The Idea of A Priori Revisited

Sanjit Chakraborty

Abstract: In this article I would like to discuss the concept of a priori mainly focusing on Kant’s Copernican revolution. How is metaphysics at all possible and how a priority takes place in Kantian metaphysics are the questions that I have addressed in the first part of my article. In this context, I have explained analytic, synthetic distinction from epistemological, metaphysical and semantical perspectives and I want to show how the concept of a priori and other associated notions are derived from this different perspective.

            I end my discussion with an account of some fundamental objections raised by Quine and Kripke against Kantian analysis. Quine mainly wants to deny the traditional analytic and synthetic distinction. Besides, Kripke is trying to show that it is not true that the traditional account of the intimate relation between a priori and necessary proposition must be true, i.e., it is mistaken that a priori proposition must be necessary, it may be contingent. I would like to analyse and explain a few obstacles of Quine and Kripke following Putnam’s (externalist approach) and Quassim Cassam’s (non-empirical approach on a priority) line of different arguments.

pdf  PDF Key words: APriori, Kant, Quine, Putnam, Necessity, Law cluster concept, Semantic

The view of philosophy, which I have discussed now I think in Kant’s terminology ‘a priori’, has relation to Kant’s Copernican revolution in epistemology. In Critique of Pure Reason, we may find as Kant claims that there is not a single metaphysical question unsolved in the Critique. The central question of Critique is — ‘how could metaphysics be at all possible?

            For Kant, if metaphysics is possible at all then such a metaphysical knowledge must be-

(i) A priori and
(ii) Necessarily true.

The classical equation of necessity and a priority is shown in the following theses:

(a) If it is necessary that P, then P can be known a priori. And
(b) If P can be known a priori, then it is necessary that P.

Metaphysical knowledge is no doubt a knowledge about the things in the world which is obtained a priori, i.e., independently of all experience. But Kant admits that all our knowledge begins with experience. Then we may put the question — how can one know anything in advance about the object, i.e. a priori to experiencing it?

            We should give an account of the mind knowing its object in such a way that metaphysical knowledge fulfils the above-mentioned conditions and such knowledge becomes possible. There are two alternative ways in which this relation between the object and the knowing mind can be construed —

First, All our knowledge must conform to objects and
Secondly, Objects must conform to knowledge.

Before Kant, the traditional opinion emphasized that our knowledge of an object is determined completely by the nature of objects. If we admit that our knowledge only mirrors reality, we must reject the possibility of necessity and a priority. Kant rejects the first view. To him the ultra-realistic view would not help us to explain how a priori knowledge about things becomes possible. Therefore, Kant assumes that the nature of an object is determined by the knowing mind. Some features of the objects are dependent on the mind and we can have a priori knowledge about them. Kant was really proposing a kind of revolution in our thought like Copernicus had in astronomy by presenting a bold hypothesis about the movement of planets. We can know a priori features of things that we ourselves contribute to them. Kant does not think that metaphysical knowledge is innate rather he thinks that the metaphysical knowledge analyses objects through a priori synthetic judgment.

            Now let me clarify the term a priori and the associated terms like a posteriori analytic, synthetic etc. from Kant’s point of view. Kant believes that ‘Necessity and strict universality are thus sure criteria of a priori knowledge and are inseparable from one another.’[1] He claims all the propositions in pure mathematics and natural science is a priori. If we want to know the nature of a priori, we should first turn to discuss some relevant contexts where one can talk of judgments. Normally the concept of a priori knowledge arises in the following sets of distinctions, viz. the epistemological distinction between a priori and a posteriori knowledge; the metaphysical distinction between necessity and contingency and the semantical distinction between analytic and synthetic judgment. In the following pages, I will discuss briefly these distinctions as applied to Kant’s philosophy and I shall point out some views of other philosophers that rest upon the above considerations.

The Epistemological Distinction

Kant makes a distinction between a priori and a posteriori knowledge. A posteriori or empirical knowledge must be derived from our experience. If we acquire a piece of knowledge through sense-experience, it would be a posteriori knowledge. ‘The man is tall’ is such a kind of knowledge. However, Kant thinks a priori knowledge is untainted by experience. If a judgment is, a priori, then it must be logically independent of all judgments, which describe the experience. According to Kant, as mentioned earlier, “Necessity and strict universality are the sure criteria of a priori knowledge”. The term ‘necessity’ and ‘universality’ bear close a resemblance; even so, they are not same. A judgment is necessary if and only if the opposite of it is inconceivable. We take an example ‘All men are mortal’. Here, we have to understand inconceivability not as some subjective incapability of a particular subject rather in principle it is impossible for anyone to think of the opposite. We can indeed say that the concept of the necessary is metaphysical (or, ontological), while that of the a priori is epistemological.

            We know that a judgment is universal when an exception of it is impossible. All men are rational is an instance of strict universal judgment as the idea of humanity and idea of irrationality conflict with each other’s.

            Now Kant says, ‘If we have a proposition which in being thought is thought as necessary, it is an a priori judgment; and if, besides it is not derived from any proposition except one which also has the validity of a necessary judgment, it is an absolutely a priori judgement.’[2] 

The Metaphysical Distinction

In metaphysical distinction, we may find a distinction between necessity and contingency propositions. We may find some propositions, which are bound to be true, and there are some other propositions, which are not so bound to be true. A proposition is necessarily true if and only if it is impossible for the proposition to be false. Leibniz tries to make the notion of necessary truth somewhat accessible by characterizing necessarily true propositions as propositions that are true in all possible worlds. Let us consider the case where we believe that ‘everything that has shape has size’ — here we find that it holds true always and in all possible worlds. Leibniz thinks, ‘When a truth is necessary, its reason can be found by analysis, resolving it into more simple ideas and truth until we come to those, which are primary.’[3] Philosophers use this term ‘Necessity’ in their technical discussion.

            The Philosopher has some intention to use necessary proposition as logical necessity and logical necessary propositions are a proper sub-class of the analytic propositions. To them a proposition is necessary if and only if it is analytic. This is a traditional opinion. Kant somehow supports this type of thinking. Many recent philosophers oppose this kind of view.[4] He gives us an example about it. The number that is the number of the planets, is greater than six. Here the number, which is the number of the planets, in fact indicates the number nine, and it is not coherent to think that nine is not greater than six.

            In general, we may find three kinds of necessity, i.e., the psychological, logical and synthetical necessity. By a psychological necessity, we mean a kind of causal necessity. Thus, Hume thinks that it is psychologically necessary for us to believe in the existence of external objects and Kant thinks that it is psychologically necessary for us to believe in the existence of external objects and Kant thinks that it is psychologically necessary for us to perceive objects as spatially located and related. These necessities are psychological in the sense that they instigate in the structure of our cognitive faculties.

            Logical necessary propositions are those that are logically true and opposite of it makes self-contradiction. For instance, P. ~ P would be false forever.

            The third kind of necessity that is called synthetic necessity claims that we can have a type of relational insight into real connection, which is essentially synthetic; this view clearly relies on a theory of intellectual intuition.

The Semantic Distinction

The distinction between an analytic proposition and a synthetic proposition is of crucial importance in Kant’s Philosophy. Both Leibniz and Hume escort a similar distinction between analytic and synthetic judgments. Leibniz distinguishes between two types of truth — truths of reason and truths of fact. Truths of reason are something, which can be true in all possible worlds. They are logically necessary and their negations involve a contradiction. Leibniz argues, ‘Hence, any truth which is not susceptible of analysis and cannot be demonstrated by reason, but receives its ultimate reason and certainty from the divine mind alone, is not a necessary truth. All the truths of this kind I call truths of fact. This is the root of contingency and so far as I know no one has hitherto explain it.’[5] Whereas, Hume tells us that ‘All the object of human reason or enquiry may naturally be divided into two kinds, to wit, Relations of Ideas, and Matters of Fact.’[6] All mathematical science like Algebra, Geometry, Arithmetic and also every affirmation which is intuitively certain are concerned with the first like three times five is equal to the half of thirty, expresses a relation between these numbers. We discover the truth of this kind of a proposition by mere operation of thought. Besides, matters of fact cannot be ascertained in the same manner. The contrary of every matters of facts is possible, because such a contrary never implies a contradiction. One can assume as if Hume that the sun will not rise tomorrow is no less comprehensible a proposition, and entails no more contradiction than the affirmation, that it will rise.

