A Unified Representation for Continuity and Discontinuity: Syntactic and Computational Motivations

2025-06-06

AI Summary

Phrase structure grammar (PSG), dependency grammar (DG), and categorial grammar (CG) lack a unified representational framework for modeling both continuous and discontinuous syntactic structures—e.g., cross-clausal dependencies in Turkish—hindering theoretical integration and cognitive plausibility. Method: We propose the first cross-formal-grammar unified syntactic representation framework, jointly driven by syntactic theory and computational complexity. Grounded in a correspondence principle, it systematically integrates PSG, DG, and CG across structural mapping, derivation mechanisms, and complexity assessment. Discontinuity is explicitly internalized—not treated as an exception—via formal modeling, cross-grammar structural alignment, and asymptotic complexity comparison. Contribution/Results: The framework demonstrably reduces syntactic processing load at the neurocognitive level, as validated through formal analysis and complexity-theoretic evaluation. It establishes a novel paradigm for unifying formal grammars and advances computational linguistic modeling by bridging theoretical syntax, formal language theory, and cognitive plausibility.

Technology Category

• Natural Language Processing: Syntax — Tagging, Chunking & Parsing
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Symbolic Representations

Application Category

• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

This paper advances a unified representation of linguistic structure for three grammar formalisms, namely, Phrase Structure Grammar (PSG), Dependency Grammar (DG) and Categorial Grammar (CG) from the perspective of syntactic and computational complexity considerations. The correspondence principle is proposed to enable a unified representation of the representational principles from PSG, DG, and CG. To that end, the paper first illustrates a series of steps in achieving a unified representation for a discontinuous subordinate clause from Turkish as an illustrative case. This affords a new way of approaching discontinuity in natural language from a theoretical point of view that unites and integrates the basic tenets of PSG, DG, and CG, with significant consequences for syntactic analysis. Then this paper demonstrates that a unified representation can simplify computational complexity with regards to the neurocognitive representation and processing of both continuous and discontinuous sentences vis-`a-vis the basic principles of PSG, DG, and CG.

Problem

• Unified representation for PSG, DG, and CG formalisms
• Addressing syntactic and computational complexity in linguistics
• Integrating continuity and discontinuity in natural language analysis

Innovation

• Unified representation for PSG, DG, and CG
• Correspondence principle integrates grammar formalisms
• Simplifies computational complexity in syntax processing