Word and Object

by Willard Van Orman Quine

Language consists of dispositions, socially instilled, to respond observably to socially observable stimuli. Such is the point of view from which a noted philosopher and logician examines the notion of meaning and the linguistic mechanisms of objective reference. In the course of the discussion, Professor Quine pinpoints the difficulties involved in translation, brings to light the anomalies and conflicts implicit in our language's referential apparatus, clarifies semantic problems connected with the imputation of existence, and marshals reasons for admitting or repudiating each of various categories of supposed objects. He argues that the notion of a language-transcendent "sentence-meaning" must on the whole be rejected; meaningful studies in the semantics of reference can only be directed toward substantially the same language in which they are conducted.

Genres: Philosophy, Language, Nonfiction, Linguistics, Metaphysics, Science, Logic

309 pages, Paperback

First published January 1, 1960

About the author

"Willard Van Orman Quine (June 25, 1908 Akron, Ohio – December 25, 2000) (known to intimates as "Van"), was an American analytic philosopher and logician. From 1930 until his death 70 years later, Quine was affiliated in some way with Harvard University, first as a student, then as a professor of philosophy and a teacher of mathematics, and finally as an emeritus elder statesman who published or revised seven books in retirement. He filled the Edgar Pierce Chair of Philosophy at Harvard, 1956-78. Quine falls squarely into the analytic philosophy tradition while also being the main proponent of the view that philosophy is not conceptual analysis. His major writings include "Two Dogmas of Empiricism", which attacked the distinction between analytic and synthetic propositions and advocated a form of semantic holism, and Word and Object which further developed these positions and introduced the notorious indeterminacy of translation thesis." - http://en.wikipedia.org/wiki/Willard_...

Reviews

[Scot · Rating 4 out of 5]

Got Quine to sign my copy. He wasn't too thrilled to be asked, but he signed it nevertheless.

[Neil LoBracco · Rating 4 out of 5]

Rigorous is a good term to describe 'Word and Object'. This is one of those books that makes you wonder if the author was beaten as a child. Essentially all of his claims are unassailable, yet at the end you're not sure if it's at all right - but it will force you to think about the consequences of what he says. The book starts off with pure linguistics, moves to philosophy, and ends with semantics (OK, the whole book is semantics). If you make it to the later chapters, you might go pages without having any clue what Quine is talking about, only to see some truism you can grab onto that lets you think you're participating.

[Joshua Stein · Rating 4 out of 5]

Word and Object is a masterpiece in modern philosophy and, as I worked through it, I was struck by how much more contemporary discussions in the domain make when I have Quine for context. The book sets up a number of the problems that Quine poses for historical views in philosophy of language, like problems with vagueness and translation, and then Quine offers an alternative account of how to develop a (highly naturalized) philosophical programme. The book has passages that are a bit dated, where good responses to Quine have been developed, but that is pretty normal, and the fact that those responses are available, and have implications for the views they were meant to defend is very important and very useful.

Many people are ultimately just not able to accept Quine's positions, or even his critiques of many arguments, but the reasons that we find ourselves unable to accept portions of Quine unacceptable are important too. Are we committed to the idea that translation between languages and idiolects has to be possible? Why? In finding our reasons for rejecting parts of Quine, we learn a lot about the role of our positions in our larger views about philosophy.

The book is excellently written, though a bit convoluted in some places and a bit excessively technical in others. The majority of the book is lucid, insightful and engaging, even to those outside of philosophy. Those interested in linguistics and the role of philosophy in interacting with semantics and approaches to translation will get a lot out of Quine, and hopefully understand some of the eccentricities of modern philosophical approaches to semantics and their relation to logic, features of the philosophical landscape that may seem odd given their stark contrast to conventional views in linguistics advanced by Chomsky, and even a number of philosophers (like Montague) later.

I recommend the book to those interested in philosophy of language and metaphysics, though the angle on language is especially useful. It should probably be used as an introductory text for those getting into the discipline, despite the philosophical jargon that rides heavily on the later sections.

