Maverick Philosopher

At he end of this paper I will have something to say about the latter reading of Nietzsche. But for now I am prepared to run both risks and impute to him as his central epistemological doctrine (or at least working hypothesis) the thesis of perspectivism. At a first approximation, the idea is that truth is always tied to definite perspectives in such a way that there is, and can be, no such thing as an absolute truth (truth-in-itself), or even an intersubjectively binding truth-for-us, whether now or at some Peircean limit of inquiry. But what makes this doctrine distinctively Nietzschean, and not just another form of relativism about truth, is the axiological corollary: absolute truth not only does not exist, it would not be good for us if it did exist. Truth is not a value or ideal.

The question I shall pursue in this paper is whether perspectivism can be coherently articulated. The conclusion will be negative. Since the salient comes out by contrast, it is natural to begin by comparing perspectivism with pragmatism.

1. Why Nietzsche is not a Pragmatist

It is tempting to read Nietzsche as a sort of pragmatist about truth: what we call 'truths' are conventions established in accordance with the pragmatic or instrumental considerations that make up the conditions of our life. (See R. Schacht, Nietzsche, RKP, 1983, pp. 72-82.) Numerous passages can be cited in support of this pragmatic interpretation, e.g., "The criterion of truth resides in the feeling of power." (WP #534) This could be construed as saying that a belief is true if and only if it is life-enhancing or life-promoting, where "life-promoting" does not mean merely life-preserving, but something like life- augmenting. Nietzsche frequently remarks that the will to power is not a mere will to survival. Hence on this criterion he would be saying more than that true beliefs are true in virtue of their survival value. But if this is a sort of pragmatism, it is not a pragmatism along the classical lines of William James or C.S. Peirce. There are at least three reasons for this.

There is first of all Nietzsche's repeated insistence that "untruth is a condition of life." (BG&E sec. 4) This makes it difficult to assimilate Nietzsche to the Jamesian view that "the true is whatever proves itself to be good in the way of belief..." (Pragmatism, p. 76); that the true is "only the expedient in the way of our thinking, just as 'the right' is only the expedient in the way of our behaving..." (p. 222) For Nietzsche seems to be saying precisely the opposite. The sense of "untruth is a condition of life" seems to be that the holding of false beliefs is a necessary condition of human flourishing. Thus in his polemic against Kant (BG & E, sec 11) Nietzsche does not deny that there are such synthetic a priori judgments as that every event is caused; his claim is rather that "such judgments must be believed to be true, for the sake of the preservation of creatures like ourselves; though they might, of course, be false judgments for all that!" He goes on to say that they are false, and that although we "have no right" to them, they belong nevertheless to the "perspective optics of life." Logic and mathematics are also falsifications of the world. "Logic...depends on presuppositions with which nothing in the real world corresponds, for example on the presupposition that there are identical things..." (HAH 16) The radicality of this should not be missed: Nietzsche is saying that nothing in the real world is self-identical at a time or over time. All such identities are linguistic fictions that falsify the real world. If so, the concept of number is a fiction, and with it the whole of mathematics. (HAH 22)

Nietzsche's position thus seems to be that there are certain beliefs (e.g., 'Every event has a cause,''Everthing is self-identical') that are perspectivally true but non-perspectivally (absolutely) false. The holding of such non-perspectivally false beliefs is necessary for our preservation and flourishing, and to this extent good for us. The difference from James' pragmatism should now be obvious. Whereas James identifies "the true" without qualification with the "the good in the way of belief," i.e., what it is good for us to believe, Nietzsche identifies the perspectivally true with what it is good for us to believe. This implies that, while for James it is the true that is good for us to believe, for Nietzsche it is the false that is good for us to believe. This, I take it, is the sense of "untruth is a condition of life."

A second reason why Nietzsche is not a pragmatist is that he does not share the typical pragmatist faith that objectivity is attainable at the ideal limit of inquiry. James speaks in this connection of the absolutely and unalterably true as the "ideal vanishing point towards which we imagine that all our temporary truths will some day converge." (p.222-3) Peirce waxes eloquent about the "great hope" "embodied in the conception of truth and reality," namely that truth is "the opinion which is fated to be ultimately agreed to by all who investigate..." (Philosophical Writings of Peirce, p. 38) Recently, Hilary Putnam has sounded this theme: the true is what is rationally acceptable at the ideal limit of inquiry. (Reason, Truth and History, p. 55) But surely Nietzsche has neither faith nor hope in the ultimate convergence of all perspectives at the ideal limit of inquiry. He would most likely see such an epistemological eschaton as a secular substitute for a soteriological eschaton: yet another vestige of Christianity that cannot survive the death of God.

