Philosophy of Computer Science: Online Resources

Further Readings for Chapter 18:

Philosophy of Artificial Intelligence

Note 1: Many of these items are online; links are given where they are known. Other items may also be online; an internet search should help you find them.

Note 2: In general, works are listed in chronological order. (This makes it easier to follow the historical development of ideas.)

§18.1: Introduction

Further Reading on the Philosophy of AI:

• McCarthy, J. and Hayes, P.J. (1969). Some philosophical problems from the standpoint of artificial intelligence. In Meltzer, B. and Michie, D., editors, Machine Intelligence 4. Edinburgh University Press, Edinburgh

• Sloman, A. (1971). Interactions between philosophy and artificial intelligence: The role of intuition and non-logical reasoning in intelligence. Artificial Intelligence, 2:209–225

• Boden, M.A. (1977). Artificial Intelligence and Natural Man. Basic Books, New York.

• Sloman, A. (1978). The Computer Revolution in Philosophy: Philosophy, Science and Models of Mind. Humanities Press, Atlantic Highlands, NJ.

• Rapaport, W.J. (1986b). Philosophy of Artificial Intelligence: A course outline. Teaching Philosophy, 9(2):103–120

• Boden, M.A., editor (1990). The Philosophy of Artificial Intelligence. Oxford University Press, Oxford.

• Moody, T.C. (1993). Philosophy and Artificial Intelligence. Prentice-Hall, Englewood Cliffs, NJ.

• Akman, V., editor (2000). Philosophical Foundations of Artificial Intelligence. Special issue of Journal of Experimental & Theoretical Artificial Intelligence 12(3)(July).
  • See also:
    Akman, V. (2000), Introduction to the special issue on philosophical foundations of artificial intelligence, Journal of Experimental & Theoretical Artificial Intelligence 12(3):247–250.

• Bringsjord, Selmer and Naveen Sundar Govindarajulu, Artificial Intelligence, The Stanford Encyclopedia of Philosophy (Summer 2020 Edition), Edward N. Zalta (ed.)

§18.2.2: Artificial Intelligence as Computational Cognition:

• Paton, E. and Friedman, V. (2018, 29 May). 'Diamonds are forever,' and made by machine. New York Times.
  • Observes that the distinction between synthetic and natural diamonds raises "an almost metaphysical question of what defines a diamond. Is it its chemical structure …? Or is it its provenance: created deep in the ground … rather than cooked up in a machine?".

• On AI and IQ, see:
  • My "Artificial I.Q. Test" from Rapaport 1986b,

  • Ohlsson, S., Sloan, R.H., Turán, G., and Urasky, A. (2015). Measuring an Artificial Intelligence system's performance on a verbal IQ test for young children.

  • In addition to an IQ test for an AI, a recent experiment testing ChatGPT on an MBA Operations Management exam had some striking results:
    • Terwiesch, Christian (2023), "Would Chat GPT Get a Wharton MBA? A Prediction Based on Its Performance in the Operations Management Course", Mack Institute for Innovation Management at the Wharton School, University of Pennsylvania.

• On my view that the main question of AI is "How much of cognition is computable?", compare what The London Times said in an article about Turing, titled "Calculus to Sonnet" (11 June 1949, p. 4, my italics):
  "Their research would be directed to finding the degree of intellectual activity of which a machine was capable, and to what extent it could think for itself."

  (As cited in Gonçalves, Bernardo (2022), "Can Machines Think? The Controversy that Led to the Turing Test", AI & Society, §7.)

• The nature of intelligence is beyond our scope, but there are useful general discussions in:
  • Gardner, H. (1983). Frames of Mind: The Theory of Multiple Intelligences. Basic Books (2011), New York.

  • Sternberg, R.J. (1985). Beyond IQ: A Triarchic Theory of Human Intelligence. Cambridge University Press, Cambridge, UK.

  and there are AI-related discussions in:

  • Wang, P. (2019). On defining artificial intelligence. Journal of Artificial General Intelligence, 10(2):1–37.