[Alina · Rating 3 out of 5]

It seems that Quine embarks on 4 different projects, believing that the completion of each is necessary for the other. I do not think so. The consequence of this structure is that the book feel quite disjointed, and many sections seem superfluous. In chapter 1, Quine introduces his behaviorist theory of language: linguistic items are conditioned responses to certain patterns of sensory stimulation. He proposes hypothetical principles of language acquisition that would make this theory possible. It amounts to an empirically unsupported outline of causal account of language (project #1); there is little 'philosophy' here.

Chapter 2 is the most rich and interesting chapter of the book. Here Quine argues for his influential theses of the indeterminacy of reference and the indeterminacy of meaning, which follow from his analysis of 'radical translation' (project #2). He presents a thought experiment in which a linguist tries to translate a totally alien language; her only data from which to construct a translation manual are observables, of the alien speaker's behaviors and environmental context. Quine argues that from this restricted set of data, the linguist can identify 'stimulus meanings' of the alien linguistic items. The stimulus meaning of a linguistic item is the conjunction of the set of stimuli, when paired with the linguistic item, provokes the speaker to indicate affirmation; and the negative set of stimuli, when paired, provokes the speaker to indicate dissent.

In order to construct a translation manual, the linguist would have to propose 'analytic hypotheses', which stipulate that an alien term and an English term share the same stimulus meaning. These hypotheses would serve as a foundation from which further translations could be inferred. Quine argues that it is possible for different linguists to end up with different translation manuals, which each equally account for the alien language, but which each are inconsistent with the other. This is because stimulus meanings are by nature referentially indeterminate (e.g., the stimuli of a rabbit could equally provoke the phrases "That's a rabbit!", "That's a particular instantiating rabbithood!", or "That's a collection of rabbit time-slices!"). Hence, reference is indeterminate, and so is meaning.

I think I need to read up more on Quine's argument because I see this flaw, which is too obvious to be a reasonable rebuttal: the conclusions of indeterminacy depend on the assumption that stimulus meaning exhausts the possible kinds of meaning that our linguistic items can have. This is obviously false. From our first-hand perspective, we can intend for our words to have very specific, determinate referents. Perhaps the indeterminacy of reference and meaning could be theoretically true regarding translation between languages, but it does not get us indeterminacy regarding the items of any given language itself -- it seems that Quine makes the latter, stronger claim. Maybe he doesn't though, and I've confused him as doing so because he refutes the traditional concepts of reference and meaning elsewhere (If anyone wants to help me out here, please do comment!).

In chapter 3, Quine presents an account of language acquisition in greater detail than in chapter 2 (continuation of project #1). I found this chapter superfluous and largely disconnected from the others; only some general points made here were relevant for the later chapters. In chapter 4, Quine examines how the references of certain types of linguistic items can be vague and ambiguous.

This seems like a prelude to chapter 5, in which Quine takes on his second project of "regimentation": to create a logical language that can simplify and consolidate the propositions of scientific theories (project #3). For him, this is crucial to expose the truth conditions of scientific propositions and to aid the progression of science. This is a very lengthy and tedious chapter; Quine goes through one type of syntactic structure and grammatical role after the other, evaluates competing philosophical treatments of their reduction into logical grammar, and promotes his own treatment. In only a few places does he rely on his causal account of language acquisition, to which he has dedicated previous chapters. Chapter 6 is a continuation of this project, with a particular focus on refuting the inclusion of intensional objects and propositional attitudes in logical grammar. In chapter 7, Quine switches to the task of defending a radical nominalism (project #4).

I would recommend potential readers to read chapter 1, patiently focus on chapter 2, and skim chapter 4. That is enough. Despite my disappointment with this book (which might be largely due to the intellectual-cultural difference of our time; Quine's naturalism does not seem radical from our modern eyes), I deeply appreciated some points, including his view of theoretical holism; his rejection of epistemological foundationalism; his attitude towards ordinary language; his criticism of conflating theoretical features with ontological features. I think if Quine took his own points more seriously, however, he would be more doubtful of, and even reject, the ontological and scientific claims that fill this book.

