Maverick Philosopher: Carnap

Tuesday, April 28, 2015

Carnap and Clarity

This entry is installment #2 in a Carnap versus Heidegger series. Here is the first in the series. It couldn't hurt to at least skim through it. Part of what I am up to is an exploration of the origin and nature of the analytic-Continental split. To quote from the first installment:

If I were were to select two writings that best epitomize the depth of the Continental-analytic clash near the time of its outbreak, they would be Heidegger's 1929 What is Metaphysics? and Carnap's 1932 response, "On the Overcoming of Metaphysics Through the Logical Analysis of Language." (In fairness to Carnap, let us note that his Erkenntnis piece is more than a response to Heidegger inasmuch as it calls into question the meaningfulness of all metaphysics. And in fairness to Heidegger, we should note that he thinks he is doing something more radical than metaphysics. Metaphysics for Heidegger is onto-theology. Metaphysics thinks Being (das Sein) but always in reference to beings (das Seiende); it does not think Being in its difference from beings. Perhaps in a later post I will venture to explain what that means.)

Analytic philosophers prize clarity. And rightly so. For one thing, "clarity is courtesy," as Ortega y Gasset once said. (I suppose the Spaniard would count as Continental, and not just geographically.) One more parenthetical remark before getting down to business: I wish Erich Pryzwara had received the message that clarity is courtesy. Then perhaps he would not have written anything as unreadable as his Analogia Entis. Even the charitable German Thomist Josef Pieper so characterized it.

I need waste no words defending the thesis that clarity in thought and expression are to be preferred to obscurity. Avoidable obscurity must be avoided. But there is an empty and trivial clarity. A clarity worth pursuing is a clarity with content. Clarity ought not be pursued as an end in itself or as a cognitive value that trumps every other cognitive value. While avoidable obscurity must be avoided, some obscurity is bound to be unavoidable if our inquiries are serious, sustained, and worth pursuing. Such obscurity must be tolerated.

One day in class I was praising clarity and its importance. A student responded that reality is messy. My counter-response was that, while reality is messy, it does not follow that our thinking about it should also be. On the contrary! The present point, however, is that thinking worth doing ought to penetrate as far as it can into reality, as rich, dark, and messy as it is, and if some obscurity proves unavoidable, then so be it.

Rudolf Carnap's brand of clarity is sterile, arbitrary and as artificial as fluorescent light. What he does is enforce or impose an arbitrary standard of clarity across the board without regard to differences in subject matter. We ought to say about clarity what Aristotle said about precision near the beginning of the Nicomachean Ethics: "it cannot be expected in the treatment of all subjects alike . . . ." (1094b10-15) Ethics, the Philosopher said, cannot be treated with mathematical precision. The same goes for metaphysics.

"Many words of metaphysics," Carnap tells us, are "devoid of meaning." ("The Elimination of Metaphysics Through Logical Analysis of Language" in A. J. Ayer, ed., Logical Positivism, The Free Press, 1959, p. 65.) He chooses as his first example 'principle' "in the sense of principle of being, not principle of knowledge or axiom." (65) But since 'principle' in the ontological sense is not much used these days outside of scholastic circles, let me substitute 'ground' used ontologically. Thus there is a large body of literature in which truth-makers are described as the 'ontological grounds' of truths. It is clear that Carnap's accusations of (cognitive) meaninglessness apply as much to grounds as to principles. They also apply to other words and phrases in the vocabulary of truth-maker theorists such as 'makes true' and 'in virtue of.' For example, 'Peter is smoking' is true in virtue of Peter's smoking, or the fact of Peter's smoking makes true 'Peter is smoking.' Apart from truth-maker theory, 'in virtue of' has long been a favorite of philosophers. Some think it a 'weasel phrase' best banished from the vocabulary of philosophy. I disagree.

"But these words are ambiguous and vague." (65) Thus Carnap. Insofar as they have a clear meaning, their meaning is empirical, not metaphysical; insofar as they are metaphysical, they are meaningless. I would put the Carnapian argument as follows in terms of the following dilemma:

Either metaphysical grounding (as it occurs in the putative relation between a truth-maker and a truth-bearer) is a causal relation or it is a logical relation. But it is neither. It is not a relation of empirical causation. Truth-maker theorists insist on this themselves. Nor is it a logical relation such as entailment. Logical relations hold between and among truth-bearers; a truth-maker, however, though proposition-like on some theories, is not a truth-bearer. 'Truth-making,' then has neither the meaning of 'causing' nor the meaning of 'entailing.'

Yet, no criterion is specified for any other meaning. Consequently, the alleged 'metaphysical' meaning which the word is supposed to have here in contrast to the mentioned empirical meaning, does not exist. [. . .] The word is explicitly deprived of its original meaning . . . . (65)

The word 'making' is stripped of its empirical meaning, but no new meaning is supplied. The word becomes an "empty shell." (66) The associations and feelings attached to the word used in the old empirical way remain in play. But these do not give meaning to the word used in the new 'metaphysical' way: "it remains meaningless as long as no method of verification can be described." (66)

The same holds for all specifically metaphysical terms. There are one and all "devoid of meaning." (67) Carnap mentions the following: the Idea, the Absolute, the Unconditioned, the Infinite, and "the being of being," which I take to be a reference to Heidegger's das Sein des Seienden, which is better translated as 'the Being of beings.' But also: non-being, thing-in-itself, absolute spirit, objective spirit, essence, being-in-itself, being-in-and-for-itself, the Non-Ego.

