Toby: A review of my entry to the ACX “Everything-Except-a-Book Review Contest 2025”

Spoiler alert: this entire review is one giant spoiler. If you ever plan on reading it, you probably shouldn’t read this review of it.

This review is a review of itself. It will be useful to give it a name. Indulging in a little anthropomorphism, let’s call it Toby. The task of writing an auto-review poses certain challenges to its author. A review is fundamentally an evaluative document and, typically, the author of a review has some independent material to work with. The review is then free to evaluate the degree to which that material fulfills its purposes, the degree to which it displays literary merit, and so on. In this case, however, I have no independent material to evaluate. My task in writing Toby is to construct an evaluation of Toby, to build an argument that critiques itself.

Considering the constraints posed by the nature of the project itself, Toby does display some strengths. Its reflections on irony are particularly apt and well-done (see section five). But it also has significant deficiencies. Its problems are of five kinds. First, and least severe, is the fact that it does not adequately represent the content of the document of which it is a review, viz., itself. An argument to that effect is presented in section two. The second and third problems (reported in sections three and four) are different than the first but closely related to each other. The entire review is an exercise in self-reference. That something can refer to itself is a deep and important fact and the phenomenon of self-reference has been used to derive some of the most significant results in, e.g., mathematics. But this review does not use it in deep and important ways, nor even creative ones. It uses self-reference as a cheap trick. A gimmick. I appeal to self-reference to show off how clever I am, but not for any deeper purposes. That’s the second problem. The third problem, connected to the last, is that self-reference is characteristic of postmodernism in literature, a movement that has, at least in this reviewer’s opinion, run its course. Something was needed to free literature from the grip of realism, and postmodernism answered the call. But it answered the call decades ago. Once upon a time an auto-review would have been refreshing. Now it’s just boring.

The fourth problem (discussed in section five) is not so much a deficiency in what the review says, but in what it doesn’t say. Grant that self-reference is tired and played out, and even that it is not used well in this case. Nevertheless, this review could have done something interesting, something worthwhile. But it doesn’t. It criticizes itself for being unimportant and boring (albeit with debatable sincerity) and for not even using the opportunity to make some other point. But that’s all. Even tired and played out literary movements can be employed for important non-literary purposes, but Toby doesn’t even try. It’s navel-gazing of the worst kind.

The final problem is that it fails qua review: it does not, indeed, cannot, fulfill the fundamental function of a review. Details about this problem will have to wait for section seven.

My first complaint is that Toby does not accurately represent its own content. Consider this: Toby is a review of a piece of writing. One of the things that we want from a review of a piece of writing is for it to tell us what the subject of the review says. Not, that is, to say it itself, but to say that the subject says it. For example, a review of Moby Dick should not say that a whale bit off Ahab’s leg,[1] but it should say that Moby Dick says that a whale bit off Ahab’s leg.

And here is Toby’s problem. Whatever Toby says, it does not say that it says those things. For example, consider this sentence: Toby is a review of itself (call this “the reflexive sentence”). Now, notice that Toby does not say that Toby contains the reflexive sentence. Toby is like a review of Moby Dick that simply contains the text of Moby Dick but does not say that Moby Dick contains that text.

It might seem like this is unavoidable, since the levels of self-reference ramify in something like the way of a funhouse mirror. Consider that if Toby said that it contained the reflexive sentence the same problem would recur: Toby would say that Toby says that Toby is a review of itself, but Toby would not say that Toby says that Toby says that Toby is a review of itself. And so on. No number of iterations of the ‘Toby says that’ operator will get us out of this problem.

That said, there is a way out. If Toby said: for all things that Toby says, Toby says that Toby says that, then the problem would disappear. A simple application of a universal quantifier would allow Toby to instantly express all the infinitely-many levels of reflexivity that otherwise prevent it from accurately characterizing what it says. But one of things that is particularly frustrating about Toby is that it does not make use of this technique. Indeed, it says in this very paragraph that such a technique is available and still declines to use it.

All that said, I think that this is Toby’s least serious problem. A good review would characterize the content of its subject, which Toby fails to do. But this is a problem that readers can deal with. The analogy with a review of Moby Dick that simply reprints the text of Moby Dick fails because such a review would not tell readers that that is what it is doing. But Toby does (read the first sentence of section one again).[2] A review of a piece of writing should accurately report that its subject says what it does, and Toby fails in this. It is, in that respect, deficient. But I trust that readers will not be misled by this.

