An effective investigate of semantic systems:

2656 days ago, 1072 views
PowerPoint PPT Presentation

Presentation Transcript

Slide 1

´╗┐An effective investigate of semantic systems: Jerry Fodor's atomism. Massimo Piattelli-Palmarini UofA, Cognitive Science

Slide 2

An absolute bottom thought Meaning is personally identified with truth conditions Word implications are ideas And ideas "apply to" (Things do, or don't, "fall under" them) An idea C is "valid for" something, and just of that something That's its referent (its augmentation) The referent is "reality producer" of C But this is insufficient a similar referent can be singled out by various intensions (the morning star/the night star) (Gottlob Frege On Sense and Reference 1892/1948)

Slide 3

Meanings are shared, they are open They are not the same as the "mental pictures" (or pictures, or affiliations) that a word may inspire in every individual speaker Dog implies canine, regardless of the possibility that I rationally engage the photo a poodle, while you picture a terrier. Marriage does not mean diverse things to a future lady of the hour, her dad, a minister, a legal advisor and so on. It's significant that our lexical semantics does not "cut" implications so thin But rather should slice them sufficiently thin to permit diverse intensions to choose a similar expansion.

Slide 4

Associations This applies likewise to mental "affiliations" Via a reasonable number of connections, everything is connected with everything else Cat is usually connected with mouse , drain and so on. These are not "constitutive" of the significance of feline

Slide 5

What is constitutive? The focal issue (Fodor's issue) here will be: Has the relationship amongst feline and creature An alternate semantic status than the relationship with, say, mouse, drain, yowls and so on.? Fodor's answer is: No!

Slide 6

Different qualities of the postulation Weakest : The accessible semantic systems don't speak to lexical implications Moderate : No such system could speak to lexical implications Strong : No sort of system (for example inferential systems) can speak to lexical implications Strongest : Lexical implications have no outside nor inside structure at all

Slide 7

The work of art (Aristotelian) hypothesis of ideas There are exclusively important and mutually adequate criteria (properties, traits, predicates) DOG = creature, warm blooded creature, residential, barks, and so forth. Nothing that "misses" one of these properties can "fall under" the idea DOG Nothing that has every one of these properties can neglect to fall under the idea DOG

Slide 8

Problems Many counterfactual cases (Chinese sovereigns had a race of non-woofing pooches) Putnam's Martian robot felines Some focal properties may not be open to reflection (the run of the mill winged creature is "amicable", while "crunchy" is personally connected with acoustic properties, and so on.)

Slide 9

An option: "family similarity" No property is shared by each individual from the class But numerous individuals from the classification share no less than one property, and as often as possible a hefty portion of them Properties are not disconnected, but rather come in groups p (flies/quills) >> p (flies/hide) p (nose/plumes) >> p (mouth/scales)

Slide 10

Problems Categories get to be "cloudy" Different people "signify" distinctive things by a similar word Unless Some properties are more vital (more indicative) than others But there is no principled (classification free) method for figuring out which ones You should "go atomistic"

Slide 11

Another option: Meanings are smaller scale auxiliary (the genuine criteria are essentialist) We need to utilize show qualities, yet we implicitly engage inside, vital characteristics which we can't see (DNA, sub-atomic organization, plan highlights, and so forth.) We concede to the specialists to let us know what these are

Slide 12

Problems We truly don't have the foggiest idea about the significance of the greater part of the words we utilize Most implications are incomplete, estimated, defeasible (notwithstanding for feline , canine , silver and so on.) Even the specialists concede to the consistent advance of science and grant. Numerous implications have no basic segment at all ( uncle , seat , costly ) What about verbs? ( cut , run , identify and so forth.)

Slide 13

Inferential Role Semantics (IRS) The importance of an idea is practically controlled By the surmisings it licenses, and those it prohibits, and by their quality If one is not arranged to consent to all and just the right derivations, one doesn't "have" that idea. This likewise applies to verbs, descriptors, intensifiers

Slide 14

Problems Intractable expansion of derivations Again, you need to delimit the essential ones But there is no principled (class free) basis for doing that So, perhaps atomism? That is Fodor's decision.