Thirdly, and most fundamentally, Nietzsche would deny that convergent truth and objectivity are even values, something that James, Peirce and Putnam take for granted. This is what makes his scepticism so radical and so fascinating. Nietzsche doesn't merely question whether nonperspectival truth is attainable; he questions whether the attaining of it would be good for us. His attitude toward truth is similar to his attitude toward Christianity. His point is not that truth would be a good thing if it could be had, but that it is not a good thing whether or not it can be had. The related point against Christianity is not that the values it enshrines cannot be realized, but that these values are anti-life, hence ought not be realized.

2. Nietzsche's Entanglement in Correspondence Theory

If such beliefs as that every event is caused are life-enabling falsifications of the world, this implies that there is a determinate world being falsified and a truth about this world even if inaccessible to us. For if logic, e.g., falsifies the real world, then there must not only be a real world that is being falsified, but also a way-that-it-is-in-itself. If the world in itself were structureless, it could at most have a structure imposed on it; to be falsified, it must have an intrinsic structure that is misrepresented.

Nietzsche therefore appears to presuppose non-perspectical truth in the very formulation of his perspectivism. If so, we are a long way from a coherent formulation of the doctrine that truth in its essence is perspectival.

Note also that without the presupposition of non-perspectival truth, it is not clear that there could be scepticism about the value of truth, which is the characteristically Nietzschean form of scepticism. For the value of perspectival truth is proven by its utility. If this is the only sort of truth there is, what reason would we have to doubt its value for life? Perspectival truth is by definition life-enhancing, answering as it does to our needs and interests. To doubt the value of truth is to doubt the value for life of absolute truth, the sort of truth that is true from no point of view or from God's point of view or from the point of the IRS (the ideally rational subject) at the limit of inquiry. There must be at least the possibility of absolute truth if there are to be serious doubts about its value for life. But if the existence of absolute truth is possible, then truth cannot be perspectival in its essence. Perspectivism, as a claim about the essence or nature of truth, rules out the very possibility of non-perspectival truth.

What torments Nietzsche is the suspicion that truth (absolute truth!) and life are at odds with one another. To live one must be onesided, unjust, partial; to live one must take a position and assert oneself at the expense of others. If so, justice, truth, equality and objectivity are inimical to life. Hence those who champion these ideals slander reality with its Rangordnung (order of rank) and represent a declining form of life. It is precisely the weak who, motivated by ressentiment, "thirst for justice and righteousness" and rail against the powers that be. The strong have no need of reactive attitudes and world-denial since they set the standard and define the good (the noble) (Cf. GM I). This theme runs throughout Nietzsche's writings. It makes sense, however, only on the assumption that there is a way things are independent of our perspectives: there can be no conflict between truth and life if there is no truth. Here we agree with Nehamas as against Danto, Schacht, Brezeale and others. (Nietzsche: Life as Literature, p. 53)

In sum, Nietzsche seems inconsistently both to presuppose and to reject absolute truth. If "the perspective optics of life" necessarily involves a falsification of the world in itself, then perspectivism presupposes for its very sense nonperspectival truth. Likewise, scepticism about the value of truth presupposes for its very sense nonperspectival truth.

3. Self-Referential Difficulties

Another kind of difficulty arises when we reflect on the status of Nietzsche's assertions. Are they intended to be perspectivally or nonperspectivally true?

The claim that truth-as-value, the most refined form of the ascetic ideal, is anti-life purports to be non-perspectivally true, in which case absolute truth is again presupposed. If the claim is intended to be perspectival only, then the ascetic ideal cannot be said to be anti-life inasmuch as it serves the will to power of those who labor in its shadow.

