Reading like Derrida

It is often said (and I'll crudely paraphrase the crude paraphrase here) that Derrida's mode of reading is to "close read" texts of philosophy.
This is right, but it seems to me to only get at one aspect of what Derrida is doing.
Derrida most certainly close-reads texts. That is, he pays attention to their use of words first and foremost, rather than like so many philosophers who also do pay extremely close attention to the letter of the text but whose focus remains beyond it, in the contours of the concepts being got at. It doesn't matter, for these philosophers, with what words Kant describes the transcendental deduction as much as what it is. This is not to fault this approach: this is precisely what philosophy is and should be doing, in my opinion. For it does not mean, again, that the text is dumped in service of "ideas" or something like that: it is just a matter of practical, pragmatic priority. One has to see the phenomenon, or think the concept, or reflect on the issue, otherwise one ends up in just following words around and making unimportant points. This is why so many critical theorists are, despite their best (and worst) intentions, horrible philosophers. So paying close attention to the text is the first way to accomplishing reading like Derrida.
But the second is more crucial, I think, because it is precisely Derrida's contribution to close reading, or hermeneutics more generally. Derrida transforms the idea of close reading by extending its logic farther than any human can control it. In other words, Derrida close reads beyond the close reading that anyone can actually pull off. But this precisely doesn't mean that he--Derrida--is some sort of genius. This isn't some sort of magic--again, as many fawning "decon" initiates would make you think. It is simply being mechanical. When we say he close reads beyond the human, we simply mean that he not only pays attention to the letter of the text, but also follows a letter, a word, or what have you sometimes mechanically, arbitrarily, so as to not submit it to some overarching human interpretation. In other words, he close reads beyond his control. The important contribution of this is that in itself this method prevents the close reading to end up as evidence for some theory about a text. Take Dissemination: the chapter headings say it all. They're just different forms of the word pharmacia: pharmakon, pharmakeus, pharmakos (to use the French way of writing the Greek). And, quite literally, for a large part of the time the difference between these signifiers is all that one can say is governing the interpretation at work in the book. Not even the difference between the signifiers--the way this difference is itself differing from itself, changing into a different difference between the signifiers: in other words, it is not that Derrida is tracing how the use of pharmakon in one instance compares with the use of its other form, phrarmakeus, in another; rather it is how this comparison also needs to be compared or can only be compared with the use of pharmakos in another case insofar as it differs from the use of pharmakeus, and so on. And the use of "needs" here--how one comparison needs to be compared with another etc.--is inadequate: there is no reason why one needs to be compared with the other. It is just that it is possible. This, then, is how one interprets according to différance. And it is this that really makes this less of a close reading than a semi-distant one. This is not to say that it is distant reading either. But the fact is that if you want to read like Derrida, it is not that you have to--as so many "deconstructionists" will tell you--really read what's on the page closely, or pay infinite attention to what's on the page, or whatever. It is that you submit it to a working and unworking of what's on the page beyond your (and anyone's) control, which more often than not means completely not paying infinite attention to what's on the page, or reading it closely. If this sounds insane, that's because it is--no doubt about it. It is also extremely dangerous, and risks real irresponsibility. But it is only here that there perhaps is a "text," in Derrida's use of this word (one sees then, how stupid it is for "deconstructionists" to make it seem as if paying close or very high attention to the text were the same thing as close reading--paying infinite attention to the text is precisely risking not reading it closely). It is only here that you realize a "text" is not, for Derrida, what's on the page. In the process of trying to somewhat counter what cannot here be controlled, you get a reading in a Derridian way--and in fact, only here you might get a text.
And this is really what is amazing about Derrida's way of reading, considered as a way of reading and precisely as a contribution to the idea of close reading. In the wake of all the various confusions about what the hell deconstruction is and was and will be about, this point about its hermeneutic or non-hermeneutic strategy of being sometimes extremely far from what's on the page, I think, is crucial.