SciELO - Scientific Electronic Library Online

vol.54 issue128As objeções de Nietzsche ao conceito de coisa em si author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Kriterion: Revista de Filosofia

Print version ISSN 0100-512X

Kriterion vol.54 no.128 Belo Horizonte Dec. 2013 



Nietzsche and transcendental argument



John Richardson

Professor da New York University.




My plan is to examine Nietzsche's view of (what is I think) the most characteristically Kantian kind of argument, what's now often called 'transcendental argument'. I understand this as an argument in which a concept or principle or value is justified as a 'condition of the possibility' of something indisputable (or indispensable). I will look at Nietzsche's critique of this pattern of argument in Kant, but also at the ways he still uses such arguments himself, in all three of the sectors of Kant's critique: theoretical, practical, aesthetic.

Keywords: Kant, Nietzsche, transcendental arguments, possibility-conditions, life-conditions.


Planejo examinar a visão de Nietzsche sobre o que creio seja o tipo de argumento mais kantiano, o que é atualmente chamado de 'argumento transcendental'. Eu o entendo como um argumento no qual um conceito, ou princípio, ou valor seja justificado como uma 'condição da possibilidade' de algo indisputável (ou indispensável). Aprofundar-me-ei na crítica que Nietzsche estabelece sobre esse padrão de argumento em Kant, mas também nas formas em que ele mesmo se utiliza de tais argumentos, em todos os três setores da crítica de Kant: a teórica, a prática e a estética.

Palavras-chave: Kant, Nietzsche, argumentos transcendentais, condições de possibilidade, condições de vida.



1 Introduction

Kant's use of transcendental arguments lies at the very crux of his great philosophical innovation, his 'Copernican' reversal. As I will try to show, Nietzsche crucially follows on the line of this innovation, advances it, but also works great changes in it.

Familiarly, Kant's reversal lies in his idea that 'objects must conform to our cognition' [CPR Bxvi]–not of course quite in general or completely, but for certain very basic structures of our cognition, in particular space, time, and the categories. We impose these structures on the objects of experience, and this is necessary for–is a 'condition of the possibility' of–any experience. Here it is important that 'experience' is meant in some minimalist sense, such that it is quite indispensable for us: we wouldn't recognize as ours something that lacked it. Those structures are preconditions for what we can't do without.1

Kant uses this necessity–of our taking things to be in space, for example–to justify our belief in these intuitions and concepts, as applying to things in our experience. Kant here means to answer the challenge of skepticism, of global doubt: the assertion that all we believe, even about the general character of things, either is or might be false. Kant regards as settled the fact that we are unable to answer such a doubt by a straightforward proof of the belief (e.g. in the reality of space). It's this inability–we can never close the possibility that space is not real–that dictates the transcendental form of argument that Kant innovates.2 We must employ those intuitions and concepts, if we're to have any experience of things at all.

This already suggests the special Kantian ambivalence, the way his idealism finds a middle ground between realism and skepticism: unlike the latter, it lets us know a world of things, but these are things qualified a certain way, not the 'things in themselves' we thought we had been discussing. Our statements are true, and can be known, but they're true about a different order of things than we had supposed. So Kant takes away from the nature of the things we relate to, in order to give us the ability to know more surely the truth about things. Either side of this split point can be emphasized, in response to different objections or opponents: against the skeptic it will be stressed the way we do know, against the realist, the way we know merely us-suited things.

So Kant proves something that is, prima facie, weaker than the old arguments had tried to prove. The difference might be framed either epistemically or ontologically. Epistemically (and I don't think Kant puts it this way), we can say that Kant gives up the aim of proving the truth of certain basic beliefs (or the rightness of certain basic ends) and proves, instead, that it is necessary for us to count these beliefs as true (or the ends as right); so they're 'true for us'. Ontologically, Kant gives up the aim to know objects 'in themselves' or noumena, but shows that we can know (and have the truth about) 'things for us' or phenomena.

To be sure Kant makes an effort to recover from this shortfall (in what he proves), by the further thought that the original epistemic and ontological projects–aimed at in-itself truths and values–were flawed or incoherent. The very aim to know things-in-themselves and not just things-for-us is misguided and even incoherent, so that our failure in those projects is not to be regretted. We're not falling short of anything, or getting something second-best.

But transcendental argument works not just with respect to belief, for Kant, but also with respect to the other two fundamental attitudes he distinguishes, the topics of his second and third Critiques. Certain values are justified as to-be-valued, as right–insofar as our agency deeply presupposes them. And certain feelings are justified as to-be-felt, as fine, because feeling so is necessary if we are to unify our practical and theoretical attitudes. Transcendental argument projects a three-part ideal for our mental stance.

Now Nietzsche makes, of course, many aggressive criticisms of Kant. He (plausibly) takes himself to 'know better' than Kant on several important topics, for example freedom. But there are also some very general features of Kant's position that he takes over, and relies on at the very heart of his position. Most generally, he too–I suggest–is looking for a middle ground between realism and skepticism, a mean that lets him accept parts of both realism and skepticism. He finds this middle ground, moreover, in a place he gets to through Kant. So his philosophical position is quite deeply Kantian. We can now get down to the logic of his sharing and his divergence by focusing on this use of transcendental argument.

