This is a JSON-format, machine-operable, open-licence subset of the research database of the “Dissident Networks Project” (DISSINET, https://dissinet.cz), focusing on conceptual and verbal entities which hold the database semantically together - its lexico-semantic network. DISSICON stands for DISSI(NET) and CON(CEPTUAL NETWORK). The broader project focuses on the semantic annotation and computational analysis of medieval inquisition records in a social science framework, which for the most part defines (albeit not exclusively) the focus of this lexico-semantic network. The dataset published here contains the full set of: Action entities (n=907), that is verbs and multiword verbal expressions (entities.json); Concept entities (n=5,847), that is non-verbal parts of speech (nouns, adjectives, adverbs, occasionally other types), including multiword expressions (entities.json); semantic relations (n=9,792) between those entities (relations.json), including hypernymy (n=5,310), actant role semantics / functors (n=1,989), synonymy (n=507), action/event equivalence (n=764), and others; those Reference entities that Concepts and Actions use for pointing to external resources, typically to WordNet and the Lila Lemma Collection (entities.json); Value entities mostly denoting textual values of IDs within the resources denoted in those References (entities.json). A basic, gradually extended documentation of the broader data model is available from the website From texts to structured data: Building knowledge graphs through Computer-Assisted Semantic Text Modelling (CASTEMO). The dataset’s value for social science history and data-oriented history is that it provides an in-progress, but already powerful machine-operable ontology for research into medieval inquisition records, including semantic relations such as synonymy or sometimes quite elaborated taxonomic hypernym paths. For interoperability and back compatibility reasons The value of DISSICON for lexicography is that it maps various uses of verbs and other parts of speech, with special attention paid to meanings of Latin verbs and concepts in medieval inquisition records and their semantic anchoring in English-language analytical concepts. A significant part of the entities refers to WordNet for meaning identifiers, but occasionally we also create our own meanings if WordNet did not yield a meaning with a satisfactory definition. The dataset identities are UUID-based, and our internal processes make the meanings reasonably persistent and thus referrable. The typical users of this dataset will be users of a new instance of the InkVisitor software, who will decide to pre-populate their instance with this existing data, if they find it useful. This specific version (v1) was used to populate the MedHate instance of InkVisitor. We know well that the work is ongoing and the semantic relations may be of imbalanced depth, can contain inconsistencies, and also that the mapping of this ontology upon some relevant standards, such as Ontolex Lemon for the lexicographic parts, is still due; we hope to progress on this soon. The contributions to this dataset, based on the Audits feature of the InkVisitor software (https://inkvisitor.net/) in one of whose instances the database is curated, are as follows, with contributions 5% or more amounting to authorship, and 1% or more but less than 5% to contributorship: David Zbíral 39.13 % (and data management); Robert Shaw 19.49 %; Katia Riccardo 17.67 %; Katalin Suba 14.94 %; Davor Salihović 5.65 %; Stanisław Banach 1.19 %. The creation of the dataset has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No. 101000442, project “Networks of Dissent: Computational Modelling of Dissident and Inquisitorial Cultures in Medieval Europe”).