These terms are meaningless because empirical truth-conditions of their use cannot be supplied. Hence the alleged statements of metaphysics which contain them are one and all pseudo-statements that are bare of sense and assert nothing. (67)

Perhaps the best response to Carnap and those of his ilk is brutal contradiction: "You're just wrong!" The words you would dismiss as meaningless just obviously have meaning and you're just obviously wrong to think otherwise.

For example, it is clear enough what it means to say that some truth-bearers need truth-makers, that a sentence such as 'Peter is smoking' cannot just be true but is true because of something external to the sentence, something external on the side of the object, not on the side of the subject, i.e., on the side of the one who asserively utters the sentence. And it is clear enough that this use of 'because' is not an empirical-causal use of the term. It is also clear enough what 'in virtue of' and 'making' mean in this context.

Another devastating response to Carnap is the obvious point that his Verifiability Criterion of Cognitive Significance cannot satisfy its own demand. "Every cognitively meaningful statement is either empirically verifiable in principle or a logical/analytic truth" is neither empirically verifiable in principle nor a logical/analytic truth. Therefore, the Verifiability Criterion is cognitively meaningless. So does it then have a merely emotive meaning? Is is a mere suggestion as to what to allow as meaningful? If the latter, then no thank you!

It's an easy rebuttal, but none the worse for that. Sometimes, simplex sigillum veri.

Carnap is to philosophy what a philistine is to the arts: just crude and ignorant . So I dismiss him as a philosophistine. I coined this word ten years or so ago in a polemic against David Stove another philosophistine whose crudity is on shameful display in his The Plato Cult.

My rule is: no polemics in philosophy. But if the other guy starts it . . . . Or the shade of the other guy . . . .

Posted by Bill Vallicella on Tuesday, April 28, 2015 at 03:50 PM in Analytic Philosophy Criticized, Carnap, Continental Philosophy Criticized, Heidegger, Logical Positivism | Permalink

Thursday, March 12, 2015

Heidegger, Carnap, Das Nichts, and the Analytic-Continental Schism

One of the reasons I gave this weblog the title Maverick Philosopher is because I align neither with the analytic nor with the Continental camp. Study everything, I say, and drink from every stream. Reverting to the camp metaphor, when did the camps become two? In dead earnest this occurred when Heidegger burst onto the scene in 1927 with Being and Time. I agree with Peter Simons: "Probably no individual was more responsible for the schism in philosophy than Heidegger." (Quoted in Overgaard, et al., An Introduction to Metaphilosophy, Cambridge UP, 2013, 110.) It is not as if Heidegger set out to split the mainstream whose headwaters were in Franz Brentano into two tributaries; it is just that he started publishing things that the analytic types, who had some sympathy for Heidegger's main teacher Husserl, could not relate to at all.

If I were were to select two writings that best epitomize the depth of the Continental-analytic clash near the time of its outbreak, they would be Heidegger's 1929 What is Metaphysics? and Carnap's 1932 response, "On the Overcoming of Metaphysics Through the Logical Analysis of Language." (In fairness to Carnap, let us note that his Erkenntnis piece is more than a response to Heidegger inasmuch as it calls into question the meaningfulness of all metaphysics.)

To nail my colors to the mast, I take the side of Heidegger in his dispute with Carnap and I heartily condemn the knee-jerk bigotry of the thousands upon thousands of analytic types who mock and deride Heidegger while making no attempt to understand what he is about. The cynosure of their mockery and derision is of course the notorious sentence

Das Nichts selbst nichtet. (GA IX, 114)
The Nothing itself nihilates.

This is the line upon which the analytic bigots invariably seize while ignoring everything else: its place in the essay in question and the wider context, that of Being and Time and other works of the early Heidegger, not to mention the phenomenological, transcendental, existential, life-philosophical, and scholastic sources of Heidegger's thinking.

Now, having called them knee-jerk bigots and having implied what is largely true, namely that the analytic Heidegger-bashers are know-nothings when it comes to Heidegger's philosophical progenitors, and thus having paid them back in their own coin, I will now drop all invective and patiently try to explain how and why Heidegger is not talking nonsense in the essay in question. This will require a series of posts. It will also require some attention and open-mindedness on the part of the reader as well as some familiarity with the two essays in question.

Heidegger's Alleged Violation of Logical Syntax

For Carnap it is obvious that existence and nonexistence are purely logical notions, more precisely, logico-syntactic notions. The sentence 'Cats exist,' for example, does not predicate existence of individual cats. It says no more than 'Something is a cat.' But then 'Cats do not exist' says no more than 'Nothing is a cat.' This sentence in turn is equivalent to 'It is not the case that something is a cat.'

