Talk:Analytic–synthetic distinction

To be added
1Z (talk) 20:53, 7 September 2011 (UTC)
 * Significance.
 * Kripke's aposteriori necessity (? Necessary aposteriori)
 * Katz
 * The Co-extensiveness thesis and Scott Soames.

Merge
This entry was created to merge analytic proposition and synthetic proposition. The two articles were pretty much duplicates of each other. - Jaymay 19:36, 11 August 2006 (UTC)

Clean up
I think this entry could use less on the details of the projects of Kant and the logical positivists and more on the analytic/synthetic distinction itself. For example, why do we need to have info on how Kant wanted to use the distinction in connection with the rest of his philosophy (e.g., how it relates to his project of explaining synthetic apriori knowledge)? There should definitely be more info on Quine's critique. I added a bit about it, but it should be expanded and cleaned up. There should also, I think, be a section on how previous philosophers thought that what was analytic was a priori and necessary, but that this has been widely rejected, primarily since Kripke. The entry on a priori has some info about this. - Jaymay 19:41, 11 August 2006 (UTC)

Kant
SelvoNT 21:22, 20 June 2007 (UTC) I feel that the emphasis on Kant is a biased interpretation on the distinction. Kant was preceded by both Leibniz and Hume respectively, and I'm sure there are others that I am not aware of. I feel that this needs to be redressed as well as some of the more Kantian-specific applications need to moved elsewhere (To Kant's page seems logical). Also, I do not think I am capable of altering this article personally but that someone should alter it. SelvoNT] June 19, 2007 [[User:SelvoNT|SelvoNT 21:22, 20 June 2007 (UTC)

Yes, does not Hume's famous "commit it to the flames" injunction anticipate the distinction? (It can be found on the net.) It is what I would do to much, at least, of Kant. Philosophers love him because he provides so much work for them. He just could not take Hume's skepticism. Seadowns (talk) 15:58, 22 April 2018 (UTC)


 * I agree that there may be too much detail and emphasis on Kant. But, I don't think it's appropriate to call it a bias.  Kant did invent the terms and popularized the distinction in philosophy.  The conceptual distinction was surely around before him (that is, people thought about the difference between something's being true in virtue of meaning alone), but it's hard not to emphasis Kant a bit since the actual terms originated with him. (Note: Of course I don't mean that Kant invented the words, rather that Kant invented their use for this distinction.) -- Jaymay 05:09, 12 July 2007 (UTC)

Who wrote this? Kant says that 7+5 = 12 is a synthetic statement, as is all maths. CPR pp. 15-16. This isn't even touched upon here in this section. I agree that the structure of this page is right off.

Von Mises
someone added a screed in which Von Mises is supposed to have solved the whole problem. I have tried to clean it up a bit, but I am not sure what the point is supposed to be. The possibility of an evolutionary explanation of apriori capabilities is not new, anyway. The first person to notice it was Charles Darwin.1Z 01:17, 29 January 2007 (UTC)


 * The Von Mises stuff is completely out of place. It was originally posted January 28, 2007 by a non-registered user (I.P. = 200.63.213.2).  Any major edit like this that is obviously not something that is standard on the analytic-synthetic distinction should be first discussed on the entry's talk page (i.e., here).  I'm going to take it out (although I'll leave the references as further reading).  If someone wants to put something about this Von Mises guy, it should (if included at all) first be discussed on the talk page, be minimal in length (currently it takes up about the same amount of space as the Kant stuff!), and be written in more of a prose form (not a bunch of bullet points). -- Jaymay 05:03, 12 July 2007 (UTC)

I was just reading Mises's book "Human Action" and can say the following. I don't think he tried to "solve the whole problem": he mentions the evolutionary apriori capabilities to support his own theories. The important connection with Mises is that he considers the entire field of economics to be entirely apriori. I'm not enough of a philosopher to tell whether he actually means it's a synthetic apriori or not. Chapter II of Human Action, "The Epistemological Problems of the Sciences of Human Action" is the relevant reference. Section 3, "The A Priori and Reality" especially. In fact I think this gives some flavor of what he means: "Aprioristic reasoning is purely conceptual and deductive. It cannot produce anything else but tautologies and analytic judgments...Hence, according to a popular objection, it cannot add anything to our knowledge...Action and reason are congeneric and homogeneous..." etc.

Distinction #1: LOGIC;  Distinction #2: Epistemology;    Here's the 2-by-2 Matrix...
This is how this article should be structured...