[path · Rating 4 out of 5]

“One frequently hears it urged that deep differences of language carry with them ultimate differences in the way one thinks, or looks upon the world” (77).

This was not a particularly accessible book, but it was, ultimately, pretty informative. The overarching goal of the book was to investigate how words reference reality, which matters to Quine because he sees the function of philosophy as scientific, a regimented approach to discovering what is true about the world by making sense of what we observe about it.

The challenge Quine addresses is how to understand the construction of true statements, when our only way to access that work is through the language that we use to communicate truth to others. For this reason, the vague, opaque, inconsistent, incomplete, and situated ways that language references reality is highly problematic and radically untranslatable.

Quine’s solution, expressed through his “Rabbit/Gavagai” thought experiment, where a linguist is charged with coming up with a manual for translating a totally unknown language, is to recognize that while words do designate a reality, the sentences through which those words are meaningfully assembled are conventional, universal configurations of words that represent the beliefs of discourse communities and the logic by which they build concepts from sensory inputs and relate those concepts to one another. This dissolved barrier between analytic and synthetic approaches to truth and the influence of a discourse community’s “web of belief” is one of Quine’s major contributions.

Quine focuses his energy on enumerating conventional classes of sentences by which discourse communities represent truth in accordance with a prevailing web of beliefs but also in a way that is ultimately responsible to the extensional reality that all words must reference. This presentation of sentence classes is a way of classifying systematic ways that communities express truth. By understanding the sentence classes, the underlying extensional truth claims can be clarified by conversion to a regimenting language (1st order symbolic logic, particularly) that substitutes situated, intensional meaning with verifiable observations. Only in this way can one arrive at truth — direct translation from ordinary language is impossible because it is riddled with perspectival views of the world, infinitely differentiated by time, perspective, bias, belief, etc — wholly unsuitable for seeking truth.

I’m inclined to believe that regimentation is a viable approach to the normalization of propositional knowledge, but I’m not sure I share Quine’s preference for converting sense-supported (i.e., intensional) propositions into symbolic logic because it is the best or most reliable way to determine truth. This just seems like a silly restriction. How do we account for the truth of saying that “here is the border of my state and your state” or “she is the chair of the committee” or “this note represents $50”? These states are consequential in the way that truth propositions are, but their truth does not ground out in sensory data about “borders” or “roles” or “value.” They ground out in how people treat those truths. Do they act like there is a border there or that she is the chair of the committee or that the piece of paper is worth $50? Perhaps that is also allowed in Quine’s formulation, as behavioral observation, but then that suggests that he ought not to disallow intensional meaning from his calculus of truth.

Implications for Translation

I also wonder if Quine is taking a too narrow understanding of translation as a method of converting truth of one language into the truth of another. There are elements of experience and belief that influence our actions and how we treat things as true or not. Perhaps all of this just falls into the penumbra of indeterminacy, where you could have two people responding to the world in incompatible ways (e.g., following Quine’s thought experiment): when two people in a group say “There’s a rabbit” and point at something, do they both mean the same thing or is one person referencing the rabbit as dinner and the other referencing the rabbit as an example of “Rabbithood”)? Both are true, and for the business of identification, it should be enough to get along. But if the goal is to either 1) establish that the entity identified is suitable for dinner or 2) establish that it represents an example of a species, then the proposition is insufficient and needs to be connected to other statements in order to come up with the class of the sentence and the appropriate responses.

There is a depth of meaning in English that is not totally recoverable, even to English speakers because the utterance is made by a person at a moment in time. We just default to the closest norm and hope for the best. Or we default to a classifying language (like logic) that gets to the empirical core of meaning. This is sort of an idealized understanding of meaning where all possible true perceptions eventually funnel into one. But we cannot describe that overarching meaning.

All sentences of beliefs, where predicated objects exist as signs can, with enough patience a detail be filled in with all of the relevant sensory conditions that make it an observable statement (183). Quine is after a pragmatic level of truth where the aim is to develop what is sufficient. As he sees it, this pragmatic outcome can be the only reasonable goal of translation: simplification to a common language that respects the functions of sentences across versions and defaults the nearest extensional norm with enough fidelity to maintain instrumental uptake.

