Review: Of Grammatology by Jacques Derrida

It’s very, very difficult to do a straight-up review of Derrida’s Of Grammatology because everything about the book inspires bad behavior from writers. It challenges and undermines the very structure of writing by the way it is written (come to think of it, not unlike Monique Wittig’s brilliant experimental feminist novel, Les Guérillères, undermines the very structures of storytelling (under patriarchy?)) which, as you read the book, you can’t help but absorb at least a bit. It’s also a completely infuriating book, refusing to get to the point at any point, full of excessively long sentences that, at least in my ebook reader, would literally go on for a whole page or more, refusing to be definite, refusing to define its terms, oh, and constantly quoting weird old “scientists” without ever questioning their completely made up “facts.”

And see, I’ve already done it. I’ve been infected by Derrida. Nested parentheticals, statements with question marks instead of owning my assertions, meandering sentences that don’t come to a point–I’m doing it all myself. To be fair to Derrida, meandering sentences that don’t come to a point are something of a natural feature of my writing, so let’s say he just encouraged that one.

I have a lot to say about Derrida’s ideas, but first, I want to tell you about the book, as a book, in case for some reason you want to read it. And, by the way, I would argue that if you’re a writer who engages with the Global North literary tradition, you should. You should also probably read Derrida if you are thinking deeply about computational linguistics and Large Language Models (LLMS), although I’ll say much more about that, later. I have to do a little literature review first on that topic.

So what’s Grammatology? #

Derrida actually defines it in the first few pages of the book, but you’d be forgiven for missing it because it’s tucked in the middle of a sentence, or forgetting about it by the time you get deep into the thicket of this book:

“By alluding to a science of writing reigned in by metaphor, metaphysics, and theology, the exergue must not only announce that the science of writing–grammatology– shows signs of liberation all over the world, thanks to decisive efforts.” (italics mine) p4

There you have it, grammatology is the science of writing.

Oh, and the exergue is the title of the weird introduction. The book has a number of introductions and I guess Derrida (and the translator, Gayatri Chakravorty Spivak) ran out of synonyms for introduction so they had to use a random other word. Merriam-Webster defines exergue as “a space on a coin, token, or medal usually on the reverse below the central part of the design.”

Sure, why not, an introduction to a book metaphorically positioned as an inscription on a coin. This level of weirdness right from the beginning is probably meant as a warning about what you’re getting into.

I completely do not remember reading it and only found it when I went back to the beginning after finishing the whole book. It’s possible I skipped it in the deluge of intros. I have a tendency to skip lengthy introductions, especially when a book is overburdened with other people’s readings and interpretations. I like to see what it has to say for itself. But, in the case of Derrida, Mr “Nothing outside of the text” I should have anticipated that the apparent introductions were also inside the text. Except–well, I wouldn’t have known that without first having read the text.

Damnit, I’m doing it again. Getting sucked into Derrida’s meandering.

The central problem of writing #

Derrida analyzes various historical thinkers writing about the problem of writing, and through the way in which he analyzes them, finds the inherent contradictions in their thinking. The whole book always operates on at least a double level like this. On one level, he is engaging with the ideas that these thinkers work through. On another level, by the way that he engages, he demonstrates at length the method of engaging with writing that would later be called deconstruction.

And here, I must give one of my many caveats. I’m trying to explain this extremely unclear book as clearly as I can. Not only might I be getting it wrong, but it is also likely, even quite probable, that I didn’t understand parts, forgot parts, and just straight up didn’t pay attention to parts. But I still want to be as direct as I possibly can, even if I risk being outrageously wrong.

Reading Of Grammatology, I noticed there was one problem of writing that returns over and over, and that then somehow leads to all the other more complex problems of writing. The problem is this: spoken words signify things, emotions, and ideas. Let’s call those things “the signified.” So, spoken words are signs that signify the signified. With me so far? Great.

Then, when we use a phonetic writing system, the letters represent sounds, and the sounds constitute words. Or maybe to put it more simply, a written word represents the sound of a spoken word. A written word is a sign signifying a spoken word. But a spoken word is itself also a sign.

And so we get into this convoluted chain of signifiers, where a written word is a sign that signifies another sign that finally signifies a signified. As in, a written word points to a spoken word which points to a meaning.

Only, does it? That’s pretty much the big question of grammatology and of Of Grammatology.

I mean, spoilers, for Of Grammatology, but there are some indications that the sign of the written word does not always ultimately point to an ultimate meaning behind, as it were, all this play of signification. Maybe the play of signs in a sea of signs is all there is, without an original meaning signified by a spoken word, and maybe the meaning arises out of the words, not through the words.