There are two distinctions to be made:

Distinction #1 -- BASED ON LOGIC

Distinction #2 -- BASED ON EPISTEMOLOGY (theory of how we can know)

DISTINCTION #1 -- LOGIC: ANALYTIC -vs- SYNTHETIC  Statements

A statement is ANALYTIC by this definition: the truth value (assigning TRUE or FALSE to the statement) can be done purely by knowing the meanings of the words. Thus, knowing the meanings of the words is sufficient to assign T or F. Nothing else is necessary.

A statment is SYNTHETIC by this definition: the statement is NOT ANALYTIC.

DISTINCTION #2 -- EPISTEMOLOGY: A PRIORI -vs- A POSTERIORI

A statement is A PRIORI by this definition: the truth value (assigning TRUE or FALSE to the statement) can be done WITHOUT NEEDING TO GO INTO THE WORLD AND OBSERVE.

A statment is A POSTERIORI by this definition: the statement is NOT A PRIORI. One requires going into the world and observe as a necessary condition to assign truth value.

THE MATRIX

With two, two-way distinctions, there are 2 times 2, or four, possibly existing beasts: ANAYLYTIC - A PRIORI, SYNTHETIC - A POSTERIORI, ANALYTIC - A POSTERIORI, SYNTHETIC - A PRIORI. Let's now see which of the four possibilities exists (or can exist).

ANALYTIC - A PRIORI -- If any beast such as an analytic statement exists, then the ANALYTICA - A PRIORI exists. ANALYTIC entails A PRIORI. That is, if a statement is ANALYTIC, then it MUST be A PRIORI as well. Why? ANALYTIC says meanings alone SUFFICE; nothing else is necessary. A PRIORI says it is NOT necessary to go into the world to give the truth value. Tautologies are analytic -- and therefore analytic - a priori.

ANALYTIC - A POSTERIORI -- Such a beast cannot exist because this categorization entails a logical contradiction. Like a four-sided triangle. Well not quite. The logical contradiction involves the modals of necessity and sufficiency. ANALYTIC says A is sufficient (where A = meanings of words). Saying A is sufficient implies that B is NOT necessary, C is NOT necessary, ...  A is sufficient implies an infinity of things that is not necessary. A POSTERIORI says it IS necessary to go into world and observe. That is a contradiction with "words' meanings" are sufficient.

SYNTHETIC - A POSTERIORI -- Such a beast not only exists, but predominates common discourse. My new Audi is Dolphin Gray. Synthetic says meanings alone do NOT suffice; something ELSE IS necessary. Saying "going into the world and observing" is necessary implies no logical contadiction.

SYNTHETIC - A PRIORI -- This was Kant's "categorical imperative" in the moral realm. Kant also believed that all of mathematics' statements were synthetic a priori. But first, let's see if the beast can exist without contradiction: SYNTHETIC = [A is NOT SUFFICIENT];  A PRIORI = [B is NOT Necessary]. Well, no logical contradiction. Wire cutters are insufficient to fix the ___ on the car. A screw driver is not necessary to fix the ___ on the car. No contradiction entails with the structure [A is not sufficient] AND [B is not necessary]. In the case of the car, maybe I don't need a screw driver and maybe wirecutters alone won't get the job done -- perhaps adding a soldering iron will help. But back to Kant. The moral imperative: "one should keep one's promises" or "one should not kill people" -- synthetic a priori statements. Too thorny are the issues in meta-ethics, but at least your pre-theoretical intuition tells you that (1) no, these are not true statements by the meanings of the words alone; but also, (2) neither do I feel I better go out and observe the world and experiment before I decide.

Let's leave the moral realm. Kant said of "2+3=5" that when he contemplated the ideas of '2' '3' or even the idea of '2+3' he in no way found that the idea was identical to the idea of '5'. Thus he claimed synthetic, not analytic. Yet, at the same time, Kant did not find he had to observe the world first before he could assign TRUE to '2+3=5' -- thus a priori. And thus, my friends, the synthetic a priori. The real mystery is "OK, no LOGICAL contradiction, but just how the blank do we know?" -- the meanings of the words are insufficient to know whether the statement is true or not AND it is entirely un-neccessary to go into the world and observe to know the truth of the statement, so just how the heck DO we know? At this point, things become very speculative. But now you understand the SYNTHETIC A PRIORI and Kant even!

acknowledgement: This structuring is from the teaching of logician Garrel Pottinger  -- a professor I had as an undergrad (1977-1981) at Carnegie-Mellon University. 199.196.144.11 20:58, 21 March 2007 (UTC)