But for things that are not empirical — love, piety, respect, admiration, hate — covering the experience relies on approximation. There is an impossibility of translation here because we can never be accurate in capturing the sense of a statement. And here is where there are implications in Quine’s argument for other kinds of translation, like literary translation, for example. Assuming that complete accuracy was even possible, would we even want or expect that? Are we assuming that writers speak to some ideal version of true experience that transcends their perspective? If so, the solution is not having the expectation of getting this truth from one author but rather from multiple authors, across language divides. Or we admit that this level of ideal understanding is not possible and accept that a functional and pragmatic equivalent is sufficient.

[Joe W. · Rating 5 out of 5]

A philosopher's philosopher, not for the meek. A cure for the common metaphysicians among us.

[Dionysius the Areopagite · Rating 5 out of 5]

Fuel.

[Zo · Rating 4 out of 5]

I don't think I am familiar enough with formal logic and the various debates Quine is engaged in to evaluate his claims with any confidence, but I nonetheless found this a fascinating read. The opening section on linguistics/translation is great. It is one of the more detailed and thought-out explications of a line of reasoning I am inclined to agree with, and provides a great resource for thinking about the vagaries of language. I would love to read criticisms of his conclusions there, but I think even with Chomsky considerations at play they seem solid. The middle/late sections which dive into symbolic/formal logic were challenging for me, and I am hesitant to latch onto the parts I do understand without always grasping his preceding arguments. That being said, they are damn thought-provoking conclusions! I am unsure exactly how far they go beyond Wittgenstein. How much is the difference between them more about their moral stance on the value of careful conceptual clarification surrounding abstraction? I would need to understand both better to say for sure, but I am inclined to think they don't have that much substantive disagreement. Quine's picture of philosophy seems similarly narrow as Wittgenstein's, and I would be curious to hear his arguments for *why* philosophers have the ability to offer conceptual clarification in-field practitioners do not. (Is the answer just that philosophy is the field where research can be devoted to such questions?) There are moments where his skepticism about truth/science seem radical, and I am unsure exactly what his commitments are. Does he believe "theory" improvement in logic/math/science can go on indefinitely? Or will we reach points of diminishing returns? And does that suggest we are approaching a more metaphysical "truth," or simply the limits of what the human cognitive apparatus can handle? Further, why is it that humans are able to understand what they do about science/math/logic? (This is where Chomsky/cog sci, like with Wittgenstein, could perhaps play a larger role). Anyway, I'm sure there is lots of room for argument about the various stances Quine takes on specific issues of ontology/logic, and I would love to learn more about them. I know there is also disagreement on Quine's larger positions, and I also want to understand those arguments better, because Quine seems awfully persuasive!
Edit: Upon partial second read (along with some of Quine's other papers) I felt like my understanding was much better, though still far from complete. It seems to me that Quine relies too much on behaviorist assumptions to get to his "radical" translation thesis, and the ontological relativity he develops is thereby perhaps overstated. But I still felt broadly in agreement with him, and think I have a greater appreciation for how carefully he militates evidence against systematic theorizing about meaning (I'm less sure now about what exactly he is trying to say about reference). See review of secondary source for more info.

[Ammon · Rating 3 out of 5]

I know that this is supposed to be a classic of analytic philosophy, but two things about the book really irked me. 1) The practice of so-called "radical translation," while interesting takes as a model of science a sort of laboratory condition that not only does not match actual linguistic anthropology (something Quine would no doubt grant) but cannot in principle match any anthropological practice and 2) even if one leaves aside the tremendously huge behavioristic assumptions of his account of the ontogenesis of language, one still has to grapple with the fact that Quine makes a huge leap from giving a description of how he takes it that language use has to arise to a normative account of the logical structure of language. It's this sort of sleight of hand that allows him to give his theory of language a stronger grounding than his own nominalism should allow. And analytics accuse us continentals of rhetorical cheap tricks...