By the way, I disagree with this metaphysical take, as taken to its logical limit, but in this book review (remember this is a book review) I want to present to you what the book is about. Because damn it, almost no one else ever seems to.

The chapter titles are absolute bangers #

While Derrida’s sentences are just a slog, he has a real talent for fantastic chapter and section titles. They’re so good that Delany just lifted them wholesale for some of his chapter titles. Who doesn't love “The Violence of the Letter” or “Writing, Political Evil, and Linguistic Evil” or “The Economy of Pity.”

If there isn’t a band (several bands) called “The Violence of the Letter” I will be extremely surprised. “The Violence of the Letter” is probably the most interesting and accessible portion of this book. I first read it in graduate school, and although it gave me anxiety (if words mean nothing and cannot communicate meaning, why am I even a poet?) it stuck with me enough that I wanted to read this damn book 20 years later. In “The Violence of the Letter” Derrida uses the example of Levi-Strauss to, in a long-winded way, call out the violent imperialism inherent in the way anthropology was (is?) practiced.

One only wishes he were as direct in calling out Rousseau when the guy is a complete (figurative and literal) wanker. The Rousseau sections are really not my favorite.

It demonstrates how to do deconstruction #

The book feels like you are going along for the ride with Derrida as he thinks through the texts he is reading and shows you how he does it. He reads them so closely that they start to break down. Using only what’s right there in the text, he repeatedly shows how what the authors say undermines what they say they are saying, and turns that into a critique of “logocentrism” which is mostly a weird way of saying European imperialism, cultural and otherwise. It’s stunning to behold, if you can hang on through the ride without falling asleep and falling off the horse.

On the other hand, it’s fucking obnoxious at times. This apparent commitment to only using the text to deconstruct the text seems to lead to a complete refusal to check up on the outside world. There are just bizarre assertions, and I noticed them especially from Rousseau, about language and linguistics, that Derrida lets pass without so much as a comment about their preposterousness.

It makes me appreciate someone like Kant, who obnoxiously says in the introduction to Critique of Pure Reason that the book is already too long and he doesn’t have room for examples so he’ll just present his ideas straight up. And if you want some examples you can make your own. Obnoxious, and yet, precisely because of the lack of examples Kant has aged pretty well. You don’t get distracted by him citing some outdated scientific theory or thinker to support a metaphysical claim. It’s just metaphysics straight up until you choke and go back to reading Derrida. (No, I haven’t yet finished Critique of Pure Reason, why do you ask?)

Second-hand Derrida is everywhere #

I think it’s important to read books whose influence has permeated your discipline so that you know where the ideas came from. If you’re a writer or a cultural critic, Derrida has permeated your discipline. If you don’t read Derrida, you will read like Derrida without knowing you’re doing it. You might absorb his ideas, even if they are bad ideas for you, without confronting them and having it out with them and deciding for yourself if you want them. Even if you half-ass read him, or read him as a hostile reader, and don’t understand all of what he’s saying, you’ll recognize his smell all over other people’s ideas. And I think that’s helpful.

In this moment, where LLMs trained on corpuses of written language produce swathes of written words that are only related to each other through their statistical relationships to each other and do not signify a spoken word nor a meaning (or at least not an intention) Derrida’s ideas about signs and shadow signs and the play of signification seems particularly relevant. You might not agree with him but the problems of language he was discussing have something to say, I think, to our moment.

Reading Derrida makes you a worse writer #

As you see, I’m nearly at 2,000 words and I’ve hardly made a point. No, no bots were used in the production of this blog post. This is all vintage, hand-made verbosity, like an ugly hand-knit scarf that is entirely too long. (On the other hand, fuck it, they pay me money at work to write concisely and clearly. Maybe I deserve to let loose with some parentheticals and a big ol’ waffle in my free time.)

So, yes, reading Derrida will temporarily make you a worse writer, but the effect wears off the moment you read literally anyone else or even just talk about things that aren’t Derrida. It’s like a migraine aura: a bit scary the first time but then you realize it’s only temporary.

Speech to text, text to speech #

Finally, I want to note, and possibly confess, that I read a lot of this book as text-to-speech on my Kindle. It’s how I read a lot of books and articles, because, except for science fiction, I usually get restless when I’m reading and want to do other stuff like cook or knit or draw or just walk around as I do it. So it is possible that due to my unusual reading method, I understood Of Grammatology particularly unusually. But also, I think it’s kind of funny to read a book that’s all about the way the written word affects the spoken word by having a text-to-speech machine vocalize the words. I also know some people think text-to-speech reading isn’t real reading, which only makes it funnier, since so much of this book is about the failure of that quest for the real presence behind the written word, the real thing, and here I was faking it all along.