 * Sensible and helpful as this sounds, it seems to me to miss the point as to what the article should be about - you say what you consider correct, the article describes how various philospophers have defined and used these concepts. I.e. you would undermine WP:NPOV. Moreover, I feel that it is more important to provide a wide range of information than a lot of explanation. PJTraill 22:11, 6 July 2007 (UTC)


 * This sort of matrix is fairly standard, especially in undergrad classes. But, there is so much to each distinction (analytic-synthetic and apriori-aposteriori).  That's why there are two separate articles with some stuff about the relation between the two.  We should not collapse them into one article.  What would it be titled?  Certainly not "Distinction #1:  LOGIC; Distinction #2: Epistemology".  Keep in mind that they are NOT the same distinction just in two different fields; rather, they are two different but related distinctions. -- Jaymay 04:45, 12 July 2007 (UTC)


 * As a non-philosopher coming to this article for some enlightenment, I became endarkened :-). In particular, this four-way distinction mentioned in the article and above has me confused.  It sounds at first reading like analytic = a priori and synthetic = a posteriori.  Obviously my understanding is a mis-, and I would like to know why.  Perhaps the "7+5=12" example is supposed to be synthetic and a priori, and the "tables exist" example is supposed to be a priori and analytic.  But if so, I would like an explanation; and if not, what are examples of these two cells in the matrix, and why do they fit there?  I gather there is another article about this somewhere, and a link to that would be fine; else please discuss it in this article.Mcswell (talk) 13:11, 31 October 2008 (UTC)

Two problems
Firstly, this page reads as though it was written by someone with a lot of knowledge of Kant, but not much of the rest of philosophy. The concept of analytic sentences originates in Leibniz and is simply copied by Kant directly. Kant made a lot of use of the idea, but he did not originate it and did not claim to have done so.

Secondly, the section on Two-Dimensionalism is irrelevant and has clearly been placed here by someone to promote their own favourite view. It has to do with the alignment of a priori knowledge with necessity and a posteriori knowledge with contingency. But it is only indirectly to do with the analytic/synthetic distinction, which is a division in semantics. Thus the whole section should either be rewritten so that it will be useful to someone coming here looking for information on the semantic distinction or it should be removed. I would prefer to see it removed. Anyone else?

Eluard (talk) 23:58, 3 March 2012 (UTC)


 * Agreed on both points. A brief explication of the Leibniz–Kant relation is very much worth adding (concept of pure reason & its relation to metaphysics, and truth claims generally—perhaps it is worth even linking to the principle of sufficient reason here as well).


 * Further: Hume needs to be addressed, as does at least via brief mention the concepts of cause and effect, i.e. causality—concept of the consequent (effect), the ground (cause), and containment (that the former follows from the latter). The level of depth given here is open, but if 2-Dimensionalism and logical positivism get their lengthy sections then at the very, very least this entry deserves a better introduction as to how the analytic and synthetic distinction relates in Kant to what he was actually proceeding from (Leibniz’s explications) and responding to (namely, scientific determinism, Humean skepticism (+ the religious claim that we’re all morally free, but that’s not so important here). Such an introduction will aid in placing these concepts into context (historically, philosophically, …). 60.242.104.119 (talk) 18:36, 25 September 2012 (UTC)


 * (Previous commenter again.) This, by the way, is one of my pet peeves in the description of Kant’s philosophy. Kant himself is quite muddled and even at times self-contradictory in his writing. How much this may have to do with the quality and capability of translations to translate from the original German I’m not sure (I am fluent in both, but have not read him extensively in German). I’d argue, Kant—as much as anyone else—is deserving of context. Taking a section—whether on the analytic–synthetic, or on anything else—out of context, however appealing it might be for the sake of pedagogy is, I feel, largely reductive to the point where important details are missed leaving me incapable of tracing the lineage and genealogy of thought in its criticism, agreement, forks, and so forth. — Preceding unsigned comment added by 60.242.104.119 (talk) 18:48, 25 September 2012 (UTC)

DanLanglois (talk) 08:05, 2 February 2014 (UTC)


 * I agree about removing, simply removing, the section on Two-Dimensionalism, and I've added some more extensive comments of my own to the bottom of this page (where I suggested removing this section before having skimmed all the previous comments).

A priority
Reading the section titled Analytic–synthetic distinction, I noticed a paragraph that has some confusing words. For illustration, I've emboldened these in the following quote:

While the first four sections of Quine's paper concern analyticity, the last two concern a priority. Putnam considers the argument in the two last sections as independent of the first four, and at the same time as Putnam criticizes Quine, he also emphasizes his historical importance as the first top rank philosopher to both reject the notion of apriority and sketch a methodology without it.

