Analytic vs . Synthetic , Distinction or Myth ? : Kant ’ s Kantribution , Quine ’ s Inquisition , Grice and Strawson ’ s Salvation

Philosophers have relied heavily on the distinction between analytic truths and synthetic ones for various philosophical pursuits. In this paper I explore Immanuel Kant’s explanation of the distinction, W.V.O. Quine’s qualms with it, and the attempt of H.P. Grice and Strawson at saving synonymy in order to salvage analyticity from doubts. I conclude that although valiant, the efforts put forth by Grice and Strawson fall short. I argue that this is so because they attack a weak interpretation of Quine’s contention. Maité Castillo Humboldt State University mgc188@humboldt.edu https://doi.org/10.7710/2155-4838.1175 Volume 9, Issue 1 Res Cogitans 2 | eP1175 Res Cogitans The following paper is an evaluative assessment of the distinction between analytic and synthetic judgments. To do this I have examined Quine’s “Two Dogmas of Empiricism”1 and Kant’s The Critique of Pure Reason2 to contrast the views that Quine and Kant had of this distinction. Whereas Kant was proud to make this distinction in an effort to revive the study of metaphysics, Quine eventually, after a long period of accepting the distinction, followed with a rejection of analyticity as it was used in metaphysics for establishing necessary truths; claiming that all explanations of analytic truths in that way are circular, “or something like a closed curve in space.”3 Quine asserts that ‘necessity’ in the case of analytic statements does not do the work that past philosophers, such as Kant, wanted it to. I have also explored the criticisms of Quine put forth by Grice and Strawson in “In Defense of a Dogma”4 to see if they restore the analytic and synthetic distinction. I have mainly been concerned with the philosophical problem of establishing the possibility of objective necessary truths through the lenses of Kant, Quine, Grice, and Strawson. I intend to defend Quine by showing where Grice and Strawson fall short in their endeavor. Let us begin with Kant. In The Critique of Pure Reason, Kant attempts to answer the question of how synthetic a priori knowledge is possible. I will begin by explaining the distinctions between ways of conceptualizing or making theoretical judgments of the world, according to Kant, then I will explain his view of metaphysics by way of the possibility of synthetic a priori judgments. Kant distinguishes between a priori judgments that occur unconnected of all outside experiences and pure a priori judgments that happen completely free of experience without anything empirical intermixed. To highlight this distinction he uses an example of a man whose house has a faulty foundation. Given a flaw in infrastructure one can expect without waiting to experience it that the house will fall in on itself, i.e., one can predict it a priori. Reasoning is done a priori by modus ponens to arrive at the conclusion: “If there’s a faulty foundation, then the house will collapse. I see that there is no foundation, therefore the house will collapse.” The sentences themselves that make up the premises are a posteriori. The a priori judgment is not pure since 1 W.V.O. Quine, “Two Dogmas of Empiricism”, The Philosophical Review, Vol. 60, No. 1 (Jan., 1951), pp. 20-43 2 Immanuel Kant, Critique of Pure Reason, Hackett Publishing Company, Inc., 1781


| eP1175
Res Cogitans The following paper is an evaluative assessment of the distinction between analytic and synthetic judgments. To do this I have examined Quine's "Two Dogmas of Empiricism" 1 and Kant's The Critique of Pure Reason 2 to contrast the views that Quine and Kant had of this distinction. Whereas Kant was proud to make this distinction in an effort to revive the study of metaphysics, Quine eventually, after a long period of accepting the distinction, followed with a rejection of analyticity as it was used in metaphysics for establishing necessary truths; claiming that all explanations of analytic truths in that way are circular, "or something like a closed curve in space." 3 Quine asserts that 'necessity' in the case of analytic statements does not do the work that past philosophers, such as Kant, wanted it to. I have also explored the criticisms of Quine put forth by Grice and Strawson in "In Defense of a Dogma" 4 to see if they restore the analytic and synthetic distinction. I have mainly been concerned with the philosophical problem of establishing the possibility of objective necessary truths through the lenses of Kant, Quine, Grice, and Strawson. I intend to defend Quine by showing where Grice and Strawson fall short in their endeavor. Let us begin with Kant.
