References, below, are to Nadeem Hussain’s paper “Honest Illusion: Valuing for Nietzsche’s Free Spirits,” in Nietzsche and Morality, ed. Leiter & Sinhababu (Oxford: 2007) (cited as HI) and the “Postscript” to the same paper in the same volume (cited as HIP).
Given Nietzsche’s explicit “anti-realism” about value—nothing has value “in itself” Nietzsche tells us—what exactly do those Nietzsche calls on to “create” values understand themselves to be doing? Nadeem’s interesting answer: they are engaged in a kind of make-believe—“regarding X as valuable in itself while knowing that in fact X is not valuable in itself” (HI, 166)—similar to what we find when we become engaged with an artistic work, or when children engage in play. “Nietzsche’s free spirits pretend to value something by regarding it as valuable in itself while knowing that in fact it is not valuable in itself” (HI, 170).
Crucial to the question Nadeem sets is his claim that Nietzsche thinks “all claims of the form ‘X is valuable’ are false’” (HI, 159) and thus Nietzsche is committed to “an error theory about moral claims”: “the beliefs expressed by moral judgments are false because they involve believing in moral facts when in fact there are none” (HI, 159).
But does Nietzsche think such judgments express beliefs, i.e., truth-apt propositional attitudes? That is the key question. Nadeem notes in a long footnote my view that “there are inadequate textual resources for ascribing to [Nietzsche] a satisfying answer” to questions about the semantics of moral claims and thus no “adequate grounds for ‘assigning’ Nietzsche a view on such subtle matters as whether ethical language is primarily cognitive or non-cognitive” (HI, 160 n. 6). Yet to motivate his version of the interpretive question, Nadeem needs the claim that evaluative judgments are to be treated as truth-apt: it is because they are truth-apt, and also all false, that those who create values seem to be in a peculiar situation of making evaluative judgments that they know to be false. (Nadeem’s commitment to this assumption also comes out in the fact that the only alternative readings he considers are cognitivist realisms: the “subjective realism” discussed early on, and the Will-to-Power Interpretation discussed at the end. The criticisms of both are apt, but beside the point, for reasons I’ll suggest, below.) This, in turn, generates the relevance of “the examples of art and imaginative play” which “are, according to Nietzsche, supposed to...show us the psychological possibility of regarding things as valuable even when we know that they are not” (HI, 175).
Because of the centrality to the argument of saddling Nietzsche with a semantics of judgments of value, Nadeem added an interesting “Postscript” to the published version of his article responding to my charge of anachronism (in particular, as formulated in my Stanford Encyclopedia entry on Nietzsche, which discusses Nadeem’s work). Nadeem proposes to adduce evidence of “kinds of fictionalism that were present in the nineteenth century” and also evidence that it is reasonable to ascribe fictionalism, meaning “a denial of non-cognitivism,” to Nietzsche (HIP, 179). It is the second kind of evidence that will be decisive in reply to my objection, since no one, to my knowledge, denies that there were views dubbed “fictionalist” in circulation in the nineteenth-century; the worry (as I put it) is that there are not “adequate grounds for ‘assigning’ Nietzsche a view on such subtle matters as whether ethical language is primarily cognitive or non-cognitive.” To put this in contemporary terms suggested by Matti Eklund, the worry is not about “ontological fictionalism” but about “linguistic fictionalism”: Nietzsche (certainly on my account, and most other ones) has views about the metaphysical status of moral properties, and one intelligible possibility is that he views them as having the same ontological status as fictional entities; the question is whether there is any reason to think Nietzsche has a view about the correct or preferred semantics of linguistic or mental content. Nadeem recognizes, correctly, that his interpretation depends on having adequate grounds for ascribing the latter.
Now Nadeem allows (again, at 179) that he has no evidence that Nietzsche was thinking explicitly about the cognitive versus non-cognitive character of moral discourse, but he claims “the historical evidence does suggest precisely what we need for fictionalism, in the sense that needs to be ascribed to Nietzsche, namely an attitude other than belief [e.g., “make-believe” belief] towards the same content—an attitude such that whether the content is false is no longer relevant” (179). I do not find Nadeem’s historical evidence about competing views about “attitude[s]…towards…content[s]” persuasive in the case of Nietzsche. He adduces strikingly good evidence that Bentham held a linguistic fictionalist view, but none that Nietzsche knew anything about this. When Nadeem turns to the authors Nietzsche did know about, it seems apparent that their views are, at best, versions of ontological fictionalism.
