• schopenhauer1
    11k
    But isn't it entirely possible that the little voice is a sort of back-construction, the internalisation, as it were, of our external language?Banno

    After reading the article, I am uncertain about its implications. Although it is possible that some individuals have a lower frequency of inner speech, further research is needed to investigate whether non-linguistic thoughts are indirectly connected to a language system, even if they are not explicitly experienced by those rare individuals who do not have any inner speech.

    However, it seems that many researchers believe that Chomsky's ideas about inner speech (and language in general) are unfounded and lack evidence. Chomsky's theory suggests that language evolved through an accidental exaptation that led to increased mental efficiency, rather than for communication purposes.

    Interestingly, I found this paper: https://ijds.lemoyne.edu/journal/8_1/pdf/ijds.8.1.01.wiley.pdf

    Norbert Wiley discounts Chomsky's notion thus:
    I will show that these commitments create serious problems for Chomsky’s
    linguistics. Inner speech is quite irregular, much more so than interpersonal or outer
    speech. It is also difficult to say there is a “competence” or “langue” dimension for
    inner speech. The competence aspect is primarily rules, but inner speech, being private,
    has no audience to carry or enforce the rules. In fact its major rule is efficiency,
    whatever that might imply for any given individual
    — Norbert Wiley

    It even appears that Chomsky is directly challenging Wittgenstein's concept of a private language.
    Actually you can use language even if you are the only person in the universe
    with language, and in fact it would even have an adaptive advantage. If one
    person suddenly got the language faculty, that person would have great
    advantages; the person could think, could articulate to itself its thoughts,
    could plan, could sharpen, and develop thinking as we do in inner speech,
    which has a big effect on our lives. Inner speech is most of speech. Almost all
    the use of language is to oneself. (Chomsky, 2002, p. 148)
    — Chomsky quoted by Wiley

    Wiley attempts to refute Chomsky's perspective that language was meant for internal speech first, by presenting evidence that inner speech is often a simplified form of language consisting of basic concepts, words, and imagery. This seems to contradict Chomsky's emphasis on fully syntactical language and the generative grammar module.

    Here is one of his examples of abbreviated (non-syntactic) inner talk:
    This is a waitress reporting on her thoughts going to work. Her inner speech is
    presented linguistically along with brief sketches of her imagery.
    “Only eight minutes, takes five to change. I’ve got to
    book (hurry).” Imagery: A disgustingly filthy locker
    room. Visions of me running from table to kitchen
    table. Sounds. Forks and knives scraping plates,
    customers yelling over each other. “ I have to make
    money. At least it’s not as bad as last summer.” Memory
    imagery: A tiny dumpy diner. Visions of me sweating.
    Sensations of being hot. Visions of thirty marines eating
    and drinking. Sounds: country music on a blaring
    juke box . “I’ll be right there, just a minute
    please.” Sensations of burning my arm in a pizza oven.
    Visions of dropping glasses. Sounds: Glass breaking,
    manager yelling, marines cheering. “Oh God, get me out of
    here.” Sensation: Cringe, humiliation. “I hate
    waitressing. Can’t wait to graduate and get a decent
    job.“ Visions of a paneled, brightly carpeted office with
    scenic pictures and healthy plants. Visions of me fifteen
    pounds thinner in a new skirt suit from Lord and Taylor.
    WILEY
    4
    A great-looking coworker is pouring us coffee. Sounds of
    a clock chiming five o’clock. “Sure I’d love to go out
    Friday night” (Caughey, 1984, p. 135. Italics mine.)
    — Wiley Quoting Caughey

    So he is trying to make an argument that if the internal structure of generative grammar was mainly for self-talk and mental efficiency, then this should be seen in fully syntactical sentences in inner talk before it is used as communication.

    I am sure there are arguments that can be made against this, that you don't need fully syntactic usage for generative grammar to create the mental efficiency. I'd have to see if an argument like that has been made. Too many Ph.D students for there not to be. Certainly, computers have compression abilities.

