbal classifier (V上这么/那么一量_动量)”  reflects an opposition between non-engagement and engagement. The former simply presents an action, without requiring the listener to attend to the specific circumstances of “V”, nor does it coordinate the interlocutors’ knowledge states. By contrast, the insertion of the deictic elements “zheme (这么)” or “name (那么)” in the latter structure upgrades the expression from a neutral statement to a performative one. Within the engagement framework, zheme and name display distinct functions：zheme operates as a marker of real engagement, whereas name functions as a marker of virtual engagement.

The construction “nan V”(难V) in Mandarin Chinese exhibits two productive readings: the effort reading and the effect reading. While these two patterns are highly parallel in terms of category identification, argument alignment, and gradability, they diverge significantly with respect to verb selection, syntactic transformation, and nominalization patterns, thus constituting a case of isomorphic heterogeneity. Building on recent developments in morphological theory, this paper proposes a cyclic approach to word formation in the syntax. It argues that both readings of “nan V” are packaged and categorized by a light adjective head, which accounts for their shared adjectival properties. The fundamental divergence between the two lies in the distinct cycles at which word formation in the syntax applies: the effort reading is derived at the root cycle, whereas the effect reading is formed at the phrase cycle. These different cyclic pathways involve varying degrees of cognitive packaging, which subsequently triggers a range of syntactic-semantic effects.

This paper examines the semantics and functions of the manner auxiliary verb “a/”(要) in Puxian dialect and explores its grammaticalization process. It argues that this  auxiliary verb developed through grammaticalization from verb “iu/iau”(要) , originally meaning “to take”, within the adjacent serial verb construction “NP+iu/iau(要)+VP”. Evidence for this grammaticalization pathway is further supported by the grammaticalization parallel development of the verb “lau”(留) , which occurs in the same syntactic position, as well as by theatrical examples drawn from early theatrical texts and their orthographic conventions.

In Eastern Zhou script, characters traditionally assigned to the “隶” (dài) phonetic series are frequently used interchangeably with “肆”(sì), despite the lack of phonological similarity in their reconstructed ancient pronunciations. Drawing on key graphic evidence-such as the form “” (sì) and its variant “” (sì) attested in the Qinghua Bamboo Slips, as well as the variant forms of “” (yì) that merge the “”with the "隶" graphic-this study argues  that the characters “” and “” are in fact distinct graphical variants derived from “肆” through inter-character graphical assimilation. This analysis clarifies the origin of “” and suggests that characters previously identified as “” in Western Zhou bronze inscriptions should be reconsidered. Furthermore, “” is shown to have developed from “” via simplification involving the reduction of repeated graphic elements. A parallel process of simplification affected the variant “”, producing a form closely resembling “隶”. A reexamination of textual and epigraphic evidence demonstrates that characters formerly read as “隶” in Eastern Zhou texts are more accurately understood as variants of “肆”. The graph representing “隶/逮” in its proper sense appears only later, first attested in Qin script.

Guxihuibian (古玺汇编) No.2558, a Chu seal during the Warring States previously read as “弡利”, should be reinterpreted as “佢梁” (ju liang, with 佢 equivalent to 巨), a variant of the personal names “巨梁” “渠梁” and “渠良” attested in Warring States and Qin-Han sources. In Chu materials, “利” (li) typically appears as “” while “” predominantly represents “(梁)” (liang). Therefore, when encountering “” in Chu script, the interpretation of “梁” (liang) should be prioritized over that of “利”.