            I would like to point out that Kant associates the distinction between the analytic and synthetic judgments will another distinction viz. a priori and a posteriori judgments. Kant has the firm conviction that in all judgments in which the relation of a subject to the predicate in thought, this relation is possible in two different ways. Either the predicate B belongs to the subject A as something, which is (covertly) contained in this concept A; or B lies outside the concept A. It does stand in connection with it. May be in one case I entitle the judgment analytic, but in the other synthetic.

            In an analytic judgment, we may find that the predicate is contained in the subject or is a part of the meaning of the subject. Therefore, we call it uninformative. The moment when we clarify the meaning of the subject term, we can be assumed about the predicate term also. Here the guiding principle is no doubt the law of contradiction. No sense experience is necessary for it. They are a priori and based on the principle of identity. ‘A is A’, is an instance of identity proposition which must be tautological, but ‘AB is A’ is an analytic proposition.

            One can define analytic proposition like Swinburn in the way of the reference to the notion of logical truth and synonymy. For example, Quine considers the suggestion that an analytic proposition is one which ‘can be turned into a logical truth by putting synonyms for synonyms,’ a logical truth being a statement which is true and remains true under all reinterpretations of its components other than logical particles.’ Quine of course argues that the distinction between an analytic and a synthetic judgment is not tenable. All these criteria, to Quine turn on the notion of meaning. The concept of meaning may replace by the concept of synonymy that is hopelessly vague. The meaning of a sentence is what the sentence has the common with the other sentences synonymous with it. I will get back to these later in details.

            Synthetic judgment has informative content, i.e. the predicate term is not contained in the idea of the subject. Thus the judgment like, ‘Today is Monday’ is a synthetic judgment as the predicate term ‘Monday’ is not a part of the subject term ‘Today’. It has to be noted here that a proposition may be synthetic when its negation does not involve any self-contradiction. Now to say that “Today is not Monday” may be false, but does not create a self-contradiction.

            After having discussed Kant’s distinction between analytic and synthetic judgments, now I propose to discuss the view shared by logical positivists. Schlick, in his famous article “Is There a Factual A Priori?” says, ‘A synthetic sentence that is to say, one that actually gives expression to a cognition, is always used in science and life to communicate a state of affairs and indeed, that state of affairs the cognition of which is formulated by the sentence. On the other hand, an analytic sentence, or, to put it more clearly, a tautology, has a quite different function; it represents only a purely formal transformation of equivalent expressions, and serves, therefore, only as a technical device within a proof, a deduction, a calculus. A tautology is naturally an a priori truth, but gives expression to no state of affairs and validity of a tautology rests in no way upon experience.’[7] For him, tautological proposition will be true by the virtue of their form alone and give us no information about the world. Such a view is not only hold by Schlick, but also supported by other philosophers. C.I. Lewis thinks that analytic propositions present itself as true by the virtue of definition; therefore, they are also a priori. Lewis argues, ‘The a priori is not a material truth, delimiting or delineating the content of experience as such . . . and a priori is knowable simply through the reflective and the critical formulation of our own principles of classification and interpretation.’[8] Lewis believes that the truth of a priori judgment is dependent on the criteria of classification based on the criteria of reality. Here I want to contend that as all a priori propositions are non-factual, they will be exclusively concerned with the universal and not with a particular, for the particulars are essentially sensible objects. P.K. Sen suggests that ‘It is impossible that a priori propositions should refer to a particular qua particular, but it is quite possible that it will refer to a particular qua instance of some universal.’[9]

            Let us recall the different kinds of judgment that Kant speaks of:

  1. a) Analytic a priori:

            A judgment is analytic when the predicate term is derived from the analysis of the meaning of the subject term. Therefore, once we accept the meaning of the subject term, it will be self-contradictory to deny its predicate term. An analytic judgment is actually uninformative. A priori knowledge, on the other hand is independent of all experience. Now it is possible to combine both these elements of analyticity and a priority. Kant takes a proposition to be of the subject – predicate form. Thus, this analytic a priori judgment is unmixed with sense experience. ‘All bachelors are unmarried’ is an instance of this type of proposition. In this case, the person who knows the meaning of the subject term ‘bachelor’, will know the meaning of the predicate term without any help from experience.

  1. b) Analytic a posteriori:

            We already know that a proposition can obtain from sense experience is called   a posteriori judgment, like ‘Delhi is north of Kolkata’. Experience is the source of knowledge. In Kant’s opinion, we cannot have analytic a posteriori judgment, for all analytic judgments are a priori i.e. untainted by experience. Kant shares his belief with his predecessors, both rationalists and empiricists.

  1. c) Synthetic a posteriori :

            A judgment whose predicate concept is not contained in the subject concept and which is known through experience is called synthetic a posteriori judgment. All synthetic judgments are a posteriori and all a posteriori judgments are actually synthetic. Thus the judgment ‘All bodies are heavy’, is a kind of synthetic a posteriori judgment. Here the idea ‘heavy’ is not part of the meaning of a body, the subject term. To know our judgment one has to take recourse to sense experience.

  1. d) Synthetic a priori:

            In Kant’s philosophy, synthetic a priori judgment has played a fundamental role. By a synthetic judgment Kant means that the predicate term is not contained in the subject term, i.e. the predicate represents a new idea. By an a priori judgment, he means a judgment which is perfectly certain, strictly universal and necessarily valid. All analytic judgments are a priori and all synthetic judgments are obtained from our experience. Here, Kant’s contribution is that he wants to link syntheticity with a priority and speaks of a new kind of judgment viz., synthetic a priori judgment. A synthetic a priori judgment cannot be logically independent of experience and the law of contradiction cannot validate it. To clarify, Kant’s aim is to make a judgment that will be free from all sense experience, but the truth-value of this judgment will be necessary and at the same time, the negation of this proposition would not involve any contradiction. In the introduction of Critique of Pure Reason, Kant makes a query into the source of human knowledge. He formulates the main problem as follows — now the proper problem of Pure Reason is contained in the question — ‘How are a priori synthetic judgments possible?’ Here two questions together arise. First, is it possible to make a synthetic judgment a priori? Secondly, where will we find such a judgment?

            Kant mainly gives us the answer to the second question. Answering the second question, he automatically gives an answer to the first question. The main task of Kant is to clarify three different places where such judgments are possible. Synthetic a priori judgments use as principles in mathematics, natural science and metaphysics. Here three correlated questions are raised in the Kantian philosophy: ‘How is pure mathematics possible?’ ‘How is natural science possible?’ and, ‘How is metaphysics possible?’

            Richard Falckenberg’s[10] thinks that Transcendental Aesthetic (the critique of sensibility or the faculty of intuition) answers the first question viz., Is pure mathematical judgment possible. The Transcendental Analytic (the critique of understanding) is an answer to the second question, viz. Is natural science possible at all?, and the Transcendental Dialectic (The critique of ‘reason’ in the narrower sense) and the Transcendental Doctrine of Method (Methodenlehre) is an answer to the third question viz. How is metaphysics possible?

            Kant clarifies his answer by explaining that ‘pure mathematics is possible, because there are purely an a priori intuition (space and time) and pure natural science or the metaphysics of phenomena because there are a priori concept (categories) and principles of the pure understanding’. First of all, Kant tries to show that all mathematical judgments are synthetic in character. We find two types of mathematics — applied and pure mathematics, which is concerned with truths that are necessary and so a priori. He has taken an example from arithmetic’s, viz. 7 + 5 = 12 that, according to Kant, is both a priori and synthetic in character. As opposed to alleged thinking that arithmetical statements are analytical, Kant shows that the idea of 12 is not contained in the idea of the sum total of 7 and 5. We understood from this proposition that here just combining two different numbers 7 and 5 we get another number 12. This is new information already coming from our intuition. He calls these judgments synthetic because it is possible for us to get 12 from 7 and 5 by counting five and seven in our fingers. He also adds that 7+5=12 is a necessary proposition because the opposite of it is an inconceivable idea. According to Kant, ‘The concept of the sum of 7 and 5 contains nothing save the union of the two numbers into one and in this no thought is being taken as to what that single number may be which combines both.’[11]He further argues that the fundamental judgments of geometry are synthetic a priori judgment. He has taken an example from Euclidean geometry, such as, ‘A Straight line is the shortest distance between two points’, which according to him, is a synthetic a priori judgment.

            In this judgment, we find that the subject term ‘straight line’ contains only quality, but the predicate term ‘shortest distance’ is nothing but the quantity. We cannot derive the idea of ‘being short’ by analysing the idea of ‘straight line’ i.e. in this case the predicate term is not contained in the subject. When Kant wants to discuss the geometrical knowledge, he insists that the cognitive faculty that is capable of acquiring the knowledge of such geometrical arguments is certainly pure intuition.

Secondly, according to Kant, natural science is a field where we locate a synthetic           a priori judgment as principles. Kant maintains that ‘In all changes of the material world the quantity of material remains unchanged and ‘In all communication of motion, action, and re-action must always be equal.’[12] Both are in character synthetic but also a priori. This is an   a priori proposition because it is strictly universal and necessary which has been proved by science. This proposition has although been synthetic, as here the predicate ‘unchanged’ is not contained in the subject term. By the subject term ‘matter’, we mean that something which occupies some space.

Thirdly, In Metaphysics, we are able to ask about the existence of God, Immortality and Rebirth, etc. We know that at present metaphysics is such a knowledge that discusses appearance and also the reality of our world separately. We know that for Bradley Metaphysics takes its stand on this side of human nature, this desire to think comprehends reality. We cannot scrutinize the concept of subject-predicate terms. Let us think, ‘space is infinite’ — in this argument we find that the predicate term ‘infinite’ is a new idea which is not contained in the subject. Therefore, we may describe it synthetic, beside, the truth-value of this proposition does not rely on experience, and this judgment is necessarily true and therefore a priori.

Now, let me illustrate how logical positivists and other philosophers disparage Kant’s theory, mainly his idea of synthetic a priori judgment. Kant tells us that all geometrical propositions like ‘A straight line is the shortest distance between two points’ is synthetic a priori judgments. To logical positivists, it is just a mistake. A.J. Ayer remarks, ‘All that the geometry itself tells us is that if anything can be brought under the definitions, it will also satisfy the theorems. It is therefore a purely analytic proposition.’[13] In the same vein, logical positivists contend that arithmetical propositions are not synthetic but they are analytic. Schlick in his famous article ‘Is There A Factual A priori?’ argues that the concepts of synthetic a priori propositions are logically impossible. As a matter of principle, all propositions are either tautology or synthetic a posteriori. Lewis also admits that it is impossible to make a synthetic a priori judgment, for all a priori propositions are analytic which are known to be true by virtue of definition. Broad also points out that no synthetic proposition could possibly be self-evident.

It is important to note here that even Einstein, unlike Kant, concludes that the source of our knowledge of external world lies in the objective reality. He discusses the reason why Kant comes to a priorism, in the latter, having been misled — ‘By the erroneous a priori — difficult to avoid in his time — that Euclidean geometry is necessary to thinking and offers assured (i.e., not dependent upon sensory experience) knowledge concerning the objects of ‘external’ perception. From this early understandable error he concluded the existence of synthetic judgments a priori, which are produced by the reason alone, and which, consequently, can lay claim to absolute validity.’[14] It is a fascinating idea that if we take a sentence like “no synthetic propositions are a priori”, then the sentence itself turns towards a synthetic a priori judgment.

If we closely scrutinize all these views, then we shall find that obscure and some mistakes have taken place in Kant’s epistemological thinking. Many philosophers of contemporary period have criticized his point of view. However, it is undeniable that Kant introduces a Copernican revolution in philosophy. I think we are following Kant’s footsteps in many ways. If we discover some valuable insights in Kant way, that will be our credit, though being our guidance, Kant obviously demands great respect from the world of philosophy.

So far, I have discussed Kant’s theory in order to set the table. In this paper, my main purpose is to do a critique of Quine and Kripke’s thinking from the point of view taken by Hilary Putnam. Both Quine and Kripke criticize some main tenets of Kant’s philosophy. Quine in his paper “Two Dogmas of Empiricism” denies the conventional distinction between analytic and synthetic statements. Quine thanks that one can assert to make sense of the whole of the convention[15] in a priori knowledge. Actually, the distinction between a priori and empirical begins to dither and dissolve at least as a distinction between sentences. (It could of course still hold as a distinction between factors in one’s adoption of a sentence, but both factors might be operative everywhere).’

Besides, Kripke tries to show that it is not true that the traditional account of the intimate relation between a priori and necessary proposition must be true, i.e., it is mistaken that a priori proposition must be necessary, it may be contingent. Kripke argues, ‘Something may belong in the realm of such statement that can be known a priori, but still may be known by particular people on the basis of experience.’[16] The present paper aims at clarifying the points that are raised by Quine and Kripke and this clarification is done based on Putnam’s reactions to these two philosophers.

Quine’s Entangle

We may now discuss Quines famous article “Two Dogmas of Empiricism”. The first dogma of empiricism as Quine calls it, is the cleavage between analytic truth that are grounded independently of matters of fact and synthetic truth that are grounded only in fact. The notion of analyticity came from Hume’s distinction between relations of ideas and matters of fact or from Leibniz’s distinction between truths of reason and truths of fact. Kant’s distinction between analytic and synthetic truths echoes the distinction — that I have discussed earlier. For Quine, ‘But Kant’s intent, evident more from the use he makes of the notion of analyticity than from his definition of it, can be restated thus : a statement is analytic when it is true by virtue of meaning and independently of fact.’[17] Quine shows that we cannot identify meaning with naming. To take Frege’s example, ‘Evening star’ and ‘Morning star’ refer to the same object, but their meaning is not same. In the case of abstract terms, we may not find it important to make a distinction between meaning and naming. He takes an example, the term ‘9’ and ‘the number of the planets’ refers to one and the same abstract entity, although these two expressions present the object referred to in different ways, i.e. they have different meaning. It is an important astronomical discovery to know that the number of planets is nine.

Quine thinks that the above idea of meaning comes down to us from the Aristotelian notion of essence; ‘Things had essences, for Aristotle, but only linguistic forms have meaning. Meaning is what essence becomes when it is discovered from the object of reference and wedded to the world.’[18] It is important to note here, as Quine believes that theory of meaning is sharply separated from the theory of reference. Quine distinguishes between two kinds of analytic statements. One is logically true sentence like ‘No unmarried man is married’— it is not only true as it stands, but remains true under all reinterpretation of man and unmarried. Another class of statements is analytically true sentence, like ‘No bachelor is married’. Quassim Cassam’s clarifies Quine’s position to say that ‘And a proposition is analytically true if it is reducible to a logical truth by the substitution of synonyms for synonyms. Analytic propositions are, if true ten necessarily true, even there is also a sense in which they depend on linguistic usage.’[19]

I think that Quine, who is not an empiricist in the orthodox sense, realizes that it is important to revise the traditional definition of analytic judgment[20]. Here it is important to mention that Carnap thinks that a statement is analytic when it is true under every state description. A state description for Carnap is a description of a state of affairs or group of state of affairs. It is like a possible world, of which propositions are describing something. This Carnapian account according to Quine, is an adaptation of Leibniz’s ‘True in all possible worlds’. Carnap also believes that if we take two sentences ‘John is a bachelors’ and ‘John is married’. They would be mutually independent if and only if they are atomic sentences of the language.

Some Philosophers try to show that the analytic statements of the second class (analytical truths) can be reduced to logical truths, i.e., first class. Because the definition of ‘bachelors’ no doubt refers to the ‘unmarried man’, Quine thinks that the believers of the above thesis also suggests that there is a relation of synonymy between these two forms. But we may find a variant type of definitional activity which is different from pre-existing synonymies that Carnap calls its explication. Quine argues, ‘In explication the purpose is not merely to paraphrase the definiendum into an outright synonym, but actually to improve upon the definiendum by referring or supplementing its meaning.’[21]

Now a natural response of Quine is that one could give an account of synonymy in terms of interchangeability. Hence, an argument is offered by Benson Mates to show that if the two expressions are synonymous, then they are interchangeable everywhere without changing its truth-value, in Leibniz’s phrase salva veritate. Quine also thinks that in intentional language these two synonyms ‘bachelor’ and ‘unmarried man’ are not interchangeable everywhere. He takes a counter example to show that the truths which becomes false under substitution of ‘unmarried man’ for ‘bachelor’ are easily erected with the help of ‘bachelor of arts’ or ‘bachelor’s buttons’, also with the help of quotation, thus: ‘Bachelor has less than ten letters’. Quine is not interested in psychological synonymy or poetic synonymy as these are concerned with cognitive synonymy that Quine rejected. In Quine’s words, ‘What we need is an account of cognitive synonymy not presupposing analyticity — if we are to explain analyticity conversely with help of cognitive synonymy as undertaken in §1.[22] And this synonymy can be defined in terms of interchangeability salva veritate.

The statements (1) ‘All and only bachelors are un-married men’, which is surely analytic in Carnap and Lewis point of view, can be turned into (2) ‘Necessarily all and only bachelors are bachelors’ by substituting ‘unmarried man’ for its synonymy ‘bachelors’. It is well known that necessity is applicable only to analytic statements. Now by putting ‘unmarried man’ for an occurrence of ‘bachelors’, we get (3) Necessarily all and only bachelors are ‘unmarried men’. The terms ‘bachelors’ and ‘unmarried men’ can interchangeable be used everywhere. Necessary statements are true in general. If we admit that the statement (3) is true, we must admit that the statement (1) will be analytic. So, the term ‘bachelors’ and ‘unmarried men’ are considered as stimulus synonymous. According to Quine, ‘Stimulus synonymy or sameness of stimulus meaning is as good as the standard of synonymy for non-observational occasion sentences as for observation sentences as long as we stick to one speaker.’[23] He also believes that the stimulus synonymy of sentences relate to stimulus analyticity. Davidson & Benson Mate has taken this idea of stimulus analyticity in their writing that I would not like to discuss here. In intentional language, we may find that interchangeability salva veritate is a sufficient condition for cognitive synonymy. For Quine, ‘A word is cognitively synonymous to a word or phrase if substitution of the one for the other always yields cognitively equivalent sentences.’[24] However, in the case of extensional language Quine thinks that extensional agreement falls far short of cognitive synonymy of the type required for explaining analyticity during the term of Carnapian sense.

Let us take, for an instance, ‘For all X, X has a heart iff X has a kidney’. Now if this equivalence is factually true, then we say that the predicate terms ‘has a heart’ and ‘has a kidney’ will be interchangeable in all extensional statements. Quine claims, ‘So we must recognize that interchangeability salva veritate, if construed in relation to an extensional language, is not a sufficient condition of cognitive synonymy in the sense needed for deriving analyticity in the manner of §1. If a language contains an intentional adverb ‘necessarily’ in the sense lately noted or other particles to the same effect, then interchangeability salva veritate in such a language does afford a sufficient condition of cognitive synonymy; but such a language is intelligible only in so far as the notion of analyticity is already understood in advance.’[25] This would be a circular approach to explain analyticity in terms of cognitive synonymy. In ordinary language, it is so tricky to segregate analytic statements from synthetic statements, due to the vagueness of ordinary language. Carnap is sympathetic to this outlook and so he constructs an artificial language and a set of semantical rules. Here Quine’s arguments against Carnap points out that without knowing the conception of general relative term ‘analytic for’, we cannot understand a statement in a language where both statement and language are variables. In fact, a semantical rule stipulates other statements, which are not specified but count as true. By appealing to this rule, we may explain the concept of analyticity thus: a statement is analytic if it is (not merely true but) true according to the semantical rule. Quine also admits that truth in general depends on both language and extra-linguistic fact. The statement ‘Rama killed Ravana’ would be false if the state of affairs in the world were different. It would also be false in Quine’s opinion, if the word ‘Killed’ happened rather to have the sense of ‘begat’. Now Quine is trying to clarify his thinking to argue that ‘But for all its a priori reasonableness, a boundary between analytic and synthetic statements simply has not been drawn. That there is such a distinction to be drawn at all is an unempirical dogma of empiricists, a metaphysical article of faith.’[26]

Kripke’s Amalgam

Saul Kripke attempts to show that we have found a mistake in the traditional account of the intimate relation between a priori and necessary proposition. In Naming and Necessity, he says that it is not true that an a priori proposition must be necessary. First Kripke considers Wittgenstein’s comment about the standard meter bar of Paris. He mentions that ‘Wittgenstein says something very puzzling about this. He says: There is one thing of which one can say neither that it is one meter long nor that it is one meter long, and that is the standard meter of Paris.’[27] Kripke believes that there is a mistake in Wittgenstein’s view. He asks that if the stick is a stick, for example, 39.37 inches long (One can assume that we have some different standard for inches), the question is that why is not it one meter long?

As the length of everything totally depends on the time when we measure it, it will be important for us to ask, ‘is it necessarily true that the stick S is one meter long at time t0?’

Some philosophers are trying to prove that purely meaning is constituted sentence like — ‘Stick S is one meter long at t0’ is a justified a priori statement that must be necessarily true. However, Kripke thinks that this definition is not able to define the meaning of what we call a ‘meter’ but it indicates to fix the reference. Let us see what we are doing here. We are marking out a stick of a particular length at a particular time and deciding to call the length ‘one meter’. This really marks out an accidental property and we have to make a note of this. Here, it is important to say that if we had applied heat on this stick S at time t0, at time t0 stick S would not have been one meter long. In Kripke’s own words ‘In some counter factual situation the stick might have been longer and in some shorter, if various stresses and strains had been applied to it.’[28] Kripke points out that there has been an intuitive difference between the phrase ‘one meter’ and the phrase ‘the length of S at t0’. In actual world, the phrase ‘one meter’ designates a certain length; we think it is according to Kripke’s conception 39.37 inches. Here the first phrase ‘one meter’ would be the ‘length of the stick S at time t0’. However, the second phrase ‘the length of stick S at time t0’ would not able to designate anything. Because, if we apply heat on the stick S, then it would have expended in length.

Kripke gives two different arguments from two different perspectives— epistemological and metaphysical, by which he wants to prove that there is a distinction, which fixes the references and designators that give us a synonymy. Here it is important to be clear about the concept of ‘designator’. We can consider that a designator introduces a definite description, which does not give meaning to the designator, but only fixes its reference. Now, from an epistemological standpoint, we may find that the person who first fixes the meter system by referring to a particular stick S at a time t0, i.e., the proposition ‘stick S is one meter long at t0’. This knowledge is certainly acquired by a person without further investigation. Therefore, we may say that the definition who offers this definition had a priori knowledge. However, from the metaphysical standpoint, we find that the second phrase ‘S is one long time’ though it is known a priori yet it will be a contingent statement. Because if we applied heat to the said stick or make it cool, then the length of it will change at the same time. It would be right to call it a contingent proposition.

In Naming and Necessity, we find that Kripke takes an example; if we want to know about an example of a prime number then the computer can give us an example of it. We are unable to prove a prime number of mere calculation but machines like calculators; computers have the ability to give us such a number. If we believe in the prime number, our belief will be based on our knowledge of the law of physics, the construction of the machine and so on. We also know that any belief that depends on empirical knowledge must be contingent, as there is always a possibility of it becoming false. We all know that according to Kant a priori knowledge is very independent of any experience. Now Kripke asks the question whether a particular person or knower knows something a priori or believe it is true because of a priori evidence?

Some philosophers are trying to show that a priori truth is beyond any empirical knowledge, because if something belongs to the realm of a priori knowledge, it does not have its material from sense-experience. However, Kripke urges this point as a mistake and claims that something may fit in the domain of such statements that can be known a priori, but still may be known by particular people on the basis of the sense experience. He gives us one common sense example about a prime number of computing machine, which I have already mentioned earlier.

In Kantian philosophy, we may find that necessary and a priori are identical with each other. But Kripke mainly attacks this point of view. One can ask the following two questions regarding Kripke’s position —

  1. a) Are the two concepts ‘necessary’ and ‘a priori’ identical?
  2. b) Are the two concepts co-extensive?

Kripke argues that ‘necessary’ and ‘a priori’ are not synonymous. He thinks that in Kantian terminology, the definition of the term ‘necessary’ is very different from the definition of the term a priori. According to Kant, a necessary proposition is a proposition, the contrary of which is inconceivable. But a priori is defined by Kant as that independent of any sensory experience. Besides, we know that a necessary proposition belonged to metaphysical knowledge, but a priori is no doubt an epistemological knowledge. P.K. Sen claims, ‘What is important is that it is not epistemological, not at least in the sense in which his conception of a priori can be said to be and what is still more important is that the two concepts of the necessary and the a priori are very clearly set apart by Kant, whatever may be the way in which it is done.’[29]

Here, I hesitate to discuss that the necessary and a priori could be said to co-exist. I think it is quite right that each necessary proposition is a priori in a sense, but I do not believe that all a priori propositions would be necessary[30]. We may, however take some example in which it will be evident that experience can tell us that a certain proposition is necessary. Let us take some examples-

First: It is a well-known principle that Archimedes thought, if a thing is partly or wholly sunk into the water, then the weight of this particular thing will be same with the water that overflows from the containers. Now I think, he discovered this principle of the purely empirical method. First, he makes an experiment by measuring the weight. We know he had to face some obstacles and at last, he discovered this principle with no hesitation. He knew that this principle would be true in every possible world. Therefore, we find that this principle is necessarily true, but it was coming from our empirical investigation.

Second: Suppose, I have got news that my teacher Prof. Amitabha DasGupta will visit my house soon. Let us think that I know the law of addition in elementary logic, and then I can make such a sentence that will be recognized as a necessary truth but this sentence really refers from empirical hypothesis. For an instance, either Prof. DasGupta will be visiting soon or he will not be visiting soon. Let us take also an easy example. ‘Today is very hot’. By the law of addition, I may obtain such knowledge that ‘either today is very hot or it is not very hot’. This kind of proposition that I deduced from an empirical knowledge, it is obviously necessarily true.

Third: Kripke himself takes an example, which I want to discuss here. He believes that it is possible to know a posteriori proposition that is in character necessarily true. Kripke argues that ‘Hesperus’ and ‘Phosphorus’ are the same heavenly body. In his own word: ‘Let suppose we refer to the same heavenly body twice, as ‘Hesperus’ and ‘Phosphorus’. We say ‘Hesperus’ is that star over there in the evening. ‘Phosphorus’ is the star over there in the morning. Actually, Hesperus is Phosphorus. Here it is important to say that in Kantian philosophy, all genuine identity statements are necessarily true and a statement that is necessarily true cannot be false at all. If we admit that Hesperus = Phosphorus is necessarily referred to the same object. Planet Venus, then we should also admit that our knowledge about this conclusion must be regarded as empirical too. But P.K. Sen raises an excellent argument against Kripke’s view. He argues that Hesperus = Phosphorus is necessary is deduced from Hesperus de facto identity with Phosphorus, together with the general principle that if an identity statement is true at all then it is necessarily true, which, in its turn, is revealed in philosophical analysis. So it seems that, antecedently to the deduction and the philosophical analysis, we know that Hesperus = Phosphorus without that it is necessary that Hesperus = Phosphorus.

I am trying to give a short answer from Kripke’s point of view. Kripke requests us to imagine a possible world where a man empirically defines a certain star in the evening as Hesperus and a certain star in the morning as Phosphorus. Let us also think that one of them mainly Phosphorus was not really Phosphorus in the sense of our actual world. It is a different heavenly body. By chance, it is defined as the same body. In this situation, we may call that they are not, identical object though there has been a qualitatively identical epistemic situation admitted. Now if we define them as two different objects or qualitatively identical these knowledge what even we grant will depend on our sense experience, not by any case that depends on a priori.

An Attainment

In Hilary Putnam’s My Intellectual Autobiography, he tells us that one evening during a party in Reichenbach’s house at Harvard, a graduate student asked Prof. C.G. Hempel “I grant that one cannot show a clear analytic, synthetic distinction in a natural language, but why cannot one do it in a formalized language?” Putnam reminiscences, ‘I have never forgotten Hempel’s answer: “Every formalized language is ultimately interpreted in some natural language. The disease is hereditary”.’[31]

            Putnam himself admits that before he wrote his article “The Analytic and Synthetic”, he was a follower of Reichenbach. However, in his article “The Analytic and Synthetic”, he frees himself from the repression of verificationism and Reichenbachism. Here he attempts to give an account of the analytic and the synthetic distinction both inside and outside of the physical theory. His replies to Quine mainly depend on an analysis of some classic examples. Putnam explains, ‘I am convinced that there is an analytic, synthetic distinction that we can correctly (if not very importantly) draw, and I am inclined to sympathize with those who cite the examples and who stress the implausibility, the tremendous implausibility of Quine’s thesis — the thesis that the distinction which certainly seems to exist does not in fact exist at all.’[32] Putnam believes that no doubt it is quite true Quine is wrong but the main issue is somewhat different.

            Strawson and Grice also tried to establish the traditional distinction between analytic – synthetic judgment by offering theoretical reasons. It is actually a paradoxical concept to deny that the distinction really exists. The arguments of Strawson and Grice delineated this distinction partly but Putnam thinks that there has been another kind of distinction present, which he mainly would not like to discuss in his article “The Analytic and The Synthetic”. Here Putnam refuses to admit a mere distinction between the analytic and synthetic statements simply by using the words in respect of the convention of mere dispositions, i.e. the tendencies to apply the co-related terms in some petty expressions like, ‘we are not able to understand what he is saying’ or ‘having the same meaning’. The existence of analytic and synthetic distinction does not depend on the mere using of the word ‘analytic’, ‘synthetic’, ‘meaning’, or ‘understanding’. It is of course true that there is a gross distinction between analytic and synthetic statement like any two different linguistic expressions in the world. Putnam argues that ‘What I think is that there is a ‘limited’ notion of analyticity, which applies to such trivial cases like ‘All bachelors are unmarried’ and which is philosophically unimportant and an important but ‘revisable’ notion of ‘necessity’ or ‘conceptual truth’.’[33] According to Putnam, there are some words called ‘one-criterion words’, like a bachelor, vixen etc: On the other hand, there are words which are not ‘one-criterion words’; they fall under semantic category. In fact, all nouns fall under semantic category, for instance, cats, home, etc. Putnam considers, ‘It is important to distinguish ‘analytic’ truths of the sort “all cats are animals” — from analytic truths of the sort “all bachelors are un-married”, in part because the former tend to be less necessary than the latter. It might not be the case that all cats are animals; they might be automata.’[34]

            The concept of synonymy and analyticity discovered by the philosophers is actually incorrect. Putnam suggests that the expression ‘chair’ synonymous with the description of ‘moveable seat for one with a back’, can be regarded as synonymous with any expression that one could in principle erect in the sense-datum language. This is an example of the type of ‘hidden synonymy’ or ‘philosophic’ synonymy that some philosophers have claimed to discover and that does not exist”. The lexicographer and grammarian have discovered sufficient rules of language and there is no need for a philosopher to discover further rules of language. If we consider that a model language is necessary, then it will be better to divide the idea into three kinds like, analytic, synthetic and many other things. But it will be so much fuzzy to accept only the statements that will be either analytic or synthetic. We may find some cases where we cannot place a statement into this twofold distinction. For instance, Putnam says that it is not a good question to ask whether the laws of natural science are analytic or synthetic?

            Actually ‘knowing’ is something where we are not able to give a certain theory because of our different opinion. It is in Putnam’s words like ‘uncanny facility at guessing the correct answer’. Now Putnam concludes, ‘I think the statements in that conceptual system — except for the trivial examples of analyticity, e.g. ‘All bachelors are unmarried’, “All vixens are foxes’ — fall on a continuum, a multi-dimensional continuum. More or less stipulation enters more or less systematic import.’[35] It is important to mention in the function of the meaning of an individual word on its own network and it is impossible to separate the actual use of this word from its network, which partly reflects the meaning of the word and partly reflects the concomitant information about it.

            Now Putnam clarifies his idea by using a scientific concept, viz., and the notion of a law cluster concept. He puts it like this way, Law-cluster concept are constitutional not by a bundle of properties, as are the typical general names like ‘man’ and ‘crow’, but by a cluster of laws which, as it were, determine the identity of the concept. Wittgenstein also considers that game is such a law cluster concept. Some have claimed that the concept of species is a law cluster concept. However, for Putnam, ‘energy’ is a cluster concept. In highly developed science, most of the scientific terms turn towards law cluster concepts and this case there are some doubt about the conception of their analyticity.

            Science considers kinetic energy as the energy of motion, where a motionful object whether in its vertical or horizontal state, always relates naturally to kinetic energy. The mathematical equation of kinetic is K.E=½mv2, where ‘m’ is the mass of the object and ‘v’ sounds velocity.

            If anyone wants to change the definition of energy, then it would be necessary for him to change the meaning of ‘kinetic energy’ or explain the cluster concept with ‘re-definition’. Similarly, in case of ‘All bachelors are un-married’, we cannot reject the term ‘unmarried’ unless we can make a radical change in the meaning of the term ‘bachelor’. It is also important to change the extension of the term ‘bachelor’. One may feel tempted to agree with Quine especially in the case of ‘All bachelors are unmarried’. Here, we define ‘bachelors’ as an ‘unmarried man’, but there are lots of plausible scientific explanation of the concept of ‘energy’. So, one may argue that the ‘truth by definition’ has no remarkable value in analytic judgment. Quine also believes that true, by stipulation plays an ornamental role in the history of science. Putnam mentions that, ‘Stipulation Quine says, is a trait of historical events, not a ‘lingering trait’ of the statements involved.’[36]

            Putnam is however going against Quine and claims that in natural language the conception of ‘True by stipulation’ has no important role regarding the concept of analyticity. But Putnam admits that ‘true by stipulation’ is the nature of analytic judgment only in the model, i.e. formal language model. Putnam questions ­­— ‘why should we have an analytic statement (or strict synonymies printed out by Quine) in our language?’[37]

            “All bachelors are un-married” – the statement in a hypothetical formalized language sounds as an analytic one that is immune from revision. Here the important points that emerge are:

  1. An inventor legitimately can express his commands as per his own choice in a formalized language.
  2. According to Putnam ‘bachelor’ cannot be regarded as a law cluster term. The reason is that in the statement, ‘bachelors are unmarried men’ – here the truth-value of the sentence depends on mere stipulation, which is also convenient and can be substituted by a complex expression like ‘male adult human being who has never in his life married’.

            One might ask, why ‘bachelor’ is not regarded as a law cluster term? Putnam argues, ‘I have admitted that my knowledge (or ‘State of pretty-sureness’) that ‘bachelor’ will not become a law cluster term is based upon what we might call, in a very broad sense, empirical argumentation. That there is no exceptionalness laws containing the ‘bachelor’ are empirical in the sense of being a fact about the world; although it is not empirical in the sense of being subject to confrontation with isolated experiments.’[38]

            Now Putnam tells us that we can think about a permanent intention, like “wedding”, but it is not necessary that our intention must be permanent because both the bride and the groom are aware of the concept of divorce that is legally granted in our society. Similarly, in our formalized language we can find that ‘Let every statement be subject to revision’, but it is not a permanent issue. Here we can find an exception. In the case of imagining all bachelors are suffering from ‘sexual frustration,’ here ‘sexual frustration’ turns into a criterion for distinguishing bachelors from non-bachelors. It may happen that some ‘neurosis’ who are already married have become victims of such a ‘sexual frustration’. Then our earlier stipulation that ‘bachelors’ will be synonymous with ‘unmarried men’ becomes inconsistent. Therefore, here we must admit of a radical difference between formalized language and natural language, especially when we talk about a linguistic rule or its use.

            Putnam tries to clarify the main issues from the perspective of natural language. In traditional philosophy, there is a wrong tendency to make a distinction between analytic-synthetic statement from the context of ‘intuitive’ and ‘demonstrative’ truths. Therefore, we have a fashion to lump together the analytic statement, traditionally considered as an ‘intuitive truth’ with all its consequences. From this thinking, one might obtain a rough definition: An analytic statement is a statement that satisfies the criteria to be presented or a consequence of such statement. But it is quite true that the last clause of these criteria indicates’ borderline’ cases of analyticity. For an instance, ‘girls are women’ etc. Putnam argues, ‘The trouble with the analytic-synthetic distinction construed as a dichotomy is far more radical than mere ‘borderline fuzziness’. Yet, there are border cases, and reason for their existence is that the analytic-synthetic distinction is tied to a certain model of natural language and correspondence between the model and the natural language is not unique.’[39] Even it is quite true we can find some statements in natural language, which is analytic, and the other may be construed as analytic. The same thinking is also applicable for the synthetic statements.

Now, what will be their criteria? Putnam mentions about four of the criteria. Let me explain them below—

  1. The statement has the form: ‘something (someone) is an A if and only if it (he, she) is a B’, where A is a single word.
  2. The statement holds without exception, and provides us with a criterion for something being the sort of thing, which the term A applies to.
  3. The criterion is the only one that is generally accepted and employed in connection with the term.
  4. The term A is not a ‘law-cluster’ word.

I shall start by focusing on the first criterion.

‘Something is A if and only it is a B’, e.g. ‘Someone is a bachelor if and only if he is an unmarried man’. However, one can point that there is a vicious circle. In other words, we may put it — ‘Someone is a bachelor if and only if he is an unwed man’. Here one objection arises in trying to define synonymy in terms of analyticity by saying that the synonymous expression is analytically true.

            Let us take an example: ‘Someone is a bachelor if and only if he is either an unmarried man or a unicorn’. We know that from the perspective of Noam Chomsky’s ‘transformational grammar’ the quoted sentence is sounded as an ungrammatical. Putnam thinks that we can summarize this issue as the conjunction of the following claims:

  1. The statement sounds linguistically odd and not clearly true.
  2. There is no question about its general acceptance.
  3. People do not determine the term ‘bachelor’ by its consequent connection with the term “unmarried” or “unicorn” etc.

The problem occurs because of some misunderstanding, lying beneath our theory. First of all, The English ‘or’ and ‘if and only if’ are not synonymous with logical function ‘U’ or ‘º’. Secondly, people might query about its intelligibility, or decline to accept it. Thirdly, People might refuse it since ‘quoted criterion’ is not a generally accepted criterion for someone being a ‘bachelor’. Doubtlessly, here the concept ‘criterion’ is of great importance. We should discuss it from its uses. Putnam shows that we can define the ‘criterion’ from two different positions.

First, the ‘criteria’ I am speaking of being necessary and sufficient conditions for something being an A.

Secondly, through these criteria one can determine that something is an A. ‘Bachelorhood’ may have many indicators, like being young, living alone, high-spiritedness etc. But this criterion is determined by the combination of the first and second criteria which are mentioned above. Then it follows that the term ‘bachelor’ is an analytic truth.

            However, my main concern remains unsound. Putnam himself has raised the issue of the relevance of the four criteria’s of analyticity. Putnam argues that it would be insignificant to admit that “Someone is a bachelor it and it he is an unmarried or unwed” is an exception less law. It is quite true that we cannot find any exception less law, but a ‘law-cluster’ concept that as discussed in science. However, one may ask whether there any such concept in our philosophy. And Putnam replies that such a concept in philosophy is a called ‘one-criterion’ word, like a vixen, bachelor etc. Again, ‘law cluster’ concepts or such one-criterion words that were once used as an analytic truth would become false propositions today. For an instance, “Atoms are indivisible” or “Whales are mammals”. We can see that if we admit these words as ‘law cluster’ concepts or one-criterion words, then the linguistic characteristic of the word also changes. Putnam considers, ‘To verify or confute a statement of the form “Something is an A if and only if it is a B” in this way requires that we have independent criteria for something being an A and for being a B. Moreover, since the subject concept is not a ‘law-cluster’ concept, the statement has little or no systematic import.’[40]

            For Putnam, an analytic statement may be true from the standpoint of ‘Rules of language’, ‘True by stipulation’ or even ‘True by implicit convention’, but all these expressions are only metaphorically true. Putnam emphasizes, ‘What is the reality behind the metaphor? The reality is that they are true because they are accepted as true…’[41] One might also argue that is there any rational ground? Putnam answers that in our formalized language, it would be possible for us to give a reasonable statement, which would be immune from revision by stipulation. However, he also cautions that the rule may someday be changed. But that does not change the fact that the present rule is to the effect that this is to be done under all circumstances. In the same way, a rational man may perfectly well adopt a rule that certain statements are never to be given up: he does not forfeit his right to be called reasonable because of what he does, and he can give plenty of good reasons in support of his action. One should accept the analytic statement as rational though there is no reason (evidence) to admit it. This is just a rational construction of the actual language.

            Pessimist does not believe in a priori knowledge. Quassim Cassam considers, ‘The project of identifying a priori enabling condition for the acquisition of the various different kinds of knowledge by various different means would be doomed if the pessimist is right since a priori enabling conditions are just ones which can be known a priori .That is why, for better or worse, I am committed to optimism.’[42]

            Even Cassam thinks that understanding based reflection, reasoning, and calculation are more or less a sure form of a priori knowledge. The idea of ‘understanding based reflection’ is coming from Peacocke’s thoughts. Cassam asks us to imagine a flag in which the left half is red and the right is green. Now can we really imagine a case where an entity can be red all over or green all over at a same time? The answer will be ‘no’ or we cannot think of such a thing. Here this knowledge instead of depending on experience arises from our understanding related to intuitive insight. The idea of an a priori knowledge becomes stable through reasoning. For an instance, Mrs. Mamata Banerjee is the present chief minister of West Bengal and the chief minister lives in Kolkata. Now by reasoning, I come to know that Mamata Banerjee lives in Kolkata. Cassam here enquires how does calculation play a pertinent role in a priori knowledge and asks us ‘how do I know it that 68+57 =125’? He says that calculating is a non–experiential source of knowledge which is also devoid of any kind of indivisibility requirements.

I will conclude my discussion with an important issue. Paul Boghossain renews his attack on externalism in his well-known paper “What the Externalists Can Know A Priori[43]. He suggests that ‘ghost’ or ‘phlogiston’ has an empty existence and so externalist is unable to denote such natural kind terms. The externalists do admit any content without its proper reference. Now the thousand-dollar question is – what will be the truth condition of the sentence ‘Water is wet’ in dry earth? Boghossain argues that based on introspection we may sometimes consider whether an apparent natural kind thought has any content or not. In this case self-knowledge and mainly a priori thinking assists us to infer about the content like ‘I have a toothache’ or ‘I am not feeling better’ etc. Putnam as an externalist replies that externalism hardly bothers about how to determine the meaning of the empty terms. They are only stressing on the natural kind terms or the terms associated with the external worlds.

I come across the most attractive part of semantic externalism that emphasizes on a naturalistic outlook towards human beings in the world. Externalism rebuffs any kind of mentalese approach and intends to strengthen the thoughts of admitting the content of belief that locates where the believers are. They also challenge the view of incorrigibility or intrinsic ability that attempts to see the whole debate from the skin in.[44] Semantic externalism holds that the concepts of our knowledge become meaningless if and only if it has no causal connection with the external world. I agree with Putnam that to have a concept, it is necessary to have an appropriate causal connection with an environment. Semantic externalism instigates externalism about mind; if to have a mind is to have thoughts, then to have a mind, it isn’t sufficient to have the right goings-on in the brain and the rest of the body; to have a mind you have to be hooked up to an environment in the proper way, or you have to maintain casual interactions that co-relates to the environment. Externalism believes in a causal chain that guides our belief systems. If one trace back to the origin of this belief, then we see an “initial baptism” or reference fixing of a name or natural kind term by some speakers who aim to demarcate any sort of analyticity cum a priority from the semantic venture. This semantic externalism contrasts with internalism considers that mental properties are the properties of the states (mental states) that formulates our behaviour. In short, the brain states that create some functional properties are doubtlessly the mental properties. There is a lack of causal connection between the functional properties and an agent’s behaviour. Moreover, these functional properties are in nature physical properties that efficaciously relates to the physical world as it a priorily supervenes on the external world. It is because the minds causally linked with the external world generate the internal states that actually represent the way of the world from our skin in, i.e. a priori dependence on conceptual role semantic insists on ontological a priority of conceptual competence over the linguistic competence.


Ayer, A.J. 1947. Language, Truth and Logic. Dover Publication, London.

Boghossian, Paul. 1997. Proceeding of the Aristotelian Society.

Cassam, Qussaim. 2007. The Possibility of Knowledge. Clarendon Press, Oxford.

Chakraborty, Sanjit. 2016. Understanding Meaning and World: A Relook on Semantic Externalism.: Cambridge Scholars Publishing, New Castle, London.

Einstein, Albert. 1949. ‘Reply to Criticism.’ In Albert Einstein: Philosopher, Scientist. ed. Paul A. Schilpp, The Library of Living philosophers, Evanston.

Falkenberg, Richard. 1962. History of Modern Philosophy. Progressive Publishers, Calcutta.

Hume, David. 1982. An Enquiry Concerning Human Understanding. Progressive Publishers, Calcutta.

Kant, Immanuel. 1961. Critique of Pure Reason. transl. N.K. Smith, Macmillian and St. Martin’s Press, London and New York.

Kripke, Saul, A. 1980. Naming and Necessity. Harvard University Press, Cambridge.

Leibniz, G.W. 1991. Monadology. ed. Nicholas Rescher, University of Pittsburgh Press, Pittsburgh.

Lewis, C.I. 1956. Mind and World Order. Dover Publication, New York.

Putnam, Hilary. 1962. ‘It ain’t necessarily so.’ The Journal of Philosophy. LIX, 22.

Putnam, Hilary. 1975. Mind, Language and Reality. In Philosophical Papers, Volume 2, Cambridge University Press, Cambridge.

Putnam, Hilary. 1983. Realism and Reason. Philosophical Papers. Volume 3, Cambridge University Press, Cambridge.

Putnam, Hilary. 2015. ‘My Intellectual Autobiography.’ In Philosophy of Hilary Putnam. Ed. Randall E. Auxier, Douglas R. Anderson and Lewis Edwin Hahn, Library of Living Philosopher Series, Open Court Press, Illinois.

Quine, W.V. 1960. Word and Object. The MIT Press, Cam, Mass.

Quine, W.V. 1978. ‘Use and its place in meaning.’ Erkenntnis 13.

Quine, W.V. 1980. ‘Two Dogmas of Empiricism.’ In From a Logical point of view, Harvard University Press, Cambridge. 2nd Edition.

Schlick, Morris. 1949. ‘Is There A Factual A Priori?’ eds. H. Feigl & W. Sellars, Reading in Philosophical Analysis.

Sen, Pranab Kumar. 1980. Logic, Introduction and Ontology. Jadavpur Studies in Philosophy 2.

Swinburne, Robert. 1975. ‘Analyticity, Necessity and A priority.’ Mind 84.

*My especial thanks go to my mentor Hilary Putnam and my old friend Professor Quassim Cassam for their valuable suggestions and help that inspire me to complete the paper.

[1] Immanuel Kant, Critique of Pure Reason, trans. by, N.K. Smith, London and New York: Macmillian and St. Martin’s Press 1961, p. B3-B4.

[2] ibid. B3.

[3] G.W. Leibniz, Monadology, edited. Rescher, Nicholas, Pittsburgh : University of Pittsburgh Press, 1991.

[4] Swinburne tries to define necessary proposition with a new impression. He considers, ‘A proposition is necessary if and only if it is incoherent to suppose that the individuals in fact picked out by the referring expressions in the sentence with expresses it do not have the properties and /or relations claimed by the proposition.’ Please see, Robert Swinburne, “Analyticity, Necessity and A priority”, Mind 84, 1975, pp. 225-243.

[5] G.W. Leibniz, Monadology, p.135.

[6] David Hume, An Enquiry Concerning Human Understanding, Calcutta: Progressive Publishers, 1982, p. 23.

[7] Morris Schlick, “Is There A Factual A Priori? ed. by H. Feigl & W. Sellars, Reading in Philosophical Analysis, 1949. p. 281.

[8] C.I. Lewis, Mind and World Order, New York: Dover Publication, 1956, pp. 231-232.

[9] Pranab Kumar Sen, Logic, Introduction and Ontology, Jadavpur Studies in Philosophy 2, 1980.p.15.

[10] Richard. Falkenberg, History of Modern Philosophy, Calcutta: Progressive Publishers, 1962, p. 334.

[11] Kant, Critique of Pure Reason, pp. 52-53.

[12] ibid., p. 54/B18.

[13] A.J. Ayer, Language, Truth and Logic, London: Dover Publication, 1947, 2nd edition, p. 83.

[14] Albert Einstein, “Reply to Criticism”, in Albert Einstein: Philosopher, Scientist, edit by Paul, A. Schilpp, Evanston: The Library of Living philosophers, (11. B ), p. 679.

[15] Actually positivists believe in the procedure of meaning fixation of a confirming experience based statement only be possible because of the “stipulation”. So, analytic statements that are dependent on confirming experience become true just because of meaning alone. The statements relating a priori truth is based on the stipulation or convention. Quine mainly alludes and later refutes analyticity and apriority in the sense of truth by convention.

[16] Saul, A. Kripke, Naming and Necessity, Cambridge: Harvard University Press, 1980, p. 35

[17] W.V. Quine, “Two Dogmas of Empiricism” in his book From a Logical point of view, Cambridge: Harvard University Press, 2nd Edition, 1980, p. 21.

[18] ibid., p. 22

[19] I am thankful to Quassim Cassam for this helpful clarification.

[20] Here one think need to clarify is that analytic truth that Quine tries to attack have an universal range of confirming experiences that is actually related with no matter what, a notion of apriority becomes redundant from the history of science as there I no legitimate principal of science that allow such a proposal even for non-standard logic that deals with a priori or unrevisable statement.

[21] ibid, 25.

[22] ibid, p.29. The definition of analyticity that Quine like to reject is “truth by definition”, a linguistic definition operate with the synonym for synonyms by accepting the notion of truth of logic. In the case of usefulness of logical truth of Kantian sense, Hilary Putnam clarifies that ‘ It might clarify Kant’s notion of a judgment whose truth’ to be a substitution instance of s theorem of monadic logic (i.e., logic up to but not including the undecidable logic of relations). The statements obtainable from the theorems of monadic logic and their substitution instances by putting synonymy for synonymy are probably more or less the class Kant had in mind.’ (Please see, Hilary Putnam, Realism and Reason, Philosophical Papers, Volume 3, Cambridge: Cambridge University Press, 1983, p.95.)

[23] W.V Quine, Word and Object, Cam, Mass: The MIT Press, 1960, p.46.

[24] W.V Quine, ‘Use and its place in meaning’, Erkenntnis 13, 1978, pp. 1-8.

[25] Quine, “Two Dogmas of Empiricism”, p. 31.

[26] ibid, p. 33

[27] Saul. A Kripke, Naming and Necessity, p. 54.

[28] ibid.p.55.

[29] Pranab Kumar Sen, Logic, Introduction and Ontology, Jadavpur Studies in Philosophy 2, 1980, p.220.

[30] Kripke argues that ‘The term ‘necessary’ and ‘a priori’, then as applied to statements, are not obvious synonym’s……….. (I will argue below that in fact they are not even co-extensive — that necessary a posteriori truth, both exist).’ Please see, Saul Kripke, Naming and Necessity, p. 38.

[31] Hilary Putnam, “My Intellectual Autobiography.” in Philosophy of Hilary Putnam, edited by Randall E. Auxier, Douglas R. Anderson and Lewis Edwin Hahn, Library of Living Philosopher Series, Illinois: Open Court Press, 2015, p.17. I am thankful to my mentor Hilary Putnam for given me the unpublished manuscript of this book in 2007.

[32] Hilary Putnam, “The Analytic and the Synthetic”, in Mind, Language and Reality, Philosophical Papers, Volume 2, Cambridge: Cambridge University Press, 1975, p.34.

[33] I am thankful to Hilary Putnam for this helpful clarification.

[34] Hilary Putnam, “It ain’t necessarily so”, The Journal of Philosophy. LIX, 22, 1962.

[35] Hilary Putnam, “The Analytic and the Synthetic”, p. 40.

[36] ibid, p. 55.

[37] In his well known paper “Two Dogmas’ Revisited’ (in Hilary Putnam, Realism and Reason, Philosophical Papers, Volume 3, Cambridge: Cambridge University Press, 1983, pp. 87-97.), Putnam summarized that Quine mainly attacked the base of putatively analytic statements that do not sound as analytic. The reason is here not only for meaning or truth by logic, but the logic of the world that keep a distinction with the matter of empirical fact that an agent holds. The second reason that Quine proposed is that because of the fallibilism an agent cannot provide an absolute guarantee about the rightness of logic in the case of apriority, so analytic sentence can turn wrong. Analytic truth can be only unrevisable in the case of revisable logic and language that would be a different matter.

[38] ibid.p.59.

[39] ibid, pp. 64-65.

[40] ibid, p. 68.

[41] ibid, pp. 68-69.

[42] Qussaim Cassam, The Possibility of Knowledge, Clarendon Press, Oxford, 2007, pp.194-195.

[43] Paul Boghossian, Proceeding of the Aristotelian Society, 1997, pp. 161-175.

[44] I have discussed the similar issues in my recent book Understanding Meaning and World: A Relook on Semantic Externalism, New Castle, London: Cambridge Scholars Publishing, 2016.

Philosophy 17/2017, pp. 70-100