In contemporary artificial intelligence, the challenge is making intricate connectionist systems—comprising millions of parameters—more comprehensible, defensible, and rationally grounded. Two prevailing methodologies address this complexity. The inaugural approach amalgamates symbolic methodologies with connectionist paradigms, culminating in a hybrid system. This strategy systematizes extensive parameters within a limited framework of formal, symbolic rules. Conversely, the latter strategy remains staunchly connectionist, eschewing hybridity. Instead of internal transparency, it fabricates an external, transparent proxy system. This ancillary system’s mandate is elucidating the principal system’s (...) decisions, essentially approximating its outcomes. Leveraging natural language processing as our analytical lens, this paper elucidates both methodologies: the hybrid method is underscored by the compositional vector semantics, whereas the purely connectionist method evolves as a derivative of neural semantic parsers. This discourse extols the merits of the purely connectionist approach for its inherent flexibility and for a pivotal delineation: segregating the explanatory apparatus from the operational core, thereby rendering artificial intelligence systems reminiscent of human cognition. (shrink)

This paper explores a unified approach to linguistic reference and the nature of objects, addressing both abstract and concrete entities. We propose a method of redefining ultra-thin objects through a modified abstraction principle, which involves two distinct computations: subsemantic computation processes direct physical input, while semantic computation derives the semantic values of a sentence from the meanings of its constituents. These computations take different inputs—one physical and one semantic—but yield identical outputs. Among these, the subsemantic computation is more accessible. This (...) approach facilitates a consistent treatment across various types of objects, including mathematical, concrete, social, and mental entities, thereby eliminating the need for domain-specific justifications. We advocate for this innovative perspective and address potential objections related to idealism and the utility of introducing objects. Our proposal advances the discourse of the nature of objects and linguistic reference, providing a comprehensive framework for understanding the existence and reference of objects across diverse discourse domains. (shrink)

W. V. Quine suggests that meaning derives from the stimulus, the effects of the outside world on the subject’s nerve endings. However, the idiosyncrasy of the stimulus poses challenges to the intersubjectivity of meaning. Gary Kemp proposes the stimulus field approach as a solution. The stimulus field approach focuses on the forces affecting the subject, rather than the effects on the subject. In this article, I critique Kemp’s solution. Furthermore, I argue that the Quinean approach can be refined in a (...) way that two agents may converge on the same meaning even without sharing the same stimuli or forces affecting them. (shrink)

In his development of an ontology for mathematics in line with Fregean abstractionism, Øystein Linnebo proposes criteria of identity as a mechanism for reference. This proposed mechanism appears to conflict with Saul Kripke's historical theory of names. However, Linnebo, distinguishing between semantic and metasemantic domains, asserts that no real conflict exists; while Kripke's theory is semantic, his is metasemantic. This paper posits a counter-argument, contending that Linnebo's account, contrary to his claim, indeed conflicts with Kripke's historical theory of names. The (...) paper underscores that, in Kripke's framework, the reference relation transcends the individual, inhabiting a social domain. (shrink)

The standard approach to model how human beings understand natural languages is the symbolic, compositional approach according to which the meaning of a complex expression is a function of the meanings of its constituents. In other words, meaning plays a fundamental role in the model. In this work, because of the polysemous, flexible, dynamic, and contextual structure of natural languages, this approach is rejected. Instead, a connectionist model which eliminates the concept of meaning is proposed.