The name of the game in this review is self-reference. And self-reference is, indeed, powerful, important, and the source of some of our greatest insights. Consider, for example, Gödel’s celebrated incompleteness proofs. Gödel found a way (through Gödel numbering) to encode the statements of a mathematical theory in mathematical terms, such that the theory could be used to derive results about itself. In effect, he demonstrated that, for sufficiently strong mathematical meta-languages, those meta-languages can be pulled into their object languages. And this allows for the construction of a statement in the object language that says, in the meta-language, “this statement is not provable.”

It’s hard to overstate how deep this result is. Provability is a syntactic notion. It has to do only with which symbols, in which orders, can be set next to each other according to the rules of a formal system. Truth is a semantic notion. It has to do with the correspondence between linguistic objects and the (typically) non-linguistic world. Now, mathematical truth has long struck people as suspect, and consequently there was long hope that truth in mathematics was nothing more than provability. That syntax and semantics would collapse. What Gödel proved is that this hope was ill-founded. Syntax and semantics are categorically different things. Gödel was a Platonist about mathematics, and while his Platonism predated his discovery of the incompleteness theorems, they must surely have bolstered his faith.

Consider also the halting problem. It has been shown[3] that there is not, in general, a method for determining whether a particular computer program will eventually halt, or whether it will continue running forever. The problem is that, for any algorithm that tests whether a program eventually halts, there could be a program that loops forever on the condition that that algorithm says that it halts. There are strict and rigorous ways to show this, but for our purposes it suffices to consider an intuitive illustration. Let F(x)=1 if x halts and 0 if it does not, where F(x) takes programs as arguments. A program, G, can then be constructed, which is defined in terms of the output of F(G). In particular, G loops if F(G)=1, and it halts otherwise. Thus, F(x) is unable to determine whether G halts. Notice that in this illustration part of the specification of G involves reference to G itself.

The incompleteness theorems and the halting problem represent deep and significant uses of self-reference. Toby does not. It’s a trick. Just an excuse to be clever. It gives me the opportunity to pan my own entry into the “anything except a book” review contest, which, to those with a certain kind of mind (e.g., me) is kind of amusing. But it uses something that matters as much as something that is purely intellectual can matter, for trivial ends. For mere amusement. Consider an analogy. Kurt Vonnegut once said that all music is holy. And, while anyone who lived through Nu-Metal knows that that is an exaggeration, I am comfortable in saying that, in some admittedly metaphorical but still rather serious sense, Pachelbel’s Cannon in D is holy.[4] But now imagine using Cannon in D for trivial purposes. Imagine that someone used it as the soundtrack to a Doritos commercial. Using something holy to shill junk food. That’s akin to what is going on here.

In my defense, every text has an intended audience and is written for that audience. And so, to some extent, grading on a curve is necessary. When evaluating a piece of writing, you need to ask who it was written for.[5] This review was written for an “anything except a book” review contest on a rationalist blog and written for the kind of people who read “anything except a book” review contest entries on rationalist blogs (i.e., you). And self-referential essays are catnip to people like you. In fact, I would not be surprised if there were multiple reviews that are reviews of themselves entered in this contest. So, perhaps a certain amount of charity is warranted here. It may be a bad review, but at least it answers to the interests of its intended audience. On the other hand, since readers of rationalist blogs are, on average, more tolerant of too-cute-by-half self-referential essays than they really deserve, and since you are a reader of a rationalist blog, however bad you think Toby is, it is probably worse than that.

Ernest Hemmingway cast a long shadow. Each literary generation needs to find some way to distinguish itself from its predecessors. ‘Postmodernism’ is a fraught word, but for our purposes it will do to think of it as referring to a literary tradition whose texts are aware of themselves as texts.[6] It was this maneuver that allowed literary authors to stake out a new identity despite growing up in the shadow of Hemmingway. Nabokov made an early movement in this direction with Pale Fire. When Vonnegut made himself (qua author!) one of the characters in Breakfast of Champions he perhaps reached the apex of text-aware-of-itself-as-text.[7]

Thereafter, postmodernism looped back in on itself. Consider the Eschaton section of Infinite Jest. (Obviously, Infinite Jest spoilers follow.) Infinite Jest is a manifestly postmodern novel (Scott has called it “the most postmodern novel”), and the Eschaton section is about postmodernism itself. Eschaton is a game that is supposed to mimic the balance of strategic nuclear power, and, as luck would have it, the conduct of a nuclear war. The players each represent nuclear armed nations and occupy positions on a tennis court that corresponds to the relative positions of their respective nations. Thus, for example, the player representing the USSR stands to the east of the player representing the USA. Each player is equipped with tennis balls, each of which represents a strategic nuclear weapon. To conduct a nuclear strike, they are to launch a ball at another player’s territory. Things go awry when it starts snowing. The character Penn argues that the snow has changed the environmental conditions and so should affect things like fallout patterns, even though there is no rule in the game connecting real environmental conditions to those in the game. DFW is not shy about bringing out the meta-narrative significance of this fact: “’It’s snowing on the goddamn map, not the territory, you dick!’ Pemulis yells at Penn, whose lower lip is out and quivering.” Which is a little bit on-the-nose, admitting-that-you-are-a-postmodern-novel-wise.

