Philosophy of Language and Computation I, Spring 2026
ETH Zürich: Course catalog
Course Description
This graduate class, partly taught like a seminar, is designed to help you understand the philosophical underpinnings of modern work in natural language processing (NLP), most of which is centered around statistical machine learning applied to natural language data. The course is a two-semester-long journey, but the second half (PLC II) does not depend on the first (PLC I) and thus either half may be taken independently. In each semester, we divide the class time into three modules. Each module focuses on a philosophical topic. In the first part, we will discuss logicist, structuralist, and generativist approaches to language, while in the second part focuses on language games, information theory, and critical perspectives on meaning. The modules will be four weeks long. Half of each module will be devoted to reading and discussing theoretical texts and supplementary criticism. In the other half, we will read recent NLP papers and discuss how they relate to philosophical insights into our conception of language—perhaps implicitly or unwittingly.
The course is designed to foster fruitful exchanges between students from different disciplinary horizons, especially between the STEM and the Humanities. As such, no prior knowledge of CS/AI/NLP or Philosophy is assumed. We particularly encourage humanities students to participate in this course, as we plan to pair students in cross-disciplinary tandems.
The course will require a weekly reading of around 30 pages (with relatively high variance at times) and weekly tasks related to the reading (free-form response or in-text comments), which are to be completed online. The weekly tasks are short and not graded; however, to pass the class, at least 70% of the tasks must be completed. See this document for the guidelines on the weekly tasks. The final grade will be based on one class presentation (in teams) and one individual oral evaluation, supported by a short text (1-2 pages maximum), at the end of the semester. The oral evaluation will include a presentation of a topic of the student’s choice, followed by an open discussion with the teaching team. The topic must correspond to a different module than the one already addressed in the class presentation, and the students will be expected to explore the relation of the topics discussed in class to work not presented in class, focusing on the connection between the philosophy of language and NLP. For example, discussing how recent NLP papers implicitly assumed or contradicted a structuralist or a logicist perspective on language would be a good topic.
Since one of the main aims of this course is to foster dialogue between different disciplines and perspectives on language, class presentations will, whenever possible, be given by tandems composed of one student with technical background and one with a background in the humanities. These interdisciplinary tandems will be given priority when reserving presentation slots during regular class sessions. Groups that are not formed as tandems will, of course, also have the opportunity to present, though their presentations may be scheduled outside of regular class time when necessary.
Grading
Marks for the course will be determined by the following formula:
- 35% Class presentation
- 65% Oral evaluation (15% supporting text + 50% oral presentation and discussion)
More information about the presentation and the oral evaluation expectations can be found here:
Organization
Lectures: Tue 18-19, ML D 28. The lectures will be given in person. This recurring Zoom meeting (ID: 632 1400 8040) will be used throughout the semester for people who want to tune in remotely. However, given the discussion based character of this course, in person participation is strongly encouraged. The password can be found on the course Moodle page. The Zoom recordings will be made available on the course Moodle page.
Discussion Sections: Tue 19-20 ML D 28.
Communication Moodle will be the main communication hub for the course. You are responsible for receiving all messages broadcast in Moodle
Class Materials
News
17.2 Class website is online!
Syllabus and Schedule
| Week | Date | Module | Topic | Reading |
|---|---|---|---|---|
| 1 | 17.02.26 | Introduction | Images of Language | Nietzsche (1873), Nietzsche (1873) in German Borges (1940a), Borges (1940a) in Spanish, Borges (1940b), Borges (1940b) in Spanish, Borges (1942) Borges (1942) in Spanish |
| 2 | 24.02.26 | Language and Logic: Formal Representations of Meaning |
Meaning: Between Sense and Denotation | Frege (1892), Frege (1892) in German, Russell (1905) |
| 3 | 03.03.26 | Formal and Natural Language | Carnap (1955), Montague (1970) | |
| 4 | 10.03.26 | Semantic Compositionality in Distributional Settings | Baroni et al. (2014) | |
| 5 | 17.03.26 | Logic, Compositionality, and Neural Nets | Zettlemoyer and Collins (2005), Andreas (2019) | |
| 6 | 24.03.26 | Structuralism: The Search for the Right Units |
Structural Dualities | Saussure (1916), Saussure (1916) in French, |
| 7 | 31.03.26 | Structure and Meaning | Harris (1954), Greimas (1966), Greimas (1966) in French | |
| 8 | 14.04.26 | The Right Linguistic Units | Goldsmith (2001), Kim et al. (2015), Sennrich et al. (2016), Vania et al. (2018), Nawrot et al. (2023) | |
| 9 | 21.04.26 | Arbitrariness of the Sign | Gutiérrez et al. (2016), Pimentel et al. (2019), Pimentel et al. (2021) | |
| 10 | 28.04.26 | Recursive Structure: The Computational Nature of Language |
Syntax vs. Distribution, Logic, and Behavior | Chomsky (1953), Chomsky (1955), Chomsky (1959) |
| 11 | 05.05.26 | Syntactic Structures | Chomsky (1957), Chomsky (1956), | |
| 12 | 12.05.26 | Syntactic Parsing—NLP’s Original Leaderboard | Charniak (2000), Hall et al. (2014), Vinyals et al. (2014), Dyer et al. (2016), | |
| 13 | 19.05.26 | Computational Approaches to Human Syntactic Processing | Hale (2016), Hale (2001), Levy and Jaeger (2007), Delétang et al. (2022) | |
| 14 | 26.05.26 | Conclusion |
Secondary Literature
| Module | Topic | Reading |
|---|---|---|
| Language and
Logic: Formal Representations of Meaning |
Philosophy | Ajdukiewicz (1936), Carnap (1947), Lambek (1958), Davidson (1967), Lewis (1970) |
| NLP | Steedman (1996), Wang and Eisner (2016), Lake and Baroni (2018), Ravfogel et al. (2019), White and Cotterell (2021) | |
| Structuralism: The Search for the Right Units |
Philosophy | Hjelmslev (1943), Harris (1951), Firth (1957), Hocket (1960), Gastaldi (2021) |
| NLP | Mielke and Eisner (2019), Nikkarinen et al. (2021), Mielke et al. (2021) | |
| Recursive
Structure: The Computational Nature of Language |
Philosophy | Piantadosi (2023), Chomsky et al. (2023) |
| NLP | Klein and Manning (2003), Meister et al. (2021), Kitaev et al. (2022) |
Useful Resources
- Manning (2022). Human Language Understanding & Reasoning
- Sanderson (2017-2025). 3blue1brown: Neural Networks. The basics of neural networks, and the math behind how they learn.
- Karpathy (2022). nanoGPT.
- Wolfram (2024). What Is ChatGPT Doing … and Why Does It Work?
Contact
You can ask questions on the Moodle forum. Please post questions there, so others can see them and join the discussion. If you have questions which are not of general interest, please don’t hesitate to contact us directly, i.e., email the lecturers with the TAs cc-ed.
