@@KaneB yah, your content is absolutely amazing, you really have helped me to understand many things that I didn't understand, and i wish you can make a video or a series of videos!😅 about pragmatism from Peirce to Rorty
@@KaneB Thanks yourself Kane. Had a chance to rewatch key sections Overall no issues with the quality - this seems more of a what is this - I was hoping for a response at the end Kinda a cliffhanger lol I guess I will cover this in my video when I can get to it and credit you and give the vid a signal boost but I like to provide some historic context I'm surprised you don't get into some of Quine's motivations - his atheism, conservatism, commitment to pragmatism, etc. One major quibble - one minor quibble the minor one is a bit of a goof in the phrasing probably a small stylistic thing but I'll try to find the timestamp when I can as for the major issue/problem/dilemma Your presentation seems fine as far as it goes from the underdetermination issue in the final sections I think Quine hints thought underdetermination might not be a major issue I assume Quine can be conceived as a verificationist [‘Should the unwellcomeness of the conclusion persuade us to abandon the verification theory of meaning? Certainly not. The sort of meaning that is basic to translation, and to the learning of one’s own language, is necessarily empirical meaning and nothing more.' Epistemology Naturalized I would assume Quine would say physical theories will not help us much with translation problems HOWEVER we do still get brute sensations due to our bodies his answer might be radical translation will always be a problem but our sensory faculties provide some (shaky) foundation for meaning and we proceed from there Usually underdetermination assumes observations are necessarily theory-ladden I'd have to double-check if Quine agrees - if so he might agree underdetermination is a major issue but it is separable from observations theory-ladden and some observations can be accepted as basal So radical translation remains a problem but we can assign (tentative) meanings via brute sensations Don't know if you are aware the verificationist issue or the video was long enough and things were cut for time Excellent video just playing devil's advocate how Quine might refine his position Additional quote from Epis Natur: ‘Two cardinal tenets of empiricism remained unassailable, however, and so remain to this day. One is that whatever evidence there is for science is sensory evidence. The other … is that all inculcation of meanings must rest ultimately on sensory evidence.’ Hope this helps thanks
Quine’s argument for the indeterminacy of translation is a philosophical thesis that challenges the possibility and objectivity of meaning. He claims that there are no objective truth conditions for sentences in a foreign language, because there are always multiple ways to translate them that are equally compatible with the available evidence. He illustrates this point with his example of radical translation, where a linguist tries to translate the utterances of a native speaker without any prior knowledge of their language or culture. He shows that there are many possible translations that fit the same behavioral data, but none of them can be verified or falsified by further observation. He concludes that translation is not a matter of discovery but of invention, and that there is no fact of the matter about the semantic content of sentences. He makes an ontological claim, not an epistemic one. He does not say that it is difficult or impossible to know what sentences mean, but rather that they do not have any determinate meaning at all.
Isn’t this just very similar to searle’s Chinese room argument? He says that you cannot get semantics (meaning) from syntax. And Quine is saying that you cannot get a specific meaning from yo, gavagai. Hevet.
Great presentation, I'll agree with Chomsky that this one of the most irrational arguments in intellectual history. Words aren't even a theoretical term to start but in so far as they exist they are a collection of their features their semantic ones as well
I'm not convinced. Even if we accept that it is impossible to accurately determine the meaning of someone else's sentence, that seems mostly separate from whether the sentence has meaning to the person themselves. It reminds me of the old question of "is my red the same as your red?" But even if they are different (which they are, since I'm actually colorblind), that doesn't change the fact that "red" means something specific to me. I also think it's interesting to think of a high-dimensional latent space for meaning, like AI researchers sometimes use. An untranslatable word might have a very specific meaning but just not be well aligned with any words in the target language. And if we give up on pinpointing the meaning exactly I think this should solve a lot of the underdetermination issue. Each word would start with a big blob of possible points it could map to, but every time you ask a question you should be able to narrow it down a bit until it is specific enough that you're satisfied. And again, going from the position in this latent space back to a suitable translation in a different language should really be thought of as an entirely separate task.