The first term looks as if it's referring to priority ("A thing that is regarded as more important than another."), but I believe it is talking about an association with a priori. The second one is similar, but without the space. I'm not sure what's the best way to fix this, as I'm nowhere near knowledgeable in Philosophy, but I figured if I brought it up, you fine folks could find a better way to present this information. I usually italicize a priori (a priority?), but I'm not sure if that's acceptable according to the MoS, or if it even helps the problem at all. Perhaps removing the space from the first term (so it looks like the second) will be a good enough solution? Thank you for your consideration! -- MisterDub (talk | contribs) 00:03, 5 December 2012 (UTC)


 * Until a better solution is found, I've changed the second term to read the same as the first. -- MisterDub (talk | contribs) 15:21, 5 December 2012 (UTC)

WIkilinks and hits
This edit removed the interwiki links to Analytic–synthetic distinction. Since that time the average number of hits on this page have dropped from about 250 hits/day to 50 hits/day. That seems to suggest that the transfer of these links to Wikidata is not bringing activity to this page, or possibly that Grok no longer counts these interwiki hits. Which is it? Brews ohare (talk) 15:47, 19 May 2013 (UTC)

Frege's Critisicm of Kant is not unanimously accepted
The line which asserts the analytically of mathematics does not reflect an unanimous epistemological opinion:

"Thanks to Frege's logical semantics, particularly his concept of analyticity, arithmetic truths like "7+5=12" are no longer synthetic a priori but analytical a priori truths in Carnap's extended sense of "analytic". Hence logical empiricists are not subject to Kant's criticism of Hume for throwing out mathematics along with metaphysics"[2]

Frege's logicism and rejection of Kant's synthetic, has many lines of criticism. I think the article needs contributions which stand against Frege's critcism of Kant. — Preceding unsigned comment added by 72.238.115.40 (talk) 20:02, 24 December 2013 (UTC)
 * It's also quite a shame that the article lists Frege under logical positivism, to which he wasnt related, although they used some of his ideas. He also did not deny the existence of synthetic propositions a priori (he did, in fact, acknowledged geometry as synthetic a priori), he only critizised Kant for restricting the analytic propositions too much. I'ld say that this part of the article needs a rework, and that Frege needs his own section. --178.1.3.228 (talk) 16:34, 17 March 2014 (UTC)

Tries to do too much
I think this article might begin, with a section on the rather informal, or 'intuitive', distinction, such that seems unproblematic and easy to understand, if possible. That is, begin with the *appearance* of the analytic. 'All ophthalmologists are eye doctors' can be known to be true just by knowing the meanings of the constituent words, right? Well, that's 'analytic' judgment. Pediatricians are doctors for children, yes? Are grandfathers parents of parents? Is sauntering a kind of movement? Is food, stuff that nourishes living things? Is pain a mental state? Yes, and this is 'analytic' judgment. Santa Claus has a beard. It's implausible to suppose that there's *nothing* people are getting at in these judgments.

That's how *I* would be begin..but, the article starts by describing the analytic–synthetic distinction as 'a conceptual distinction, used primarily in philosophy to distinguish propositions (in particular, statements that are affirmative subject-predicate judgments) into two types..'

Here, I read that the distinction in question is a 'conceptual' distinction. Actually, I'm not enamored of that, as I incline to reflect on what other kinds of distinctions there are? Like, ontological distinction, terminological distinction, vacuous distinction, etc. What is 'conceptual' distinction contrasted with, here? In truth, I've never liked the phrase 'conceptual distinction'. Compare: 'conceptually cogent distinction, but empirically important as well'. What is a 'conceptual distinction'? I do realize that this 'analytic/synthetic distinction' page is listed in the 'conceptual distinctions' category, on Wikipedia. There are two subcategories here, 'dichotomies', and 'difference'. In the case of 'difference', or, 'Category:Difference', there is, specifically, 'distinction (philosophy)'. Where I read, that 'Distinction, the fundamental philosophical abstraction, involves the recognition of difference.' That is to say that a 'distinction (philosophy)' is a *kind of* 'conceptual distinction'..? Now, we can unravel that in our leisure time, but I don't hesitate to criticize the description of the 'analytic/synthetic distinction' as a conceptual distinction. I'd suggest starting the article out by referring to this distinction as 'the distinction'. Now, it's been called a logical distinction, and also it's been called a grammatical distinction, and at least these terms mean anything to me.