[Audrey Saltarelli · Rating 4 out of 5]

Very good just very dense

[Larry · Rating 3 out of 5]

Churchland's preface sucks. Follesdal's touching. The whole thing is a bit boring to read from cover to cover, but it's the kind of stuff you work on, rather than dive in.

The very beginning makes an interesting point against (presumably) early Carnap's reduction of physics to a language of sense-data: a memory is not a memory of an impression, but of a conceptualization, and so reference to objects is actually what keeps the stream of consciousness together. Then there are illuminating points about metaphors and explanation in science (if I explain light with waves, I'm kind of moving amidst implicit definitions; cf. §4). And then there's the ultrafamous pages about radical translation. A few things to remember: no amount of pointing is going to help the linguist bootstrap his way out of indeterminacy, and any string of sounds or letters can 'stand for' a sentence, or a word. All of which amounts to: the experience of meaning, presented in its truth thanks to the circumstances of radical translation, is theoretical. (Here, Boghossian's criticism fails: B argues Quine is trying to have his cake and eat it with this way of arguing for indeterminacy (through confirmation holism), because we don't get a clear sense of how something can both be indeterminate, yet require nothing less than the entire web of belief for its meaning: precisely because it is indeterminate, nothing less than the maximum amount of evidence will do, because, for want of an intension, we need an approximation.)

In ch 3 I think it's too bad there's no confrontation with Chomsky's view. The all-out behaviorism, framed in quality space talk, about language learning, sounds implausible today.

It gets better about referential opacity and singular terms. Quine's strategy is to find a way of knowing in what contexts ill-advised quantification yields nonsense, and he pins it down to intentional and modal contexts (intensions in general, therefore understood as what the opaque terms, understood in bulk, *name*). One way of doing without singular terms, which introduce the nonsense when substituted for in said contexts, is to manoeuvre them on the right-hand side of the sign '=', or turn them into general terms (the only 'singular terms' being the variable on the other side). As I had pointed out in my MA's thesis, even Socrates turns into a general term, viz. one of which only one thing, Socrates, is it; and Socrates is thereby recast as a singular term (see §39, end). That's not a problem, because Socrates qua singular term has still been eliminated. The confusion can be cleared up if I quote Quine's definition of elimination: “By the very nature of the elimination technique, propositions stayed on as denizens of the universe alluded to in the 'everything' and 'something' of '(x)' and '∃x'; as values, in short, of the variables.” (§40, 176) The problem of quantification in modal contexts is solved if we thus eliminate singular terms; we do lose a sense of the modal distinction between contingency and necessity, or essence and attribute per se, but that's OK (better: that's part of the point).

[Allan Olley · Rating 5 out of 5]

This is an important and much referenced work in mid 20th century analytic philosophy. It presents Quine's arguments for the breakdown of traditional concepts of meaning, including philosophical concepts like the distinction between analytic and synthetic reasoning. Quine attempts to demostrate this inadequacy by giving a general reconstruction of language in more behaviouristic terms and pointing to the limitations of such an effort as limitations not just of language but of thought itself. Most famously he argues for the indeterminacy of translation that two competent translators might not translate the same word the same way making their translations equally adequate by altering other parts of the translation between languages. He continues the argument to say that there is no robust meaning in the intuitive sense behind words and language and instead a looser association exists outside of our basic formulation of physical law. Most if not all interpretation of others and one's own thought becomes a kind of reinterpretation. He also deflates empiricists notions that might take sense experience or sense data as an underlying structure since he argues much empirical reasoning is based in the gloss of stimulus by linguistic structures.

The style of the arguments broken up into dozens of subsections across 7 chapters, is lively and persuasive. At times it can be hard going as his logical reconstructions of language become intricate, but interesting ideas are broached and explained that depend less on the technical matter.

Ultimately I wonder if his constructions can avoid either falling into naive relativism or expanding back into something resembling the intuitive semantics and meaning that he rejects. Although I would probably need to reread and consider this book a time or two before this would be more than a vague doubt.