    The embeddedness of language concepts can work as such. As Wiley notes from Vygotsky,
    For Vygotsky the syntax of inner speech is, in his words, “predicated”
    (Vygotsky, 1987, p. 267). By this he does not mean the predicate of a sentence in the
    usual sense. He means the thought which answers a question and supplies only the
    needed information. If the question concerns a time of departure, the predicate might be
    “eight o’clock.” That would be the whole sentence. If one said (to oneself) “the best
    time to leave would be eight o’clock” the first seven words would be unnecessary.
    — Wiley quoting Vygotsky

    He mentions Saussure:
    Saussure’s associative axis is helpful here (1959, pp. 122-127). He had two axes
    for a sentence. The one he called syntagmatic was merely the syntactical unfolding of a
    sentence, going from subject to predicate. But what he called the associative axis was
    the set of meanings that might be suggested by the actual words in a sentence, even
    though these words were not chosen and remained in the background. This axis was a
    collection of related meanings, i.e. both similar and contrastive, that hovered over a
    sentence’s core meanings. He thought only in terms of similar meanings, those that
    could be substituted for the meanings actually used. But I think contrasting or opposite
    terms also belong on this axis. “I’m tired and want to go to bed” could have an
    associative axis in which words like “weary, exhausted, beat and bushed” might
    surround the word “tired.” Also such contrasting words as “energetic, alive and fresh” might be present as opposites. This embedding gives the inner speech semantics a
    fluttery, epistemologically labile quality
    — Wiley Quoting Saussure

    This theory indicates that on top of syntax is concept formation and their embeddedness in a network. This Saussure idea of language possessing both a syntax and conceptual axis also could be the missing link I wasn't seeing when Chomsky was discussing a "conceptual" system that was learned and outside the scope of his generative grammar. That being said, that is just evidence against Chomsky’s theory of generative grammar origins in self talk but possibly points to the non-Chomskyan domain of concept formation evolving for self talk.
  • Dawnstorm
    249
    And there's this.Banno

    Now there's an interesting link. I've always been confused about why people think language and thought are tied together as much as they seem to think. I do have verbal thoughts, but only if I'm explicitly forumulating, and when I do, there's always some sort of non-verbal stream of thought in the background that checks whether what I'm saying internally is what I'm actually thinking, or if I need to start anew. For me, anything verbal that's interior is *clearly* at least partly derived from social language. I often have verbal blank-outs: I know intuitively what I want to say, but there aren't any words. I need to find a way to approximate this with language. It's different from the tip-of-the-tongue experience, where I know there's a word I seem to have misplaced. It's an intuition that I don't know how to formulate.

    Interestingly, the article also links not having an "interior monologue" to aphantasia, which I think I have, too (at least I'm more on a page with accounts from people with aphantasia than I am with people who puzzle over them).

    Indeed, i find it hard to understand what an analytic statement would be like in an I-language...

    I'm not an expert on Chomsky, so my immediate question would be whether Chomsky sees the judgement of a statement as analytic or synthetic as a task for the i-language. Chomsky's difficult, since he revised his theory a lot. From what I remember, if you think a line like "Snow is white," that line isn't what i-language is about; it's already internalised e-language maybe? I don't know. I'd have to read up.

    I'm fairly sure the i-language, though, is supposed to be some human universal, so that "Snow is White," (Enlish), "Schnee ist weiß," (German) and "Yuki wa shiroi" (Japanese, if I didn't mess up) are the same i-language sentence that generates a different surface structure for each language. A common deep structure that results in different surface structure via different generative rules. (At least that was generative grammar; I'm not sure how much of this still applies to his minimalist program.)

    Chomsky's core interest, if I'm not mistaken, was always how sentences were formed rather than what they mean or if they're ture. (I think. As I said, I'm no expert on Chomsky.)
  • schopenhauer1
    11k
    I know intuitively what I want to say, but there aren't any words. I need to find a way to approximate this with language. It's different from the tip-of-the-tongue experience, where I know there's a word I seem to have misplaced. It's an intuition that I don't know how to formulate.Dawnstorm

    Yet, this might point more to Chomsykean ideas of a "mentalese" (I-language), not that there is no language at all. It complicates things that Chomsky seems agnostic about concepts and very in favor of a generative grammar. As far as concepts, do animals that don't have language have concepts? What is a non-linguistic concept? A dog associating a leash with a walk, is that a concept? It's association sure. Concepts seem to be something beyond just association. Concepts seem to join with a mechanism whereby they are "used" and that can be something akin to a grammar. I can see I-languages being a mentalese that accounts for our shorthand internal language maybe.

    At least that was generative grammar; I'm not sure how much of this still applies to his minimalist programDawnstorm

    The minimalist program is just "merge" now I think.
  • Janus
    16.5k
    What is a non-linguistic concept? A dog associating a leash with a walk, is that a concept? It's association sure. Concepts seem to be something beyond just association.schopenhauer1

    Would you count seeing something as something as a kind of conceptualization? The ball stands out for the dog; it's a gestalt. This is cognition and re-cognition. Then the dog sees the ball as a to-be-chased; would you say this as a conceptualization?