Next, the page needs to prioritize. What is its ideal length? It tries to explain Kant, and several comments to this talk page concern whether this is the most relevant thing. I do feel that Kant is relevant, it's 'his' distinction, he made it popular, in more or less the usage that is still current. I see the points about Leibnitz had some such distinction, but he didn't coin the jargon 'analytic/synthetic', in this usage --although a distinction between analytic and synthetic methods is often made in geometry. Now, I say that I find Kant relevant, but if you want to explain Kant, it's not enough to try and fail. The page needs to succeed at what it tries to do, even if this means lowering our sights. Naturally, I have my criticisms of the discussion of Kant, but let's stick w/what are our priorities. I think the page can give some stuff about Kant, actually, and also certainly some stuff about Quine's famous paper disputing the analytic/synthetic distinction. But when we get into Quine, we are getting into 'analytic' philosophy, by which I mean 'anglo-american analytic' philosophy, in which there is lots of philosophy of language. The sort of obsessions that go back to Frege, major figure in analytic philosophy. And, I have this point, that here, well, I'll start over, I started by saying that:

"The article starts by describing the analytic–synthetic distinction as 'a conceptual distinction, used primarily in philosophy to distinguish propositions (in particular, statements that are affirmative subject-predicate judgments) into two types..'"

And, my point is that the term 'propositions' is a heavy heavy term in analytic philosophy. Frege commenced this mainstream traditional obsession. A proposition, this is a thing that is both very important in analytic philosophy, and also isn't really the focus previously, in philosophy. It's like 'propositions' didn't exist, even if the term was part of the language. Here, I read about Kant, who predates analytic philosophy of course, that he:

'..defines "analytic proposition" and "synthetic proposition" as follows:

analytic proposition: a proposition whose predicate concept is contained in its subject concept

synthetic proposition: a proposition whose predicate concept is not contained in its subject concept..'

Now, Kant never defined 'analytic proposition' or 'synthetic proposition'. He didn't use this jargon. None of this is Kant quotes, and it's not very trustworthy as a paraphrase, and I will again emphasize that the whole 'proposition' trip is extremely important in analytic philosophy, but Kant is innocent of the entire subject, he's talking about something else. Perhaps I seem to enjoy quibbling, but indeed, Kant never used the term 'predicate concept', or 'subject concept' (while, again, not also having ever used the terms 'synthetic proposition' or 'analytic proposition').

And, as it were, he didn't have the 'concept' of a proposition, at all, --that is, it's not that simple and convenient for people who have had some exposure only to analytic philosophy, when they try to go read some Kant. One wants to 'cotton on', and this non-Kantian jargon isn't a big help.

Now, here are some actual relevant Kant quotes:

"If I am to go beyond the concept A in order to cognize another concept B as combined with it, I rely on something that makes the synthesis possible:..'

Now, this is about the analytic/synthetic distinction, but w/great care Kant does not simply assert that 'experience' provides the thing that makes the 'synthesis' possible, in the case of synthetic judgment (this, although, as it turns out, he says that judgements of 'experience' are wholly synthetic). Rather, he relies on 'something' that makes the synthesis possible. And, if synthetic judgment is, like, synthesis, then what about analytic judgment? Well, '..analytic principles are indeed exceedingly important and needed, but only for attaining that distinctness in concepts which is required for a secure and extensive synthesis that, as such, will actually be a new acquisition..'

He's talking about a 'new acquisition' of cognition, that is. Then, analytic judgment is attaining distinctness in concepts. That's it? Yes, and so the analytic/synthetic distinction is based on logic, while analytic judgment itself is based entirely on logic, and is *what is* based on logic. Suppose, that what is going on in your mind is not utter primordial chaos, and this is how you like it, you like to have an organized mind, furnished with organized concepts. You do some housekeeping, you use logic to do it. You say that a bachelor is a married man. This is analytic judgment. W/out analytic judgment you are a raving lunatic. We cannot take sanity for granted, and this is based on analytic judgment. Coming to know anything new, learning, changing one's mind, is all synthetic judgment. Changing one's mind is in fact a logical contradiction. You can't therefore use analytic judgment to change your mind. Kant talks, as well, of ampliatory versus explicatory. That is, by an analytic judgement is meant a judgement in which the predicate is merely an explication of what is already contained in the subject. A synthetic judgement is a judgement in which the predicate is an amplification of the subject -- a clear addition to the content of the concept.

Anyways, next, I read about how the logical positivists took interest in Kant and his distinction, and revised the Kantian definition. There is this: 'This expanded idea of analyticity was able to show that all Kant's examples of arithmetical and geometrical truths are analytical a priori truths and not synthetic a priori truths.'