I'll start by filling in Kant's use of transcendental arguments a bit more.


2 Transcendental argument in Kant

Here my interest is not in the details of Kant's position, but in certain general features of his use of this argument, which I will be relating to Nietzsche. Highlighting features that bear on this comparison puts Kant's position in a light somewhat different than usual, I think.

The point of transcendental argument is to validate, to single out, to promote. It is a particular strategy for validating some X, by showing X requisite for something one does and must accept. This strategy is applied in three different domains, laid out by our three basic orientations, and treated in Kant's three Critiques. Our basic stances are towards the true, the good, and the fine. Transcendental argument validates certain things as true, as good, as fine, singling out certain beliefs, certain aims, and certain feelings as requisite. Believing or aiming or feeling this way has some privilege, relative to the alternatives, and also relative to the position that no way of believing or aiming or feeling has any privilege.

The general structure of a transcendental argument is:

i. we agree that X is indisputable/indispensable,

ii. but X is only possible on the basis of C (its condition),

iii. therefore we should not dispute C either.

Let's call X and C the 'elements' of the argument, which aims to establish C on the basis of X. So its being the condition of X, justifies the C. Consider some features of this argument.

1) X is the starting-point of transcendental argument, what we agree with/about. This beginning must be something so minimal that any interlocutor will have to agree with it. (So, in a way, the argument tries to 'start at the beginning' as Descartes does.) The weaker and less disputable this starting-point is, the stronger the argument will be. Kant's starting-points are certain essential features of our experience, without which a skeptic's own doubts could not occur. The 'slender premise' that Kant begins his Transcendental Deduction from is 'the self-attributability of mental items, apperception'. By contrast in the Refutation of Idealism the argument begins with our awareness of a temporal order.

Now what kind of 'belief' in these Xs does Kant think is required? On the one hand, there is the overt assent by me (his reader) to a description of my experience, e.g. that in it I count all thoughts as my own. On the other hand, there's what I do within that experience, as I so count my thoughts. Does Kant think that this 'taking as my own' is something that happens overtly or consciously as well? Kant may indeed think that I am conscious of each of my thoughts as my own. In this case the assent asked from the reader is something he/she already affirms all the time. But there's another way of understanding this 'taking as my own', as it works in my everyday experience: that it operates as a background structuring principle, implicit in each and every thought, but rarely becoming explicit (hence 'self-attributable' not 'self-attributed'). In that case, what's needed is a 'deep' acceptance of the point and not an 'overt' one. Indeed, it seems quite possible for me to deny that we do have mental unity, as of course Nietzsche does; he wouldn't also deny that we take ourselves to have such unity. Even when denying that I have it, I still take this thought as 'mine' in that implicitly structuring way.

2) C is the condition we reason to, as being what is required for X. It's what the skeptic is shown he must accept, given that he already accepts that X. In the Transcendental Deduction, the C is a set of a priori concepts, the categories, including substance and cause. In the Refutation of Idealism the C is knowledge of the external world: such knowledge is a condition of the possibility of my awareness of a temporal order.

Now notice, about the C, that Kant takes it to have a different epistemic status from the X, at least initially. It is, after all, something one might try to deny. It's something that skeptics do deny. It's not presumed that these skeptics must accept C by its own claims; this is why the transcendental argument needs to reason them into it, from that firmer beginning they do accept. The argument shows that it's indispensable after all, though not as obviously or immediately so as the X. There should even be a certain surprise (as not for the X), in being shown that C is necessary. C's necessity is hidden in a way the X's is not.

Notice, next, a certain oddness the argument shows, when seen in this light. The argument tries to show that the belief in C is necessary, as a condition for X, which we all agree with. But it needs to show this precisely because C can be, and indeed is, denied by the skeptic in particular. The skeptic denies this and yet it doesn't prevent him from thinking all his thoughts as his own, or from organizing his experience temporally. How can belief in C be necessary for something we constantly do, yet it be possible to dispute about it?

We can answer this puzzle by recalling those deep and overt forms of 'belief' just mentioned. What's necessary, as a condition of the possibility of the unity of apperception, is a 'deep' and structuring use of the categories, not overt awareness of them, or explicit belief in their validity. The skeptic is shown, in effect, that he must already believe in e.g. causality at this deep level (otherwise he wouldn't experience his thoughts as all his). He is encouraged, in consistency, to affirm it overtly as well.

3) Now what is the relation between C and X, exactly? Kant's famous phrase is that C is a 'condition of the possibility' of X. Clearly, he is not talking here about a causal condition or a causal kind of possibility. C isn't needed merely in order to bring X about. Besides the illegitimacy of relying on a causal relation to justify belief in causes, such a causal dependence would be hard to know with the certainty Kant seeks. Instead, the relation of C to X must be somehow logical or constitutive: what it is to treat all of one's thoughts as one's own, is to attribute them to oneself as a subject applying the categories.

4) Because X, thus, depends logically (not causally) on C, our epistemic route from X to C is a priori. There's no need to refer to any empirical evidence, and the dispute with the denier of C does not hang on any scientific findings. It depends on reasoning alone, by a path the interlocutor (the skeptic) can't dispute, Kant thinks.