The problem here is that if (human) life is the standard of evaluation, no value distinction can be made between ascending and descending life. Life is life, whether seething with ressentiment or dancing in Dionysian affirmation. Kierkegaard's life is no less lebendig than Nietzsche's. If life is the sole measure of value it becomes as impossible to distinguish higher and lower forms of life as it was for the "blockhead" (Nietzsche's epithet) John Stuart Mill to distinguish higher and lower forms of pleasure on the assumption that pleasure is the sole measure of the good. ("Man does not strive for pleasure, only the Englishman does." TI, "Maxims and Arrows" #12) To the extent that Mill succeeds in distinguishing higher from lower pleasures, he must invoke a standard of better and worse independent of pleasure. (See here for an exfoliation of this problem in Mill.) The same holds, mutatis mutandis, for Nietzsche. The genealogical critique of all values from the point of view of life must presuppose a value standard apart from life; it must presuppose that certain forms of life are objectively better than others. Why is the life of a Goethe better than the life of ascetics, executioners and other subterraneans? The reason cannot be that pessimism and world-denial lack survival-value. Even if that were true, survival is not the issue. Survival has at best instrumental value; it is not end in itself as Nietzsche would admit. Why, to ask a hard question, is life-affirmation better than life-denial? To appreciate the difficulty one must bear in mind that both are forms of life, both are forms of the will to power which cannot do otherwise than will itself. Thus the problem is not to give a reason why life is better than death, but a reason why yea-saying life is better than a life of Christian renunciation or Buddhist non-attachment. The meditating vegetarian may well live longer than he who lives dangerously, building his cities on the slopes of Vesuvius (cf. GS #283) — which underscores that survival and longevity are not what is at stake. My claim, then, is that the genealogical critique of values rests on an absolute value judgment immune to genealogical debunking: the judgment that the life of affirmation is better than the life of denial. So the validity of the genealogical critique presupposes the absoluteness of at least some values. And of course the judgment that affirmation is better than denial is justified only if Buddhism, Christianity, and other forms of "life-denying" metaphysics are false. (Note also that these systems are not life-denying if true; for if true there is a life beyond this life.) Thus Nietzsche is committed to their being non-perspectivally false. He therefore presupposes absolute truth once again in maintaining that this world is all there is. In sum, Nietzsche presupposes both the absoluteness of value and the absoluteness of truth.

4. The Paradox of Perspectivism

The paradox, then, is that Nietzsche is simultaneously committed to both a perspectival and a non- perspectival (absolute) theory of truth. This can also be seen as follows. In the preface to BG & E, after the mischievous suggestion that truth might be a woman, Nietzsche censures Plato for standing the truth on her head by denying "perspective, the basic condition of all life..." But surely it is not perspectival truth that is being turned on her head, assuming that Nietzsche is right about Plato. It is precisely absolute truth that Plato is violating, in which case absolute truth is being presupposed. In other words, the proposition expressed by "Perspective is the basic condition of all life" is intended by Nietzsche to be non-perspectivally true, and Plato's denial non-perspectivally false.

5. Perspectivized Perspectivism

To avoid the paradox of self-reference, we must push the death of God idea all the way to the point where absolute truth and its correlative ethical ideal, truthfulness, are eliminated. The result would be a logically consistent perspectivism according to which there are no facts, only interpretations, and this is applied to the thesis of perspectivism itself. As Nehamas correctly notes, perspectivized perspectivism escapes self-referential inconsistency. (p.66) But what Nehamas fails to realize is that perspectivized perspectivism falls victim to other damaging objections.

Let p be the thesis of perspectivism: All truths are perspectival; equivalently: there are no facts, only interpretations. The consistent perspectivist must say that p is itself perspectivally true. Call this p'. Since p' is true, it must in turn be perspectivally true. Call this p''. It is clear than an infinite regress ensues. But of course not every such regress is vicious. Some regresses are, if not virtuous, at least benign. The corresponding regress of absolute truths is an example of a benign regress: q; it is true that q; it is true that it is true that q;and so on.

Here there is no problem since we need not embark on the regress. q is absolutely true; it stands on its own, so to speak, and the iterations are just along for the ride. But in the regress of perspectival truths, no term stands alone; each refers to the next for completion. Thus, p; it is true for N that p; it is true for M that it is true for N that p; and so on.

In other words, "All truths are perspectival" is elliptical for "All truths are perspectival from N's perspective" just as "It is hot in here" is elliptical for "It is hot in here from N's perspective." Thus the infinite regress of perspectival truths is vicious. For what explains the truth of the first term is the truth of the second term, but the truth of the second term, since it is not absolute, requires explanation in terms of the truth of the third term, ad infinitum. By contrast, the regress of absolute truths is such that no term's truth is needed to explain the truth of any other term. One might wonder about the case in which N = M in the perspectival regress. In this case the regress terminates, but in circularity.

There is a second consideration. Perspectivized perspectivism, since it doesn't assert itself as absolutely true, cannot foreclose on the possibility of there being absolute truths. Given this possibility, perspectivized perspectivism, if true, is at best contingently true, true from some, but not all, possible perspectives. But then perspectivized perspectivism tells us nothing about the essence or nature of truth. But if perspectivism is not a claim about the nature of truth, what is it?