Okay, in passing, I note that 'analytical a priori truths' is not the same jargon as 'analytical a priori propositions', and neither one of these terms is from Kant. And, I feel that there is too much here, about the logical positivists, because really, we only need enough to be able to get into Quine, Quine is what is famous and important, when the matter is the analytic/synthetic distinction. The two-dimensionalism, 'Water is H2O', can be cut out, and does not at all dovetail nicely into the next section, "Quine's criticisms". Whether I have opinions about this section, I think it's important, so let's backtrack to Kant.

Kant does not say 'analytic proposition' he says 'analytic judgment'. Same thing? Close enough for gov't work? Well, I don't expect it to seem obviously plausible when I say no, but would it kill us to include some more verbatim Kant, here? Such as:

'Analytical judgements (affirmative) are therefore those in which the connection of the predicate with the subject is cogitated through identity;..'

And here is the rest of that sentence:

'..those in which this connection is cogitated without identity, are called synthetical judgements.'

And, Kant's explication has been 'improved' to such a degree in this article, that the page has no mention anywhere of 'identity' at all. Now, speaking of identity, when we later get into Quine, here, there is this:

'To summarize Quine's argument, the notion of an analytic proposition requires a notion of synonymy, but establishing synonymy inevitably leads to matters of fact - synthetic propositions. Thus, there is no non-circular (and so no tenable) way to ground the notion of analytic propositions.'

And (of course) there is some relationship between the 'synonymy' that is mentioned here, in summarizing Quine's argument, and the 'identity' on which Kant relies.

I haven't contributed much to Wiki, and I have not even considered modifying this article myself. I'm free w/my criticisms, but I now realize that actually writing a good article on this would be lots of work.

There can be, I think, a section on Kant, and a section on Frege (or general pre-Quinean analytic philosophy), and how their attitudes contrasted. Frege wanted the distinction to perform crucial epistemological work, where, for Kant, this distinction is not epistemological ('epistemological' distinction, a distinction based on epistemology, or, specifically, not logic; now, I don't know how it could be controversial to assert that for Kant the distinction is based on logic, or perhaps can be found in both logic and grammar, as in 'grammatical distinction'). It's not easy to wade into Frege, and again, we commence a virtual obsession, here, in the world of analytic philosophy, with so-called 'propositions'. The idea that the distinction can perform epistemological work was applied in particular to science and mathematics. And, with Quine, we get his criticisms (essentially of this 'epistemological work' idea, more than of the distinction itself as it is offered in Kant), and several generations of philosophers then despair of the analytic-synthetic distinction. They're also caught up in despairing of the category of a priori knowledge entirely. This is Quine's real target, he hates him some a priori knowledge.

And, there has been a wide variety of responses to Quine's attack, but we can put a pin in that. Chomsky is an of course important figure, and has commented on the analytic/synthetic distinction, so that could be included, but I don't want to *add* to the page, I think it's already long.

Now, what about Quine's attack. Well, in analytic philosophy, post Frege, “the analytic/synthetic distinction” refers to a distinction between two kinds of truth. Synthetic truths are true both because of what they mean and because of the way the world is, whereas analytic truths are true in virtue of meaning alone. What happens when we say 'true in virtue of meaning alone', is that the existence of analytic 'truths' becomes controversial. This is because something being 'true' has to have its epistemological justification, especially if you are an empiricist. We observe the world, the empirical world of experience, to decide what is 'true'. And, analytic judgement does not employ observing the world, analytic judgement is always 'a priori'. The relevant Kant quote here, though he's not labeled an empiricist generally, is 'All knowledge comes from experience'.

And, Quine 1951 is by far the most widely read paper objecting to the analytic/synthetic distinction. Here is his position. Suppose, as an empiricist, and this paper is 'Two dogmas of Empiricism', after all, Quine is an empiricist, so suppose as an empiricist that there are no a priori truths about the world. Then, if analytic judgements are a priori truths about the world, then there are no analytic judgments. There is no such thing as an analytic judgment. It's important here, though, to note that for Quine, the assertion is actually that there are not analytic 'propositions'. What's the difference between a proposition and a judgment. This is, informally, pretty much the same thing, more/less, isn't it? In brief, no, because every 'proposition' is taken, by Quine, in particular, to be sensitive to empirical experience.