You're not convinced because you aren't the same kind of physicalist that Quine is. Or at least, you can imagine better than Quine how we could come up with a physicalist account of subjective experience, therein making possible a physicalist account of meaning (as experienced by subjects) discoverable by observation.
@@_notion2976 What was the argument supposed to be about then? I admit I was listening while doing other stuff so I might have missed something, but that certainly seemed to be what he was doing...
I suspect that what you call "physical theory" in the "argument from above", I would actually call a _metaphysical_ theory. Example: physical theories tell us what the wavefunction _does,_ but they will never tell us whether the wavefunction is _a real entity_ (as opposed to a useful fiction, or an evil demon's simulation, et cetera). Wavefunction realism is a _metaphysical_ theory. In this case, maybe Quine has actually rediscovered (or even, rehabilitated) the Logical Positivist thesis that _metaphysical claims specifically_ don't have any meaning, precisely because they are not empirically distinguishable? Are there philosophers who interpret Quine in that way?
I don't think there is any difference in the uses of the terms. Physical theories are the theories proposed by physicists, such as the theories that tell us what the wavefunction does. I think Quine would agree that, in themselves, these theories do not tell us anything about whether the wavefunction is real. Quine's metaphysical theory is that we ought to be committed to all and only those entities that are quantified over in our best physical theories.
I am bothered by the assumption that there are separate sets of native speakers. We are aware that sets of native speakers can overlap: people can be fully bilingual, who are in a position to adjudicate on the accuracy of translation. It is possible to try to argue that all communications are indeterminate and involve acts of translation, but other than helpfully encourage us to be aware that we can always be prone to misunderstand, it is an observation of limited value when we are more interested in how it is we more often than not do understand each other.
Im very sympathetic to the conclusions, but for slightly diffrent reasons. But isn't it much easier to confirm statements that aren't correct? Because then you will get a 'no' every time? Kinda how we usually think of scientific experiements as disproving existing theories, while it being underdetermined which theories are proven by any given experiment?
It doesn't seem like getting a "no" will help much if the statement or the question are open to alternative translations. What are the natives denying?
@@KaneB maybe your right, but i was thinking the uniformaty of the responses will be enough. So for example If i ask If something is a "fox" and point to diffrent things (chairs, TVs, turtles ect.) , you will presumably say "No" to alot more things then those that are related to foxes (its parts, its shadow, its fur ect. Things i cant destinguish from the fox) where you will say yes, but i wont be able to destinguish which part you mean. So No will be heared alot more, giving me information little by little. But never quite adding up to "truth" of what fox means. Very much like the scientific method
Why the limitation to verbal responses? why is the translator the only one allowed to point? why can't the translator try to teach vocabulary to the speakers of L? it seems like some progress could be made by asking the natives to point to a gavagai from a selection of objects, including rabbits with missing tails and ears. It seems like some progress could be made by asking the natives to draw a gavagai. Or at least Quine would have to work much harder to make alternative translations that match all evidence if we allow more than yes/no questions.
he allows for any and all responses in radical translation. The yes/no just has to do with finding out what assenting to or dissenting from by speakers of the foreign language even looks like, which is the kick-off for presenting hypotheses, testing out their words yourself under conspicuous circumstances. Radical translation isn't merely drawing conclusions from elicited yea/no answers.
It's not limited to verbal responses; all behaviour can be considered when constructing a translation manual. Quine's view though is that there will always be interpretations of the sentences or questions that are consistent with the alternative translation manual. Suppose we translate "gavagai" to mean "undetached rabbit-part". Now we instruct one of the natives to draw a gavagai. What would we expect them to draw? Well, an undetached rabbit-part... which would look exactly like a rabbit. If they were to draw e.g. just the ears, this would be a detached rabbit-part, not an undetached rabbit-part. A drawing of an undetached rabbit-part has to show it attached to the rabbit. So it has to show the rabbit.