    My own experience of inner dialogue is that much of it goes on in terms of images, not words. But then perhaps the earliest forms of communication of ideas were in images and gestures delineating forms, rather than in words. Problem is we cannot be there to observe.
  • Banno
    25.3k
    ...no two people's understanding of "grass" will be the same,RussellA
    This is to suppose that there is a thing, which is someone's understanding of grass; as if to understand "grass" were to have a certain box in one's mind; so that your box can be different to my box.

    That's the image that is to be rejected. Understanding grass is not having a thing in one's head, but being able to roll on it, pick it, plant it, mow it, set out it's parts in a botanical essay...

    You give an oddly subservient view of the way we might use words. A triangular view, inflexible, fixed. Boxes in one's mind.
  • Dawnstorm
    249
    Yet, this might point more to Chomsykean ideas of a "mentalese" (I-language), not that there is no language at all. It complicates things that Chomsky seems agnostic about concepts and very in favor of a generative grammar.schopenhauer1

    Well, Chomsky's mainly concerned with syntax, not semantics. (I'm aware that the distinction is not unproblematic.) That I sometimes don't know how to say things because I lack words, doesn't mean that I have problems with the grammar; I just have no words to arrange and modify according to those "transformation rules". So, if deep-structure thought (I-language thought?) can proceed without words, what stands in for words in such a situation? What's the relationship between wordless thought and the constructed word-sequence? I mean, sure, if there's an I-language that's different from an e-language, such wordless thought-stream controlling the e-language output could easily be described as an i-language. That would be some sort of retro-engineering, no? I'm not sure I ever really understood Chomsky.

    As far as concepts, do animals that don't have language have concepts? What is a non-linguistic concept? A dog associating a leash with a walk, is that a concept? It's association sure. Concepts seem to be something beyond just association. Concepts seem to join with a mechanism whereby they are "used" and that can be something akin to a grammar.schopenhauer1

    How do you explain association without concepts? A dog associates what with what? A leash and a walk need to be something associatable; I'm fine with using the word "concept" for that. I don't think it's all that different from a person demonstrating knowledge about role of chairs in waiting rooms by sitting down on one. (You don't need to think the word "chair" to do that.) Whether or not the association itself is also a concept, I don't know. Maybe the dog sees it as some sort of ritual? Likely not, but how would you rule this out?

    I can see I-languages being a mentalese that accounts for our shorthand internal language maybe.schopenhauer1

    I somehow get it and somehow don't. If I were to put it in terms that are more intuitive for me, I'd say that Chomsky posits a grammar of thought, and a grammar of language, and a connection between the two. So we have this grammar module in our head that's pretty much the same for everyone, but doesn't determine what language emerges as output (since that's partly social). And Chomsky seems to want to talk about it all in terms of language.

    I mean, I'll just point out here that Chomsky thought "Colourless green ideas sleep furiously," is a grammatically well-formed sentence, even though it makes no sense. Grammar is about how we form thoughts, not whether they make any sense, much less about wither they're true. That's where some his pupil parted ways with him (google generative semantics, a shortlived movement, but it did lead to other theories, like cognitive linguistics [via Lakoff, I think?]). I may be wrong about that, too; it's been a while.

    The minimalist program is just "merge" now I think.schopenhauer1

    Heh. Well, there's certainly more to the methodology, like, say, X-bar theory. I always thought Generative Grammar gave us quite an interesting set of methodology to work with, but I never quite bought into the language faculty stuff.

    At any rate, I don't know how analiticity relates to all that. I'd probably expand the question to:

    Do analytical sentences exist, and if so are they a feature of the I-language, or are they judgements we port over from non-linguistic cognition to fully formed e-language sentences? (I might need to read more Chomsky to phrase this properly.)
  • Banno
    25.3k
    It even appears that Chomsky is directly challenging Wittgenstein's concept of a private language.schopenhauer1
    I'll take some small issue with this. Wittgenstein's private language is used to refer to supposedly private sensations, to that feeling you have when your blood pressure is high, to that pain. That's different to what is being described in your quote.

    The last speaker of a natural language, and Robinson Crusoe, do not provide examples of such a private language.

    Could a Boltzmann Brain develop a language? Perhaps, if it divided itself against itself.
  • Banno
    25.3k
    It occurs to me that since the characteristic of language most central to Chomsky's approach is that language is compositional, hence permitting a small finite range of words to provide innumerable structured sentences, that it is compositionally that might best serve to remove a dog's associating a leash with a walk from what we might commonly call conceptualisation.

    Concepts can be merged to construct new concepts.