Now, the page offers Quine's arguments but I'll offer an alternative summary. The first argument, I will call Quine's circularity argument. To make this argument, the first move that Quine makes is to say never mind about logical truths, so-called. This is, for example, P=P. Also, no unmarried man is married. For Quine, a 'not', or an '=', is a logical term. And, logically true propositions only depend on logical terms, for their 'truth'. They are, that is, truth no matter how you re-interpret the other words. In this fashion, Quine actually simply sets aside the whole category of Kantian analytic judgements. --never mind those. Then, besides logical truths, there is also what else, that we call (contra Kant) an 'analytic proposition'? 'No bachelor is married' is, for Quine, a case of synonymy. It's like saying 'no unmarried man is married', which is a logical truth. But 'no bachelor is married' is not, itself, a logical truth, unless you believe that employing synonymy here is unproblematic. Can you replace words with their synonyms, are there no pitfalls here? This doesn't change the meaning? Well, this is where I cash in my emphasis on the difference between 'judgments', as Kant calls them, and 'propositions', in the usage of analytic philosophy..

Suppose that I assert, says Quine, that A is synonymous with B if one is defined as the other. Then, how do I know that A is defined as B? I can make a judgement, but we don't have a judgement anymore, we're not Kantians and all we have is a proposition. Such as, 'no bachelor is married'. So, to know that A is defined as B, I mean if I'm just reading this sentence, and wondering what it means (not making a judgement for myself), I might look it up in a dictionary. In this fashion, I may learn that lexicographers believe that A and B are considered synonymous. Well, how do they know, do they look it up in the dictionary? What if the dictionary is not correct? So, this is circular. Now, that's no problem for Kant, because he's just worried about 'explication', or, attaining distinctness in concepts. Which is only a matter of logical truth. You can not know *anything*, all of your notions can be false, and you still can work w/'logical truth' all day long, doing, as I call it, 'housekeeping' on your totally inadequate concepts. Maybe you think Socrates is a man and maybe you don't, maybe you think all men are mortal and maybe you don't, you can do logic w/it either way. Which, Quine has already stipulated.

However, next Quine suggests that somebody might point out how when we look up a word in the dictionary, we get its definition, not merely an assertion that the word is a synonym. What about the definition of the word, then, that is, A and B are synonymous if A = B is true by definition. But, with a definition, we create a new meaning. And, this would be adequate for a definition of logical truth, but Quine (unlike Kant) is looking for something more than explication. He's looking, that is, for extensions of terms. This requires getting into Quine's notions, a bit, he's a logician.

I'll give it a shot. For starters, if I say that 'necessarily all and only Fs are Gs', then this would be a statement that captures synonymy. Suppose, though, that maybe it is a fact, that all and only creatures with hearts are creatures with kidneys. And you believe, shall we say, that fish are creatures with hearts. And, this is true. But you also believe, that fish are creatures with kidneys. This, is false. Then, it is not the case that necessarily all and only creatures with hearts are creatures with kidneys. Here, saying that 'necessarily' etc., is using analytic judgement. And, using analytic judgement preserves meaning. But, meaning is sensitive to empirical experience. By 'meaning', here, I mean, we're in the world of analytic philosophy, where there is a great preoccupation with 'propositions'. The propositions do a great deal, though it's kind of funny what happens if you ask for a formal definition of what they are. Well, they have words in them, always, I think. Propositions are not sentences, it's not an accident that we don't simply talk about sentences. Propositions have lots of jobs to do. Are propositions statements, then? Well, a 'proposition', mostly tends to have to include a relation to experience, if it is to have 'semantic content'. And the truth of a proposition tends to be supposed to have to be confirmed, or learned, by means of experience. And if you have a belief in a 'proposition', then this belief mostly is supposed to need to be justified in a particular way. It must be supported by experiential evidence. This is my breezy summary of highpoints (I have favored the fallacies) of what I will call classical semantic tradition. The most important thing about it, is that whatever it may be worth, one cannot simply gaze out from deep in the thicket of analytic philosophy's preoccupations, with its intension and extension, its meaning and reference, its non-Kantian notion of 'apriority', and the rational criterion thereof, and hope to understand Kant's spirit easily. It is perhaps not all about Kant anyways. But, the analytic/synthetic distinction is viable in Kant, and it's worth pointing out that it's his. Subsequent 'improvements' broke it.

Anyways, there are, then, for Quine, no 'analytically necessary' truths. Again, Quine simply makes an exception for truths about the kind of necessity that flows from the nature of concepts, which are Kant's analytic judgments. Quine's point is that there is not a priori knowledge of non-logical truths, or 'substantive' truths. Though I emphasize that this is not really a point against Kant, Quine wasn't talking to Kantians, and so this made a big stir. I started with best intentions, but I realize that I could explicate Quine more patiently. This is probably already too long. It's a dilemma..and probably, me arguing that analyticity of analytic elementary logical truths is perfectly intelligible at the end of the day, and that 'Kant was right', is what is called original research.