There is some dispute between characters over this: “’Except is the territory the real world, quote unquote, though!’ Axford calls across to Pemulis.” Axford is trying to blur the distinction between the thing represented and the thing doing the representing. Which is, of course, exactly what Infinite Jest does. The book is about a video called “Infinite Jest” that is so amusing that if you start watching it, you’ll get sucked into it forever. The Eschaton section is something like a thesis statement for the book. Infinite Jest is itself a map/territory confusion, and the Eschaton section tells you how it works.

Toby takes this progression to its logical conclusion. Even Infinite Jest, although about itself, is also about other things: Québécois separatists, junior tennis academies, microwaving one’s own head. Toby is not. It is only about itself; its invocation of other works is not even commentary on those works, the entire purpose of such invocation is to call attention to the literary tradition to which it belongs and to emphasize the degree to which it is only about itself. If postmodern literature is text that is aware of itself as text, Toby is maximally postmodern. All it is, is its awareness of itself as text.

If I was writing in the shadow of Hemmingway, this would be original and refreshing. But I’m not. Hemmingway is long dead, both in flesh and in spirit. Instead, I am writing in the shadow of Nabokov, Vonnegut, and DFW. They are dead in flesh but not in spirit. And in their shadows, Toby does not stand out. Far from being original and refreshing, it is derivative and boring. This is a problem with conceptual art in general. If another artist had managed to recreate the ceiling of the Sistine Chapel, both the original and the copy would be magnificent. But the second guy who writes “this is not a pipe” on a picture of a pipe isn’t a genius, he’s just a jerk.

All that would, perhaps, be forgivable if there was some larger point. As it is a review of a review, Toby could have been used to examine the standards for a good review. What makes a review good as opposed to bad? What are the virtues of a review qua review, that distinguish its characteristic mode of excellence from that appropriate to, say, an essay or a work of fiction? What we are told—that it should not be derivative or trivial—is as true of any other form of writing as they are of a review. These are not points that you need an auto-review to make. It doesn’t explore interesting questions about what makes for a good review, except, in a weird way, by exploring it through criticism of this very point. Now, admittedly, there is one exception to this claim. Toby does say one thing about what makes a review qua review a good one. But I am going to exercise some authorial discretion, to build suspense, as it were, and make you wait. We’ll get to it eventually.

There are, of course, other important things that I could have done with Toby. But, again, I mostly fail to do them. Consider the following. Toby could have been a meditation on irony. Because, in case it wasn’t clear before: I’ve slathered the irony on here pretty thick. This could have been a useful, and maybe even insightful, thing to have done. I could have made points like the following. The ironist is a coward. Ironic writing is defensive writing. It allows the author, when criticism comes in, to say “but I didn’t really mean it!” I could have illustrated this fact using Toby as an example. Like, I could have said that perhaps that’s all that’s going on here; perhaps I’ve hidden behind so much irony because I am afraid of criticism from ACX readers. When they come up with criticism of my piece, I can advert to the wink and nod with which it was written and so disclaim any of the bits that they object to.

But, in any case, there are two reasons that this “meditation on irony” reading will not redeem Toby. The first is that it isn’t really a meditation on irony. Toby is more than 5000 words long and it has only a few paragraphs on irony. “The ironist is a coward” is a good line (I’d like to put it on a t-shirt),[8] but Toby just isn’t mostly about irony. The other problem is that, because of the many layers of self-reference involved here, it’s not entirely clear just how ironic I am being. I am criticizing my own writing, of course, but to the extent that those criticisms hold merit, Toby isn’t really that bad. So, do I mean it or not? I’ve said that I’m being ironic (“I’ve slathered the irony on here pretty thick”), but the whole thing about ironists is that you can’t trust them. Moreover, even granting that I’m writing ironically, it’s hard to tell which parts are ironic. In particular, is the claim that I am writing ironically itself ironic? It could be that I present the criticisms of Toby straight but make the claim that I am writing ironically with a wink and a nod. On this reading, the claim that I am writing ironically is the only bit that is written ironically, and so the arguments against Toby are earnest, and so I cannot hide behind the shield of irony when their flaws are exposed (including when I am the one exposing them, since much of what Toby consists in is an exposition of its own flaws).