  • Smith, B.C. (2019). The Promise of Artificial Intelligence: Reckoning and Judgment. MIT Press, Cambridge, MA.

  • Landgrebe, Jobst; & Smith, Barry (2023), Why Machines Will Never Rule the World: Artificial Intelligence without Fear (New York: Routledge), esp. Chs. 3 and 4

§18.3: The Turing Test:

• Encyclopedia articles on the Turing Test include:
  • Rapaport, W.J. (2006b). The Turing test. In Brown, K., editor, Encyclopedia of Language and Linguistics, 2nd Edition, Vol. 13, pp. 151–159. Elsevier, Oxford.

  • Oppy, Graham and David Dowe, The Turing Test, The Stanford Encyclopedia of Philosophy (Winter 2021 Edition), Edward N. Zalta (ed.)

• Several anthologies of essays on the Turing Test have appeared:
  • Akman, V. and Blackburn, P., editors (2000). Alan Turing and Artificial Intelligence. Special issue of Journal of Logic, Language and Information 9(4) (October).

  • Moor, J.H., editor (2000–2001), The Turing Test: Past, Present and Future, two special issues of Minds and Machines:
    • Vol. 10, No. 4 (November 2000):461–595
    • Vol. 11, No. 1 (February 2001):3–93

  • Moor, J.H., editor (2003). The Turing Test: The Elusive Standard of Artificial Intelligence. Kluwer Academic, Dordrecht, The Netherlands.

  • Shieber, S.M., editor (2004). The Turing Test: Verbal Behavior as the Hallmark of Intelligence. MIT Press, Cambridge, MA.
    • See also:
      Rapaport, W.J. (2005c). Review of Shieber 2004. Computational Linguistics, 31(3):407–412

  • Marcus, G., Rossi, F., and Veloso, M., editors (2016). Beyond the Turing Test. Special issue of AI Magazine 37(1)(Spring).

• Wilkes, M. (1953). Can machines think? Discovery, 14:151. Reprinted in Proceedings of the Institute of Radio Engineers 41(1953)(10) (October):1230–1234.
  • Contains speculations by one of the pioneers of computers on the Turing Test, learning machines, and the role of external input.

• The differences between various versions of the Turing Test are discussed in:
  • French, R.M. (2000). The Turing test: The first fifty years. Trends in Cognitive Sciences, 4(3):115–121.

  • Piccinini, G. (2000). Turing's rules for the imitation game. Minds and Machines, 10(4):573–582.

  • Rapaport 2006b

• Argamon, S., Koppel, M., Fine, J., and Shimoni, A.R. (2003). Gender, genre, and writing style in formal written texts. Text & Talk, 23(3):321–346.
  • Provides evidence that women can be distinguished from men on the basis of their writing style.

• Piccinini 2003 is primarily about Turing's views on AI, but also discusses his theory of computation and the role of "oracle" machines.

• Aaronson 2006 discusses the Turing Test in the context of quantum hypercomputation.

• Shieber, S.M. (2007). The Turing test as interactive proof. Noûs, 41(4):686–713.
  • Proposes an interpretation of the Turing Test as an interactive proof.

• McDermott, D. (2014, 31 December). What was Alan Turing's imitation game?.
  • A critique of Turing 1950 written by a well-known AI researcher as background for the film The Imitation Game.

• There have been several programs that are thought by some people to have passed a Turing Test:
  • "Eliza" (Weizenbaum 1966, Weizenbaum 1967, Weizenbaum 1976) was a natural-language processing program designed to show how the test could apparently be passed with no real understanding of natural language.

  • "Parry" was an Eliza-based program designed to simulate paranoia:

    • Colby, K.M., Weber, S., and Hilf, F.D. (1971). Artificial paranoia. Artificial Intelligence, 2:1–25.