In The Critique of Pure Reason, Kant attempts to answer the question of how synthetic a priori knowledge is possible. I will begin by explaining the distinctions between ways of conceptualizing or making theoretical judgments of the world, according to Kant, then I will explain his view of metaphysics by way of the possibility of synthetic a priori judgments.
Kant distinguishes between a priori judgments that occur unconnected of all outside experiences and pure a priori judgments that happen completely free of experience without anything empirical intermixed. To highlight this distinction he uses an example of a man whose house has a faulty foundation. Given a flaw in infrastructure one can expect without waiting to experience it that the house will fall in on itself, i.e., one can predict it a priori. Reasoning is done a priori by modus ponens to arrive at the conclusion: "If there's a faulty foundation, then the house will collapse. I see that there is no foundation, therefore the house will collapse." The sentences themselves that make up the premises are a posteriori. The a priori judgment is not pure since commons.pacificu.edu/rescogitans eP1175 | 3 prior to, one must have had to experience the concept that bodies have weight and therefore fall when their base is compromised. 5 This example shows a belief that is reached without immediate experience, but involves justification (of heavy bodies) that one can only arrive to by use of the senses. As a pure a priori judgment, Kant refers to the example that all effects have a cause.
In the endeavor of justifying our beliefs about the world we use assertions in different forms. Kant distinguishes between two kinds; the first I will discuss is the analytic. Analytic assertions are ones in which the predicate is contained in the concept of the subject. Kant writes, In all judgments in which the relation of a subject to the predicate is thought (if I only consider affirmative judgments, since the application to negative ones is easy) this relation is possible in two different ways. Either the predicate B belongs to the subject A as something that is (covertly) contained in this concept A; or B lies entirely outside the concept A, though to be sure it stands in connection with it. In the first case, I call the judgment analytic, in the second synthetic. 6 Kant also writes, For [analytic judgments] do not through the predicate add anything to the concept of the subject; rather, they only dissect the concept, breaking it up into components which had already been thought in it (although thought confusedly). 7 Moreover, these assertions would be contradictory when negated. Take for example, the assertion, "Every cat is feline," 'Cat' is the subject that includes the predicate idea of 'feline-ness', which I can know without referring to every cat in the world. If I were to assert, "Not all cats are felines," I would contradict the concept of "cat," since being a cat implies also being feline. The idea is that analytic judgments are ones of clarification, so they merely spell out something already contained in a concept. If I said I found a cat that is not a feline, you would reasonably say that what I found is not a cat because the two ideas are not so easily separable.

Res Cogitans
Synthetic judgments, on the other hand, give us new information about the world. This is the case because, unlike analytic statements, the predicate is not contained in the subject. These judgments combine information and explain more than just the definition of a concept. If they are true, they amplify our knowledge by giving us new information, as opposed to clarifying details that are found in the investigation of a subject or concept. They can amplify the knowledge of someone who knows the meanings of the words, because they determine some state of affairs to be the case and excludes other states of affairs from being the case. Analytic statements do not determine anything, and are not informative to someone who knows the meaning of the words. For these types, Kant offers an example of physical bodies being heavy; whereas 'bodies' are the subject and 'heavy' is the predicate. By definition (at least for Kant), bodies are not necessarily heavy, so to deny that they are would not lead to a contradiction of what it means to be a body; that is, extended in space: Metaphysics is not at all concerned merely to dissect concepts of things that we frame a priori, and thereby elucidate them analytically. Rather, in metaphysics we want to expand our a priori cognition. In order to do this, we must use principles which go beyond the given concept and which add to it something that was not contained in it; and, by means of such synthetic a priori judgments, we must presumably go so far beyond such concepts that even experience can no longer follow us [...] 8 Metaphysics is a matter of the possibility of synthetic a priori judgments because, first, metaphysics must tell us something new about the world hence, synthetic; second, we cannot know it through experience. Thus, we must know it a priori. So, we must have some non-trivial explanation of the world that amplifies truths that we can know without relying on sense-experience. Kant did not question the possibility of synthetic a priori judgments, because he believed that their possibility was already shown and contained in mathematics and all theoretical sciences of reason.
Since metaphysics is considered one of these sciences, it follows that those synthetic a priori judgments are contained in this study as well.
