more on this theme
|
more from this text
Single Idea 2992
[filed under theme 18. Thought / A. Modes of Thought / 1. Thought
]
Full Idea
We are on the verge of solving a great mystery about the mind: how is rationality mechanically possible?
Gist of Idea
We may be able to explain rationality mechanically
Source
Jerry A. Fodor (Psychosemantics [1987], p. 20)
Book Ref
Fodor,Jerry A.: 'Psychosemantics' [MIT 1993], p.20
A Reaction
Optimistic, given that AI has struggled to implement natural languages, mainly because common sense knowledge seems too complex to encode. Can a machine determine logical forms of sentences?
The
170 ideas
from Jerry A. Fodor
6650
|
Fodor is now less keen on the innateness of concepts
[Fodor, by Lowe]
|
12615
|
Mental representations are the old 'Ideas', but without images
[Fodor]
|
12614
|
I prefer psychological atomism - that concepts are independent of epistemic capacities
[Fodor]
|
12613
|
Empiricists use dispositions reductively, as 'possibility of sensation' or 'possibility of experimental result'
[Fodor]
|
12617
|
Associationism can't explain how truth is preserved
[Fodor]
|
12616
|
English has no semantic theory, just associations between sentences and thoughts
[Fodor]
|
12618
|
It is essential to the concept CAT that it be satisfied by cats
[Fodor]
|
12619
|
We have no successful definitions, because they all use indefinable words
[Fodor]
|
12620
|
If 'exist' is ambiguous in 'chairs and numbers exist', that mirrors the difference between chairs and numbers
[Fodor]
|
12621
|
Definable concepts have constituents, which are necessary, individuate them, and demonstrate possession
[Fodor]
|
12623
|
The theory theory can't actually tell us what concepts are
[Fodor]
|
12622
|
Many concepts lack prototypes, and complex prototypes aren't built from simple ones
[Fodor]
|
2465
|
Maybe explaining the mechanics of perception will explain the concepts involved
[Fodor]
|
2469
|
The world is full of messy small things producing stable large-scale properties (e.g. mountains)
[Fodor]
|
2467
|
Functionalists see pains as properties involving relations and causation
[Fodor]
|
2468
|
Type physicalism is a stronger claim than token physicalism
[Fodor]
|
2471
|
Are concepts best seen as capacities?
[Fodor]
|
2472
|
For Pragmatists having a concept means being able to do something
[Fodor]
|
2473
|
Analysis is impossible without the analytic/synthetic distinction
[Fodor]
|
2470
|
Transcendental arguments move from knowing Q to knowing P because it depends on Q
[Fodor]
|
2474
|
It seems likely that analysis of concepts is impossible, but justification can survive without it
[Fodor]
|
2475
|
Don't define something by a good instance of it; a good example is a special case of the ordinary example
[Fodor]
|
2476
|
The goal of thought is to understand the world, not instantly sort it into conceptual categories
[Fodor]
|
2477
|
If to understand "fish" you must know facts about them, where does that end?
[Fodor]
|
2484
|
The theory of the content of thought as 'Mentalese' explains why the Private Language Argument doesn't work
[Fodor]
|
2482
|
It seems unlikely that meaning can be reduced to communicative intentions, or any mental states
[Fodor]
|
2483
|
Mentalese doesn't require a theory of meaning
[Fodor]
|
2486
|
Content can't be causal role, because causal role is decided by content
[Fodor]
|
2481
|
Despite all the efforts of philosophers, nothing can ever be reduced to anything
[Fodor]
|
2485
|
Do intentional states explain our behaviour?