'Nothing,' then, is not a name, but a mere bit of logical syntax. Carnap calls it a "logical particle." (71) And the same goes for 'something.' If I met nobody on the trail this morning, it does not follow that I met somebody named 'nobody.' (Bad joke: I say I met nobody, and you ask how he's doing.) If nothing is in my wallet, that is not to say that there is something in my wallet named 'nothing.' It is to say that:

It it not the case that something is in my wallet
It is not the case that, for some x, x is in my wallet
For all x, x is not in my wallet
~(∃x)(x is in my wallet)
(x) ~(x is in my wallet).

The above are equivalents. It should be obvious then, that in its mundane uses 'nothing' is not a name but a logico-syntactic notion that can be expressed using a quantifier (either universal or particular) and the sign for propositional negation. By a mundane use of 'nothing' I mean a use that presupposes that things exist. Thus when I assert that nothing is in my pocket, I presuppose that things exist and the content of my assertion is that no one of these existing things is in my pocket. (Don't worry about the fact that it is never strictly true that there is nothing in my pocket given that there is air, lint, and space in my pocket.)

I think we can all (including Heidegger) agree that in their mundane uses, sentences of the form 'Nothing is F' can be translated, salva significatione, into sentences of the form 'It is not the case that something is F' or 'Everything is not F.' The translations remove 'Nothing' from subject position and by the same stroke remove the temptation to construe 'nothing' as a name. Not that Heidegger ever succumbed to that temptation.

But now the question arises whether every use of 'nothing' fits the deflationary schema. Is every meaningful use of 'nothing' the use of a logical particle? Consider ex nihilo, nihil fit, 'Out of nothing, nothing comes.' The second occurrence of 'nothing' readily submits to deflation, but not the first. Suppose we write

It is not the case that something comes from nothing.

This removes the quantificational use of 'nothing' in 'Out of nothing, nothing comes' but leaves us with a 'substantive' use. Of course, 'nothing' cannot refer to or name any being or any collection of beings. That is perfectly evident. And Heidegger says as much. But 'nothing' does appear to refer to, or name, the absence of every being. The thought is:

Had there been nothing at all, it is not the case that something could have arisen from it.

The 'at all' is strictly redundant: it merely serves to remind the reader that 'nothing' is being used strictly. Now could there have been nothing at all? Is it possible that there be nothing at all? More importantly for present purposes: Is this a meaningful question? 'Possibly, nothing exists' is meaningful only if 'Nothing exists' is meaningful. So consider first the unmodalized

There is nothing at all

Nothing exists.

These are perfectly meaningful sentences. That is not to say that they are true, nor is it to say that they are possibly true. Suppose they are not possibly true. Then they are necessarily false. But if necessarily false, then false, and if false, then meaningful. For meaningfulness is a necessary condition of having a truth-value. 'Nothing exists,' then, is a meaningful sentence, and this despite the fact that 'nothing' cannot here be replaced by a phrase containing only a quantifier and the sign for negation.

For Carnap, however, the above are meaningless metaphysical pseudo-sentences because they violate logical syntax. If you try to translate the second sentence into logical notation, into what Carnap calls a "logically correct language"(70) you get a syntactically meaningless string:

~(∃x)(x exists).

This is meaningless because 'exists' cannot serve as a first-level predicate in a logically correct language. Existence is not a property of individuals. 'Exist(s)' is a quantifier, a bit of logical syntax, not a name of a property or of any entity. Therefore, 'Nothing exists' is as syntactically meaningless as the ill-formed formula

~(∃x)(∃x(. . . x . . .)).

Two Interim Conclusions

The first is that Heidegger commits no schoolboy blunder in logic. He does not think that a use of 'Nothing is in the drawer' commits one to the existence of something in the drawer. He cannot be charitably read as assuming that every use of 'nothing' is a referring use. The second conclusion is that Carnap has not shown that every occurrence of 'nothing' can be replaced by a phrase containing a quantifier and the sign for negation. He has therefore not shown that a sentence like 'Nothing exists' is a syntactically meaningless pseudo-sentence.

Heidegger Partially Vindicated

But now the way is clear to ask some Heidegger-type questions.

I showed above that 'nothing' has meaningful uses as a substantive, uses that cannot be eliminated by the Carnap method. And I suggested that 'nothing' could name the total absence of all beings. If this total absence is a possibility, as it would be if every being is a contingent being, then Nothing (das Nichts) would have some 'reality,' if only the reality of a mere possibility. It could not be dismissed as utterly nichtig or nugatory. Nor could questions about it be so dismissed.

One question that Heidegger poses concerns the relation of negation (Verneinung) as a specific intellectual operation (spezifische Verstandeshandlung) to Nothing:

Gibt es das Nichts nur, weil es das Nicht, d. h. die Verneinung gibt? Oder liegt es umgekehrt? Gibt es die Verneinung und das Nicht nur, weil es das Nichts gibt? (GA IX, 108)

Is there Nothing only because there is the Not and negation? Or is it the other way around? Is there negation and the Not only because there is Nothing?

I grant that with questions like these we are at the very limit of intelligibility, at the very boundary of the Sayable. But you are no philosopher if you are not up against these limits and seeking, if possible, to transcend them.