In his paper "Two Dogmas of Empiricism," Quine began with an explanation of what Kant meant by an analytic truth; that is, one of clarification, true as a consequence of their meanings, and separate from facts. Unlike Kant, Quine believed that there was no distinction to be made, or to be understood, between synthetic and analytic statements, because statements of the analytic sort could not be sufficiently defined. He claims that all attempts to adequately define analyticity are circular. He claims this because, as he explicates in the first sections of his paper, attempts to explain analyticity lead to the reliance on other ideas that are in just as much need of clarification as analyticity. The criteria are just as nebulous as what they seek to identify.
Quine makes a major distinction between meaning and naming, stating that one is not to be confused with the other. Quine uses the example, "the terms '9' and 'the number of planets' they name one and the same abstract entity but presumably must be regarded as unlike in meaning […]." 9 He goes on to distinguish between two kinds of analytic statements: ones that are logically true, and ones that can be turned into logical truths by the use of synonyms. a. An ophthalmologist is an eye-doctor.
b. A vixen is a female fox. c. A cat is a feline.
The first are considered logically true because they remain true under any replacement of 'X', 'man' and 'married'(here we can omit the reinterpretation of logical particles, or operators like, 'not,' 'if then,' 'all,' 'or,' 'no,' 'some,' etc.). Analytic statements of the second type might be formed into statements of the first by substituting 'bachelor' for 'unmarried man' (in 1a.). But this process is still unclear because it relies on synonymy, which is lacking in a "formal explanation" like analyticity. Statements of the second type are the ones Quine is concerned with because of this unclear notion of synonymy, and, by that means, analyticity, too. 10 He goes on to describe how definitions, though seemingly to console the worry about synonymy, actually rely upon and presuppose synonymy also. Instead of pointing out that 'unmarried man' is synonymous with 'bachelor', one might say that one is the definition of the other. However, as Quine points out, the only way one can be 6 | eP1175 Res Cogitans sure that one defines the other is if we rely on synonymy. This is so because even if we take the lexicographer's account of definitions, the lexicographer, as an empirical scientist, creates his definitions on the basis of a pre-existing relation of synonymy among words. This can be said to be true of anyone who attempts to define one word with another, as it is not just the lexicographer that creates definitions, but philosophers and scientists alike.
Quine points out at that there exists an extreme case of definition in which "[the definition] does not hark back to prior synonymies at all; namely, the explicitly conventional introduction of novel notations for purposes of sheer abbreviation." 11 A case of this form would be one such that the word to be defined is synonymous with the word defining it. 12 This kind of case would consist of me making up a word; let us say, "Slatiblatfist." Now I will ascribe the combination of "Zb3" to my new word "Slatiblatfist." Quine wants to say that there was no need to presuppose synonymy here because I have only shortened and represented "Slatiblatfist" by convention. For this reason Quine claims this as a "transparent case of synonymy created by definition". 13 More simply, this is a case in which synonymy arises from definition, and not viceversa. Better examples can be found in math or logic; we write π instead of 3.14... But it is just a convention to use Greek letters in math. This is also contingent since all the mathematicians could have a meeting and change it if there was a better, more conventional symbol to use instead.
Quine's criticism is not merely that the distinction is hopeless or unsatisfactorily explained. He goes a step further to say that the distinction is erroneous on part of the philosopher. The most prominent point for Grice and Strawson (hereafter GS) was to illustrate that Quine's attacks on the proper characterization of the analytic/ synthetic distinction do not adequately warrant a complete denial of it. The lack of clarification of the distinction, GS want to say, is true of other distinctions that we utilize as well, but this does not make the distinctions illusory. Just because there is not a formal explanation of analytic statements does not lay good enough ground for the conclusion that they are fake-notions.