5) Consider further how, at least, the C and, probably, the X, as well, occur in both overt and deep forms in this argument. I think this is a crucial aspect of Kant's argument, which is not often noticed, but that emerges through the comparison with Nietzsche, who puts much more explicit weight on the distinction. The purpose of the argument–answering skepticism or doubt–requires this distinction in levels: the doubt in C is overt and is overtly answered by showing that the deep belief in C is indispensable. It's indispensable because it's required for (I suggest) the similarly deep belief in X, which we acknowledge we can't do without.

We should notice a certain limitation to Kant's transcendental arguments, which will be important to us later. This limitation was pointed out by Stroud in a well-known paper [1968]. Stroud argues that transcendental arguments are unable, on their own, to defeat external-world skepticism, because the necessary condition it justifies is not the existence of external things, but our belief in their existence. Stroud suggests that a transcendental arguer can bridge the gap between belief and things only by a verificationism or an idealism. Kant's way of bridging it is idealism, as we've noted: our deep beliefs structure the things we experience, so that the latter are phenomenal not noumenal, they are 'things of experience'.


3 Nietzsche's revision of transcendental argument

Nietzsche operates within a very broadly Kantian framework. He is a modern, in the philosophical age founded by Kant. So he is constantly struck by how we 'construct' the things we experience and make theories about. Indeed, Nietzsche is notable for how vastly he multiplies the ways we so make things. His wide-ranging psychology diagnoses that, one way after the other, our view of things has been steered by interests other than truth and objectivity. They suggest that, in wide-ranging respects, our 'truths' are 'for us' or subjective.

Sometimes these diagnoses seem to lead him to skepticism. But more often, he continues to express his own views about things–about human things in particular. We must suppose that he states his own views 'in the light of' the subjectivizing diagnoses. Nietzsche's truths are also meant as subjective, not reaching things as they are in their own right, but treating things 'for' a certain perspective. He's talking about 'phenomena not noumena'–but which phenomena? How are they different from Kant's phenomena?

We can answer these questions by examining Nietzsche's revision to Kant's transcendental arguments. And I think our quickest way into his shift is by examining how Nietzsche converts Kant's possibility-conditions into life-conditions.3 Nietzsche makes this change with reference to those very categories that are Kant's Cs: instead of being possibility-conditions, beliefs in substances and causes are existence-conditions for us. He harps on this point, so often, that it's a plausible clue to what he thinks his own innovation is. Consider e.g. KSA.12.352 (WP.507):

preservation- and growth-conditions [Erhaltungs- und Wachsthums-Bedingungen] express themselves in valuations / all of our knowledge-organs and –senses are developed only with regard to preservation- and growth-conditions / trust in reason and its categories, in dialectics, therefore the valuation of logic proves only their usefulness for life, proved through experience: not their "truth".4

There are a number of terms he uses for this point; I think they are, for the most part, interchangeable, though we'll need to come back to talk about nuances. Let me list a few, with occurrences in his published works: Life-condition [Lebens-Bedingung]: BGE.188, A.25. Lebensbedingung: GS.110, BGE.4, BGE.62, BGE.268, BGE.276. Bedingung des Lebens: GS.110. Existence-condition [Existenz-Bedingung]: GS.1, GS.7, GS.335, EH.iv.4. Existenzbedingung: GM.i.10, EH.iv.4. Preservation-condition [Erhaltungs-Bedingung]: A.26, EH.iii.BT.3. Erhaltungsbedingung: A.16. Growth-condition [Wachsthums-Bedingung]: BGE.188, A.25. These uses of these terms in his books are the iceberg-tip to a great many more uses in his notebooks.5

Let's go back through the 5 features we treated in Kant's transcendental arguments, to see how Nietzsche's shift to life-conditions changes them.

1) X, what the condition is of. For Kant, this was some pervasive and indispensable feature of our experience, something even the skeptic must acknowledge: e.g. that I take my experience to be unified, or that I structure it in time; the skeptic can't be a skeptic without doing this too. For Nietzsche, I suggest that this X is 'life'. This is, in different ways, the starting-point for all of his arguments. It is, he thinks, our basic allegiance, which we can't forswear. So it's what clinches most of his arguments, for or against: that the thing advances or hinders life.

As we will see, Nietzsche builds a lot into his idea of life. Different aspects of it come into play in his different versions of transcendental argument. But we should notice, from the start, that he thinks of life as, in a certain way, reflexive or self-referring: life essentially aims to preserve itself, or, Nietzsche thinks better, to enhance itself. Life wants more life. Of course he calls this its 'will to power'. Living is not just a matter of surviving, of maintaining the status quo, but it is a competitive flourishing that grows.

Sometimes, Nietzsche's arguments rest on life in the abstract, but more often he has in mind some particular kind of life, seeking to preserve, but even more, to enhance itself. The conditions are factors that allow a particular kind of life to flourish and grow (and not merely to survive, to keep barely alive). As we'll see, Nietzsche thinks these kinds of life have both ontological priority for us–they're what we essentially are–and also valuative priority–they set our ultimate ends.

2) C, the condition. Given what we've seen of X, this C is no longer (as I'll put it) a possibility-condition, but now a life-condition. It is what makes possible a particular kind of life–what enables it to flourish. Nietzsche is not interested in physical structures such as teeth or muscles, but in behaviors, and our dispositions to them, which he calls 'drives'. Above all, he is interested in the 'learned' dispositions inculcated in us by culture and upbringing. We acquire these especially by absorbing 'values' in linguistic formulations. Still more narrowly, among values he is most interested in 'virtues', i.e. values setting our conception of the self we're trying to be.6 Certain virtues are requisite for certain kinds of lives–this is the conditionality he's most interested in.