    Hence the dog might understand that it's master will take it for a walk, but not that its master will take it for a walk next Tuesday.
  • Banno
    25.3k
    The question for Professor Chomsky is at https://thephilosophyforum.com/discussion/comment/804011

    Draft form. Suggestions welcome.
  • RussellA
    1.8k
    This is to suppose that there is a thing, which is someone's understanding of grass; as if to understand "grass" were to have a certain box in one's mind; so that your box can be different to my box.Banno

    I believe that i) grass doesn't supervene on its properties, a typically short plant, etc. and ii) "grass" doesn't supervene on its properties, "a typically short plant, etc."

    To start, there is a diference between what exists in language and what exist in the world. In language, "grass" is defined by the properties "a typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop". In the world, grass includes the properties a typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop.

    I see something in the world that has the properties typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop. It would be unwieldy in conversation to say "yesterday I mowed a typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop". It is far easier to define "grass" as "a typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop" in order then to more simply say "I mowed the grass"

    Once "grass" has been defined as the set of properties "a typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop". the statement "grass has narrow leaves" is true by definition and is therefore an analytic statement.

    Grass in the world may have numerous properties, only some of which are included in the definition of "grass". In part for the reason that some properties may only be discovered in the future, and in part that some properties may not be useful in daily conversation. It is also true that the definition may change with time.

    Definitions aren't perfect, but they are better than the alternative of not having definitions. Without definitions communication would break down. Especially if you said "I mowed the grass yesterday", and my concept of "grass" was "a thickset, usually extremely large, nearly hairless, herbivorous mammal of the family Elephantidae having a snout elongated into a muscular trunk and two incisors in the upper jaw"

    As regards boxes in my mind, I agree that my concept of grass is private and inaccessible to anyone else, but my concept of "grass" is accessible to others as it is has been publicly defined.

    Communication using language would break down without definitions. One consequence of definitions are analytic statements.

    Analyticity is the acknowledgment that the whole doesn't supervene on its parts.
  • RussellA
    1.8k
    So, if deep-structure thought (I-language thought?) can proceed without words, what stands in for words in such a situation?Dawnstorm

    Thoughts and concepts stand in for words

    From SEP - Analayticity and Comskyan Linguistics
    He sharpens this distinction in his (1986, pp. 20–2) by distinguishing what he regards as essentially the ordinary, folk notion of external, what he calls “E-languages,” such as English, Mandarin, Swahili, ASL and other languages that are commonly taken to be spoken or signed by various social groups, vs. what he regards as the theoretically more interesting notion of an internal “I-language.” This is not a “language” that is spoken at all, but is an internal, largely innate computational system in the brain (or a stable final-state of that system) that is responsible for a speaker’s linguistic competence

    As I see it, in the mind there are two aspects. There are conscious thoughts about concepts, in that I can have the conscious thought that the tree is to the left of the river, and there is an underlying subconscious structure that enables me to have these conscious thoughts about concepts.

    I can have conscious thoughts about concepts without the need of words. When driving through a busy city street, if I had to put a word to every thought or concept, I would have crashed in the first five minutes.

    Once I can have conscious thoughts about concepts, an E-language can be developed. My conscious thought that the tree is to the left of the river may be expressed as "the tree is to the left of the river".

    In a sense, our conscious thoughts about concepts is a language, but this is secondary to the underlying subconscious structure that enables thoughts about concepts. It is this underlying structure that is common to different peoples.

    In an E-language, if "grass" is defined as "a typically short plant with long, narrow leaves", the statement "grass has narrow leaves" is analytic.

    In the mind, if I have the thought that there is something that is typically a short plant with long, narrow leaves, then I know that this something that is typically a short plant with long, narrow leaves has narrow leaves, and again, this is analytic.

    IE, in the mind, we can have conscious thoughts about concepts, but underlying this is a deeper subconscious structure that is common to different peoples that enables us to have these thoughts about concepts in the first place.
  • RussellA
    1.8k
    Suggestions welcome.Banno

    Chomsky says that concepts cannot exist without language. See Noam Chomsky on the Big Questions (Part 4).