The quotes from Bentham (HIP, 180) are so striking precisely because Bentham distinguishes explicitly “the grammatical form of the discourse employed” from the ontological question of what actually exists, and he endorses, again explicitly, the idea of treating the syntactic entites—e.g., “the noun-substantive”—as genuinely referring expressions, but referring to “fictitious” entities. Here is a kind of self-consciousness about distinguishing the meaning and nature of linguistic items (e.g., are they genuinely referential? If so, to what do they refer?) from metaphysical questions about what really exists. Unfortunately for Nadeem’s argument, there is no evidence—Nadeem does not claim otherwise--that Nietzsche had any familiarity with Bentham’s prescience on this score. More problematically, when Nadeem turns to the 19th-century authors Nietzsche did know something about, they display none of the Benthamite prescience that would warrant ascribing to them linguistic fictionalism.
For example, in the case of David Strauss, Nadeem has good grounds (HIP: 181-183) for saying that Strauss believed religion should be treated as a “myth,” since its claims were false but had some significance when construed metaphorically or mythically. But that view is clearly compatible with differing semantics for religious discourse. One could, for example, think religious discourse is non-cognitive, but that it admits of a metaphorical construal which admits of a cognitive interpretation that does not entail error theory. Or one could think that religious discourse is cognitive, and thus systematically false, but that its metaphorical content is not systematically false. There is nothing in the texts of Strauss to decide between these two subtly different options.
Nadeem’s evidence about Feuerbach and Lange is no better—indeed, some of his evidence creates problems for his preferred view. For example, Nadeem quotes (HIP: 186) Nietzsche reading Lange as follows: “Art is free also in the domain of concepts. Who would refute a phrase by Beethoven and who would find an error in Raphael’s Madonna?” But surely the implication of this comment is that we should treat music and art non-cognitively, as expressing attitudes or feelings of some kind, and thus not susceptible to refutation or error, as they would be if cognitive. When Nadeem turns, finally, to Vaihinger, he effectively acknowledges that Vaihinger has no coherent view about the semantics, when he mentions all the various (and inconsistent) locutions Vaihinger employs (HIP: 187). Nadeem tries to elide this by saying that all the locutions reflect a “concern…to ensure that by changing our attitude we avoid having a false belief.” But that is far too weak for linguistic fictionalism, since there are two ways to avoid having a “false belief” on offer: first, by having a belief that is true (e.g., because it picks out a metaphorical meaning); and second, by not having a belief at all. Vaihinger says things consistent with both possibilities, not surprisingly.
Would it suffice if Nadeem can establish Nietzsche’s commitment to ontological fictionalism? I do not see that it will, since Nadeem consistently casts his interpretive thesis in a semantic idiom and, moreover, he describes his account as attributing revolutionary fictionalism to Nietzsche. But revolutionary fictionalism recommends a revolution in how we conceive of the discourse, i.e., the semantics of the discourse, in order to save it from error theory, and thus ward off the prospect of eliminativism. (If the discourse is cognitive and systematically false, why not just get rid of it altogether?) Given that there is no real evidence that Nietzsche has a clear view about the semantics, it seems extraordinary to think he was recommending a revolution in how we conceive of it.
Nietzsche’s lack of clarity about the semantic content of our judgments about value is reflected in Nietzsche’s texts too, and to an extent that is not really acknowledged by Nadeem. This comes out perhaps most clearly in Nadeem’s critical discussion of Reginster’s book (“Metaethics and Nihilism in Reginster’s The Affirmation of Life”, hereafter, M&N, available on his homepage: http://www.stanford.edu/~hussainn/StanfordPersonal/Online_Papers_files/
HussainReginsterv17.pdf). Nadeem thinks that Reginster’s “nihilism of disorientation”—the disorientation that flows from realizing there are no objective values—involves committing Nietzsche to an error theory, which (as Nadeem candidly admits at p. 7) “is a combination of a semantic claim about what evaluative language purports to be about, namely, objective value facts, and an ontological claim that denies such facts.” Reginster does not frame the nihilism of disorientation this way, and I think he is right not to do so (whatever his unclarities about the notions of “objective” and “subjective” value, with which Nadeem correctly takes some issue). Indeed, Nadeem can adduce no clear textual evidence for ascribing the error theory, and, as in HI, he is silent on the textual evidence in tension with this reading.