[Fodor]
|
2487
|
Mentalese may also incorporate some natural language
[Fodor]
|
2480
|
Language is ambiguous, but thought isn't
[Fodor]
|
2489
|
Why bother with neurons? You don't explain bird flight by examining feathers
[Fodor]
|
2490
|
Modern connectionism is just Hume's theory of the 'association' of 'ideas'
[Fodor]
|
2491
|
Modules have encapsulation, inaccessibility, private concepts, innateness
[Fodor]
|
2492
|
Experience can't explain itself; the concepts needed must originate outside experience
[Fodor]
|
2493
|
According to empiricists abstraction is the fundamental mental process
[Fodor]
|
2494
|
Rationalists say there is more to a concept than the experience that prompts it
[Fodor]
|
2496
|
Blindness doesn't destroy spatial concepts
[Fodor]
|
2498
|
Modules make the world manageable
[Fodor]
|
2497
|
Something must take an overview of the modules
[Fodor]
|
2495
|
Obvious modules are language and commonsense explanation
[Fodor]
|
2499
|
Modules analyse stimuli, they don't tell you what to do
[Fodor]
|
2500
|
Babies talk in consistent patterns
[Fodor]
|
2502
|
How do you count beliefs?
[Fodor]
|
2501
|
Berkeley seems to have mistakenly thought that chairs are the same as after-images
[Fodor]
|
2504
|
Rationalism can be based on an evolved computational brain with innate structure
[Fodor]
|
2503
|
Empirical approaches see mind connections as mirrors/maps of reality
[Fodor]
|
2508
|
The function of a mind is obvious
[Fodor]
|
2507
|
Rationality rises above modules
[Fodor]
|
2505
|
Turing invented the idea of mechanical rationality (just based on syntax)
[Fodor]
|
2509
|
Modules have in-built specialist information
[Fodor]
|
2506
|
If I have a set of mental modules, someone had better be in charge of them!
[Fodor]
|
2432
|
Is content basically information, fixed externally?
[Fodor]
|
2434
|
Broad semantics holds that the basic semantic properties are truth and denotation
[Fodor]
|
2433
|
For holists no two thoughts are ever quite the same, which destroys faith in meaning
[Fodor]
|
2436
|
It is claimed that reference doesn't fix sense (Jocasta), and sense doesn't fix reference (Twin Earth)
[Fodor]
|
2435
|
Psychology has to include the idea that mental processes are typically truth-preserving
[Fodor]
|
2437
|
XYZ (Twin Earth 'water') is an impossibility
[Fodor]
|
2438
|
In the information view, concepts are potentials for making distinctions
[Fodor]
|
2439
|
Semantic externalism says the concept 'elm' needs no further beliefs or inferences
[Fodor]
|
2440
|
Propositional attitudes are propositions presented in a certain way
[Fodor]
|
2441
|
Truth conditions require a broad concept of content
[Fodor]
|
2442
|
Inferences are surely part of the causal structure of the world
[Fodor]
|
2463
|
A standard naturalist view is realist, externalist, and computationalist, and believes in rationality
[Fodor]
|
2458
|
Theories are links in the causal chain between the environment and our beliefs
[Fodor]
|
2446
|
Cartesians consider interaction to be a miracle
[Fodor]
|
2459
|
Externalist semantics are necessary to connect the contents of beliefs with how the world is
[Fodor]
|
2451
|
To know the content of a thought is to know what would make it true
[Fodor]
|
2461
|
An experiment is a deliberate version of what informal thinking does all the time
[Fodor]
|
2443
|
I say psychology is intentional, semantics is informational, and thinking is computation
[Fodor]
|
2453
|
We are probably the only creatures that can think about our own thoughts
[Fodor]
|
2457
|
If meaning is information, that establishes the causal link between the state of the world and our beliefs
[Fodor]
|
2462
|
Control of belief is possible if you know truth conditions and what causes beliefs
[Fodor]
|
2454
|
We can deliberately cause ourselves to have true thoughts - hence the value of experiments
[Fodor]
|
2455
|
Interrogation and experiment submit us to having beliefs caused
[Fodor]
|
2460
|
Participation in an experiment requires agreement about what the outcome will mean
[Fodor]
|
2450
|
Rationality has mental properties - autonomy, productivity, experiment
[Fodor]
|
2445
|
Semantics v syntax is the interaction problem all over again
[Fodor]
|
2447
|
Hume has no theory of the co-ordination of the mind
[Fodor]
|
2452
|