    7min "Even the simplest concepts tree desk person dog, what ever you want , even these are extremely complex in their internal structure . If such concepts had developed in proto human history when there was no language they would have been useless. They would have been an accident if developed and quickly lost as you cannot do anything with them. So the chances are very strong that the concepts developed within human history at a point where we had computational systems which satisfy the basic property"

    Perhaps Chomsky would say that as concepts cannot exist without language, if there is analyticity in language then there must also be analyticity in concepts.
  • schopenhauer1
    11k
    So, if deep-structure thought (I-language thought?) can proceed without words, what stands in for words in such a situation? What's the relationship between wordless thought and the constructed word-sequence?Dawnstorm

    Good question, which is why it's interesting to me that Chomsky doesn't seem as interested in concept formation (the content or representation of what is being generated). Because the assumption seems to be that if concepts are in place, and if inner talk is the primary target of language, the internal grammar should be consistently used for that purpose. This article and others seem to indicate that internal talk is varied and oftentimes truncated and only formalized when externalizing it in external language. So yeah, I'm just trying to throw some ideas out defending inner talk as the target whilst not seeming to have generative grammar internally. I can only think as a defense, that grammar is there in internal thoughts but it's hidden and working in the background. It's running in the background, but the person doesn't realize its efficacy in mental efficiency until it is fully actualized in external language perhaps? Again, this would be much easier if Chomsky's focus was conceptual networking and association, but it is generative grammar which is rule-based. So that makes it harder to defend I would think. However, I still think there is actually a case for conceptual representation and networking being something much more efficient in human brain structures that is innate and targeting inner mental activity and not simply a hodgepodge of other mechanisms or targeting communication. But that is pure conjecture on my part.

    How do you explain association without concepts? A dog associates what with what? A leash and a walk need to be something associatable; I'm fine with using the word "concept" for that. I don't think it's all that different from a person demonstrating knowledge about role of chairs in waiting rooms by sitting down on one. (You don't need to think the word "chair" to do that.) Whether or not the association itself is also a concept, I don't know. Maybe the dog sees it as some sort of ritual? Likely not, but how would you rule this out?Dawnstorm

    Yeah, concepts are tricky. I think in fact, much of analytic philosophy's general confusion (starting with people like Frege, Russell, and Wittgenstein and going from there) goes back to this problem of concepts and being baffled by what exactly concepts are. The ancient Greeks of course had their notions- Plato had pre-existing Forms, and Aristotle had essences. But I would gather to say that something that ties these "concepts" together is a sort of abstraction; it isn't just recognizing a pattern (i.e. associative learning), but having a level of remove from the association whereby it becomes "tokenized" like a mental "object" that one's memory can refer back to. And it is precisely the nature of this "tokenization" that creates the question of whether some sort of linguistic ability has to be there for conceptual thought to take place. In other words, it begs the question of whether concepts entail language. It might not be equivalent, but perhaps where you see smoke (concepts) you see fire (language).

    Do analytical sentences exist, and if so are they a feature of the I-language, or are they judgements we port over from non-linguistic cognition to fully formed e-language sentences? (I might need to read more Chomsky to phrase this properly.)Dawnstorm

    I think he doesn't have much to say on it, as that's not what he's interested in. Rather, I think people like Kripke wrote some influential theories in the philosophy of language community, specifically regarding the idea of "causal link theory" whereby someone "baptizes" a word and there is a link of that name to that object that obtains in all possible worlds. Mind you, I believe this theory was meant for proper names and expanded to scientific kinds and some other instances. Perhaps it only applies to objects that are actual instances and not universals. So, BrownMug is what I dub this mug next to me. It is thus causally linked as long as it is continually being used in the community. BrownMug is always linked now to that particular mug. However, I don't think this is the same for the universal "mug", as "mug" could have been something else in all possible worlds.

    This relates tangentially to @Banno OP.
  • schopenhauer1
    11k
    I'll take some small issue with this. Wittgenstein's private language is used to refer to supposedly private sensations, to that feeling you have when your blood pressure is high, to that pain. That's different to what is being described in your quote.

    The last speaker of a natural language, and Robinson Crusoe, do not provide examples of such a private language.
    Banno

    So I think it's half-and-half. In terms of grammar generation, Chomsky seems to believe language can be wholly internal (I-language not E-language) and therefore, some sense private (if we mean innate by this) and does not need a language community to generate the rules. For E-languages, sure, the conventions have to be there. As to the conceptual/semantic aspect, that doesn't seem to be the focus for Chomsky either way, so not sure about that one.