@@KaneBHe would necessarily have to circle the part he’s referring to or otherwise draw attention to it and emphasize that it is undetached, in that case
I wonder why Quin believes that languages must be meaningful or have a list of determined meaning. Ultimately if two interloceters communicate in such a way that all (or even most) of there behaviors are as expected be each other then I argue the language purpose has been achieved. Let w be a word. And U(w) be all the utterances that can be made in a language L with w. Now let A_me(u), u in U(w), be true if I agree with u, also let A_you(u) be true if you agree with u. I argue that if A_me is approximately equal to A_you then me and you agree on the meaning of w. Further I argue that if G is a grammar of terms in L contains only and all (or most) values in A_you then dictionary is a translation module for L_you, L as spoken by you. Of course there may be an infinite number of such grammars but so what, each can translate you and each approximates your meaning such that I can approximately agree with you. Such that the purpose of language to communicate or make concrete ideas is achieved.
The idea of "theory" - illustrated here as a line through many points, and the idea that there are infinite lines that pass through all points, etc. - this definition of "theory" is better stated as a "model" and that there are infinite "models" which yield lines passing through all points. Yet, a model and a theory are two different things. You might call the proper "theoretical backdrop" of all these "models", i.e., lines passing through points, is that of a function. All of the models are functions such that for every x there is one and only one y. Further, I would resist viewing this relationship between theory and model as something that is static. Rather, perhaps it functions more like the way I understand (perhaps mistakenly) genus and species do for Aristotle such that genus and species are not static taxa which occupy a particular position within a particular taxonomy, but instead are two aspects of a dyad which are found at all levels of a taxonomy. For example, phylum and class are particular taxa of a particular taxonomy where in the phylum is the genus and the class is the species. It doesn't help that genus and species are also particular taxa within this particular taxonomy wherein genus is the genus and species is the species.
With data fitting, we avoid adding unjustified bumps. This is like occram's razor. So we start with a straight line. If we want to bend the curve that goes through the data, we need to have an explanation as to why the curve bends. In the video, the curve was always increasing, but marginally decreasing - it was always going up, but going up less each increment. We would need an explanation like "you get benefit from each apple that you receive, but you get less benefit from the 100th apple than the first apple". If you want some polynomial that bumps and twists it's way through all of the points, you need explanations for all the bumps. Not only is this tricky, but it is likely to be falsified when new data comes in. Such is the way with tacking on "and God exists" or other such extemporaneous assertions onto translations of sentences. Why would one think that "schnee ist veiss" means "snow is white and God exists". There's no where for it to be grammatically because we see the word schnee in "the snow is on the mountain" and we see veiss in "the cloud is white". When words are under-determined, this is no problem. There are these things called logical operators. One of these operators is called "or". We can use this for when things are under-determined. Logos means "words" OR "speech" OR "language" etc, depending on context. Quine should have been ridiculed out of academia.
Kane said rightly that this isn't an epistemic argument, in other words, this isn't about the evidence and ways available to the linguist underdetermining a translation. The indeterminacy is not an underdetermination. Furthermore, on your own solution to underdetermination (which is irrelevant to the supposed indeterminacy) merely adding disjuncts just creates a disjunction problem. I'm honestly struggling to see how what you said is relevant to what's being discussed.
@@_notion2976 I'm merely pointing out that additional ng ad-hoc meanings to translations is unnecessary and counter productive. All of Quine's arguments rely on adding ad-hoc meanings. His indeterminacy seems to be derived from under-determinancy as under-determinancy seems to keep coming up in the arguments.
Have we read the same Quine? I read his _Word and Object_ and found, mostly, a lot of dogmatic and naive Skinnerian behaviourism. Quine claims children are taught their native language by rewarding them for making true utterances and punishing them for making false ones (his prime example for that is the utterance "Ouch"). Apparently, children have to be punished for engaging in make-believe play, or else they will never acquire a language. Yeah, right. With that sort of theoretical assumptions, I must say nothing that follows can be given much credit. Another major problem with Quine (which he shares with most of the field) is that "meaning" for him has to take the form of word-object pairings (and the usual nonsense about "logical form" when considering sentences). That's simply not how language works.
