Are Language Models More Like Libraries or Like Librarians? Bibliotechnism, the Novel Reference Problem, and the Attitudes of LLMs

Are LLMs cultural technologies like photocopiers or printing presses, which transmit information but cannot create new content? A challenge for this idea, which we call bibliotechnism, is that LLMs often generate entirely novel text. We begin (Part I) with a sustained defense of bibliotechnism against this challenge showing how even entirely novel text may be meaningful only in a derivative sense, and arguing that, in particular, much novel text generated by LLMs is only derivatively meaningful. But we argue (Part II) that bibliotechnism faces a different, novel challenge, stemming from examples in which LLMs generate "novel reference", using novel names to refer to novel entities. Such examples could be smoothly explained if LLMs were not cultural technologies but possessed a limited form of agency (beliefs, desires, and intentions). According to interpretationism in the philosophy of mind, a system has beliefs, desires and intentions if and only if its behavior is well explained by the hypothesis that it has such states. So, according to interpretationism, cases of novel reference provide evidence that LLMs have beliefs, desires, and intentions. Given that interpretationism is a live hypothesis about the nature of these states, we suggest that cases of novel reference provide evidence that LLMs do have beliefs, desires, and intentions.