Semantic prosody describes the affective meanings or connotations of a linguistic element, which are commonly understood in terms of positive and negative alignment. Under the view of Construction Grammar, not only individual words but larger linguistic structures may carry semantic and pragmatic meaning, including positive or negative semantic prosody. The present pilot study takes this approach as a point of departure to investigate LLMs’ sensitivity to subtle pragmatic meanings in the form of semantic prosody. 4 state-of-the-art LLMs as well as a human participant group were tasked with rating the connotation and pleasantness of the go-around-Ving construction said to carry negative semantic prosody as well as two syntactically parallel constructions with neutral/positive semantic prosody. Results showed that while LLMs exhibit some sensitivity to constructional semantic prosody, their rating behavior differed significantly from humans when pooled into one group. Compared to humans, LLMs gave higher median and mean connotation and pleasantness ratings while also assigning less neutral ratings across constructions. Although not generalizable due to the exploratory character of this study, the results demonstrate that subtle pragmatic phenomena like semantic prosody represent a promising research area when it comes to discerning LLM language abilities and how they compare to humans’.

Large language models ( LLMS) have made language technologies widely accessible to humanities scholars, but they have also intensified concerns about transparency, reproducibility, and interpretive responsibility. This article argues that graph-based meaning representations, especially Abstract Meaning Representation (AMR) and Uniform Meaning Representation (UMR), can function as interpretive infrastructure for humanities research in the age of LLMS. Meaning graphs are “thin” by design in that they deliberately encode a constrained set of semantic distinctions. That selectivity is not a weakness for humanistic inquiry; rather, it enables a disciplined workflow in which researchers can separate (i) the semantic commitments that a text licenses (events, participants, temporal and modal dependencies) from (ii) richer interpretive claims (stance, ideology, affect, narrative framing) that can be layered on top. I review AMR and UMR at a level accessible to humanities audiences, discuss what changes in the LLM era (including both opportunities and limits of using LLMS for semantic parsing), and propose humanities-centered workflows and research questions. Several compact sample analyses illustrate how meaning graphs can support interpretive tasks in historiography, narrative analysis, and translation studies. A final section explicitly lists resources (datasets, tools, and guidelines) to support reproducible experimentation.