    • Colby, K.M., Hilf, F.D., Weber, S., and Kraemer, H.C. (1972). Turing-like indistinguishability tests for the validation of a computer simulation of paranoid processes. Artificial Intelligence, 3:199–221.

    • Colby, K.M. (1981). Modeling a paranoid mind. Behavioral and Brain Sciences, 4:515–560.

  • The Loebner Prize competitions — originally intended to be a real Turing Test — have devolved into competitions for Eliza-like "chatterbots":
    • Shieber, S.M. (1994a). Lessons from a restricted Turing test. Communications of the ACM, 37(6):70–78.

    • Loebner, H.G. (1994). In response [to Shieber 1994a]. Communications of the ACM, 37(6):79–82.

    • Shieber, S. M. (1994b). On Loebner’s lessons. Communications of the ACM, 37(6):83–84.

    • Rees, R., Crawford, C., Bowyer, A., Schubert, L., and Loebner, H. (1994). Forum: Notes on the Turing test. Communications of the ACM, 37(9):13–15.

  • No current natural-language-understanding computer has achieved the "understanding" exhibited by the fictional computer HAL in the movie 2001, not even Apple's Siri or Amazon's Alexa.

    On HAL, see:

    • Stork, D. G. (1997). HAL's Legacy: 2001's Computer as Dream and Reality. MIT Press, Cambridge, MA.),

  • The question of whether some robots and softbots may have passed a kind of Turing Test is raised in:
    • Kahn, A. (2014, 3 November). Rover disguised as penguin chick does research better than scientists. Los Angeles Times Science Now.

    • Hill, D.J. (2016, 11 May). AI teaching assistant helped students online—and no one knew the difference. SingularityHUB.

  • Walsh, T. (2016). Turing's red flag law. Communications of the ACM, 59(7):34–37.
    • Proposes a "Turing Red Flag law: An autonomous system should be designed so that it is unlikely to be mistaken for anything besides an autonomous sysem, and should identify itself at the start of any interaction with another agent."

• Do programs like ChatGPT pass a Turing Test? This question is explored in:
  • Metz, Cade (2023, 20 January), "How Smart Are the Robots Getting?", New York Times

  • See also Terwiesch 2023 and draw your own conclusions!

• Berkeley, I.S. (2020). AI: A crowd-sourced criterion. A commentary on Pei Wang's paper "On defining Artificial Intelligence". Journal of Artificial General Intelligence, 11(2):24–26.
  • Compares Turing's "use of words and general educated opinion" to Dennett's intentional stance.
    See also:
    • Berkeley, I S. and Rice, C. (2012). Machine mentality? In Müller, V.C., editor, Philosophy and Theory of Artificial Intelligence, pages 1–15. Springer, Heidelberg

  • Similar views are discussed in:
    • Proudfoot, D. (2013). Rethinking Turing's test. Journal of Philosophy, 110(7):391–411.

    • Danziger, S. (2018). Where intelligence lies: Externalist and sociolinguistic perspectives on the Turing test and AI. In Müller, V., editor, Philosophy and Theory of Artificial Intelligence, pages 158–174. Springer, Cham, Switzerland.

• I say that Turing's Imitation Game can really be played and is "not a mere thought experiment" (emphasis added). I still believe that, but the following paper argues that it is (also) a thought experiment:
  • Gonçalves, Bernardo (2023), "The Turing Test is a Thought Experiment", Minds and Machines 33: 1–31.

• For more on philosophical zombies, see:
  • Kirk, Robert (2023), "Zombies", The Stanford Encyclopedia of Philosophy, Edward N. Zalta & Uri Nodelman (eds.).

• For more on metaphors and metaphorical extension, see:
  
  • Simpson, John (2013), "The Oxford English Dictionary and Its Philosophy", Philosophy and Technology 26:341–348, esp. §3 ("The Historical Principle in the Entry for Culture"), pp.342–343.
    

