{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

PH100 Lecture Notes

But if computersmachines can convince us that they

Info iconThis preview shows pages 3–5. Sign up to view the full content.

View Full Document Right Arrow Icon
BUT, if computers/machines can convince us that they are conscious, then we have no reason to think that they are not conscious. (5) Arguments from Various Disabilities Machines/computers cannot think for they cannot perform some of the things that humans can. 1. Machines cannot make mistakes. Why is this a problem? Couldn't a machine be programmed to make mistakes? > response: “programmed mistakes” are not actual mistakes. Turing says there are two types of mistakes that machines can make: a. Errors of functioning – mechanical failure b. Errors of conclusion – gives the wrong answer 2. Computers cannot be the subjects of their own thoughts 3. Machines do not learn from experience Turing: If learning = changing one's behavior according to stimuli, then machines can learn (6) Lovelace's objection Machines can't do anything novel Turing: Machines can do something that will surprise us. > Is there any originality to be found in human beings? art – our creative self variation – how we live our own lives evolution free will September 11, 2012 1. Can computers think? 2. Are minds computers? - MIND IS THE SOFTWARE & BRAIN IS THE HARDWARE (computational interpretation) > mind = thinks, feels, has desires Haugeland Three dilemmas: (1) How can psychology of thinking be scientific? - If minds are computers, it has to follow certain rules; to study the mind, we only need to study how the software is being run – what are the rules that it's following? (2) How do you explain rule-following behavior? - Correct application of a rule implies that that rule has to have meaning . (3) How does the mind interact with the body (brain)?
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
What are computers? > Haugeland says: automated formal systems Defining a formal system 1) define the tokens 2) define starting position 3) define what the legal moves are Formal systems are medium-independent If minds are formal systems (which are medium-independent), then minds can be made out of anything 3 fixtures of a formal system 1) Self-contained - Nothing from the outside world matters 2) Definite - We know precisely what the legal moves are – no “maybe” 3) Finitely checkable - We can check whether a move is legal Semantics vs. syntax syntax → specifies rules/symbols - “grammar” - properties of formal systems semantics → relation to the world - “meaning” - interpretation of formal systems Formal systems “live a double life” - Haugeland (pp. 205) If you have syntax, you get semantics → if you take care of the rules and the grammar, you take care of the meaning (If by semantics you mean preservation of truth) start with something that is true, then you get a conclusion that is also true starting position MUST have truth Haugeland says: syntax → truth, but will not get you everything that means “making sense” ex) What is the capital of Illinois? Chicago. (Meaningful, but false) September 13, 2012: Computers, minds, and the Chinese Room argument REVIEW OF SEPTEMBER 11 “Semantic Engines” J. Haugeland Important distinction!
Background image of page 4
Image of page 5
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}