This is not to say that the points Quine raises are not the beginning steps on the path to finding an adequate account of the analytic and like concepts. However, it is to 11 Ibid,26 12 Ibid, 26 commons.pacificu.edu/rescogitans eP1175 | 7 say that the points raised by Quine are not enough to prove that there is no such distinction. The distinction relies upon philosophical and ordinary usage. GS describe Quine's rejection as a "philosopher's paradox" in that the philosopher ignores an exploration into the usage of the notion of meaning the same, but instead the philosopher sets a standard by which to measure the clarity of the notion. Then, since the notion fails the measurement, the philosopher rejects the existence of the notion and proclaims it illusory. 14 GS explain that Quine's analytic circle is such that, there is a group of phrases or concepts in which the term "analytic" is a member along with "synonymous," "definition," "necessary," "self-contradictory," and "semantical rule," etc. 15 Of all of these members, if one should be adequately clarified, then the rest would be as well. GS note that Quine often refers to making "satisfactory sense" of expressions at this point. An exploration into what that would consist of could be of some use in our understanding of his circle. GS formulate what would constitute "satisfactory sense" and hence, the escape from the circle with two things: 1. It would involve providing an explanation that does not incorporate any expression belonging to the family circle.
2. It seems that the explanation provided must be of the same general character as those rejected explanations that do incorporate members of the family circle.
In other words, "It should take the form of a pretty strict definition but should not make use of any member of a group of interdefinable terms to which the expression belongs." 16 GS combat this predicament of a difficult undertaking in two ways; the first is to claim that this criterion is not a necessary condition for a concept to make sense. They show this by giving examples of words that, although we have yet to precisely clarify them, we still make sense and use of. Words such as "morally wrong," "blameworthy," "breach of moral rules," "true," "false," "statement," and "fact," to name a few. GS expressed that it would be absurd to claim that since these words have yet to be given a formal explanation, we should therefore give up on their providing any sense for us. They go on to say that just because an expression cannot be explained by Quine's standard does not mean it cannot be explained in less formal ways. The second way they combat Quine's standard is by giving an informal explanation of a member of the group, due to its complexity, for now we will say of it that GS break out of the "circle" by explaining a 'logical impossibility' without an appeal to other ideas in the family circle. 17 One of the key points of concern for GS is Quine's claim of the extreme definition form. GS assert that if we accept Quine's claim, pertaining to the case of 'transparent synonymy,' "[...] his position as a whole is incoherent." 18 To take their famous words on this, It is like the position of a man to whom we are trying to explain, say, the idea of one thing fitting into another thing, or two things fitting together, and who says: 'I can understand what it means to say that one thing fits into another, or that two things fit together, in the case where one was specially made to fit the other; but I cannot understand what it means to say this in any other case.' 19 In other words, GS question what correlation between concepts this process establishes, and why Quine thinks it is unfathomable to attribute the same (or similar) process, or correlation, to situations of synonymy that are not so transparent. It appears unclear to GS that such a case of transparency cannot be applied to other cases because synonymy by convention seems to require the notion of synonymy by usage. tion since we have long standing philosophical usage of them.
It is clear to see problems with the lack of clarity of the distinction so there is room to continue to strive for a formal explanation of analytic truths. Simply because we do not have the strict definition of analytic statements does not mean we should throw them by the wayside until we find one. We should continue to use those truths in our language and debates about them, in ways they have made sense for us in the past. Analytic truths, and thereby necessary truths, can still be made sense of even if we lack the exact sense of them.
Although GS have given theoretical explanations for the existence of the distinction they do not get any clearer on the distinction. They do not directly answer the point of Quine's contention. GS claim that the case of 'transparent synonymy,' taken seriously, shows Quine's project to be inconsistent. I argue here that they are misguided in this claim. Just because Quine shows one case of synonymy does not mean that the mysteries of the kind of synonymy he questions are solved. Quine could not deny synonymy, especially in such antiquated examples as "a bachelor is an unmarried man," and I do not think he wished to. In the case of 'transparency,' the introduction of notation is by convention and does not rely on meanings. Whereas in the case of a 'bachelor' being an 'unmarried male' the meanings of each term is important in their being synonymous. To show that there is a distinction is one thing; to show how it comes to be and it's nature is another. The analyticity, and thereby synonymy, that Quine argues against are the kind that attempt to assert universal necessary truths. The point is that saying a truth is universal and necessary would mean that it could be no other way. However, Quine claims that these too are contingent and in being so cannot be universal and necessary. So, although GS recognized and showed the existence of the distinction, they did not pursue inquiry into the nature of such a distinction.