  • Xu, Yang; Malt, Barbara C.; & Srinivasan, Mahesh (2017), "Evolution of Word Meanings through Metaphorical Mapping: Systematicity over the Past Millennium", Cognitive Psychology 96:41–53

  • Ramiro et al. 2018

  
  

• When my son was a little over 3½ years old, he told me that his grandfather "is going to fly" to visit us. To him, this metaphoric use of the word 'fly' was indistinguishable from its non-metaphoric use in "the bird is going to fly".
  

• For an interesting take on the importance of "the human ability to learn" by Kurt Gödel, see the following essay on Gödel's belief in the afterlife(!):
  Englert, Alexander T. (2024, 2 January), "We'll Meet Again", Aeon

• Too much has been written on the CRA to cite here, but you might start with these:
  • Cole 1984

  • Rapaport, W.J. (1986c). Philosophy, Artificial Intelligence, and the Chinese-room argument. Abacus: The Magazine for the Computer Professional, 3:6–17. (Correspondence, Abacus 4(Winter 1987):6–7; and 4(Spring)5–7).
    • An overview written for a semi-popular computer magazine.

  • Rey, G. (1986). What's really going on in Searle's 'Chinese room'. Philosophical Studies, 50(2):169–185.
    • An excellent overview.

  • Rapaport, W.J. (1988b). To think or not to think: Review of Searle's Minds, Brains and Science. Noûs, 22(4):585–609.

  • Hauser, L. (2001). Chinese room argument. Internet Encyclopedia of Philosophy.

  • Preston, J. and Bishop, M., editors (2002). Views into the Chinese Room: New Essays on Searle and Artificial Intelligence. Clarendon Press, Oxford.
    • See also:
      Rapaport, W.J. (2006c). Review of Preston and Bishop 2002. Australasian Journal of Philosophy, 84(1):129–133.

  • An online series of articles providing background for the movie The Imitation Game contained two critiques of the CRA:
    • Cole, D. (2014). Alan Turing & the Chinese Room Argument

    • Horst, S. (2014). The computational theory of mind: Alan Turing & the Cartesian challenge

  • Cole, David, The Chinese Room Argument, The Stanford Encyclopedia of Philosophy (Winter 2020 Edition), Edward N. Zalta (ed.)

• In my description of the CRA, I said (p. 419, line 4) that "the rule book of instructions is a complete natural-language-understanding algorithm for Chinese". (I also used the italicized word in that context in §18.7.2, p. 422, 2nd full paragraph, line 2.) But to call it that is a misnomer of the kind discussed in:
  • McDermott, Drew (1980), "Artificial Intelligence Meets Natural Stupidity", in John Haugeland (ed.), Mind Design (Cambridge, MA: MIT Press): 143–160.

  as well as an assumption that begs the question of whether the algorithm (or anything else affiliated with the Room) really does "understand".

  I should have said that the rule book is a complete natural-language-processing algorithm for Chinese.

• And for a description of "the" Chinese language (which shows how complex such a Chinese natural-language processing algorithm would have to be), see:
  • Wang, William S-Y. (1973), "The Chinese Language", Scientific American 228(2, February):50–60 (with a bibliography on p.124 that is not included in the online PDF)

• Weinberg, J. (2019, 15 May). Did a story about a computer made of humans scoop Searle's "Chinese room" by 20 years? Daily Nous.
  • Argues that A.P. Mickevich's 1961 science fiction story "The Game" anticipated the CRA.
  • However, that story is actually closer to the philosopher Ned Block's 1978 "Chinese nation" thought experiment:
    • Block, N. (1978). Troubles with functionalism. In Savage, C., editor, Minnesota Studies in the Philosophy of Science, Vol. 9, pages 261–325. University of Minnesota Press, Minneapolis

• Is ChatGPT a Chinese Room? For an example conversation that might shed light on this question, see Terwiesch 2023

§18.7: The Argument from Biology

• For an excellent analysis of the argument from biology (and comments relevant to a related argument in Landgrebe & Smith 2023), see:
  
  • Dennett, Daniel C. (1987), "Fast Thinking", in Daniel C. Dennett (ed.), The Intentional Stance (Cambridge, MA: MIT Press), Ch. 9, pp. 323–337.