Anyways, we need to achieve a low common denominator, when it's this technical subject. I think some discussion needs to happen, before *I* take a swipe at this..?DanLanglois (talk) 07:58, 2 February 2014 (UTC)


 * I don't want to sound mean, but I suggest you take a short break, re-read the text you have just written, and then ask yourself "am I the man to write a clear, concise, layman's description of the analytic-synthetic distinction?". -Ashley Pomeroy (talk) 20:54, 26 August 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Analytic–synthetic distinction. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20130228133303/http://faculty.unlv.edu:80/jwood/wm/Quine.pdf to http://faculty.unlv.edu/jwood/wm/Quine.pdf

When you have finished reviewing my changes, please set the checked parameter below to true or failed to let others know (documentation at ).

Cheers.— InternetArchiveBot  (Report bug) 09:02, 12 October 2016 (UTC)

Odd edit adding reference to Sloman (1965) while summary says it removes part of one.
The edit https://en.wikipedia.org/w/index.php?title=Analytic%E2%80%93synthetic_distinction&curid=6395956&diff=871109468&oldid=869858714 of 2018-11-29T00:59:50 by claims in its summary "Removed spurious portion of the last reference", but in fact adds a reference to

a relevant article on the genuine OUP site. Is anything odd going on? PJTraill (talk) 12:25, 29 November 2018 (UTC)

Remove or displace Peikoff's Criticism
Peikoff's argument is bad. In addition, the publication where this was published is a biased Rand Association publication. — Preceding unsigned comment added by Sorbonneparis (talk • contribs) 18:22, 13 March 2019 (UTC) Peikoff wrote Analytic-synthetic distinction; both the article and the author are noteworthy of inclusion on Wikipedia making the criticism worthy of inclusion, too.  Orville1974  (talk) 23:20, 23 May 2019 (UTC)

It's absolutely absurd to include the section on Piekoff. He's a complete nobody in philosophy, and the argument is ridiculously poor. The quoted section alone is seriously misleading. — Preceding unsigned comment added by 109.144.219.118 (talk) 19:07, 4 November 2019 (UTC)

Recent changes to section: Kant's version and the a priori / a posteriori distinction
I reverted one recent change by User:NONIS STEFANO for what seems to me a clear enough reason (given in edit summary, not to mention that it hijacked a reference). Incidentally, I also found it unclear how that reference (A Proof that Synthetic a priori Propositions exist) supports either the description of the contents or the dependence of the possibility of metaphysics on the decidability of synthetic a priori propositions; that might, however, be because I could only see the first half-page at JSTOR.

Moreover, I also wonder about that user's previous edit, which also strikes me as possibly questionable. They replaced:
 * This question is exceedingly important, Kant maintains, because all important metaphysical knowledge is of synthetic a priori propositions.

with
 * This question is exceedingly important, Kant maintains, because all scientific knowledge (for him Newtonian physics and mathematics) is made up of synthetic a priori propositions.

I took a look at Critique of Pure Reason, but there was too much for a quick conclusion: could someone who knows the subject please deal with this?

— PJTraill (talk) 17:17, 11 July 2019 (UTC)


 * I have access to and read the entirety of "Proof that Synthetic a priori Propositions exist". In it, C. H. Langford claims that the following are examples of synthetic a priori propositions: "a cube has twelve edges", "anything that was red must necessarily by colored", and "If anything is orange in color, then it is intermediate in color between red and yellow." There is no explicit mention of "metaphysics" (nor "science"), though the arguments could be considered to be about both, in some sense. Solomon Ucko (talk) 00:45, 19 September 2023 (UTC) P.S. It also never mentions Kant; the only other person is mentions is philosopher G. E. Moore.

All bachelors are unmarried?
A problem with classifying this as the prototypical analytic sentence is that one must already have a pre-existing notion of a bachelor; otherwise, one must consult a lexicon, which defeats the definition of a priori. An intellectual challenge would be to derive the concept of a bachelor independently, without the use of "cheats". Heavyarms2025 (talk) 17:54, 27 December 2020 (UTC)

Wiki Education assignment: PHIL 390 Contemporary Philosophy
— Assignment last updated by RudyCarnap (talk) 10:45, 19 April 2023 (UTC)