The ebook version of this I read (an epub borrowed from a library) did not play well with my ereader and so several lines that included logical symbols did not render. This further impaired my reading of the work. However it seemed relatively rare and I was reading mostly for fun so I did not reread those sections.

[T. Tokunaga · Rating 5 out of 5]

【Word and Object / W. V. O. Quine】

--Now there is also a decidedly difference domain that lends itself directly to radical translation: that of truth functions such as negation, logical conjunction, and alternation. (P57, Chapter 2, Section 13)

Or even usage of synonyms can be translations, according to Quine. This realistic (and positivist) theory, probably succeeded from Bertrand Russell and honed by Wittgenstein and Carnap, is showing its efficiency in his investigation on infantile language acquisition too (see Chapter 3, Section 19 and 20) - "how" before "why," context over truth (as a truth function, it's necessarily so. Even in Russell's later essays, he opted out to be dependent on its own contexts).

However, he was more accurate than his older fellows, because he put a certain effort in becoming a polyglot - which was shown in his "letting analytics' myth that 2+3 is not 3+2" by pointing out Russian has, for example, a drastically different identity system in grammar (P118, Section 24, Chapter 3). Also, the problem of subjunctive which would nestor anyone learning a romance language is anatomized in Chapter 5 meticulously (especially in P168, Section 35, but in the English "latent" subjunctive using "that"), so if you're struggling with language learning, this would be a great read as well.

A lot of discussions in this extensive study would be found rather dull - personally, it's full of basic realization reviving to me, whose first language is not English.

Let me share one more quote from this book, which would be effective in tackling "false information" in general.

We recall of indirect quotation that it has no general translation into quoted speech, though each specific true case has a quota ke actual utterance behind it (P222, Section 46, Chapter 6).

[Corbin · Rating 3 out of 5]

Quine is a sharp thinker who probes thoroughly into the consequences of various conceptual and formal commitments--it's very helpful and provocative. However, sometimes he'll point out a dilemma and presume that one side is much preferable to the other (usually to defer to vernacular usage or theoretical simplicity) but seemingly judging that simplicity by preservation of the status quo rather than on true parsimony. (Sometimes parsimony is counterintuitive, especially when trying to explicate such complicated issues, and Quine's endorsements of extensionalism and object-centric ontology disregard the ways that process-oriented metaphysics and intensional normativity can thin out theories by embracing the tacit pragmatics of reasoning over paradox-inducing semantic explication.) Still, by showing what's at stake when he presents the dilemmas, and by showing how formerly accepted answers don't actually resolve the underlying issues, Quine reveals deep structures of various frameworks and methodologies. And the issues raised by radical translation prompt us to see how interconnected logic, metaphysics, and subjectivity really are.

Not, however, for the uninitiated. He blows through the formal issues and refers to prior debates in philosophy of language that would be lost on someone who had never studied any early analytic philosophy. A decent familiarity with Frege, Russell, and Strawson (minimally) is probably prerequisite for appreciating what Quine is trying to do and why it matters.

[Richard · Rating 4 out of 5]

Everything is this book is *fire emoji*, except for Quine's argument against modality which is uncharacteristically awful. However, he is highly motivated to have some sort of argument, because trying to find a naturalistic explanation (in keeping with the spirit of the rest of the book) of what makes statements about necessity and possibility meaningful is very difficult.

This book is hard work, and unless you are already familiar with the contours of debate in philosophy of language a lot of it will seem very banal. The only Quine I read at uni was Two Dogmas of Empiricism, and I would have definitely benefited a lot from reading this at that time. But not one for the casual reader.

[Leonardo]

La insuficiencia del comportamiento para discriminar entre significados discriminables fue demostrada por Quine. Quine no consideraba que el argumento fuera un reductio ad absurdum de las descripciones conductistas del significado.

La mente Pág.124

[Daniel Solomon · Rating 3 out of 5]

I have mixed feelings about this book, which is seen as perhaps the most important book of the most important analytic philosopher of the 2nd half of the 20th century.

