SHARP Project

SHARP investigates the role of definitions as tools for conceptual clarification, theoretical construction, and explanation. Its research programme is organised around three interconnected Research Questions (RQs), addressing the epistemic, logical, and explanatory dimensions of mathematical definitions. Each Research Question is articulated into three Research Lines (RLs), combining philosophical analysis, historical investigation, formal methods, and interdisciplinary case studies.

RQ1 — EXPLICATE

What is the epistemology of mathematical definitions?

RQ1 investigates how mathematical concepts are clarified, defined, and transformed in the development of mathematical knowledge. It examines the epistemic principles and methodological choices guiding the transition from informal mathematical notions and practices to precise theoretical concepts, with particular attention to the role of definitions in conceptual change.

RL1.1 — The Methodology of Conceptual Clarification

This research line examines the diversity of methods employed to clarify and define mathematical concepts, including recursive, inductive, implicit, and axiomatic definitions. Through historical and philosophical analysis, it aims to identify the epistemic constraints and methodological principles governing different definitional practices.

RL1.2 — The Transformation of Concepts: Explication Revisited

Building on Carnap’s notion of explication, this research line investigates how informal or pre-theoretical concepts are transformed into precise theoretical notions. It examines how conceptual transformations are guided by epistemic virtues such as fruitfulness, explanatory power, applicability, and conceptual fit, while balancing continuity with existing practices and theoretical innovation.

RL1.3 — Definitions from Practice to Theory

This research line explores how mathematical concepts emerge from concrete practices of reasoning, proof, classification, and modelling, and subsequently become stabilised within formal theories. Through selected historical and contemporary case studies, it investigates how definitions mediate between mathematical practice and theoretical systematisation.

RQ2 — REGIMENT

What is the logic of mathematical definitions?

RQ2 investigates the logical structure of mathematical definitions and their interaction with the logical frameworks in which they are formulated. It examines how different foundational assumptions influence conceptual regimentation, mathematical construction, and the inferential commitments associated with mathematical theories.

RL2.1 — Regimentation of Abstraction

This research line investigates abstraction principles and their role in the logical foundations of mathematics, focusing on neo-logicism, impredicativity, and quantification. Particular attention is devoted to Hume’s Principle, Frege’s Basic Law V, and alternative logical frameworks for abstraction, including dynamic and potentialist approaches.

RL2.2 — Construction of Infinity

This research line explores the logical and conceptual construction of mathematical infinity through different foundational programmes, including those of Poincaré, Weyl, Russell, and Zermelo. It examines the interaction between predicativity, quantification, and inductive definitions in classical and constructive approaches to infinite mathematical structures.

RL2.3 — Implicit Commitment of Mathematical Concepts

This research line investigates how mathematical concepts and theories may carry commitments extending beyond their explicit definitions and axioms. Drawing on Gödelian incompleteness, reflection principles, and theories of truth, it examines the relationship between formal systems, implicit commitments, and the inferential and semantic resources required to articulate mathematical knowledge.

RQ3 — EXPLAIN

How do definitions explain in mathematics, science, and AI?

RQ3 investigates the explanatory functions of mathematical and scientific concepts. It examines how definitions contribute to mathematical understanding, scientific modelling, and explanatory practices, extending the investigation to the role of conceptual structures in artificial intelligence and human–AI interaction.

RL3.1 — Mathematical Explanation through Concepts

This research line investigates how mathematical definitions contribute to explanation by revealing structural dependencies, organising mathematical knowledge, and establishing explanatory connections. Through historical and contemporary case studies, it examines how conceptual articulation can provide mathematical understanding beyond the inferential structure of individual proofs.

RL3.2 — Explicating Scientific Concepts

This research line examines how mathematical and scientific concepts are constructed, refined, and transformed within scientific theories and modelling practices. Focusing on concepts such as probability, information, entropy, and robustness, it explores the relationship between conceptual explication, mathematical formalisation, explanation, prediction, and practical applicability.

RL3.3 — Conceptual Explanations in AI

This research line investigates the role of conceptual structures in artificial intelligence and explainable AI. It explores how conceptual organisation, mathematical reasoning, and epistemic interaction contribute to explanations generated by AI systems, with particular attention to belief revision, higher-order epistemic states, and the conditions for meaningful human–AI explanatory interaction.

***

The three Research Questions are closely interconnected. The epistemological investigation of conceptual clarification (RQ1) informs the study of logical regimentation (RQ2), while both contribute to understanding the explanatory roles of definitions in mathematics, science, and AI (RQ3). Conversely, logical and explanatory case studies provide essential resources for evaluating and refining the project’s general account of mathematical definitions.

Through this integrated approach, SHARP aims to develop a comprehensive framework for understanding definitions as conceptual technologies that shape mathematical knowledge, scientific reasoning, and artificial intelligence.