    So it seems for public E-language, the community is needed for convention, but I-language could suffice without a community. What this means for inner talk, I'm not sure though. It still seems to use E-language, just internally, so would probably not count as a private language. I'm not sure though to what extent I-language can be considered a private language prior to its usage in generating E-language.
  • schopenhauer1
    11k
    7min "Even the simplest concepts tree desk person dog, what ever you want , even these are extremely complex in their internal structure . If such concepts had developed in proto human history when there was no language they would have been useless. They would have been an accident if developed and quickly lost as you cannot do anything with them. So the chances are very strong that the concepts developed within human history at a point where we had computational systems which satisfy the basic property"

    Perhaps Chomsky would say that as concepts cannot exist without language, if there is analyticity in language then there must also be analyticity in concepts.
    RussellA

    This very much aligns with what I was discussing here:

    Yeah, concepts are tricky. I think in fact, much of analytic philosophy's general confusion (starting with people like Frege, Russell, and Wittgenstein and going from there) goes back to this problem of concepts and being baffled by what exactly concepts are. The ancient Greeks of course had their notions- Plato had pre-existing Forms, and Aristotle had essences. But I would gather to say that something that ties these "concepts" together is a sort of abstraction; it isn't just recognizing a pattern (i.e. associative learning), but having a level of remove from the association whereby it becomes "tokenized" like a mental "object" that one's memory can refer back to. And it is precisely the nature of this "tokenization" that creates the question of whether some sort of linguistic ability has to be there for conceptual thought to take place. In other words, it begs the question of whether concepts entail language. It might not be equivalent, but perhaps where you see smoke (concepts) you see fire (language).schopenhauer1
  • Banno
    25.3k
    ...supervene...RussellA

    An odd word, now becoming surprisingly common. What could it mean to have properties supervene onto individuals... green supervene on grass... that the green "occurs as an interruption" to the grass? Hu?

    And what does it relate to what I have said?

    Communication using language would break down without definitions.RussellA
    Why? A child knows its mother, despite not being able to provide a definition. And so on for the vast majority of words. I think you are here just wrong.

    The following argument is stolen from Austin:
    Look up the definition of a word in the dictionary.

    Then look up the definition of each of the words in that definition.

    Iterate.

    Given that there are a finite number of words in the dictionary, the process will eventually lead to repetition.

    If one's goal were to understand a word, one might suppose that one must first understand the words in its definition. But this process is circular.

    There must, therefore, be a way of understanding a word that is not given by providing its definition.

    Now this seems quite obvious; and yet so many begin their discussion with "let's first define our terms".
    Banno
  • Banno
    25.3k
    Thoughts and concepts stand in for wordsRussellA

    Is this your opinion, or your view of Chomsky, or both?
  • Banno
    25.3k
    analyticity in concepts.RussellA

    Analyticity without language? What could that be? I can make sense of two lexical elements standing for the same concept, but of two concepts standing for the same concept? How could that work?
  • Banno
    25.3k
    So here is where connectionism comes in to the discussion, with the possibility of cognition without (at least local) representation.

    A neural network need not, and usually does not, work things out using symbols to represent the things on which it is working.
  • schopenhauer1
    11k
    A neural network need not, and usually does not, work things out using symbols to represent the things on which it is working.Banno

    Yes this has been a classic debate of connectionism versus computationalism which this article provides a good overview of: https://plato.stanford.edu/entries/computational-mind/#ArgForCon.

    It seems that Chomsky thinks grammar more computationalist, and conceptualization can be either. According to Chomsky, grammar is too hierarchical for it to be distributed.
  • unenlightened
    9.2k
    When one shouts "Fire!" in, say, a theatre, one does not mean merely to refer to "the rapid oxidation of a material (the fuel) in the exothermic chemical process of combustion, releasing heat, light, and various reaction products."

    Rather, it is a call to action in a matter of life and death. One means 'evacuate immediately, bring an extinguisher, call the fire brigade, wake up and stop watching the play, Sauve qui peut.

    In the trenches, it means something else again.

    In another thread, https://thephilosophyforum.com/discussion/14286/a-potential-solution-to-the-hard-problem, the beginnings of consciousness are posited as being in evaluating sensations. I propose that not naming, but evaluation is the beginning of language; the first word was something like a thumbs up or a thumbs down.

    I would ground meaning and language in the same giving-a-fuck-icity. The Boy who cried 'Wolf!' is not a tale of someone describing the fauna, but of someone calling falsely for help, and how that falsehood undermined himself as a communicating member of society.

    And when we hear that "the Assyrian came down like a wolf on the fold" we should understand that the word 'wolf' is being used in the same sense, with the same urgency of meaning, and not merely "... the largest extant member of the family Canidae, and is further distinguished from other Canis species by its less pointed ears and muzzle, as well as a shorter torso and a longer tail. Wiki.

    Analiticity surely comes much later, when wolves are not much problem any more, and we can start measuring the length of their tails. Certainly one does not begin with Euclid's Elements.