• On the brain's interpretation of the external world being more important than the actual state of the external world, see Yong 2022

• On not having to investigate where the brain's inputs come from, see Chalmers 2022

• Question to Think About:
  I said that Searle assumes that computers have no links to the external world. What about interactive computing, as we discussed in §11.8?

• For more details on syntactic semantics, see the Further Readings for Chapter 16.

• Any study of semantics in linguistics — in the sense of a study of the meanings of linguistic expressions — that focuses only on relations either among the expressions or between the expressions and mental "conceptual structures" — and not on relations between expressions and aspects of the external world — is a syntactic enterprise. This is the nature of semantics in, for instance, cognitive linguistics:
  • Lakoff 1987

  • Langacker, R. (1999). Foundations of Cognitive Grammar. Stanford University Press, Stanford, CA.

  • Talmy, L. (2000). Toward a Cognitive Semantics. MIT Press, Cambridge, MA.

  • Jackendoff, R. and Audring, J. (2020). Morphology and memory: Toward an integrated theory. Topics in Cognitive Science, 12:170–196

• For more about Cassie, see:
  • "A 'Conversation' with Cassie"

  • Shapiro, Stuart C. and Rapaport, W.J. (1987). SNePS considered as a fully intensional propositional semantic network. In Cercone, N. and McCalla, G., editors, The Knowledge Frontier: Essays in the Representation of Knowledge, pages 262–315. Springer-Verlag, New York.

  • Rapaport 1988a

  • Shapiro, Stuart C. (1989). The Cassie projects: An approach to natural language competence. In Martins, J.P. and Morgado, E., editors, EPIA 89: 4th Portugese Conference on Artificial Intelligence Proceedings, pages 362–380. Springer-Verlag Lecture Notes in Artificial Intelligence 390, Berlin.

  • Shapiro, Stuart C. and Rapaport, W.J. (1995). An introduction to a computational reader of narratives. In Duchan, J.F., Bruder, G.A., and Hewitt, L.E., editors, Deixis in Narrative: A Cognitive Science Perspective, pages 79–105. Lawrence Erlbaum Associates, Hillsdale, NJ.

  • Shapiro, Stuart C. (1998). Embodied Cassie. In Cognitive Robotics: Papers from the 1998 AAAI Fall Symposium, pages 136–143. AAAI Press, Menlo Park, CA. Technical Report FS-98-02,

  • Rapaport 2006a

§18.8.4: Points of View

• For arguments why AI should take the first-person point of view, see:
  • Maida, A.S. and Shapiro, Stuart C. (1982). Intensional concepts in propositional semantic networks. Cognitive Science, 6:291–330, esp. p. 296. Reprinted in Ronald J. Brachman & Hector J. Levesque (eds.), Readings in Knowledge Representation (Los Altos, CA: Morgan Kaufmann, 1985): 169–189.

  • Rapaport, W.J. and Shapiro, Stuart C. (1984). Quasi-indexical reference in propositional semantic networks. In Proceedings of the 10th International Conference on Computational Linguistics (COLING-84, Stanford University), pages 65–70. Association for Computational Linguistics, Morristown, NJ.

  • Rapaport, W.J. (1986d). Logical foundations for belief representation. Cognitive Science, 10:371–422.

  • Wiebe, J.M. and Rapaport, W.J. (1986). Representing de re and de dicto belief reports in discourse and narrative. Proceedings of the IEEE, 74:1405–1413.