We should notice that Nietzsche also makes another use of 'condition': sometimes he uses Bedingung to refer to 'environmental' circumstances, including especially the strengths of enemies or rivals. And his main point here is the way these determine what virtues a group or kind of person–with its distinctive 'life'–requires. These environmental conditions set virtues their assignment, as it were. It's the struggle against adverse conditions that most strengthens virtues: BGE.262: 'A kind comes to be, a type becomes fixed and strong, through the long fight with essentially constant unfavorable conditions.'7

3) C's relation to X. Now in what way is a life-condition a 'condition'? What is the nature of the step back from the X to this kind of C? The relation is not logical but biological, and Nietzsche understands this in broadly Darwinian terms. So life-conditions are the drives and values that enable an organism to thrive; they are there in the organism because they so allow it. Thus, the relation is not just causal, but doubly causal:

i) the condition tends to cause/promote life

ii) its causing/ promoting life causes it to be there

That is, the C is a 'function' not just in the propensity (forward) sense, but in the etiological (backward) sense too: it explains why the feature is there. Nietzsche usually means that the drive or value evolved in order to play that role.

Now is the C still 'necessary' for X? In that case a life-condition would be just a kind of possibility condition: without it, the X could not be. The issue of necessity is important and difficult. It's clear, at least, that the necessity is not unconditional, but relational to a particular kind of life.8 And in many cases of biological function, even conditional necessity does not hold. In other cases, it's unclear, and experiment may be needed to test whether some C is necessary, or whether different drives or values might enable a kind of life to flourish as well or better.

4) The route (argument) from X to C. Since the step back from life to its condition purports to be doubly causal, it can't be established by logical argument, and can't be a priori. It needs support by evidence of this causality. Since Nietzsche is most interested where the Xs are person-types and the Cs are certain ways of viewing/valuing, the causal relations are principally psychological, and need to be grasped first-personally. Nietzsche thinks he can carry out experiments, and gather evidence, all by himself, by 'entering' ways of living imaginatively and seeing how they require certain view/values. So these Cs can be discovered and studied almost as much 'from the armchair' as Kant thought his possibility-conditions could be.

5) The distinction between deep and overt levels. Like Kant's possibility-conditions, Nietzsche's life-conditions are deep, structural ways of comporting; they are ways we believe/value/feel implicitly and more or less constantly. It is precisely the deep forms of these beliefs that matter, since these (and not any overt beliefs) are requisite for those indispensable Xs. It's the way we are believing 'all along', in our deep assumptions of e.g. space, time, and the categories, that is a possibility-condition (for Kant) or a life-condition (for Nietzsche).

By contrast, the challenge of skepticism concerns what we should believe in a different way: within our avowed and conscious viewpoint, in how we think and speak. As the example of the skeptic shows, belief in categories such as causation is not necessary at the overt level. But the skeptic continues to believe in causes deeply, where he counts himself a self or agent who holds all these skeptical views. How does Kant convince the skeptic to make his overt beliefs match his deep ones? Since his transcendental argument reveals only the deep belief as a possibility-condition for apperception (e.g.), he needs a supplemental argument to pass from the necessity of the deep belief to the preferability of the overt belief.

My sense, which is correctable, is that Kant is not much concerned about this step; he hardly notices it. And we tend not to notice it either, which is one of the advantages of the comparison with Nietzsche, who is very concerned with it. I think there are a couple assumptions that make the step look short and easy to Kant. A first assumption is that the skeptic will wish to be consistent: if he can't but believe at the deep level, shouldn't he match this in his overt beliefs? A second and more important assumption: Kant thinks the transcendental argument shows that deep belief in causes etc. is true–because it's about not things-in-themselves, but about phenomena. And, of course, the skeptic wants his overt beliefs to be true, as well.

But, for Nietzsche, the gap between deep and overt belief is much more problematic. He expresses doubts about both of Kant's reasons to bridge it. He rejects the demand for consistency, it seems, and doesn't much care whether one contradicts oneself. Why not believe different things overtly than we must believe deeply? Why need these match up? Nietzsche stresses that we can't help but depend on lies. For those deep beliefs that are life-conditions for us, are false for all that. So why can't we continue to take all these things as true implicitly or subliminally (if we must), but explicitly recognize their falsehood? Why should we believe them overtly, if it's only deep belief in them that is necessary? It is an important issue for him how insulated from one another our deep and overt beliefs can/should be; he calls this the problem of 'incorporation'.


4 Argumentative force

But we have not broached a crucial question about Nietzsche's change to the Kantian argument: how does it affect its argumentative force? When he converts Kant's possibility-conditions into life-conditions, does this leave it still an argument, and if so an argument for what? And if it is still an argument is it still 'transcendental'?

This depends, of course, on what we'll call an 'argument'. I suggest: a multipart and sequential reason for believing something (or for willing something or feeling something).