Knowing the cause of a thought is almost knowing its content
[Fodor]
|
2464
|
Type physicalism equates mental kinds with physical kinds
[Fodor]
|
3114
|
Concepts aren't linked to stuff; they are what is caused by stuff
[Fodor]
|
11143
|
If concept-learning is hypothesis-testing, that needs innate concepts to get started
[Fodor, by Margolis/Laurence]
|
8090
|
Since the language of thought is the same for all, it must be something like logical form
[Fodor, by Devlin]
|
12624
|
Only the labels of nodes have semantic content in connectionism, and they play no role
[Fodor]
|
12630
|
If concept content is reference, then my Twin and I are referring to the same stuff
[Fodor]
|
12628
|
Knowing that must come before knowing how
[Fodor]
|
12625
|
Pragmatism is the worst idea ever
[Fodor]
|
12629
|
For the referential view of thought, the content of a concept is just its reference
[Fodor]
|
12627
|
Before you can plan action, you must decide on the truth of your estimate of success
[Fodor]
|
12626
|
Cartesians put concept individuation before concept possession
[Fodor]
|
12632
|
In the Representational view, concepts play the key linking role
[Fodor]
|
12633
|
Definitions often give necessary but not sufficient conditions for an extension
[Fodor]
|
12634
|
'Inferential-role semantics' says meaning is determined by role in inference
[Fodor]
|
12635
|
Having a concept isn't a pragmatic matter, but being able to think about the concept
[Fodor]
|
12636
|
Mental states have causal powers
[Fodor]
|
12651
|
Some beliefs are only inferred when needed, like 'Shakespeare had not telephone'
[Fodor]
|
12649
|
We think in file names
[Fodor]
|
12648
|
Names in thought afford a primitive way to bring John before the mind
[Fodor]
|
12650
|
'Paderewski' has two names in mentalese, for his pianist file and his politician file
[Fodor]
|
12652
|
Concepts have two sides; they are files that face thought, and also face subject-matter
[Fodor]
|
12647
|
Mental representations name things in the world, but also files in our memory
[Fodor]
|
12637
|
Frege's puzzles suggest to many that concepts have sense as well as reference
[Fodor]
|
12638
|
If concepts have sense, we can't see the connection to their causal powers
[Fodor]
|
12639
|
Belief in 'senses' may explain intentionality, but not mental processes
[Fodor]
|
12642
|
Co-referring terms differ if they have different causal powers
[Fodor]
|
12640
|
Associative thinking avoids syntax, but can't preserve sense, reference or truth
[Fodor]
|
12641
|
Connectionism gives no account of how constituents make complex concepts
[Fodor]
|
12643
|
Ambiguities in English are the classic reason for claiming that we don't think in English
[Fodor]
|
12644
|
Who cares what 'philosophy' is? Most pre-1950 thought doesn't now count as philosophy
[Fodor]
|
12645
|
Semantics (esp. referential semantics) allows inferences from utterances to the world
[Fodor]
|
12646
|
Semantics relates to the world, so it is never just psychological
[Fodor]
|
12653
|
There's statistical, logical, nomological, conceptual and metaphysical possibility
[Fodor]
|
12654
|
You can't think 'brown dog' without thinking 'brown' and 'dog'
[Fodor]
|
12655
|
Frame Problem: how to eliminate most beliefs as irrelevant, without searching them?
[Fodor]
|
12656
|
P-and-Q gets its truth from the truth of P and truth of Q, but consistency isn't like that
[Fodor]
|
12657
|
Abstractionism claims that instances provide criteria for what is shared
[Fodor]
|
12662
|
We have an innate capacity to form a concept, once we have grasped the stereotype
[Fodor]
|
12658
|
Nobody knows how concepts are acquired
[Fodor]
|
12659
|
Maybe stereotypes are a stage in concept acquisition (rather than a by-product)
[Fodor]
|
12660
|
One stereotype might be a paradigm for two difference concepts
[Fodor]
|
12661
|
The different types of resemblance don't resemble one another
[Fodor]
|
12664
|
A truth-table, not inferential role, defines 'and'
[Fodor]
|
12663
|
We refer to individuals and to properties, and we use singular terms and predicates
[Fodor]
|
12631
|
Compositionality requires that concepts be atomic
[Fodor]
|
2597
|
Contrary to the 'anomalous monist' view, there may well be intentional causal laws
[Fodor]
|
2598
|
Lots of physical properties are multiply realisable, so why shouldn't beliefs be?