@WhyCan’tIRemainAnonymous?! “…found, mostly, a lot of dogmatic and naive Skinnerian behaviourism”, this is a common criticism of Quine, though, I believe, a mistaken one. I would direct you to Sander Verhaegh’s article “The Behaviourisms of Skinner and Quine (2019)” which serves to show, surprisingly, how little they influenced one another. Moreover, I’d argue against your assertion that Skinner or Quine held varieties of behaviourism that were dogmatic or naïve; although, I can sympathise with that ascription to Skinner, particularly the radical variety of behaviourism he espoused in his later years. Indeed, Quine described Skinner’s radical behaviourism-in his “Sellars on Behaviourism, Language and Meaning” (1980)-as “dogmatic”. Re: your third point. I believe this criticism is an example of “talking past” the object of criticism i.e. the stances on meaning you ascribe to Quine are not views he holds. Other philosophers? certainly. Quine, like J.L. Austin, doesn’t subscribe to Wittgenstein’s “meaning is use” doctrine-language is far too messy, it’s uses far too diverse and contrary (not that this is a bad thing, mind). Again, I sympathise with your attitude towards the use of “logical form” by philosophers as a kind of cure all for all problems, Austin roundly criticised such facile talk, the obsession with cer in Sense and Sensibilia
Accidentally posted this too quickly (the joys of typing on a phone!). Here’s what I intended to add: To the last paragraph, line 18, I planned to add the following [the obsession with certain buzzwords etc.,], and [Depending on how interested you are in this topic (I’m assuming quite a lot here about your motivation; I’d like to think I’m being charitable ;)) I’d recommend looking into Quine’s debate with Chomsky and the secondary literature concerning that debate, since many seek to use Quine as a foil for Chomsky. For instance, it might surprise you to learn that Quine and Chomsky are much closer as concerns language, than many have held them to be (see Alexander George’s “Whence and Wither the Debate Between Quine and Chomsky”). Anyway… much of what I said might seem tangential, granted, but I feel that it highlights the systematic nature of the older generation of philosophers’ work-of Quine’s, Carnap’s, Sellars’, Russell’s and Mach’s work-to name but a few luminaries. One can’t appreciate one work (e.g. Word and Object), or one sentence in isolation (to use a Quinean metaphor), for they are each connected in significant ways to others within that philosophers corpus; themes, theories and ideas aspire to something grander.
@@michaeldalscais2538 Quine was openly enamored with Skinner, at least in the relevant period of his life (I recall Hillary Putnam give his own impression about that at a lecture I once heard; Putnam and Quine worked in the same department, so that's first-hand testimony 🙂). I do agree that his debate with Chomsky is overstated, though; the two fundamentally share the same conception of language. And it is this conception of language that I find to be fundamentally conservative (and also just plain false). In a sense, it's little wonder that Quine is skeptical about meaning (and the same goes for his disciple Davidson's rekindled skepticism from the 1980s on), because if "meaning" is understood in the rigid sense he insists on, it is indeed the case that nothing remotely resembling it exists in reality. The problem with Quine (and with Davidson too, and with Chomsky, for that matter) is that these observations never for a moment lead him to reconsider his notion of meaning. And that sort of conceptual rigor mortis is in fact an affliction that has smitten almost the entire field of philosophy of language (at least its Analytic branch). You see philosophers of language being willing to ascribe absolutely ridiculous interpretations to people's utterances because their model says so. If anything, the willingness to ditch the model in the face of clear contrary evidence is denounced as fallacious.