A first hard problem is to say what a 'reason' is. I suggest, for a start, that a reason is addressed to an entity that is considering or trying to decide what to believe in; it is offered for that entity's inspection and assessment, as trying to believe 'as it should'. Argument differs from propaganda, which tries to bypass this reason-weighing capacity, and to induce belief by using/appealing to different factors in the other, such as his/her unconscious drives. And the reason is transcendental if it claims the belief to be required by the very structure of the experience of any hearer/reader.

A second hard problem is whether an argument must–to be an argument–commend its to-be-believed idea as true, i.e. (as I'll put it) must offer an epistemic reason for these beliefs. The central issue with Nietzsche is whether he gives such a reason. At first appearance, his main point about life-conditions seems to be that they lack epistemic reasons and rest only on (what we might call) pragmatic ones.

For, on a first view the impact of Nietzsche's change, it looks obvious: it's to spoil Kant's transcendental arguments, destroying their ability to justify those beliefs (e.g. in causation), at least as true. When possibility-conditions become life-conditions, they are revealed as mere contingent causes of particular forms of life. The categories, for example, are just what we have had to believe in order to develop our distinctive human life. But, Nietzsche insists, this does not show these beliefs to be true. GS.121: 'Life, not an argument. –We have arranged for ourselves a world in which we are able to live–by positing bodies, lines, planes, causes and effects, motion and rest, form and content; without these articles of faith no one could endure living! But that does not prove them. Life is not an argument; the conditions of life might include error.'

Indeed, if anything, Nietzsche suggests, this conversion into life-conditions reverses the argument's force, making it count against the C's truth. By diagnosing these beliefs as life-conditions, it shows that we don't hold them on epistemic grounds; it exposes our lack of reasons to believe them. It explains these beliefs as established in us for quite non-truth-seeking motives, and this tends to deflate them.

This verdict against the Kantian argument-form begins with Nietzsche's recognition of Stroud's point (in §2) that the argument can carry us only to belief: it shows we must believe in e.g. the categories, not that they're true. 9 It leaves the categories quite unproven. KSA.11.168: 'The law of causality a priori –that it is believed–, can be an existence-condition of our kind; thereby it is not proved.' And BGE.11:

it is high time to replace the Kantian question, "How are synthetic judgments a priori possible?" by another question, "Why is belief in such judgments necessary?"–and to comprehend 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! Or, speaking more clearly and coarsely: synthetic judgments a priori should not "be possible" at all; we have no right to them; in our mouths they are nothing but false judgments. Only, of course, the belief in their truth is necessary, as a foreground belief and visual evidence belonging to the perspective optics of life.

Here the point seems to be not that categorial judgments are false but that they might be.

But Nietzsche further holds that these beliefs are not just unproven, they are indeed false. BGE.4:

we are fundamentally inclined to claim that the falsest judgments (which include the synthetic judgments a priori) are the most indispensable for us; that, without accepting the fictions of logic, without measuring reality against the purely invented world of the unconditional and self-identical, without a constant falsification of the world by means of numbers, man could not live–that renouncing false judgments would mean renouncing life and a denial of life. To recognize untruth as life-condition ...

He gives a positive account of the true character of the world the categories are false about. KSA.12.353 (WP.507): 'we have projected our preservation-conditions as predicates of being in general / because we need to be stable in our believing, in order to thrive, we have made the "real" world one not changing and becoming, but being.' He purports, it seems, to know this different character of the world on independent grounds.

All of this suggests that Nietzsche thoroughly rejects and abandons transcendental argument, and yet, this is not the case. Although he, sometimes, takes beliefs to be thus de-validated by being shown to be life-conditions, at other times he thinks this validates them. And we must count this as more important than his merely critical point. Indeed I think this argument via life-conditions may even underpin his positive account of the world as becoming, so that his grounds for this aren't fully independent after all.

At issue is whether Nietzsche thinks life-conditional beliefs are, as such, true. It's clear enough that he thinks many such Cs are to-be-believed on pragmatic grounds, i.e. as needed for a certain kind of life (that we want). But this is quite consistent with the argument also devalidating these beliefs epistemically. Indeed, these together compose Nietzsche's well-known view that it may be advisable and even necessary to believe falsehoods.

So the question is whether Nietzsche also uses this argument to justify some Cs as true. I think he does, and indeed on a similar ground to Kant's. Kant supposes that what we must believe in order to have experience at all amounts to a truth about things–that is, about things 'of experience'. If certain beliefs are basic enough conditions–conditions for having beliefs at all–they belong as it were not to the content of our views but to their very form or structure: they help constitute what it is to be an object (of experience) at all. To be sure, this is a secondary or limited kind of truth, because it's about a special kind of object, an object 'for us'. But it's a truth nevertheless, and Nietzsche, following Kant, does sometimes hold that life-conditional beliefs are thereby true.

But how can Nietzsche have it both ways? How can these arguments both validate and de-validate their Cs as true? He would need to be operating with different notions of truth, but what are these, and which is primary for him?

Clark's well-known position [1990] is relevant here. (I will offer perhaps only a caricature of it in order to locate it quickly within my problematic.) She argues that Nietzsche abandons his early denial of truth because it has rested on a 'metaphysical correspondence theory of truth', which understands truth as correspondence to things-in-themselves. He gradually realizes that the notion of a thing-in-itself is incoherent, and that truth is dependent on our 'cognitive interests'. Clark calls this Nietzsche's 'neo-Kantian position on truth', and thinks that he expresses it metaphorically in his so-called 'perspectivism' [1990 135].

