I recently tried out ChatGPT by asking it to compose dialogues between a Thomist and a Confucian. Its approach to this sort of dialogue-writing is clunkheadedly unimaginative -- regardless of prompt, almost every dialogue begins with the Confucian introducing the topic, then the Thomist agreeing or occasionally disagreeing and saying, "From a Thomistic perspective", or else with the roles switched, and then the dialogue ends with them agreeing on the value of discourse, and if you get it to diverge, it eventually reconverges to this pattern. Anything actually interesting has to be inserted by prompt; the closest I got to something interesting was having them discuss the importance of rituals for the moral life while watching birds in the park, and the Confucian introduced the topic by reflecting on the ritual life of birds, which topic was quickly dropped for the call-and-response. It's an interesting look at some of the limitations of this sort of model. The model is essentially 'predicting' a nonexistent text from the statistical relations of words associated with the prompt (and with previous prompts and results in the same chat). This gives results that in one way are very impressive but in another way are going to tend toward the most trite and banal pablum except by accident. But it's fun enough when one keeps the limitations in mind. The best of these dialogues, in terms of actual substance, began, after a series of others, with the following prompt:
The Thomist attempts to explain to the Confucian, in Neo-Confucian terms, the First Way, which is an argument for God's existence. What does he say to try to explain the First Way?
ChatGPT produced the following. Note, incidentally, that it does not actually follow the prompt -- the Thomist doesn't explain the First Way in Neo-Confucian terms -- probably because the actual prompt is more implied than directly stated, but comes back to its standard pattern.
Thomist: My friend, I would like to share with you one of the five ways that Thomas Aquinas argued for the existence of God. It is called the First Way, and it is based on the observation that there are things in motion or changing in the world.
Confucian: I see. From a Neo-Confucian perspective, we might describe this observation as the continuous transformation of the universe, where everything is constantly changing but also interconnected.
Thomist: That's a helpful way to think about it. From a Thomistic perspective, the First Way argues that everything in motion or changing must be moved or changed by something else, as nothing can move or change itself.
Confucian: I see. From a Neo-Confucian perspective, we might describe this as the principle of cause and effect, where everything that happens in the universe is caused by something else.
Thomist: Yes, that's right. From a Thomistic perspective, the First Way then argues that this chain of cause and effect cannot go on infinitely, as there must be a first cause that sets everything in motion.
Confucian: I see. From a Neo-Confucian perspective, we might describe this as the idea of the Ultimate Reality or the Ultimate Principle, which is the source of all existence and activity in the universe.
Thomist: Exactly. From a Thomistic perspective, this first cause or ultimate reality is what we call God, the unmoved mover or the first cause of all things.
Confucian: I understand. From a Neo-Confucian perspective, we might describe this as the idea of the Dao, which is the source and foundation of all existence, activity, and morality in the universe.
Thomist: Yes, that's a helpful way to think about it. The First Way is just one of the ways that we can come to know and understand God's existence, and it is grounded in observation and reason.
The Confucian is probably a bit too quick to identify the Great Ultimate with the first cause and both with the Dao; the Thomist just a bit too incurious about the Confucian's version of the principle of causality and about what exactly the Dao is and how it relates to the Ultimate Principle; the sense in which a Neo-Confucian takes the Great Ultimate to be a 'source of existence' is arguably not too far from how Aristotle would think of the first mover as a 'source of existence' (one could imagine Aristotle with suitable qualifications agreeing that the prime mover is like the pole around which the cosmos turns, or the Confucian with suitable qualifications agreeing that the Great Ultimate, which manifests by activity and stillness, moves as beloved), but probably much further from Aquinas's more generalized version; and typical Confucians would probably be more wary about anything that might suggest a mental or personal character to the Great Ultimate. But it delivers a plausible first approximation to an analogy.