@@whycantiremainanonymous8091 Quine and Skinner were close friends at Harvard, yet surely, surely you're not suggesting being friends they must have had the same beliefs and views? Equally, Davidson was a close friend of Quine's, maybe even a disciple, but his understanding of Quine was, by some accounts (see Robert Sinclair's article on Davidson's "Third Dogma", for instance) inadequate, in spite of which Davidson held himself, emboldened by his friendship, as a leading authority on Quine "...because if "meaning" is understood in the rigid sense he insists on, it is indeed the case that nothing remotely resembling it exists in reality"----I agree, in the sense that it is in accepting from the outset certain standards of what constitutes understanding something----in this case the understanding the word 'meaning'----, standards of evidence, standards that invoke behaviourist language at the exclusion of any intensional notions, that makes it almost inevitable that they don't find anything resembling meaning in reality. To be sure, the most effective criticism, by my lights, of Quine should focus on the core assumptions that motivate his thesis of indeterminacy. I believe those "core assumptions" to be: his behaviourism, his extensionalism and its rejection of intensions and intentionality, and his holism. Interestingly, each of these assumptions are the result of Quine's dialogue with Carnap (which tells you a lot about the extent Carnap's influence upon Quine). "The problem with Quine (and with Davidson too, and with Chomsky, for that matter) is that these observations never for a moment lead him to reconsider his notion of meaning", correct me if I'm wrong, but your tone in this sentence suggests to me that you believe that Quine, knowing the implications of his views for 'meaning', that is, knowing that meaning would likely be eliminated if his arguments are sound, should see this as reason for questioning his entire argument; that is, since this is a conclusion unpalatable to man's traditional conception of the world, he should reconsider his thesis.... but this is simply the fallacy of appeal to consequences. If an argument might be true we don't rescind it on the basis that it might have consequences we don't like. Then again, I shouldn't ascribe a fallacy to my interlocutor, should I? Davidson would be most displeased!
@@michaeldalscais2538 On Quine-Skinner, and Davidson-Quine too, actually, what I heard from Putnam was not about personal friendship, but about expressed loyalty to theoretical positions. Obviously, this does not imply actual identity in their views, but does testify to motivation.
@@KaneB It’s exasperating. Especially when you writes a paragraph! I thought I had it fixed: I ticked a box saying allow all comments. But alas…. Maybe it was the link? I provided a link to Putnam’s old blog “Sardonic Comment”.
Okay I know you hate these, but I cannot restrain myself. Is there a fact of the matter about the exact meaning of Quine's "indeterminacy of meaning"? #selfrefutation
Thanks for the response. So why take him more seriously than any other guy who make generalizations like: "All sentences are false and / or meaningless"? I'm genuinely curious, not trolling.
@@mjuky First, I take the people who make those kinds of statements very seriously. I'm interested in radical skepticism, including meaning skepticism, and related positions such as trivialism and alethic nihilism. In my view it would be high praise to treat Quine as a fellow traveler to that crowd. Second, it's not clear how radical Quine's conclusion is. Quine doesn't say that all sentences are false or meaningless. He thinks that sentences are meaningful; it's just that their meaning is indeterminate. Does this commit us to some sort of global skepticism? Well, Quine himself certainly wasn't a skeptic. Perhaps all that follows from his thesis is that some traditional ways of thinking about meaning, concepts, and language have to be revised.
@@KaneB Thanks again! I was curious myself because I've taken Wittgenstein's earlier work about the senslesness of philosophical claims seriously, although I agree with many commentators that its self refuting. But it still seems to me that Quine's argument is self-refuting. He proposes that all meaning is indeterminate, including his own propositions. How could you ever agree or disagree with a position that has no determinate meaning? Surely, you cannot agree with his EXACT argument. On the other hand, all measures taken to avoid self-refutation reduce the radicality that you mentioned in the beginning of the video.
The question I'd have: if the two 'translations' entail the exact same observations such as to be indistinguishable even in theory, in what way can we say they actually are two different translations rather than just a single one? In general scientific inquiry the most common approach when you have two theories that make identical predictions is to treat them as statements of the same underlying truth, not to reject the entire idea of scientific inquiry. I'm struggling to see why meaning in language would be any different.
The two translations are in the same language. We can read them and just see that they contradict each other. English speakers can see that "rabbit" and "undetached rabbit-part" refer to different things. So an English translation of L that takes "gavagai" to refer to rabbits is a different translation from one that takes "gavagai" to refer to undetached rabbit-parts. (On the other hand, why are we so sure that "rabbit" and "undetached rabbit-part" refer to different things? Indeterminacy of meaning holds for English as well...)
@@KaneB Typical Quine-endorsed test for co-referentiality would be intersubstitutability salva veritate. Try the terms on sentence like “Jane owns exactly one rabbit” and “undetached rabbit part” fails.