    The sound of the dinner gong does not indicate a concept, it is a call to arms.
  • schopenhauer1
    11k
    Analiticity surely comes much later, when wolves are not much problem any more, and we can start measuring the length of their tails. Certainly one does not begin with Euclid's Elements.unenlightened

    The idea becomes muddled as analyticity is often associated with a prioricity. The idea being that logic is both analytic and a priori (needs no investigation of instances of the world to be true). For example, if something is fully X then it is not not X.
  • unenlightened
    9.2k
    For example, if something is fully X then it is not not X.schopenhauer1

    Where X, with deep irony, stands for anything at all. And what is this "not"? It must be an unsaying, like the all clear after an air-raid warning. Panic over!
  • schopenhauer1
    11k
    Where X, with deep irony, stands for anything at all. And what is this "not"? It must be an unsaying, like the all clear after an air-raid warning. Panic over!unenlightened

    It could be the case that all analytic statements were simply one-time synthetic statements that were conventionalized.
  • RussellA
    1.8k
    When one shouts "Fire!" in, say, a theatre, one does not mean merely to refer to "the rapid oxidation of a material (the fuel) in the exothermic chemical process of combustion, releasing heat, light, and various reaction products.".......Rather, it is a call to action in a matter of life and death.unenlightened

    True, the same word may be defined in many different ways. The Merriam Webster dictionary for "fire" lists almost 42 different uses.

    Though the same principle applies, in that is is more convenient and quicker to say "fire" than "I am making a call to action in a matter of life and death."
  • unenlightened
    9.2k
    True, the same word may be defined in many different ways. The Merriam Webster dictionary for "fire" lists almost 42 different uses.RussellA

    I can well imagine an audience of philosophers looking up the word in Merriam Webster and discussing back and forth which of the 42 definitions applies in the particular case while the auditorium burns around them. Not.

    First language, then definitions. Let us talk about language before definition, before dictionaries, for a moment. The language of a child. For example: my daughter would hear us saying things like "Can you do that on your own?" And being independent minded, she soon started to demand, "Let me do it on my rown!." Now you will not find "rown" in Merriam Webster, but we knew what she meant, as does everyone reading this.
  • RussellA
    1.8k
    Suggestions welcome.Banno

    Chomsky likes Locke.

    From IEP - Locke:Epistemology
    Locke defines knowledge as the perception of an agreement (or disagreement) between ideas (4.1.2). This definition of knowledge fits naturally, if not exclusively, within an account of a priori knowledge. Such knowledge relies solely on a reflection of our ideas; we can know it is true just by thinking about it. Some a priori knowledge is (what Kant would later call) analytic.

    Would it be possible to get Chomsky to talk about the analytic in reference to Locke, someone he likes.
    ===============================================================================
    An odd word, now becoming surprisingly common. What could it mean to have properties supervene onto individuals... green supervene on grass... that the green "occurs as an interruption" to the grass? Hu? And what does it relate to what I have said?Banno

    In my mind is a box that includes both the knowledge of my private experience of grass, something not available to anyone else, and the knowledge of the word "grass", which is publicly available to everyone else, as described by Wittgenstein in PI para 293.

    As some philosophers, including Searle, believe in supervenience, the concept should be taken into account.

    If there is no superveninece, then the word "grass" is simply the set of words "a typically short plant with long, narrow leaves, growing wild or cultivated on lawns and pasture, and used as a fodder crop".

    As the SEP - The Analytic/Synthetic Distinction notes
    “Analytic” sentences, such as “Paediatricians are doctors,” have historically been characterized as ones that are true by virtue of the meanings of their words alone and/or can be known to be so solely by knowing those meanings.

    In this case the meaning of "grass" is fully known by knowing the meaning of the words "a typically short plant, etc". In this event, as the SEP notes, analytic.

    However, if there is supervenience, then the word "grass" is more than the set of words " a typically short plant, etc". In this event, the meaning of grass cannot fully be known just by knowing the meaning of the words "a typically short plant, etc". IE, the expression cannot be analytic.

    IE, whether an expression is analytic or not partly depends on one's attitude to supervenience.
    ===============================================================================
    Why? A child knows its mother, despite not being able to provide a definition. And so on for the vast majority of words. I think you are here just wrong.Banno

    We are communicating using written language. I have no other clues to your meaning other than the words on my screen, such as tone of voice, facial expression or bodily movements.

    You have written the word "mother". You may not believe me, but I don't know what this word means. Within the context of the paragraph it could mean, "a child knows its neighbourhood", "a child knows its school", "a child knows its pet", etc.