So, for example, in M&N (p. 9), Nadeem points to a line from Twilight of the Idols in which Nietzsche declares “there are altogether no moral facts.” Yet, in context, it is clear that Nietzsche is not denying the objectivity of value per se, but rather denying certain descriptive presuppositions about human agency that he takes certain kinds of moral judgments to require. There are “no moral facts” means, e.g., that there is no such thing as a will that is causal, which there would have to be, he thinks, for ascriptions of moral responsibility to be justified. Even if we give that claim a semantic gloss, it would not give us a general error theory about judgments of value.
Nadeem goes on to quote another part of the same passage—in which Nietzsche says “Moral judgments are therefore never to be taken literally: so understood they always contain mere absurdities”—and then says (M&N, 9) these represent “the typical semantic claims of the error theorist.” Yet the claim that “moral judgments are therefore never to be taken literally” could just as well serve as the slogan for the non-cognitivist as the cognitivist error theorist. That Nietzsche is clearly not thinking of the semantic issues seems apparent when we notice that the sentence in question is followed by the claim (not quoted by Nadeem) that the “semiotic” value of moral judgments is as symptoms of “cultures and interiorities.” Indeed, in the very first chapter (“The Problem of Socrates,” section 2) of the same book, Nietzsche tells us that, “Judgments of value…can, in the end, never be true: they have value only as symptoms,” which suggests, to my ear, not error theory, but non-cognitivism.
Consider, too, that in the same and surrounding sections of Twilight of the Idols, Nietzsche repeatedly compares morality to religion, on the grounds that both are committed to the existence of “imaginary causes.” Even if we want to gloss that as an error theory about moral and religious discourse, it would be incredible to think that fictionalism is Nietzsche’s response in the religious case, as opposed to eliminativism. Yet why would the two cases come apart like this? It is more plausible to my mind that Nietzsche simply had not thought, and so had no clear view, about the semantic implications of his thesis.
We may come at the textual problem confronting Nadeem’s ascription of fictionalism to Nietzsche a different way. Nietzsche’s texts, including many that Nadeem quotes, also suggest that the “free spirits” don’t think “having value” means “having value in itself.” The Nietzschean discovery, in other words, is that nothing has value in itself (as he puts it in both The Gay Science [sec. 301] and in Dawn [sec. 3]), but that things do have value, namely, whatever value we project upon them. So evaluative judgments might be cognitive, but they are not false, because they do not involve a commitment to believing that things have value in-themselves. In the Nietzschean world, every evaluative judgment contains within it recognition of what value actually is: namely, a projection.
But we might press this alternative reading in a different direction, one also suggested by the texts. Perhaps the relevant semantics for evaluative judgments in a projectivist world really should be non-cognitivist: moral judgments express our attitudes towards things, and those who “create values” are those who succeed in “projecting” their attitudes on to things (the way, e.g., the slaves in the Genealogy succeeded in projecting their estimation of the masters to the point that even the masters accepted it [to be sure, neither the slaves, nor the masters, recognize the projective nature of value in this story--but unlike “free spirits,” they presumably also don’t recognize that the world in-itself is valueless]). This story certainly resonates with one of Nietzsche’s favorite metaphors for value creators, namely, that they are legislators (Gesetzgeber):
[T]rue philosophers are commanders and legislators: they say “That is how it
should be!”…Their “knowing” is creating, their creating is a legislating, their will
to truth is—will to power. (BGE 211)
Legislators, those who say “this is how it should be,” presumably are not systematically in error: they do not have a false belief that the world conforms to their legislation, but rather express the desire that it should conform to their legislative act—their legislative act expresses their will to power, their will to make the world be as they command it. So legislation might give us a simple, non-cognitivist model of “value creation.”
As far as I can see, the textual evidence adduced by Nadeem is equally well-suited to this reading of the metaethical framework in Nietzsche; indeed, this alternative reading may have the advantage of fitting the “legislation” metaphor more successfully. Because Nadeem is committed, without convincing evidence, to the claim that Nietzsche believes evaluative judgments express beliefs, however, the only alternatives to his reading he considers are cognitivist realisms (such as the “subjective realism” discussed early on in HI). But the real challenges, the one he needs to take up, will come from readings which reject the assumption that Nietzsche is committed to an error theory, either because he is not a cognitivist or because evaluative judgments don’t involve erroneous realistic assumptions about value.
So I am not persuaded, obviously, that Nietzsche is a fictionalist in Nadeem’s sense, but he has posed a powerful challenge to anyone who wants to resist that reading and he has focused scholarly attention on an important interpretive issue that had been relatively neglected in previous work. For all these reasons, HI is one of the most important papers in Nietzsche studies over the last decade.