@@PeteMandik Okay but is there a determinate fact of the matter whether two expressions are intersubstitutable salva veritate? It seems like there will be alternative translation manuals on the basis of which we will give different answers here. (Of course, a translation manual on which "rabbit" and "undetached rabbit part" are intersubstitutable salva veritate won't perfectly preserve the stimulus meanings of those terms, but I take it that no translation ever perfectly preserves stimulus meanings.)
@@KaneB No, there's no such determinate fact of the matter. Quine's interlocutor in the dialectic, who I thought you were temporarily adopting the persona of, thinks there is such a determinate fact. And that interlocutor says well, 'maybe the "udetached" manual is just a notational variant of the "rabbit" manual' to try to undercut Quine's attack on there being determinate reference facts. So if the interlocutor is conceding indeterminacy, our man Willard Van gets to do a victory dance.
Quine can be very frustrating in his apparent reluctance to accept the nature of language as a social construct designed as a tool for communication as opposed to a physical phenomenon to be explored. This seems to be way he refuses to accept that the meaning of language might be in people's minds instead of being discoverable by observation. Unfortunately Quine is just wrong about that. The meaning of language is in the minds of the speakers and the listeners. When Alice says sentence X to Bob, X can have two meanings: the meaning that Alice intended when she said X and the meaning that Bob interpreted when he heard X. Observing their behavior may give us clues to help us guess these meanings, but the actual meanings are in their minds. As a physicalist Quine should be content to accept that even people's minds will somehow reduce to the physical, and therefore we can in principle observe the meaning of language by observing the physics that underlie the minds that contain the meaning, such as with some sort of brain scanning machine. The brain scanning machine should be able to reveal everything within a person's mind, including the meaning of "gavagai" as this person understands it. If the machine says that "gavagai" means rabbit in this person's mind, then that settles it. Unfortunately, because Quine rejects that meanings are within people's minds, he is left mystified as to where to find meanings. Having rejected the truth, Quine is forced to conclude that there is no fact of the matter.
Not trying to ruffle your feathers, but say we have two scientists and two watercolor artists behind those scientists, in sequence. We run this machine through a cycle and the scientists read the data to the artists so they may prepare two pieces for a press release. Are there five meanings in total then? If we say a fact is something that may be acknowledged then Quines is correct as each meaning is kept to their own and even with a brain scanning machine the meaning of the scan will be in each their own.
@@KaneB : The idea of the brain scanning machine is that we would exploit the reduction of a mind to a physical process to read a person's mind by studying that person's brain. By looking at her brain and understanding its mechanisms, we see her thoughts. Of course Quine would deny that the meanings of words are in minds, but if the meanings of words were in the minds of people, then under physicalism we should in principle be able to directly observe those meanings by looking at people's brains. It would be like tracing the circuits of an electronic device and recognizing that a particular LED will turn on under particular circumstances. If any part of that would be properly called "translation" then I'm not sure which part that would be.
@@dostoyevsky1222 : But if physicalism is true, then the mind must be reducible to the physical. That means every thought we ever have is accessible through examining the physical structure of our brains. The only thing that prevents us from accessing meanings is our lack of technical ability to observe brain states. If it were not for all the practical obstacles like the skull and the vast complexity of the neurons, then the thoughts in a person's head would be every bit as accessible as the workings of an automobile engine. Sometimes we might wonder if the color that I call "red" looks the same to me as it does to you. If we had a perfect brain scanner, then we would be able to answer that question with objective measurements. We could trace the signals from the eyes and through the neurons and find where in the brain the signals becomes a sensation of color and we could technically describe every aspect of that color's appearance within a person's subjective experience. And then we would know exactly what the word "red" means.
Hasn’t this straight up been falsified by the development of large language models which have managed to understand meaning directly from text absent human guidance
quine failed to realise hes simply and obviously wrong about everything he says. ridiculous! how is this guy taken seriously?! obviously we can moorean shift to state that there are determinate meanings 🤦♂️🤦♂️ this guy man
