"What a sentence! Is it a sentence?"

It seems to me that Derrida is paranoid about tone (a writing's attitude towards its readers). His entire method of inquiry, insofar as it can be rigidified into "moves," or "rules," or (in a sightly richer formulation of the same thing) ethical demands for specific actions (see The Gift of Death for an articulation of how tough these alleged "moves" can be), could be said to be the emptying out of tone. "Emptying out" means here however, at the same time, an infinite expansion of the reliance on tone: tone is everything and nothing at once. In other words, Derrida's project can be seen as a giving of thickness to the idea of tone, and trying to overcome our reliance on it as a positive tool that forwards (or--and this is infinitely more pressing for Derrida--shuts down or dismisses the other's) discourse.
For example, the question of Derrida when reading Beyond Good and Evil is first and foremost one that problematizes or develops tone. "What a sentence!" he exclaims in The Politics of Friendship. But then he asks and adds (that is, supplements) the really crucial part: "Is it a sentence?" In other words, is this passage in Nietzsche merely a fragment and nothing more? Is it--and here the real inquiry begins if one follows Derrida carefully--merely an emphatic gesture, hot air, meant to rile up one's spirits more than communicate or remain communicable? Or is it precisely thereby communicating? In other words, is it resisting being said, being stated in a sentence--is it just remaining a fragment of tone--or is the tone precisely what makes it a sentence, brings it into the field of the possibility of representing? Here is the full passage (the, indeed amazing, Nietzsche included):

The transmutation to which Nietzsche submits the concept of virtue--sometimes, as has often been remarked, also in the Machiavellian sense of virtù--shudders in the tremor of this perhaps. This is something other than a reversal [or, as Foucault would say, a sacrifice--mj]. The famous passage on "Our virtues" (para. 214) from the same book [Jenseits--mj] turns resolutely towards us, towards ourselves, towards the "Europeans of the day after tomorrow" that we are, and, first of all, towards the "first-born of the twentieth century." It invites us, we the "last Europeans," to be done with the pigttail and the wig of "good conscience," the "belief in one's own virtue (an seine eigne Tugend glauben)." And here again, the shudder of the sentence, the shudder of an arrow of which it is still not known where and how far it will go, the vibration of a shaft of writing which, alone, promises and calls for a reading, a preponderance to come of the interpretative decision. We do not know exactly what is quivering here, but we perceive, in flight, at least a figure of the vibration. The prediction: "Alas! if only you knew how soon, how very soon, things will be -- different! --" ( -- Ach! Wenn ihr wußtet, wie es bald, so bald schon -- anders kommt!).
What a sentence! is it a sentence? Do we know that -- that things will be different; and how very soon things will be different? Do we not already know that? Can that be measured by knowledge? If we knew that, things would no longer be different. We must not totally know this in order for a change to occur again. So, in order for this knowledge to be true, to know what it knows, a certain non-knowledge is necessary. But the non-knowledge of the one who says he knows that we do not know ("Ah if you only knew," a ploy or a figure which is neither a question or an affirmation, not even a hypothesis, since you are going to know very soon, starting at the end of the sentence, that which you would know if you knew, and that therefore you already know: "Ah if you only knew!") -- to wit, what the person signing the said sentence (which is not a full sentence, but only an incomplete subordinate) cannot state without attributing to himself knowledge concerning what the other does not yet know, but already knows, having learnt it in this instant -that is, instantaneously, and so soon (so bald) that it will not wait until the end of the sentence.
The acceleration in the change or the alteration which the sentence in suspension speaks (wie es bald, so bald schon --anders kommt!) is in truth only its very rapidity. An incomplete sentence rushes to its conclusion at the infinite speed of an arrow...
-The Politics of Friendship, 31

One can see (and Bernard Stiegler brings this out best) that Derrida is a thinker only of speed--the only thinker to think speed. For he is able to duplicate this speed of Nietzsche, albeit differently, in that amazing remark: "Quelle phrase! Est-ce une phrase?"
This is deconstruction in a nutshell. But what can be lost is that Derrida's sentence is really developing that space in which tone operates positively--positively, that is, as a the purest space of a difference. In other words, its own work of affirmation (but is it affirmation? and thus is it not negation?) is located here.
What the problem is, as I see it, is that this space is perhaps infinitely problematized or given its infinite thickness (as the purest space) too quickly for us: in its gauging how pure this difference is, we forget already that it is a pure difference--the pressure of its purity already deferring all our calculations as to the level of its difference. We're thus left with the task of catching up with Derrida (as he catches up to himself), constantly, rather than feeling all this as an incitement to precisely see how this space of difference (to be deferred already) is a space of tone, as we asserted. This is not actually his fault entirely: he is preparing us for moving within the rhythms of this forgetting and remembering, i.e. allowing us to exercise ourselves so we may perhaps attain the right speed. But still, Derrida allows us (and a whole generation of thinkers has been plagued by this, dealing with it in more or less productive and provocative ways) to doubt sentences and language as a whole, and yet not feel that we have an obligation to express through it. In other words, in his haste to point towards the the deferring (or the differing, it does not matter which comes first) he does not make us feel as well as he should this doubt as the very condition of (an unconditioned) ethics. No doubt he tries--strenuously, his entire life. But because in his haste to pass through that point where difference becomes deferred (already), he leaves us merely with a demand and not a field or formation of impossibilities or possibilities--except in what Americans call the "readings" of Nietzsche (and so many others) that he "produces" in this haste. One then makes the mistake of seeing what Derrida advocates as the endless production of readings (which is precisely to ignore what he calls above the "vibration of a shaft of writing which, alone, promises," my emphasis): text is everywhere, without also being (and this is the only Derridian point) nowhere. This objection is not one (like Gillian Rose's) about the difficulty or impossibility of expanding on what the demand is--indeed we're familiar since Heidegger (if not Kant) of the problems one encounters when one thinks that one can quickly and easily do this. It is rather an objection (and Rose also begins to hit on this at the same time) to Derrida's situating this demand always within the speed of his analysis--of too quickly folding it back into his text.
This means, with respect to tone, that it is vital to look closer at that point (or points) where difference defers itself and expand it in terms of how it belongs (or already belonged) or does not belong to tone. That is, (rather than emptying it out--and this would be really emptying it out--into logical "moves," which is how that explication of Nietzsche by Derrida so stupidly is begging to be read, but in fact cannot be read qua merely "moves"--it needs to be read as precisely the following)--that is, at this point there must be established a field in which the difference to be broken down (and how, we cannot determine in advance) will be constituted, in which it is articulating itself and (in fact) in which the possibility (or impossibility) that it has already deferred itself is developed (and indeed, unlike the majority of American theorists, philosophers, and critics, this is what some truly amazing souls actually do mean when they talk of Derrida's "moves"--more power to them!). One might call this, then, lending expression to the possibility or impossibility of the deferred in the different--and this, as I see it, is the task of Foucault.
If we become familiar enough with Foucault's task (and it might be that also of Deleuze--and Deleuze and Guattari--though, alas!, I am not yet authorized to say), we can begin to specify this possibility--and specify it as a determinate, quantifiable field (i.e. as a discursive formation, or as a field of "moves" taken in the good, developed sense)--with respect to tone in how it posits a space in which arguments can turn on themselves. I'll look at this more in detail later with respect to Matthew Arnold and I.A. Richards. For now, I'll just let (if I can) the necessity of becoming familiar with this task disseminate itself.