Abstract While linguistic research on sign languages is getting larger and larger impact in the linguistic community, much less is known about languages in the tactile modality, namely the languages used by DeafBlind signers. These languages are originally adapted from their visual counterpart, although after that, they witness independent evolution. In this work, we present a detailed overview of the properties of tactile languages showing their degree of autonomy from visual sign languages and how these properties allow to re-assess some foundational aspects of human language, including the role of native users in driving linguistic change, and the impact of iconicity in transmitting meaning. The evidence discussed in the paper will give the tools to provide an informed answer to the question whether languages in the tactile-gestural modality should be considered as an autonomous externalization of human language.
External arguments have long been argued to hold a special status in the literature on argument structure. Recent works have converged on the view that their thematic interpretations are post-syntactic derivatives. However, most of these studies focus exclusively on arguments serving as the highest arguments within certain domains, leaving the thematic interpretations of intermediate external arguments understudied. This study aims to address this research gap by exploring the different causee interpretations in the Teochew mue-causative and ke-causative. I demonstrate that, like other external arguments, intermediate ones also have their thematic interpretations contextualized through syntactically-oriented event structural interpretation as post-syntactic derivatives. However, intermediate external arguments, such as causees, require a more complex two-step contextualization mechanism. In this mechanism, the complement-oriented approach, widely adopted to analyze the highest external arguments, provides only the initial interpretation, which is further modified during subsequent syntactico-semantic derivations.
This paper examines Mandarin nominal headless relatives, with a focus on how they obtain their lexical categories. I argue that Mandarin headless relatives represent a distinct subtype of [-wh] headless relatives, characterized by two key properties: (i) relativization involves movement, akin to their [+wh] counterparts, and (ii) an additional nominalizer is required, distinguishing them from their [+wh] counterparts. Building on the (re)labeling analysis of headless relatives, I propose that, because dislocated null-operators do not move along with their categorial feature [N], Mandarin [-wh] nominal headless relatives obtain their lexical categories through an external nominalizer that supplies the [N] feature. This contrasts with [+wh] nominal headless relatives, where the overtly fronted wh-word directly provides the [N] feature to the headless-relative node. Besides, I explore how the semantic asymmetries between [+wh] and [-wh] nominal headless relatives can be explained under the proposed analysis.
This paper defines the necessary and sufficient conditions for the licitness of indefinite subjects in Mandarin Chinese. The primary thesis is that their distribution is governed by two distinct but interacting principles. First, we posit, drawing upon Chierchia (1995. Individual-level predicates as inherent generics. In G. N. Carlson & F. J. Pelletier (eds.), The generic book, 176-223. University of Chicago Press), that Davidsonian arguments - both in stage-level and individual-level predicates, can be formally constrained to function as existential stage topics or generic situation topics. The presence of such a topic can license an indefinite subject, satisfying the discourse requirement for a sentence to have a topic as its starting point. Second, we further argue that the felicity of a Chinese indefinite subject also depends on a legitimate topic-comment structure, as defined by the topic- licensing condition from Pan and Hu (2008. A semantic-pragmatic interface account of (dangling) topics in Mandarin Chinese. Journal of Pragmatics 40(11)). This condition requires a non-null intersection between the sets denoted by the topic and the comment. The well-formedness of a Chinese indefinite subject is thus shown to be contingent not only upon the availability of a suitable discourse topic but also upon the licit semantic integration of the topic-comment structure itself.
This study employs Large Language Model (LLM)-based translation as a diagnostic probe for generative predictions regarding Agree and phi-feature valuation in the context of discourse-driven underspecification. The translation direction from Korean - a topic-prominent, radical pro-drop language lacking overt gender marking - to Spanish - a language requiring morpho-syntactically explicit realization of person, number, and gender - compels a process of feature reconstruction rather than direct transfer. Using controlled context-window probes on a 22,800-sentence parallel corpus, the study analyzes how non-biological learners reconstruct missing features. The results show that expanding the discourse context significantly mitigates person mismatches via antecedent accessibility, yet fails to eliminate gender mismatches. This asymmetry reflects the structural scarcity of gender cues in the Korean input, consistent with Minimalist expectations: discourse context facilitates antecedent selection, yet the reliable valuation of specific lexical phi-features necessitates richer formal specification than the source provides. Furthermore, model architecture crucially determines the reconstruction strategy, ranging from conservative non-commitment to the over-generation of spurious feature values. By treating LLMs as learners devoid of Universal Grammar (UG) but shaped by architectural biases - including the attention mechanisms, positional encoding schemes, and fine-tuning objectives that constrain how each model integrates contextual information - this study demonstrates how computational probes can test whether grammatical generalizations depend on structurally specified input or can be recovered from distributional context alone.