There is a third point. Under dialectical pressure, relativists and perspectivists may retreat to some relativized or perspectivized position; but it is surely not their original intention to put forth some relativized/perspectivized form of relativism/perspectivism. Nietzsche, for example, thinks that he has achieved an insight that Plato and others have failed to achieve, an insight that transcends time and circumstance — and perspective. It cannot have been his intention as a philosopher to say something of the form: "This is my opinion, but since it is merely my opinion, you need not take it seriously." So even if perspectivized perspectivism were free of such logical difficulties as vicious infinite regress, it would be just too uninteresting a thesis to merit philosophical debate.

The upshot is that absolute perspectivism, which tries to say something about the essence of truth, is self-referentially inconsistent, while perspectivized perspectivism, which fails to say anything about the essence of truth, is viciously regressive.

One might try the following maneuver to avoid both the inconsistency of absolute perspectivism and the problems with perspectivized perspectivism. Make a level distinction: truths on the first level are perspectival, those on the meta-level are absolute. Thus one could count "All truths are perspectival" and "There are no facts, only interpretations" as non-perspectival meta-truths. This maneuver is an ineffectual shuffle. For if there are absolute truths on the meta-level, then there is nothing in the concept of truth to prevent there from being some on the first level. If the cut between different levels is to be more than an ad hoc response to logical embarassment, then there must be some principled way of showing that only meta-level truths can be absolute. It is not clear how this could be done. But the problem is even deeper than this. For what do we do with the supposed truths expressed by "God is dead" and "The world is the will to power and nothing besides"? These, if true, are not meta-truths but first-level truths: they are about the world, not about our talk about the world. How can they be segregated in some principled way from perspectival truths? But even more fundamentally, the idea that truths come in two kinds, absolute and perspectival, is arguably nonsensical, involving as it does an equivocation on "true." You may as well say that leather comes in two kinds, genuine and artificial. Artificial leather is not a kind of leather in the way that cowshide is a kind of leather; similarly perspectival truth is not a kind of truth at all. It is obviously not a kind of absolute truth; and there is no genus of truth of which absolute and perspectival truths are species. Here we should be good nominalists: the fact that the word "truth" appears in both "absolute truth" and "perspectival truth" should not lead us astray, anymore than the word "leather" in "artificial leather" and "genuine leather."

6. Perspectivism and Selectivism

The foregoing arguments demonstrate in a negative way the incoherence of perspectivism. If I can defuse the positive arguments in its favor, then my refutation will be complete. What I want to claim is that the positive arguments, at least as reconstructed by Nehamas, are based on a confusion of the false and deservedly controversial thesis of perspectivism with the true but trivial thesis of what I shall call "selectivism." This sort of pattern occurs all the time: by confusing the trivially true with the non-trivially false, one thinks one has gotten hold of a thesis that is both true and non-trivial.

The following passage from Nehamas serves as an excellent characterization of (what I am calling)selectivism.

To engage in any activity, and in particular in any inquiry, we must inevitably be selective. We must bring some things into the foreground and distance others into the background. We must assign a greater relative importance to some things than we do to others, and still others we must completely ignore. We do not, and cannot, begin (or end) with 'all the data.' This is an incoherent desire and an impossible goal. (p. 49)

The very next sentence in the same paragraph begins the slide from selectivism to perspectivism: "'To grasp everything' would be to do away with all perspective relations, it would mean to grasp nothing, to misapprehend the nature of knowledge." (p. 49)According to selectivism, all finite knowing is inevitably selective. If I want to concentrate on the chess game, I must tune out the background noise. If I want a close-up of the moon, I can't train my telescope on my neighbor's bedroom window. So far, banalities that do not in the least support the following claims that are constitutive of perspectivism: (1)It is in the "nature of knowledge" that there can be no grasp of everything; (2) There is no best perspective (p. 49); (3) the "...many points of view cannot be smoothly combined into a unified synoptic picture of their common object." (p. 49)

The negation of each of (1)-(3) is compatible with selectivism; hence selectivism does not entail perspectivism. Surely each of these points is false if there is an omniscient god; and selectivism as a claim about finite knowers is compatible with the existence of such a god. But even if there is no omniscient god, it is possible that there be. Hence each of (1)-(3) is possibly false, even if selectivism is true. It follows that selectivism does not entail perspectivism.

Let us briefly consider (1). What it says is that the "nature of knowledge" rules out the possibility of omniscience. This is far from transparent, and appears to be nothing more than a question-begging assumption. Why should all knowing be patterned on finite knowing? No doubt the nature of finite knowing rules out the possibility of omniscience, but that is trivial. In any case, perspectivism is a claim about truth, not knowledge. Finally, it is unclear how a consistent perspectivist can speak of the nature or essence of anything.