A new Turing test: metaphor vs. nonsense

AI and Society 36 (3):677-684 (2021)
  Copy   BIBTEX

Abstract

My basic argument is that a computer cannot distinguish between metaphor and nonsense. This would be my new “Turing Test.” I was very fond of a particular Italian poem, but I was told by an Italian friend that it was a hackneyed poem of little worth. I then taught myself to experience the poem alternately, as real poetry and as the silly nonsense that my friend claimed it really was. Having done so, I realized that I could do the same with any metaphor, such as “having a green thumb.” Thinking about the nature of this switch, from the literal to the metaphorical, I also realized that it was the sort of change that could not be prescribed or even described: this, the basic aesthetic gesture, remains beyond the boundary of logical definition. It then dawned on me that it might provide a test for the validity of a “Turing Test,” by any definition. Can a computer track a mind as it goes through this transformation? I could not envisage such a possibility. This would be a “Turing Test” based on the discipline of aesthetics rather than on technology. It may even be argued that the ability to experience metaphor is the very definition of the human.

Links

PhilArchive



    Upload a copy of this work     Papers currently archived: 93,779

External links

Setup an account with your affiliations in order to access resources via your University's proxy server

Through your library

Similar books and articles

Analytics

Added to PP
2021-07-20

Downloads
24 (#647,262)

6 months
7 (#594,125)

Historical graph of downloads
How can I increase my downloads?

Citations of this work

No citations found.

Add more citations