Grant Academy Project Network Institute

In Modelling Perspectives a MA student in philosophy and a MA student in computational linguistics take the first steps toward developing a sound method to extract and interpret information about perspectives as expressed in philosophical texts in a computational way. We investigate appropriate calibration of an already existing model combining sentiment/opinion mining and event factuality (Van Son et al. 2014) for application to philosophical texts. We focus on an English corpus including one of the most influential philosophical books of the 20th century, Word & Object (1960) by W. V. O. Quine.

Philosophical texts aim mainly at exposing perspectives, i.e. evaluating the correctness of certain statements or the fitness of certain concepts. The knowledge we possess today on important concepts such as capitalism, tolerance or meaning is based on such perspectives. Perspectives include judging a statement’s truthfulness or a concept’s fitness, or expressing positive or negative sentiments about it. Both of these aspects of perspectives are valuable information when trying to interpret information from philosophical texts.

In this exploratory project we want to combine two areas of philosophy and computational linguistics in a novel synthesis, and obtain tools to explore philosophical texts computationally that can valuably complement and enrich traditional methods. To our knowledge, neither NLP-based factuality nor sentiment analysis has ever been applied to philosophical texts. The project will be carried out across two Universities (UvA/VU), and will run parallel to BA- and MA-level philosophy courses during which students are taught to apply traditional methods to the same corpus. We aim at producing at least a common paper or one MA thesis by one of the students, a(n annotated) corpus and scripts/tool prototype apt for philosophical exploration, and have the MA students present their results periodically to the two other groups of students working traditionally.