Clark thus distinguishes two notions of truth which she claims Nietzsche holds in sequence. I concur that there are two notions, though I think Nietzsche holds both in mind right to the end. I also concur with Clark that these notions interpret truth, respectively, as being about an in-itself, and as perspectival. But I believe I read Nietzsche's point about perspectives as less metaphorical than Clark treats it. I think he has an idea of 'perspectival truth', whose logic gives its own ground for thinking that even our best truths are, nevertheless, still false.

In his use of the term 'truth' as with many others, Nietzsche shifts freely between hearing it to mean what it has meant, and what he thinks it should mean. The so-far meaning of 'truth' does take it to be of or about an 'in-itself': things as they are in their own right and independently of us and our viewpoint on them. Nietzsche, we've seen, makes the Kantian move of analyzing our beliefs to depend, one and all, on certain structuring beliefs that all our truths presuppose. So our truths are all internal to our human perspective. The new notion of truth will do justice to this perspectivity.

And this is the new positive role of his arguments about life-conditions: they show what's 'true for' certain perspectives. When we see that certain beliefs are necessary for a certain kind of life, this shows it as perspectivally true, as true for that perspective, but at the same time as not true. It is not true not just because it is not about an in-itself (an illegitimate though prevalent standard), but by the standards of the new perspectival truth itself. For this truth-for-us is also false-for certain higher perspectives, which we must always remember there are. As we might put it, truth is now a scalar notion, not all-or-nothing.

It's important to remember that these life-conditions are deep and structural beliefs, not overt ones. This is an important continuity with the Kantian argument-form. Nietzsche's descendent argument justifies the 'lived' belief in C–a belief incorporated into our routine bodily responses. One's theoretical position–what one overtly thinks and affirms–can be anything at all, so long as it does not spoil the deep faith in C, as it covertly structures one's overt beliefs. It's an important question for Nietzsche, in what cases our conscious beliefs, the ones we put into words to ourselves and others, can indeed affect–and perhaps undermine–deep beliefs opposed to them.

I think there are two rather different ways in which Nietzsche uses this argument about life-conditions. I'll call these 'confirming' and 'revising'.

a) Confirming: the condition is justified by being requisite for a kind of life we can't help but live. Here the X is some very general kind of life, such as human life, or even life itself. And the C is something we've believed all along, what we can't help but believe. This form comes closest to the Kantian argument format.

b) Revising: the condition is justified by being requisite for a kind of life we aspire to. Here the X is the higher human, conceived in particular (when we regard belief) as having as much of the truth about things as possible. And the C is a condition that we try to foster, for the sake of this advance. This form looks more like a pragmatic argument, but it keeps the focus on 'deep' beliefs from the Kantian pattern.

a) We've seen that Nietzsche rebuts Kant by insisting that the categories, exposed as life-conditions, are false. But I think this verdict partly relies on a deeper use he makes of Kant's argument-type, a use in which those deeper life-conditions are validated as true. Kant's Cs, the categories, are conditions for our specifically human life–for our subjectivity or agency, and its ultimately scientific stance towards the world. Nietzsche stresses the falsity of these deep beliefs, but they are false not in relation to an independent reality, but by their conflict with still deeper structuring beliefs–built into us as animals and organisms. It's these more deeply necessary posits that are true–and true because they're so deeply structuring.

Our deepest structuring beliefs are those that belong to us simply as alive. Nietzsche puts great weight on this point of view of (mere) 'life', and what its conditions are. Life (any living thing) understands itself as a will, in a world of other wills with which it competes. This understanding lies at the bottom of us, as a condition for being alive at all. It's a condition for having a perspective at all, Nietzsche thinks, and hence too for anything–any kind of world–appearing to a perspective. And so this is what we are (I claim he thinks): what we fundamentally take ourselves to be.

This I suggest is the essential structure of the world as Nietzsche conceives it: a world of competing wills, each with its interested perspective on the others, and standing in certain balances of power towards them. The categories are false insofar as they conflict with this deeper constitutive understanding. They are false to the world as life must find it; they are contradicted by 'the perspective of life'. So his positive picture of the world rests on a kind of transcendental argument. Nietzsche is driven to such argument by his own strong skeptical tendencies, which rule out (as they did for Kant) a more direct access to the character of things.10

But Nietzsche has a further thought regarding the categories. Although they are false in relation to this privileged perspective of life, they are also 'true for' the human perspective they help to constitute. And such 'truth for' is, after all, the only kind of truth there is–a perspectival truth about things structured or constituted by perspectives. Truth is (as it were) 'internal' to perspectives. And although what is true for our agency may be false for our bodily life, it is obviously an enormously important truth, still. It underlies all of the more particular conditions for particular kinds of human lives.

b) Let's turn now from these 'universal' life-conditions–beliefs that are requisite for any human life–to the Cs for particular kinds of human lives. Nietzsche has a different kind of interest in these, because they are, potentially, revisable. Regarding these conditions for some human lives and not others, it matters above all what kind of life the deep belief is a condition for. And it matters whether that kind of life is (or should be) our goal or is, instead, undesirable. (EH.ii.BT.2: the lie is one of the preservation-conditions of decadents.)