But, then again, it could be that the claim that I am writing ironically is to be read straight and the criticisms of Toby are meant ironically. If my criticisms are meant ironically, it would seem to indicate that I actually think that Toby is a pretty good piece of writing and my claims to being an ironist should not be believed. However, if my claim that I am writing ironically should not be believed, then the criticisms of Toby are in earnest after all. Now, it could also be that both parts are ironic – that I don’t mean any of it. But this isn’t really a stable reading. If both parts are ironic then the criticism isn’t meant seriously, but then neither is the claim to be speaking ironically. And if that claim isn’t meant seriously, then the criticism can’t be ironic, and must be taken straight. But then it turns out that I wasn’t being ironic about the criticism.

Finally, maybe this whole discussion takes talk of ‘irony’ too seriously. I am writing ironically, but perhaps by that I don’t mean “falsely,” maybe I just mean something like “with a smug indifference towards the truth.” Although this is neither the literal sense of ‘ironic,’ nor the Alanis Morrisette sense, it does seem to be a fair meaning of the word. Consider, for example, the way in which Richard Rorty talks about values. He recognizes that all values are contingent and historically conditioned (this is why he is so often accused of being a cultural relativist), but he also realizes that he can’t help but employ the values of his own culture. There’s no value-neutral perspective that he can step out into. There is, therefore, always something arch about Rorty’s moral claims. When he says that cruelty is wrong, it’s not that he’s telling you something that he doesn’t believe, he’s not lying, but he is engaging in a different kind of insincerity. In using moral vocabulary “ironically” he treats moral claims as a kind of a game; the moves in the game are made in earnest, but it is always understood that they are a part of a game. Rorty’s subtle ironist contrasts with the rube, who, when he says the cruelty is wrong, means it with every part of his being. Interpreting Toby as being ironic in the literal sense is fraught with problems (it’s not even clear to me that such a reading is internally consistent), but interpreting it as being ironic in the Rortian sense is not. It’s a game: I wrote a review that is a review of itself because it is clever, it’s the kind of thing that people who like to think of themselves as subtle and sophisticated (e.g., me) are attracted to,[9] and I made it a negative review because it’s funnier and wittier than writing a review that lauds its own greatness.

The cost of engaging in Rortian irony, however, is that you don’t get the advantages of regular old irony. Ordinary irony protects you from criticism. You can always say “I didn’t mean it.” And while the Rortian ironist can say that too, he can only say it because he doesn’t really mean anything. And he is criticizable on those grounds. The regular ironist is sticking their neck out, just not in the direction that you’d think. The Rortian ironist isn’t. He’s just playing a game, laughing at the rubes behind his hand. He’s immune to criticism for what he says but is always criticizable for how he says it. If Toby is just an exercise in cleverness and wit, and I mean neither the claims that I explicitly make nor their denials, then it’s so much hot air. It’s bullshit, in the Frankfurtian sense. It consists of layers upon layers of self-reference, that repeatedly curve back on themselves, to no purpose. It says nothing, it’s just an excuse for me to show off and wink to those in on the joke. And that is bullshit, in a sense somewhat less polite than the one that Harry Frankfurt had in mind.[10]

Second thoughts. Maybe the criticisms of Toby contained in the previous four sections are unfair. To begin with, the complaint that self-reference is “holy,” somewhat like the way that Cannon in D is holy, is somewhat strained. It’s true that very important things can be done with self-reference, but “holy”? C’mon. The criticism would be fair if it was merely that what I am doing with self-reference here is less significant than what Gödel did with self-reference, instead of relying on the claim that self-reference is holy, or that it matters in something like the way that holy things do. But then, “as meaningful as Gödel’s incompleteness theorem” is a high standard for meaningfulness. No review is going to live up to that, regardless of the use it makes of self-reference.[11] So, while it’s true that Toby is less meaningful than the incompleteness proofs, it’s not clear that this is a criticism of it. It can be perfectly worthwhile while being less meaningful than Gödel’s work.