[Fodor]
|
2599
|
Either intentionality causes things, or epiphenomenalism is true
[Fodor]
|
22186
|
Mental modules are specialised, automatic, and isolated
[Fodor, by Okasha]
|
3975
|
Folk psychology explains behaviour by reference to intentional states like belief and desire
[Fodor]
|
3976
|
Intentional science needs objects with semantic and causal properties, and which obey laws
[Fodor]
|
3977
|
Laws are true generalisations which support counterfactuals and are confirmed by instances
[Fodor]
|
3980
|
Intentional states and processes may be causal relations among mental symbols
[Fodor]
|
3978
|
Associations are held to connect Ideas together in the way the world is connected together
[Fodor]
|
3981
|
Most psychological properties seem to be multiply realisable
[Fodor]
|
3982
|
How could the extrinsic properties of thoughts supervene on their intrinsic properties?
[Fodor]
|
2604
|
We must have expressive power BEFORE we learn language
[Fodor]
|
5498
|
Mind is a set of hierarchical 'homunculi', which are made up in turn from subcomponents
[Fodor, by Lycan]
|
7326
|
Intentionality doesn't go deep enough to appear on the physicists' ultimate list of things
[Fodor]
|
7014
|
A particle and a coin heads-or-tails pick out to perfectly well-defined predicates and properties
[Fodor]
|
15494
|
We can't use propositions to explain intentional attitudes, because they would need explaining
[Fodor]
|
2990
|
Contrary to commonsense, most of what is in the mind seems to be unlearned
[Fodor]
|
2991
|
Hume's associationism offers no explanation at all of rational thought
[Fodor]
|
2992
|
We may be able to explain rationality mechanically
[Fodor]
|
2993
|
Any piece of software can always be hard-wired
[Fodor]
|
2994
|
In CRTT thought may be represented, content must be
[Fodor]
|
2995
|
Supervenience gives good support for mental causation
[Fodor]
|
2996
|
Mental states may have the same content but different extensions
[Fodor]
|
2998
|
Grice thinks meaning is inherited from the propositional attitudes which sentences express
[Fodor]
|
2999
|
Obsession with narrow content leads to various sorts of hopeless anti-realism
[Fodor]
|
3000
|
Meaning holism is a crazy doctrine
[Fodor]
|
3001
|
Behaviourism has no theory of mental causation
[Fodor]
|
3002
|
If mind is just physical, how can it follow the rules required for intelligent thought?
[Fodor]
|
3003
|
Very different mental states can share their contents, so content doesn't seem to be constructed from functional role
[Fodor]
|
3004
|
The meaning of a sentence derives from its use in expressing an attitude
[Fodor]
|
3005
|
'Jocasta' needs to be distinguished from 'Oedipus's mother' because they are connected by different properties
[Fodor]
|
3006
|
Whatever in the mind delivers falsehood is parasitic on what delivers truth
[Fodor]
|
3007
|
Many different verification procedures can reach 'star', but it only has one semantic value
[Fodor]
|
3008
|
Evolution suggests that innate knowledge of human psychology would be beneficial
[Fodor]
|
3009
|
Sticklebacks have an innate idea that red things are rivals
[Fodor]
|
3010
|
Belief and desire are structured states, which need mentalese
[Fodor]
|
3011
|
Causal powers must be a crucial feature of mental states
[Fodor]
|
3012
|
Do identical thoughts have identical causal roles?
[Fodor]
|
2988
|
Folk psychology is the only explanation of behaviour we have
[Fodor]
|
2983
|
Maybe narrow content is physical, broad content less so
[Lyons on Fodor]
|
3135
|
Is thought a syntactic computation using representations?
[Fodor, by Rey]
|
15473
|
How does anything get outside itself?
[Fodor, by Martin,CB]
|
2981
|
Is intentionality outwardly folk psychology, inwardly mentalese?
[Lyons on Fodor]
|
2985
|
Are beliefs brains states, but picked out at a "higher level"?
[Lyons on Fodor]
|