The argument is no longer transcendental, in that it no longer deduces what we must believe for a kind of experience we do and must have. Still, it retains something from Kant's argument-pattern: the focus on a deep and structural kind of positing, and the effort to align our overt and conscious beliefs to it. But now this deep belief is something we're hoping to inculcate, and overt belief aligns with it in order to bring it about.

Nietzsche thinks that these life-kinds are determined by structurings that are principally society-wide. These structuring beliefs for life-kinds are reflected in the moralities preached and practiced in societies.11 GS.116: 'since the conditions for preserving one community have been very different from those of another community, there have been very different moralities'. A morality is an attempt to embed certain practices in the community's members' deep structure. And to the extent that it does, it becomes a structuring truth for their kind of life. In being passed from generation to generation, Nietzsche thinks, a morality is embedded more and more deeply in the very bodies of members. GS.110: 'the strength of knowledge lies not in its degree of truth, but in its age, its embeddedness, its character as a condition of life [Lebensbedingung]'.

So a morality expresses, and tries to embed, the conditions for a particular kind of life. But this morality can be transmitted to persons who either can't or don't want to live that kind of life. As it persists over generations, and is embedded more deeply, it paradoxically becomes less and less apt for changing circumstances; its values 'outlive their usefulness', and lose their fitness with life-conditions.12 KSA.12.468 (WP.110): 'feelings about values are always behind the times, they express preservation-and growth-conditions from a much earlier time; they resist new existence-conditions with which they cannot cope'. This way a morality can cease to reflect life-conditions is a key problem for Nietzsche.

Because the kind of human life we live may be alterable, it's important to understand not only the Cs presupposed by the life we do live, but those that would make possible a life better than this one. Because these conditions are deep beliefs (and aims and feelings), they can't be simply adopted by choice, but must be somehow trained and inculcated in us. We can acquire certain values as explicit rules, but the challenge is to 'incorporate' these values by building them into our implicit responses. This is, indeed, the task of the new philosophers. KSA.12.424 (WP.898): 'That which partly necessity, partly chance has achieved here and there, the conditions for the production of a stronger type, we are now able to comprehend and consciously will: we are able to create the conditions under which such an elevation is possible'.13

So even here we find Nietzsche using a descendent version of Kant's transcendental arguments. Now the X is not something already in place–our 'life', or our 'human life'–but a life we aspire towards, a 'higher life'. KSA.12.462 (WP.866) speaks of the aim 'to bring to light a stronger species, a higher type that has different Erstehungs- and Erhaltungsbedingungen than the average-human.

But we don't necessarily set our sights first on a 'higher life' and reason back from it to the conditions it requires. Our idea whether a life is 'higher' will partly depend on what its conditions are. It will depend in particular on whether its life-conditional beliefs are true. So EH.iv.4 says that to evaluate a type of human one must know his Existenzbedingungen, but that the condition of the (morally) good is the lie. The strongest are those who are able to bear to see the most clearly, and who are able to 'incorporate' these truths into their deep viewpoint. Of course Nietzsche also expects there are limits to just how much of the truth can be thus incorporated. But he admires the effort to push up against this limit again and again.


5 Applications to belief, value, feeling

The more interesting of Nietzsche's uses of Kant's argument-pattern are the revising ones. This is where he tries to change us (rather than just to show us what we most deeply must be). He tries to change us in all three of the dimensions we've seen Kant applies his transcendental arguments to: (what I'm calling) believing, aiming, feeling. I've focused so far on how the argument applies to belief, but this isn't the most important aspect of us, according to Nietzsche. Let's consider–and I'll be very brief–how he uses this argument in all three domains. The point is to incorporate a new deep structuring not just of our beliefs, but of our aiming and feeling–what Nietzsche calls our 'will' and our 'affects'.

Let's remind ourselves how Kant's arguments work in these three dimensions. Each of the three Critiques tries to privilege a particular way of standing in its domain. So the first Critique tries to justify thinking in terms of an ontology of substances, causes, etc., as requisite for any empirical knowing. The second Critique justifies aiming/acting according to the categorical imperative, as necessary for agential freedom. And the third Critique justifies our feeling for beauty and sublimity, as necessary to integrate our theoretical and practical commitments. In each case justification works through the claim that thinking-aiming-feeling (at the deep level) so is a 'condition of the possibility' of having that experience at all.

Nietzsche wants to be not just a 'critic', and ultimate confirmer, as he thinks Kant is. He wants to find not just what we must deeply think/will/feel, but what we, improvingly, can. Philosophy's task is to 'create values', and it does so with the hope that these values will be 'incorporated' at the deep level of our aiming. Philosophy's values are experiments whether this new structuring of our willing makes humans stronger and better. Values are so important to Nietzsche because they are means of effecting change in our willing/aiming, our most important dimension.

What deep aim does Nietzsche propose the higher human needs? First, I suggest, a deep will for his/her own power, recognizing this as a self-overcoming that leaves behind lower stages of itself. KSA.12.278 (WP.326) says that virtues should be adopted 'as conditions of, precisely, our own existence and growth, which we recognize and acknowledge independently of whether other men grow with us under similar or different conditions'. But Nietzsche has a particular idea how a higher human will pursue power: by maximizing his/her freedom, through self-diagnosis and self-overcoming. We may understand this as a variant on the freedom Kant thinks we find through acting from the categorical imperative.