    In order for me to fully understand what you are saying, what does "mother" mean ?
    ===============================================================================
    The following argument is stolen from Austin: Look up the definition of a word in the dictionary. Then look up the definition of each of the words in that definition.Iterate.Banno

    There are two types of concepts, simple and complex (there may be better terminology).

    As I wrote before
    The gavagai problem may be solved by taking into account the fact that there are simple and complex concepts, and these must be treated differently. In language, first there is the naming of simple concepts, and only then can complex concepts be named, such as the complex concept "gavagai".
    Simple concepts include things such as the colour red, a bitter taste, a straight line, etc, and complex concepts include things such as mountains, despair, houses, governments, etc.


    Dictionaries allow us to learn complex concepts, where a complex concept is a set of simple concepts. But we cannot learn simple concepts from the dictionary

    In Bertrand Russell's terms, the dictionary allows us knowledge by description but not knowledge by acquaintance.

    For simple concepts, we need knowledge by acquaintance, achievable using Hume's constant conjunction of events.
    ===============================================================================
    Is this your opinion, or your view of Chomsky, or both?Banno

    True, I should have made it clearer when I wrote " As I see it".

    @Dawnstorm asked "So, if deep-structure thought (I-language thought?) can proceed without words, what stands in for words in such a situation?"

    It seems inconceivable that we use words such as "road", "traffic light", "pedestrian", "blue sky" in an I-language.

    As I wrote:
    When driving through a busy city street, if I had to put a word to every thought or concept, I would have crashed in the first five minutes.

    I see no alternative to the idea that in an I-language thoughts and concepts stand in for words.
    ===============================================================================
    Analyticity without language? What could that be?Banno

    Starting with E-language, if "grass" has been defined as "vegetation consisting of typically short plants with long, narrow leaves, growing wild or cultivated on lawns and pasture, and as a fodder crop", then the statement "grass has long narrow leaves" is analytic.

    There is a direct analogy between analyticity in E-language and analyticity in I-language.

    Within the I-language, there are no words such as "vegetation, consisting, etc", but rather thoughts about concepts.

    It follows that within the I-language, if I know that grass is vegetation consisting of typically short plants with long, narrow leaves, growing wild or cultivated on lawns and pasture, and as a fodder crop, then I must also know that grass has long narrow leaves, which is also analytic knowledge.
  • RussellA
    1.8k
    Now you will not find "rown" in Merriam Webster, but we knew what she meant, as does everyone reading this.unenlightened

    Would you know what a child meant if they said something more complicated, such as
    "habari za asubuhi, habari gani" without having to look in a dictionary.
  • Jamal
    9.9k
    It could be the case that all analytic statements were simply one-time synthetic statements that were conventionalized.schopenhauer1

    Was there a time when it wasn’t common knowledge that all bachelors were unmarried men? You know, before it was discovered?

    Maybe “all bachelors are unmarried men” seems synthetic when it informs someone who doesn’t know what a bachelor is. So it could be reworded to show that the statement in this case is about the word rather than about bachelors: “‘bachelor’ means ‘unmarried man’”. This is synthetic (as I’m supposing all definitions are) and it follows from it that “all bachelors are unmarried men” is analytically true.
  • schopenhauer1
    11k
    Was there a time when it wasn’t common knowledge that all bachelors were unmarried men? You know, before it was discovered?

    Maybe “all bachelors are unmarried men” seems synthetic when it informs someone who doesn’t know what a bachelor is. So it could be reworded to show that the statement in this case is about the word rather than about bachelors: “‘bachelor’ means ‘unmarried man’”. This is synthetic (as I’m supposing all definitions are) and it follows from it that “all bachelors are unmarried men” is analytically true.
    Jamal

    Yes! Exactly what I am suggesting. And I know others have posited something like this, like Quine. It seems to have been a holdover from 17th and 18th century statements about truths to make the distinction so clean cut. Kripke was getting to its nature by emphasizing necessity above all else. That is to say, a triangle necessitates it being three sides and 2 +2 =4 is true in all possible worlds. However, the discovery of this truth is in some way synthetic when first discovered. The passing on of this discovery as a convention that we learn very early on, makes it "analytic", but this is only the way we discover the information. Logic truths might be entailed by necessity, but these are still things for which have to be discovered. That is the main point. Someone worked that out through observation and computation and inference and comparison and all that.
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment

Welcome to The Philosophy Forum!

Get involved in philosophical discussions about knowledge, truth, language, consciousness, science, politics, religion, logic and mathematics, art, history, and lots more. No ads, no clutter, and very little agreement — just fascinating conversations.