  • Shapiro and Rapaport 1987

  • Shapiro and Rapaport 1991

  • Chalmers, D.J. (2019). The meta-problem of consciousness. Journal of Consciousness Studies, 25(9–10):6–61.

• On a theory of consciousness that is consistent with the first-person point of view, see:
  • List, Christian (2023), "The Many-Worlds Theory of Consciousness", Noûs 57: 316–340.

• On antecedent understanding, see:
  • Rosenblueth and Wiener 1945, p. 317.

  • Rapaport, W.J. (1996). Understanding understanding: Semantics, computation, and cognition. Technical Report 96-26, SUNY Buffalo Department of Computer Science, Buffalo, p. 30.

  • It also plays an important role in Michael Dummett's theories about understanding the meaning of linguistic expressions:
    • Dummett, M.A. (1975). What is a theory of meaning? In Guttenplan, S., editor, Mind and Language, pages 97–138. Clarendon Press, Oxford.

• On syntactic understanding, see Rapaport 1986a

• On meaning holism, see:
  • de Saussure, F. (1959). Course in General Linguistics. Philosophical Library, New York. Charles Bally, Albert Sechehaye, & Albert Reidlinger (eds.).

  • Fodor, J.A. and Lepore, E. (1992). Holism: A Shopper's Guide. Basil Blackwell, Cambridge, MA.

  • Rapaport 2002

  • Jackman, Henry, Meaning Holism, The Stanford Encyclopedia of Philosophy (Winter 2020 Edition), Edward N. Zalta (ed.)

§18.9: Leibniz's Mill and Turing's "Strange Inversion":

• For a science-fiction treatment of Leibniz's mill, see Asimov 1987

• Question to think about:
  Dennett (in the quotation on p. 432) talks about an "Intentionally characterized problem". That sounds like the "G" of §16.5's formulation "To accomplish goal G, do algorithm A". So, perhaps "To G, do A" could be called an "Intentionally characterized solution". In Dennett's "reduc[tion] to functionaries 'who can be replaced by a machine'", what happens to "G"?

• Here's another quotation from Dennett about Leibniz's Mill:
  "A good theory of consciousness should make a conscious mind look lika an abandoned factory (recall Leibniz's mill), full of humming machinery and nobody home to supervise it, or enjoy it, or witness it. … I don't maintain, of course, that human consciousness doesn't exist; I maintain that it is not what people often think it is."
  — Dennett, Daniel C. (2005), Sweet Dreams: Philosophical Obstacles to a Science of Consciousness (Cambridge, MA: MIT Press), pp. 70–71.

• Here are two comic strips that illustrate the idea of homunculi:
  
  

• For more on what Searle-in-the-room might know, see:
  • Cole, D. (1991). Artificial intelligence and personal identity. Synthese, 88(3):399–417.

  • Rapaport, W.J. (1990). Computer processes and virtual persons: Comments on Cole's "Artificial intelligence and personal identity". Technical Report 90-13, SUNY Buffalo Department of Computer Science, Buffalo.

  • Rapaport 2000

  • Rapaport 2006a, pp. 390–397.

  • Suits, D.B. (2005). Out of the Chinese room. Computers & Philosophy Newsletter, (4:1+4:2):1–7.
    • A science-fiction version of the CRA.

§18.10: A Better Way:

On what might be in the Chinese Room rulebook, see:

• Levesque, H.J. (2009). Is it enough to get the behaviour right? In IJCAI'09: Proceedings of the 21st International Joint Conference on Artifical Intelligence, pages 1439–1444. Morgan Kaufmann, San Francisco.
  • Agrees that "in the end, it all depends on the [rule] book. … [W]e need to ask what a real book would have to be like … ".
  • See also: Levesque 2017.

• For another list of things that a (computational) cognitive agent must be able to do in order to understand natural language, see:
  • Landgrebe, J. and Smith, B. (2019, 2 December). There is no artificial general intelligence
    • But their conclusion is much more pessimistic than mine!