On one hand, I generally sympathise with Quine's naturalist view of philosophy, in which philosophy is basically a more theoretical branch of science (both deductive and empirical). I also sympathise with the general idea of regimented theory up to some point, trying to build a more formal and precise representation of scientific theories.

Unfortunately, Quine's attempt to naturalise philosophy and especially epistemology is founded on an outdated and superseded psychological behaviourism. Writing in 1960, Quine misses the boat on the cognitive revolution that was just starting and the emergence of modern cognitive science. Following Quine's own commitment to science being the ultimate source of knowledge about reality, one is better off skipping his attempts to explain the learning of language in favour of modern linguistics/cognitive science explanations that put much more emphasis on partly innate mental and language concepts. Quine's indeterminacy of translation from meager stimulus-response observations is usually reduced in most realistic situations precisely by extra mental and social information that goes beyond just observations of behaviour, but this seems to be omitted in Quine's account of language learning and translation.

His rejection of the analytic-synthetic distinction is very much based on this old fashioned behaviourism and is therefore unjustified. Subsequent treatments of this distinction like Putnam (1962) or Bhogossian (2018) are more balanced and provide a useful version of it. At any rate, under certain conditions of the kind used in developing formal languages of logic-mathematics one can develop precise definitions and rules, one can apply determinate rules of synonymy/identity of concepts and one can therefore derive propositions that rational readers would see as essentially necessarily true by definitions/rules. While it's possible for any proposition to be rejected under some conditions (sure, we could imagine a world where people decide to stop using classical logic and redo all science and thinking in e.g intuitionistic logic-though very unlikely, and linguistic meanings of words like bachelor may change), the notion of analytic propositions is at least a useful idealised approximation, distinguishing certain propositions from others that are much more dependent on empirical evidence. Once equipped with this notion, we can link many analytic sentences to things like innate universal grammar or language of thought concepts and stable combinations of neurons' firing in response to certain stimuli such as x + y = y + x. In short we can probably distinguish psychologically/cognitively analytic propositions (I think Chomsky and some of his followers in linguistics go in that direction). A recent survey of philosophers/philosophy students suggests that at least in the Anglo-Saxon Analytic philosophy discipline, around 2/3 of philosophers in the early 21st century think that the analytic/synthetic distinction is useful and meaningful. In this regard, Quine's position remains an influential but minority view.

Quine's formal language and ontology of regimented theory throws out a lot of useful concepts and restricts itself to first order logic and set theory, making it impractical for actual scientific modeling and theorising. I checked logic texts and summaries of philosophical analysis after Quine: fortunately, propositions, properties, propositional attitudes are all very much used and discussed in modern philosophical and logical discourse. Fortunately, because practical applied scientists and common sensical people use these notions all the time and find them extremely useful in building and discussing theories, models and representations of the world. Open any mathematics text or even rather formal social science article/text and you see regular implicit quantification over sets, functions and properties which Quine rejects. Try to state any reasonable theorem, and you find yourself de facto informally using the second order logic that Quine dislikes. Modern formal linguistics (e.g Montague semantics) and philosophy of science analyses (e.g Ramsey sentences) are full of properties and higher order logic. Similarly, while Quine rejects modal concepts such as possibility or necessity, modal logic has flourished post-Quine and has proved useful in fields such as computer science and AI. And state of the art mathematical logic investigates combinations such as second order modal logic.

I'm left with mixed feelings. I like Quine's positioning of philosophy as part of a continuum from more applied to more theoretical science, e.g merging epistemology into cognitive science, linguistic analysis into actual linguistics etc...But I reject what I perceive as a dogmatic empiricism, and excessively restricted language for logical analysis of science/regimented theory. It's as if Quine had the right ideas on the very general framework of analysis as science (and common sense thinking which Quine sees as a starting point for science), but then was wrong in his choice of scientific theories and excessively restricted language for doing science. Quine may have had some good ideas about philosophical naturalism, but he should have left the cognitive science and linguistics to the actual scientists.