Slide 15

Fodor on ideas The cosmology of ideas (word implications) is personally tied with the issue of what it is to have ideas (and implications), and to obtain them. He is unequivocally against the possibility that the epistemic issue of ownership and recognizable proof is before, and ontologically overwhelming over, the issue of what ideas and implications are .

Slide 16

Contra Wittgenstein and family likeness Concepts are open ; they are the sorts of things that bunches of individuals can , and do, share Concept "comparability" (whatever that may mean) won't do. To be reasonable, it must clarify and safeguard the invariance of deliberate clarifications, yet it must not presuppose a "vigorous thought of substance character". No hypothesis of reasonable similitude has possessed the capacity to do both. Something, at base, must be actually shared (for example a conviction), regardless of the possibility that all you need is to adjust degrees of closeness.

Slide 17

Identity is the thing that we require If you have criteria for strict sharing (for the personality of ideas and convictions), then you have a hearty thought of what considers "open". You needn't bother with similitude. There are independently extraordinary degrees of power of a similar conviction. There are distinctive mental pictures or relationship for a similar idea (say, DOG) But idea (and substance) personality is dependably (implicitly) presupposed . In the event that you don't, than relativism is relentless. In this way, how about we go for personality.

Slide 18

What is (and what isn't) constitutive of importance There are magical associations between implications, since properties are frequently associated with different properties, presuppose different properties, involve different properties and so on. You need to have counter-real supporting criteria for new and conceivable utilizations of the idea. No such association, presupposition or entailment, be that as it may, is constitutive of the significance of an idea.

Slide 19

What is essential It's not dialect utilize, it's not limits or capacities that are essential. In the event that having ideas were having limits (to perceive, to sort, or to draw sound derivations), then ideas would not be mental particulars, they would not be things by any means, and along these lines a fortiori they would not be mental things. Be that as it may, they are!

Slide 20

Fodor contra Inferential Role Semantics First : Buying IRS would prompt to circularity : "I can't both recount a computational tale about what deduction is and recount an inferential anecdote about what substance is." ( Concepts , 1998. p.13) Second : "an inferential part semantics has all encompassing ramifications that are both unavoidable and insufferable" (on the same page.) (see additionally his book with LePore)

Slide 21

Fodor contra Inferential Role Semantics Third : atomism. " Satisfying the powerfully vital conditions for having one idea never requires fulfilling the supernaturally important conditions for having some other idea." (p.14) No deduction can be constitutive of the significance of an idea. It is supernaturally possible that a brain can exist that has just the idea (the significance) DOG, and that's it (no "feline", and, the greater part of all, no "creature" idea). Frogs and sticklebacks are flawless instantiations

Slide 22

Fodor contra Inferential Role Semantics "A significant part of the life of the mind comprises in applying ideas to things" (p.24). Ideas have their fulfillment conditions basically, however it doesn't take after that "the affirmation states of an idea are among its key properties". (p.25) Satisfaction conditions are powerful , while affirmation conditions are epistemic .

Slide 23

Fodor contra Inferential Role Semantics Confirmation may well be (and now and again is ) an all encompassing endeavor (activating pertinent pieces of information and inferential aptitudes from all that you know), however idea fulfillment is not all encompassing. Manners to draw derivations, to sort things skillfully, to make associations and so forth are the result of knowing the significance of ideas Not the a different way.

Slide 24

Fodor's atomism A word means the property that the idea it communicates is bolted to. Mental states and procedures are commonly types of relations to mental representations, of which last ideas are normally the parts. Considerations are mental representations practically equivalent to shut sentences, while ideas (their constituents) are mental representation similar to the comparing open ones.

Slide 25

Fodor contra iconicity "The thought that there are mental representations is the possibility that there are Ideas less Ideas are images" ( Concepts 1998, p.8, see additionally Hume Variations, 2003) Thought is a considerable measure like dialect , and ideas are a great deal like mental lexical passages (if primitive), or mental phrasal constituents (if made). This is splendidly OK. We need them two (thought and dialect) to be profitable and orderly.

Slide 26

Fodor's atomism DOG implies puppy PUT implies put KEEP implies keep and so forth (a Tarskian hypothesis) What you have on the privilege are mental particulars (atomistic mental representations of characteristics). No decay

Slide 27

No disintegration Kill implies KILL Not "bring about to-end up not-alive" Nor anythin