Perhaps more significantly, it’s also not clear to what extent the criticisms of Toby’s employment of postmodern literary devices is on target. Of course, Toby is about itself. But quite a lot of it is also, for example, ruminations about what makes a piece of writing meaningful, a summary of Infinite Jest, a warning about the dangers of irony, a claim about the purpose of a review (you’re not there yet, but that part is coming). That is, there’s a lot of material here that is not obnoxiously recursive, that is, much of it is just ordinary, object-level, this-is-a-review-of-a-written-document stuff. If Toby was a review of some other review, it may well contain ruminations about what makes a piece of writing meaningful, and so on. So far so good. Not all of Toby is obnoxiously recursive. But this will only take us so far. The problem is that this paragraph certainly is obnoxiously recursive. It’s included as an attempt to rebut the charge that Toby is obnoxiously recursive, by pointing out that much of Toby isn’t, but the longer it goes on, the weaker the point that it makes gets, since any attempt to rebut that charge will itself be obnoxiously recursive. It is, to a degree, a self-refuting argument. The case that it makes would have been strongest if I had never made it.

Finally, I have second thoughts about the arguments in this section (except for this one). If the preceding arguments of this section are successful, it follows that they are not. Toby pans itself. But the arguments of §6 (except for this one) purport to show that the arguments against Toby are no good. That is, they purport to show that Toby isn’t so bad, despite what the preceding arguments say. But since Toby is those arguments, to show that they are no good is to support Toby’s conclusion. It is to say that Toby is pretty bad after all, despite (because!) the arguments in Toby’s favor are so strong.

Enough. Life is short, rather than demand any more of your time, I am going to wrap things up now. There are some good things about Toby. It says that it makes only trivial use of self-reference, and, while others may care less about this point, I agree. And its observation that it is derivative and boring is spot-on. It gets points, then, for accuracy. Unfortunately, what it is accurate about are its own flaws.

Moreover, the purpose of a review of a piece of writing is to aid readers in deciding whether to read the piece being reviewed. It is, in a way, an exercise in triage: readers have only so much time and attention and a review’s function is to help readers decide where to spend their time and to what to direct their attention. Let me, then, give you my verdict on Toby. It is a bad review. It is a bad review for the reasons just mentioned, but also for one more. The central reason that it is a bad review, the one that shows that it fails qua review, is that it is unable to fulfill this triaging purpose. It is a bad review and so you should not read it. The problem is that you just did.

[1] In case this isn’t obvious: Ahab never existed, and so no whale ever bit off his leg. So, a review of Moby Dick should not say in its own voice that Ahab lost a leg to a whale. But it can (and should) attribute that claim to the novel, since the novel does say that a whale bit off his leg.

[2] Notice that I can point readers to that first sentence, but I cannot say that Toby says what the first sentence says. If I do, then the claim, from a few paragraphs back, that Toby does not say that Toby says what it says, becomes false.

[3] The literature on this topic is complicated and it is not entirely clear to me where credit should be assigned. At the very least, Turing’s work on undecidability from 1936 deserves mention (although many other peoples’ work does as well).

[4] I once cut Cannon in D into pieces and spliced it back together over techno beats. When my father heard it, he said: “you’re going to hell.”

[5] You also need to ask what purpose it was written for. Although, in this case, it’s not clear. If I was trying to win this contest, presumably I’d have written a better review. But it’s also not clear that I care about winning the contest. If, somehow, I do win, I’m not going to keep the prize money. I’ll just donate it to GiveWell’s top charities. (On second thought, it’s better if Scott donates it for me. I don’t itemize my taxes, but I bet he does, so he can get a tax deduction for the donation. So, I’ll ask him to donate the prize money, plus the amount of the deduction he’d get for the donation, plus the amount of the deduction he’d get for donating the money he saved through the tax deduction, and so on, until we’re within pocket change of whatever the limit of this function is.)

[6] This definition is only very approximate. The framing that Tolkien uses in The Lord of the Rings is that he is merely translating an ancient manuscript (this comes out most clearly in the appendices). So, The Lord of the Rings is a text that is aware of itself as a text, but it is no one’s idea of a postmodern novel.

[7] This isn’t the only place that he uses this trick, but I like it better than his more widely read uses of it.

[8] 2XL, black text on light grey or white background. Shipping address upon request.

[9] If I was writing about anyone other than myself here, I would be more polite.

[10] As always: since this is a criticism that I wrote about myself, one must ask how earnest it could possibly be. If I was serious, presumably I wouldn’t have written the thing that I am criticizing. But, if it is not meant in earnest, then this criticism itself is exactly the kind of Rortian irony that it criticizes.

[11] Chomsky’s review of Skinner might be the most significant review of all time, and while dedicated partisans of mentalistic psychology might say that it compares favorably, in terms of meaningfulness, to Gödel’s incompleteness theorem, I’m not persuaded.