University of Sussex
Browse
physical-omniscience.pdf (215.81 kB)

Artificial consciousness, meta-knowledge, and physical omniscience

Download (215.81 kB)
journal contribution
posted on 2023-06-09, 23:19 authored by Ron ChrisleyRon Chrisley
Previous work [Chrisley & Sloman, 2016, 2017] has argued that a capacity for certain kinds of meta-knowledge is central to modeling consciousness, especially the recalcitrant aspects of qualia, in computational architectures. After a quick review of that work, this paper presents a novel objection to Frank Jackson’s Knowledge Argument (KA) against physicalism, an objection in which such meta-knowledge also plays a central role. It is first shown that the KA’s supposition of a person, Mary, who is physically omniscient, and yet who has not experienced seeing red, is logically inconsistent, due to the existence of epistemic blindspots for Mary. It is then shown that even if one makes the KA consistent by supposing a more limited physical omniscience for Mary, this revised argument is invalid. This demonstration is achieved via the construction of a physical fact (a recursive conditional epistemic blindspot) that Mary cannot know before she experiences seeing red for the first time, but which she can know afterward. After considering and refuting some counter-arguments, the paper closes with a discussion of the implications of this argument for machine consciousness, and vice versa.

History

Publication status

  • Published

File Version

  • Accepted version

Journal

Journal of Artificial Intelligence and Consciousness

ISSN

1793-8430

Publisher

World Scientific Publishing

Issue

2

Volume

7

Page range

199-215

Department affiliated with

  • Informatics Publications

Research groups affiliated with

  • Centre for Cognitive Science Publications
  • Sackler Centre for Consciousness Science Publications

Full text available

  • Yes

Peer reviewed?

  • Yes

Legacy Posted Date

2021-03-15

First Open Access (FOA) Date

2021-08-06

First Compliant Deposit (FCD) Date

2021-03-15

Usage metrics

    University of Sussex (Publications)

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC