IIIM Magazine Online, Volume 2, Number 36, September 4 to September 10, 2000


by Marty Fields


To any student of the history of philosophy it becomes quite evident, after only a brief introduction, that the quest for certainty dominates the landscape. After all, how can we be sure that any of our resolutions to the various problems of philosophy have any merit at all unless the propositions that we use to describe our "solutions" are true? To be sure, there have been no few attempts to show how one might be able to identify the truths that are true in all instances and circumstances. In philosophical terms, the question is, "What truths can I count on?" "What truths are immune to revision?" "What truths cannot, under any circumstances, be false?" The history of philosophy is replete with suggestions, but the most common and popular solution in recent years has been to identify truths by what is called the "Analytic/Synthetic Distinction." The influence of this point of view has been enormous. It has not only infiltrated virtually every philosophical method since the time of David Hume, but it has also infiltrated Christian philosophical method, exegesis, and apologetics.

Let's get right to it. According to this distinction there are two kinds of "truths." The first, called "analytic," are truths which are necessary and immune to revision. They apply equally in all instances and circumstances. The truth-seeker rejoices at this category and wants to know more, but the more she learns, the more she is disappointed, for there is a price to be paid for this certainty or "revisionary immunity." According to the Analytic/Synthetic Distinction, if you want certainty you have to sacrifice the informative power of your statements. Here's an example of an analytic truth according to this model:

The Red Barn is Red.

This is a true statement. Its truth can be determined solely in terms of the contents of the statement itself. The concept of the predicate is contained in the subject. It is true; it is necessary. Unfortunately, it doesn't tell us anything significant about the real world. It is simply a statement of logical identity (A=A). It does not need to be experienced to be true; it is known apart from experience (i.e. a priori). So, while it doesn't tell us anything significant, it is a statement that we can believe with all our hearts. It will never be false.

So, perhaps the saving grace will be the so-called "synthetic" truths of this distinction. In contrast to an analytic truth, a synthetic truth would look liked this:

The Barn is Red.

Now we have something significant; we have a statement that tells us something about the real world. But there is a price to be paid here as well. Unfortunately, because they are propositions about the real world, they are only probable (we are told) since they are subject to empirical methods of confirmation; they are known by means of experience (i.e. a posteriori). In other words, if I want to verify this statement, I have to go to the barn and see if it is red. The problem is that it could really be blue, while I might perceive it to be red because I'm developing cataracts. In other words, synthetic truths are subject to the frailties of sense-perception as the standard of verification. Therefore, while synthetic truths tell us something about the real world, their dependence upon sense-perception and induction do not allow us to know with certainty that they are true.

So the issue is this: If the truths we believe are analytic, they are certain, but they are useless since they are only statements that translate into logical truths; the concept of what is predicated is already stated in the subject. They may be immune to revision, but they are trivial; they tell us nothing significant, and make no claim. If we make claims about the world, then those statements are synthetic and must be verifiable by sensory means. All other statements that do not either translate into a logical truth, or state something that can be verified by empirical methods, are meaningless. We will discuss this further as we progress.

It does not take long for one to see the implications of this position on the Christian worldview. If the Bible contains propositions then they must be divided up as either analytic or synthetic. If the Biblical proposition is deemed analytic (a priori and necessary), then it tells us nothing significant. It may be immune to revision, but it says nothing substantial. If it is synthetic (a posteriori and contingent), then it must be tested according to empirical methods and destined to the uncertainty of induction. For example, the proposition "God exists" cannot be analytic (we are told) because it makes a statement about the real world, about history. So, it must be synthetic, and to determine its truth one must test it empirically. However, since no one has ever seen God, we have no way of empirically verifying this statement. At best, as many Christian apologists who subscribe to the Analytic/Synthetic Distinction have said, we can only conclude that God probably exists (as we set forth our claim).1 In short, no claim of the Christian faith can be known for certain. However, we as Christians claim that the Bible contains true and necessary propositions about the world, many of which are non-empirical. But the price of revisionary immunity, the philosophical salesman tells us, is the total surrender of the meaningfulness of our statements.2

Therefore, we find it necessary to examine this supposed distinction to see if it is tenable. A study of this will also help us: gain insight into world views and their claims; test claims of a world view; and show that empiricism as the be-all and end-all test of meaningful statements is unjustified. This paper will explore the historical background that gave rise to the Analytic/Synthetic Distinction, examine and critique the distinction, and propose a presuppositional explanation of analyticity and syntheticity as a solution to the problem.

The Elimination of Metaphysics

From the time of Plato to Descartes, the study of speculative philosophy, or metaphysics, was a respected discipline. It was a regular practice of traditional philosophy to contemplate questions about ultimate reality. The doctrines of God, interactions, and fate, to name a few, were common areas of study for the philosopher. Much research, thought, and formulation on these subjects were contributed to the marketplace of ideas. However, with the dawn of the eighteenth century, many questions about the methods and propositions of traditional philosophy were raised. Philosophers began asking questions about the meaningfulness and certainty of the propositions that metaphysicians had been suggesting. No public procedure existed for testing these propositions, therefore there was no way to determine whether or not any of these propositions corresponded to reality. It did not simply matter if was rational (logical),3 although that was important, but it also mattered if the propositions could not be tested, not only by logical coherence, but also according to some empirical procedure of verification. This shift, then, isolated two kinds of statements: formal (logic and mathematics); and matters of fact (those pertaining to matters of experience). All other matters were to be rejected. This eighteenth-century dichotomy and anti-metaphysical attitude can be seen in the following statement from none other than David Hume:

If we take in our hand any volume; of divinity, or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? Commit it to he flames. For it can contain nothing but sophistry and illusion (emphasis in original).4

From the perspective of Hume and other empiricists, metaphysical inquiry is meaningless and must be abandoned. The only statements that have any relevance are those propositions that are either tautological or empirical, which form the entire class of significant propositions.5

Hume's strong critique of metaphysics influenced a great number of philosophers. Among these was Auguste Comte (1798-1857). Comte's system is often called "Positive Philosophy." Although Comte was attempting to deal more with sociological problems, his positivism came to have a significant influence on philosophy. According to Comte any sentence (or proposition) that is not either a tautology or an assertion that has empirical significance is unintelligible.6 Comte saw metaphysical study as outdated by science. He believed that philosophers should abandon metaphysical speculation and devote their energies to constructing a unified encyclopedia of the sciences.7

With the dawn of the twentieth century, a new interest in Hume's empiricism and Comte's positivism came to new horizons.8 This new school of thought was known as "Logical Positivism." This school was a member of a larger school known as "Analytical Philosophy."9 The latter proposed the idea that philosophical problems can be solved by an analysis of the usage of language. The new role of the philosopher was not to continue to devise complex philosophical systems, but to rather analyze the statements or propositions of language. Logical Positivism agreed with the above, but stated that analysis was restricted to empirical matters. The most prominent example of this school was the continental group called "The Vienna Circle." Among the group were such renowned philosophers as Moritz Schlick (around whom the group centered), Rudolf Carnap, and Herbert Feigl. The renowned linguistic philosopher, Ludwig Wittgenstein, though not a member of the circle, wrote a book that defined much of their outlook. In his Tractates Logico-Philosophicus, Wittgenstein said that "whatever can be said can be said clearly," and "whereof one cannot speak, thereof one must be silent." From this perspective, all statements about the metaphysical subject are attempts to say what cannot be said.10 Combining the extreme empiricism of Hume and the positivism of Comte, the Logical Positivists had established that the only meaningful statements must be analytic (a priori and necessary) or synthetic (a posteriori and contingent). Analytic propositions are simply tautological and are not tested (since their necessity is determined by their logical relation alone), but how are the synthetic truths tested? As we have seen, empiricism demands that sensation be the ultimate criteria. To this the Logical Positivists promulgated the "Verification Theory of Meaning." This was the theory that the meaning of a statement is the method of its verification. This would require the proponent to specify what kinds of observations would count for or against its truth.11 From this it is clear that metaphysical propositions could not pass the test of meaningfulness. Positivists delighted in showing that theological, metaphysical, and ethical statements failed to meet their criteria of meaningfulness. Statements like "God exists," being neither "analytic" nor "synthetic," were meaningless to them.12

Logical Positivism is no longer considered a tenable position. However, with the other forms of empiricism that gave rise to it, it set the trend that still dominates much of philosophy today: the only statements that have any meaning are either analytic or synthetic. It should be noted that not all philosophers, as will be seen, adhere to this position. It is safe to say, however, that the common attitude towards metaphysical propositions is at best skeptical. Most philosophers, it seems, are either empiricist or quasi-empiricist, and it is characteristic of empiricist to eschew metaphysics on the ground that every factual proposition must refer to sense experience.13 Many, then, would hold to this Analytic/Synthetic Distinction between propositions. Certainty, or revisionary immunity, is only granted to propositions that are self-evident (e.g. logical, mathematical, etc.). All others (e.g. synthetic) must succumb to empirical standards to ascertain truth-value; anything that is to count for evidence for a statement must be a matter of observation or sense experience.14

Holding to the Analytic/Synthetic Distinction in the above manner is not an impartial or neutral conviction. Indeed, it commits one to a philosophical method that will have significant consequences. To treat some statements as "analytic" and others as "synthetic" is: 1) to discriminate between the statements of your belief-system and judge each claim one by one, rather than to consider and test the corporate system as a whole; 2) to hold that some propositions are such that their denial can be ridiculed as a failure of understanding; 3) to hold that not only are such statements unfalsifiable but also non-informative (i.e. make no difference to what is true about the world); 4) to "flatten out" all of the significant claims in the system, thus recognizing no privileged status for any statement (i.e. basicity, importance, immunity, etc.), and subject each and every statement to the same procedures and standards; 5) to hold that the meaningful, informative, and partially judged statements of your system have been justified in an empirical manner (experience being the test); and 6) to use this conviction's epistemological standards (i.e. the previously mentioned "philosophic methodism") to govern and critically determine what substantive beliefs about reality are endorsed by a belief system (i.e. decreeing the priority of criteria over the determination of all accepted beliefs instead of the priority of some beliefs over the determination of the criteria to be used in accepting further beliefs).15

Implications for Christian Apologetics

The Analytic/Synthetic Distinction is not indigenous to only non-Christian philosophy. Many Christian apologists adhere to this same distinction. R.C. Sproul, for example, says:

The only way that we can [sic] absolute philosophical certainty about anything is in the pure formal realm. Unfortunately, that doesn't get us into the real world.16

What he is saying is that certainty is only to be found in those statements which translate into logical truths (analytic). Nicholas Wolterstorff observes that many Christian thinkers are, unfortunately, brothers under the skin with the logical positivists.17 The reason for this negative view of the distinction is that many philosophers have held it to oppose the infallible truth of God's Word. If what the Bible says is really informative then it must be justified by empirical criteria and seen as only probable. Since not all of its claims can be confirmed in this manner, one must fideistically leap from the physical to the metaphysical. Likewise if we claim that certain propositions are immune from revision, then the distinction demands that the statement cannot be informative about the world.

This question, then, is an urgent one. If the Bible is put to such standards, then revisionary immunity will be impossible to establish in the apologetic situation. But is this distinction tenable? Can a sharp line be drawn between analytic and synthetic propositions? It is the position of this paper that it cannot. From here we will move to an examination of the proposed dichotomy to see whether or not this position is possible.

The Analytic/Synthetic Distinction18

The Analytic/Synthetic Distinction, as we have seen, was foreshadowed in Hume's distinction between the relations of ideas and matters of fact. The actual terms "analytic" and "synthetic" were really first used in this way by Kant.19 Kant defined an analytic statement as a statement that is true by virtue of the meanings and independently of fact. For any statement to be analytic, the concept of the predicate must already be contained in the concept of the subject, and the analytic proposition (or judgement, as Kant called it) must be one in which its denial would be self-contradictory.20 Philosophically, then, the attributes of a priori (non-experimental) and necessity (such that its denial would be self-contradictory) are attributed to analyticity. But how do we find these analytic truths? Since an arbitrary enumeration by some philosophical "pope" will not be acceptable, a different criteria has been proposed. It is now said that analytic truths are logical truths by means of semantic translation into a formal language.21 The logical truth of "A=A" provides the foundation.22 Therefore, the following statement is analytic:

"All bachelors are unmarried males."

It is logically true, in terms of translation. It does not require any empirical justification for its validity. Likewise, neither does it say anything significant about the world. It is true in terms of the intension of each semantic token.23 The use of the principle of identity, which provides the necessity of analytic propositions, then, presupposes that each semantic token is synonymous with the other. Very simply, this means that we could also affirm the truth of the statement:

bachelor=unmarried male

Therefore, it seems that analyticity depends upon the concept of synonymy. For a statement to be analytically true, its subject and predicate must be synonymous. By definition, "bachelor" is "unmarried man." It would seem that the problem is settled until we ask, "Exactly what is synonymy?" The problem with the above examples is that, in appealing to synonymy, they presuppose analyticity already, and analyticity is what was supposed to be explicated by synonymy. This would be flatly circular and thus unacceptable. What we need is an account of synonymy not presupposing analyticity.24 Obviously, an appeal to the dictionary will not do since the dictionary amounts to an arbitrary enumeration of analytic statements (which is unacceptable prima facie), but more importantly it should be seen that the dictionary presupposes the concept of synonymy in performing definition.25 The explication or paraphrase that follows an entry assumes synonymy already. Recognizing this problem philosophers have said that synonymy can be found when the terms in question are interchangeable salva veritate (literally "saving the truth"). That is, wherever one of the terms is used, its "synonym" can be inserted in its place in the sentence, and the truth value will not change. So, whenever a sentence contains "bachelor," "unmarried man" could be substituted and the truth value would remain the same.26

This seems to work until we come across problematic examples such as:

All "creatures with a heart" are "creatures with kidneys."

"Creatures with kidneys" is like "creatures with a heart" in extension, but its terms are not synonymous. We could perform interchangability, but the truth value would change. So we are then given a glaring counter-example. Not to be outdone, some have proposed that the criteria be expanded to incorporate the concept of "necessity" in order to salvage the technique.27 This would result:

"Necessarily all ‘creatures with a heart' are ‘creatures with kidneys.'"

This obviously eliminates the above statements as analytic, since we know that it is not necessarily the case that all creatures with hearts have kidneys. Because it happens to be a regularity of nature does not, in and of itself, demand its necessity. Thus we reform our usual proposition:

"Necessarily all Bachelors are unmarried males."

We have expanded our criteria and thus saved our definition of synonymy. By importing the concept of necessity, we exclude frivolous counter-examples and define synonymy without analyticity. Therefore, analyticity is explicated in terms of necessity. This seems to work until we note that this technique is, again, somewhat circular. Necessity is what was supposed to be explicated in terms of analyticity. Therefore to appeal to necessity as an explicator of analyticity is to beg the question.

To put it more clearly, we started with logical laws and synonymy. From there we went to define synonymy in terms of interchangeability salva veritate. This eventually required an ad hoc addition of the modal adverb ‘necessarily' into the first-order, predicate logical language. But we now realize that the notion of necessity presupposes a grasp of analyticity, which is what we were supposed to be explicating! It becomes analyticity explicating analyticity.28 The whole attempt at drawing the distinction degenerates into hopeless circularity.

Besides the problem of drawing such a line between analytic and synthetic, there are other statements that we come across in assuming this distinction that are not analytic (as is understood by this distinction) but are nevertheless considered a priori, such as "everything green is extended." The denial of this is necessarily false. There are no synonymies present. If fact, in light of the traditional understanding of the distinction, this proposition would be "synthetic."29

What we have seen in this tedious exercise is that the common notion of drawing a hard line between analytic and synthetic propositions is simply not possible. To say that some truths are empty of factual content (thus making the linguistic component everything) and that others are only "verified" by a particular set of experiences is a misguided maneuver. To do so would require one to explain how to separate a factual component from a linguistic component, and one cannot sharply distinguish language from reality, the truths of definitions from the truths of other statements.30 What has to be realized is that no statements are unfalsifiable, even those that are formal in nature. The usefulness of logical truths is dependent upon the translation technique that we employ, and that, in turn, is dependent upon the functions of language. Likewise not all claims that are synthetic in nature are necessarily verifiable in experience, yet they are meaningful.31 The result is that propositions cannot be divided up and judged one-by-one by some supposed "neutral" method of philosophy (i.e. analytic/synthetic). As we have seen such neutrality (as is assumed in the sharp dichotomy between linguistic significance in analyticity and the empirical verification of synthetic propositions) is not possible. Each test for propositions will depend upon other beliefs (i.e. world view propositions within a person's noetic structure) that are held prior to encounters with subsequent propositions that may challenge the more basic beliefs, which are significant and non-empirical. Our propositions about reality can only be examined as a body of belief. "Revisionary immunity," then, will be ascribed to those propositions that we are least likely to give up within our world view; the "analytic" statements are going to be those propositions which we are most jealous to protect. Quine diagrams this prioritizing of certain beliefs over others as a "web" in which the more basic are the more central while the non-basic are more peripheral.32

Applications for Apologetics

The question may arise at this point as to how we then judge alternate world views to the Christian faith. Since we have seen that it is not possible to divide up propositions and judge them in isolation from the others (since the distinction breaks down) we might conclude that apologetics is not possible; indeed this was Abraham Caper's position.33 What we have are antithetical world views that have different standards for testing, modification, rejection, etc. Anything that is offered as evidence for or against the world view's claims (especially those more basic) must pass through the judgement standards of the opposite position. It would seem that apologetics is futile.

In reality, however, this is not the case at all. As each position has ultimate commitments and basic assumptions, we must ask if in fact those assumptions can provide the necessary pre-conditions of intelligible experience. That is, we ask whether or not those basic assumptions can provide a basis for rational inquiry, our experience of the world, and how we should relate to the world. Once it is demonstrated that the anti-Christian position cannot, the Christian can invite the unbeliever to put himself into the Christian position to see that it alone can make sense of all three tests — it alone gives science and logic a foundation. What will hopefully happen is not simply a change in the unbeliever's attitude towards certain facts, but rather a total reconstruction of his paradigms within his world view. Of course, this whole procedure will be a reminder of the fact that the unbeliever really does believe in God, and that he has been suppressing the truth (Rom. 1:18ff.). A solid presuppositional method that bases its ultimacy in God and his revelation will have a certain foundation and devastating apologetic.


This paper has attempted to demonstrate the fallacy of the commonly held Analytic/Synthetic Distinction in modern philosophy and apologetics. To be sure, much more could be written on the many facets of this supposition and its unfortunate implication. What this paper has sought to do, however, has been to critique the distinction, to reestablish the validity of many metaphysical propositions by dethroning empiricism as the sole standard for significant propositions, and to promote a presuppositional method to provide a better understanding of "revisionary immunity."

While the Christian must realize that all of her statements are subject to falsification, she can be certain that the Christian world view will establish itself. God is the sovereign one who makes all things possible or impossible. Without God there would be no possibility of certainty. Fortunately, we can stand on our presuppositions and have confidence in the certainty of our faith.

  1. The majority method for defending the Christian faith (apologetics) among evangelicals is known as "evidentialism." At the root of this method is the assumption that the Analytic/Synthetic Distinction has partial, if not complete, validity. In this case, to prove that God exists, one must inductively make a case for God. At best, the apologist will be able to show that God probably exists. While there may be some usefulness in utilizing the arguments of this approach, there are many problems with this method, far beyond the scope of this paper. But at the very least, evidentialism says that our methods of investigation (logic, science, etc.) are more certain than God's existence, a position that the Bible clearly rejects (cf. Romans 1:18ff.).
  2. See Greg L. Bahnsen, "Revisionary Immunity."
  3. This was a decisive point of departure for the empiricist. The previous century had been concerned primarily with deduction and rational coherence. To the rationalists the mind had the ability to ascertain knowledge apart from serious consideration of sensation. Descartes is perhaps the best known advocate of the rationalist school. With the rise of Locke, Berkeley and Hume, questions about deduction and concepts as a priori methods of knowing were rejected. All of our ideas are not innate, but are rather derived through our experience by association of these experiences. Sense perception and induction came to reign in philosophy as methods of knowing. (See "Empiricism" in the Encyclopedia of Philosophy, vol. II. New York: Macmillian and Free Press, 1967, p. 499.)
  4. David Hume, An Inquiry Concerning Human Understanding (La Salle, IL: Open Court, 1935), section 12, pt. 3.
  5. See A.J. Ayer, Language, Truth, and Logic (New York: Dover Publications, (1946) 1952), p. 41. Ayer is known as the popularizer of the movement of "Logical Empiricism," which has its roots in the movement of Logical Positivism. Ayer claims that the quantity of metaphysical statements arises from defective linguistic analysis. Because we can speak about all sorts of names which do not import existence into the class mentioned, metaphysicians have supposed that there must be a special ontological mode that fills these classes. Ayer concluded that this usage is defective and that the only value of these types of statements is aesthetic or moral.
  6. See "Auguste Comte," in The Encyclopedia of Philosophy, Vol. II (New York: Macmillian and Free Press, 1967), p. 175. A full-length treatment of Comte is beyond the scope of this paper. However, his contribution to philosophical empiricism should not be belittled for it is here that the foundations were laid for logical positivism, which will be critiqued later.
  7. Brennan, The Meaning of Philosophy (New York: Harper and Row, (1953) 1967), p. 228.
  8. The reason for saying a "new" interest is that Hume's empiricism did not go unchallenged. Kant, in observing Hume's dichotomy of relations of ideas and matters of fact (or, as Kant called them: "analytic" and "synthetic"), and agreeing that knowledge of ultimate reality or "things in themselves" was not possible, attempted to argue that there are some truths that are "synthetic" and yet a priori. Kant believed that the laws of mathematics, for example, were among these kinds of truths. Likewise Hegel's idealistic "Dialectical Philosophy" attempted to revive metaphysical study and interpret universal reality as organic. (See Samuel E. Stumpf, Socrates to Sarte: A History of Philosophy, 3rd ed. New York: McGraw-Hill Book Co. (1966) 1982, pp. 291ff. and pp. 312-320; 418-419.)
  9. The rejection of metaphysics that was characteristic of Logical Positivism should not be understood of the whole school of Analytical Philosophy. Bertrand Russell and G.E. Moore, both of Cambridge, were members of this previously mentioned school and retained a profound interest in metaphysics. Russell's "Logical Atomism" was an attempt to create a new language that focused on the precessional character of language. The relation of word and object needed to be more precise. Naming was not indigenous only to empirical propositions. (Stumpf, Socrates to Sarte, p. 419; See also F. Copelston, S.J., A History of Philosophy, vol 8, part II. Garden City, NY: Doubleday/Image Books, 1967 pp. 161ff).
  10. Stumpf, Socrates to Sarte, p. 419; and F. Copelston, S.J., A History of Philosophy, vol. 8, part II, p. 259.
  11. See William P. Alston, The Philosophy of Language (Englewood Cliffs: NJ: Prentice-Hall Publishing, 1964), p. 69; and Dallas M. High, Language, Persons and Belief (New York: Oxford University Press, 1967), pp. 40-42. It should be noted that the ‘verification theory' died with logical positivism. Thinking that each isolated statement has a unique correlate in sense experience which necessarily confirms or infirms it as fallacious. This method has been named "reductionism." (Bahnsen, "Revisionary Immunity," p. 67).
  12. See Ronald H. Nash, Faith, and Reason: Searching for a Rational Faith (Grand Rapids: Zondervan/Academie, 1988), p. 53.
  13. See A.J. Ayer, Language, Truth and Logic, p. 71.
  14. Bahnsen, "Revisionary Immunity," p. 7.
  15. This delineation of results of the Analytic/Synthetic Distinction is taken from Bahnsen's "Revisionary Immunity," pp. 8-9.
  16. Tape: "The Sproul-Bahnsen Debate" (December 1, 1977. Reformed Theological Seminary, Jackson, MS). I asked Dr. Sproul in class if it was true that he held to this distinction and he affirmed it. He claimed it was a close kin to the similar understanding of A.J. Ayer about this distinction.
  17. See Nicholas Wolterstorff, Reason Within the Bounds of Religion (Grand Rapids: Wm. B. Eerdmans, 1976) p. 20.
  18. The contemporary notion of an analyticity and syntheticity are taken from John Hospers, An Introduction to Philosophical Analysis (Englewood Cliff, NJ: Prentice-Hall, (1953) (1967), pp. 160ff.
  19. See footnote # 9 above.
  20. Although not really pertinent to this paper, it should not be passed over that Kant's formulation of analytic judgements was limited to statements of subject-predicate form. If the scope were this narrow, then the law of excluded middle would not meet the criteria. Most logicians would have a broader idea of what constitutes analyticity. Likewise the notion of self-contradictoriness is also in need of clarification as is analyticity. Quine says, "The two notions are the two sides of a single dubious coin" (Quine, "Two Dogmas of Empiricism," p. 20).
  21. For something to be "logically true," it must be true in "all logically possible worlds." But the advocate of the logical necessity of formal propositions will have to explicate logical possibility without appealing to logic itself. Not to do so would result in embarrassing circularity. This, however, is not a problem for the Christian who can justify the transcendental necessity of logic by presupposing the mind of God as the necessary foundation; it is a tremendous problem for the evidentialist who denies the Christian world view and must attempt to justify logic without logic in order to escape circularity!
  22. I have intentionally skipped, for purposes of space, the step of pure logical truth seen in the statement: "No unmarried man is married." Since the debate does not center around these types of useless utilizations of identity, I have chosen to pass over the obvious. Our problem is not with these kinds of logical truths, but rather it is with those that depend upon the notation of synonymy. (See Quine, "Two Dogmas of Empiricism," p. 24.) Bahnsen points out, however, that even attempting to be purely formal is not without its problems. For example, "business is business" appears to be logically true until we notice informal instances of equivocation are possible; symbolically, the denial is self-contradictory, but denial of the ordinary use is not necessarily contradictory (Bahnsen, "Revisionary Immunity" p. 17).
  23. "Intension" is not to be confused with "extension." "Evening Star" and "Morning Star" have the same extension (Venus) but not the same intension. Likewise, Willard Van Orman Quine's example of "creature with a heart" and "creature with kidneys" also illustrates this. Many have confused intension and extension and missed the point. (See Quine's essay, "Two Dogmas of Empiricism," in From a Logical Point of View. Cambridge, MA: Harvard University Press, (1953) 1961, p. 21.)
  24. See Quine, "Two Dogmas of Empiricism," pp. 25-27. See also Quine's discussion of "stmulas analyticity" in Word and Object Cambridge, MA: M.I.T. Press, (1964) 1976), p. 62ff. Quine lists various proposals of defining synonymy, but he shows that they too wind up relying on analyticity for their foundation.
  25. Quine correctly points out that the dictionary is the work of lexicographers, which are empirical scientists. The dictionary records antecedent facts; uses of words prior to the writer's own work. (Quine, "Two Dogmas of Empiricism," p. 24)
  26. This, of course, excludes a reference to the semantic token itself. "‘Bachelor' is a word with eight letters" could not have "bachelor" exchanged for "unmarried male." Though obvious such exclusions must be made.
  27. Indeed we are confronted with another problem here. To utilize the modal adverb "necessarily" assumes that we are working with a language rich enough to include such a word. But the language we started with was of an elementary, first-order predicate nature (that includes names, descriptions, and syncategormatic logical terms) that excludes subjunctive conditionals and modal adverbs. (See Bahnsen, "Revisionary Immunity," pp. 25-26). An ad hoc rescuing device is arbitrarily introduced.
  28. What we would assume, naturally, is that when analyticity and necessity are explicated (as they are supposed to be in philosophy) independently of each other, the set of necessary truths will turn out to be the same as the set of analytic truths.
  29. This was the type of proposition that eventually caused a shift in Wittgenstein's position. The so-called "latter Wittgenstein" came to see more clearly the complex nature of actual language as complex phenomena. (See High, Language, Persons, and Belief, p. 51.) Earlier, Kant had already dealt with this problem in his theory of knowledge by ascribing the name "synthetic a priori" to certain types of statements. For Kant the so-called "categories" (which dealt with space-time, mathematics, etc.) were synthetic and a priori. (See Bertrand Russell, The Problems of Philosophy. New York: Oxford University Press, (1912) 1959, pp. 83-84.)
  30. See John Frame, The Doctrine of the Knowledge of God (Phillipsburg, NJ: Presbyterian and Reformed, 1987), p. 250.
  31. I am, of course, referring to metaphysical propositions!
  32. See W.V. Quine and J.S.D. Aeolian, The Web of Belief, 2nd ed. (New York: Random House, (1970) 1978).
  33. See Abraham Caper, Principles of Sacred Theology (Grand Rapids: Baker Book House, 1898, 1980).