[Griffin Deutsch · Rating 5 out of 5]

Only Quine could have written this book. By the end of the book I felt like we had only talked about language use and referential syntax, but somehow he manages to craft each part of his behaviorist theory of language into a sweeping argument touching on everything from underdetermination to scientific realism to analyticity. Also, it’s hard to not be an externalist after reading this book.

[Ragdoll Metaphysics · Rating 3 out of 5]

Extremely pretentious writing style I gave him some latitude when reading Mathematical Logic as he was trying to reconstruct the foundations of mathematics in the style of Russell and Whitehead who famously took like 300 pages to prove 1+1=2 or something like that but apparently verbosity is a constant in his work. I’m giving this 3 stars because I more or less agree with indeterminacy of translation I think his philosophy of science is kind of cringe to be honest. It’s interesting stuff nonetheless.

[Anna · Rating 4 out of 5]

Please don't read this without supervision. It was kind of cool though, and really interesting, even though I don't agree with his pragmatist assumptions. Hopefully, in some time I'll actually understand everything (or most of it, I mean).

[Crito · Rating 4 out of 5]

We love our flawed kings, don't we folks. We won't be hearing about the boring middle third, the lamentable association with behaviorism (which relies less on Skinner than is generally assumed), or how Quine absolutely would have rather just been writing essays instead.

[Serge · Rating 3 out of 5]

The interplay between theoretical physics and semantics was sometimes hard to follow.

[Roberto Loja · Rating 2 out of 5]

A poor autobiography of an otherwise brilliant man.

[Indigo Curnick · Rating 2 out of 5]

Philosophically interesting but Quine is a terrible writer

[Lemon Platt · Rating 5 out of 5]

Plato, Leibniz, Hegel, William James, Quine.

This > the Critique of Pure

[Tiffany Clark · Rating 1 out of 5]

Not what I thought…

Not what I thought I was getting, and very difficult to read. The philosophy did not mesh here. Too high academic for me.

[Abe Kazemzadeh · Rating 5 out of 5]

This book was quite simply amazing. Quine builds up what it means to learn language and communicate from a behavioralist perspective, yet still relates it to logic. It was not an easy book to read and it took me quite a while to finish, but I feel like in almost every chapter there is an "aha" moment where I realized something new about language, logic, and philosophy. Basically, the book starts from the premise that you go to a foreign land where you know nothing about the natives' language, so then how do you know what they mean when they say something. There is a similar topic how a child learn language as an infant. Quine takes these examples and builds up a behavior-based account of when a sentence is true, a categorization of how observational a sentence is, and relates this to both logic and subjectivity. There are a lot of issues that are brought up that might kind of seem overdone if the reader is not familiar with related research (mainly logic and the philosophy of language). Even though these issues seem overdone, it is actually useful to think about these things if you are dealing with computational linguistics or natural language processing because with computers you need to be exhaustive/thorough/rigorous about all the assumptions and specifications. One thing that I found interesting was that this book deals with very technical stuff without almost any equations. I think it attests to the supreme skill of of Quine's prose, but still the engineer in me would like some more equations (I won't deduct any stars for that though).

[Matt · Rating 4 out of 5]

Though one may find oneself disagreeing with many of Quine's positions, one cannot fail to appreciate the accomplishment here. Not only is Quine a clear and clever writer, he also provides a compelling, systematic, and beautiful understanding of understanding the relationship between the way we talk about things and the way things really are. Where this relationship is tenuous, Quine artfully maneuvers his way through the thicket.

[Christine Cordula Dantas · Rating 4 out of 5]

Exceedingly well-written as much as it is exceedingly difficult. Pre-requisites are the philosophical questions on language, meaning, proposition, analytics, epistemology and necessity. Not a book to be simply read, but returned and studied in detail. Without the pre-requisites and hard work, this book will pose a great challenge to the layman reader, as myself.

[Dave Peticolas · Rating 3 out of 5]

Quine analyzes the referential mechanisms of English. Along the way he introduces the notion of "radical translation", the attempt to translate a wholly unknown language into a known one, concluding that it is, in general, not possible.