And what deep feeling does Nietzsche think the higher human needs? I suggest: a capacity for a particular version of the creative feeling of Rausch, one that feels to be beautiful even despite the ugliness it uncovers. This is the feeling expressed in the supreme moments Nietzsche was so intent on, in which one 'says Yes', and 'loves fate', and welcomes the return of even the worst. The higher human Nietzsche anticipates needs this deep feeling for the same structural reason Kant assigns to aesthetic judgment in the third Critique: it holds together the other two deep stances. The higher human must reconcile his/her will to lay bare the psychological truth about human phenomena, with a healthy 'will to power' that is not disheartened by the ugly truths uncovered.14 Nietzsche offers that feeling of the universal fineness of things to heal the will, after the surgery of psychological diagnosis. It is a condition for living freely with the truth.

But of course there's this huge difference from Kant's transcendental arguments: Nietzsche justifies this deep thinking/willing/feeling as needed for this 'higher human', i.e. for a goal that needs itself to be justified. By contrast the transcendental argument carries its whole justification within: the deep Cs are requisites for what we are and must be. Where will Nietzsche get this independent justification for that goal from? I think he gets the elements of it from his 'universal' arguments about life-conditions. A deep will to power is built into us just as animals and organisms; a deep will to truth is built into us as humans. The tension and struggle between these is the great challenge we face. Nietzsche justifies his ideal for us as reconciling these conflicting wills as it was never possible before. And he justifies those wills by his use of a version of transcendental argument.



Clark, M. 1990. Nietzsche on Truth and Philosophy, Cambridge: Cambridge University Press.         [ Links ]

Constâncio, J. 2011. 'Instinct and Language in Nietzsche's Beyond Good and Evil', in (eds.) Constâncio and Mayer Branco, Nietzsche on Instinct and Language, Berlin: de Gruyter, 2011.         [ Links ]

Schacht, R. 1983. Nietzsche. London: Routledge & Kegan Paul.         [ Links ]

Stroud, B. 1968. 'Transcendental Arguments', in Journal of Philosophy LXV pp241-56. Repr. in (ed.) R.C.S. Walker, Kant on Pure Reason, Oxford: Oxford University Press, 1982.         [ Links ]



Artigo recebido em 12/07/2013.



1 So Pereboom on 'Kant's Transcendental Arguments' [Stanford Ency.]: such argument 'begins with an uncontroversial premise about our thought, experience, or knowledge, and then reasons to a substantive and unobvious necessary condition of this premise'.
2 Bxvi: 'try whether we do not get farther with the problems of metaphysics by assuming that the objects must conform to our cognition'.
3 Schacht [1983 162] recognizes how Nietzsche diagnoses categories as 'conditions of life for us'. See too Constancio [2011 113].
4 KSA.9.501: 'The achieved similarity of experience (about space, or the feeling of time, or the feeling of large and small) has become a life-condition of our genus, but it has nothing to do with the truth.'
5 There are over 50 occurrences of either Existenz-Bedingung or Existenzbedingung in KSA volumes 9-13.
6 A.16: 'A people that still believes in itself will still have its own god. In the figure of this god, a people will worship the conditions that have brought it to the fore, its virtues'.
7 See also GM.ii.16. KSA.12.132 (WP.856): 'war and danger the presupposition for a rank to hold fast its conditions'.
8 KSA.13.334 (WP.515): 'the categories are "truths" only in the sense that they are life-conditioning [lebensbedingend] for us: as Euclidean space is a conditioning "truth". (Between ourselves: since no one would maintain that there is any necessity for there to be humans, reason, as well as Euclidean space, is a mere idiosyncrasy of a certain kind of animal, and one among many'.
9 KSA.12.352 (WP.507): 'In valuations are expressed conditions of preservation and growth. All our organs of knowledge and our senses are developed only with regard to conditions of preservation and growth. Trust in reason and its categories, in dialectic, therefore the valuation of logic, only proves their usefulness for life, proved by experience–not that something is true' ... 'what is needed is that something must be held to be true–not that something is true' ...
10 Nietzsche uses a parallel argument to justify certain values: he argues e.g. in favor of aggression and even violence, that these are conditions built into life.
11 KSA.12.260 (WP.261) says that study of peoples shows that the criterion of a moral action is not disinterestedness or universality, but 'a belief "such a behavior belongs to our first existence-conditions"'. KSA.12.437: 'the affects and fundamental drives in every race and class express something of their existence-conditions (–at least the conditions under which they have prevailed for the longest time)'.
12 KSA.12.207 (WP.953) speaks of the change when humans acquired leisure to cultivate themselves into something higher: 'then a host of virtues are superseded that had been existence-conditions'.
13 Here Nietzsche claims another difference from Kant: he creates new values and, thereby, a new kind of life, whereas Kant was only a 'critic'.
14 Art has played this role before. GS.107: 'had we not approved of the arts and invented this type of cult of the untrue, the insight into general untruth and mendacity that is now given to us by science–the insight into delusion and error as a condition of cognitive and sensate existence–would be utterly unbearable. Honesty would lead to nausea and suicide.'

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License