Tlön, Uqbar, Orbis Tertius
"Tlön, Uqbar, Orbis Tertius" is a short story by the 20th-century Argentinian writer Jorge Luis Borges. The story was first published in the Argentinian journal Sur, May 1940. The "postscript" dated 1947 is intended to be anachronistic, set seven years in the future. The first English-language translation of the story was published in 1961. Told in a first-person narrative, the story focuses on the author's discovery of the mysterious and possibly fictional country of Uqbar and its legend of Tlön, a mythical world whose inhabitants believe a form of subjective idealism, denying the reality of objects and nouns. Relatively long for Borges (approximately 5,600 words), the story is a work of speculative fiction. The story alludes to many leading intellectual figures both in Argentina and in the world at large, and takes up a number of themes more typical of a novel of ideas. Most of the ideas engaged are in the areas of metaphysics, language, epistemology, and literary criticism.
Stats for nerds
15recommend
39mentions
34people
2010–2026years
41% recommend59% neutral
Mentions per year
What Hacker News says
- In no particular order:
Book of the New Sun series by Gene Wolfe. (Highly recommended)
A Fire Upon the Deep (Either you like the Tines or the Skroderiders)
Tlön, Uqbar, Orbis Tertius by Jorge Borges (Is reality imbued with our ideas?)
Conan the Barbarian by Robert E. Howard (Not unlike "the Greeks" of modern fic)
John Carter of Mars by Edgar Rice Burroughs (perhaps the first true modern fic)
Understand by Ted Chiang (Most of his short stories are good, or provoking)
Xeethra and the Zothique Cycle by Clark Ashton Smith
The Dying Earth by Jack Vance
- Many LLMs are surprisingly good at using specific named authors (rather than just example texts) to evoke a style, so you could try "in the style of Jorge Luis Borges" or "…Douglas Adams" or "…Robert Anton Wilson" – whose surreal/absurd/fantastic styles could be fertile seeds.
(If not already familiar with Borges, definitely check out his 'Tlön, Uqbar, Orbis Tertius' and 'Library of Babel' as inspiration.)
While "each article written once" an interesting & useful constraint, a Hallucipedia that evolves like Wikipedia, with revisions "towards" some level of inter-article agreement, or even shows scars from edit wars between competing schools of thought, might also be fun.
- While we're Borges-posting, I recommend "Tlön, Uqbar, Orbis Tertius"
https://sites.evergreen.edu/politicalshakespeares/wp-content...
- I'll add some recommendations to the author's list, as I have found that reading difficult literature (both fiction and non-fiction) has been like exercising a muscle for me. For example, I read Blood Meridian before doing this and then again after doing it for a few years, right after McCarthy passed away, and it was a night and day difference in how "difficult" the prose was.
A few things I think fit into the "short little difficult books":
Borges is not someone I consider too difficult, but many do for the same reason the author mentioned people finding Calvino to be difficult. His works require that you invest some curiosity into thinking about the scenarios in his fiction. He also plays with the nature of the narrative of the story in a sometimes postmodern way that is still accessible. None of his works are longer than 20 pages or so, so not a huge time investment. I would recommend buying the Penguin "Collected Fictions" edition. It contains a collection of books of stories, and I would recommend prioritizing reading the Fictions and The Aleph collections first. Some of his popular stories to start with would be "Tlön, Uqbar, Orbis Tertius", "The Lottery in Babylon", "The Library of Babel", and "The Immortal". If you want slightly more sentimental, "The Circular Ruins" is wonderful. If you want dryer, more satirical and postmodern, "Pierre Menard, Author of the Quixote" is hilarious.
If you want a book that plays with the basic structure of a "novel", Nabokov's Pale Fire is a great read and a good introduction into what the technique of an unreliable narrator can truly achieve. No matter how out-there you think your interpretation of it is, if you look into published literary analysis of it, the rabbit hole goes so much deeper than you might think.
Gene Wolfe is highly regarded among science fiction fans, and for good reason. While he's best known for his Solar Cycle, a set of a dozen novels, I think that the best introduction to him can be found in his short stories. They might not seem difficult at first, but some of them, especially Seven American Nights, Forlesen, and the trio of stories in The Fifth Head of Cerberus, are similar to Pale Fire insofar as they are far more than they seem on the surface. They require some degree of scrutiny and interpretation from the reader. I wish I had read them before his bigger novels, as I found reading those novels was so much more rewarding after learning how to read Wolfe from his short stories. If you want to buy a few short story collections, there's going to be some unavoidable overlap, but I would recommend "The Best of Gene Wolfe" (for The Island of Doctor Death and Other Stories, The Death of Dr. Island, Forlesen, Seven American Nights, Death of the Island Doctor), "The Island of Doctor Death and Other Stories and Other Stories" (Tracking Song, The Doctor of Death Island), and "The Fifth Head of Cerberus" for its three novellas.
If you want something much more difficult, try J.G. Ballard's The Atrocity Exhibition. He was influenced by Burroughs (Naked Lunch would be another recommendation here if you can stomach the Beatnik depravity within) and wrote a novel that will challenge its reader in many ways. It's very interesting to go straight from this into Crash, which covers a lot of the same material in a less difficult structure. Baudrillard praised Crash with "After Borges, but in another register, Crash is the first great novel of the universe of simulation", but I would say The Atrocity Exhibition deserves this praise(?) just as much, even if it's much rougher around the edges. Certainly, the political stunt pulled using one of its sections (called Why I Want to Fuck Ronald Reagan) belong more to the Situationists' games than Baudrillard's universe of simulation, but the stunt relies just as much on the hyperreal breakdown of signifiers as anything Baudrillard even pointed to.
- Gernsback Continuum is still my favorite story since Borges' Tlön, Uqbar, Orbis Tertius. Physical reality seems pretty solid, but social reality is certainly "flexible." And all the tech I work with is a continuum of different ages smushed together; javascript client apps embedding COBOL apps complied to wasm bytecodes. SQL databases whose schemas predate some of the developers working on the apps that use them. Instruction Set Architectures that were invented when we still thought Raymond Lowey-esque fins and gills were the height of material design.
In a bit of synchronicity, I found my copy of Islands in the Net last week and am re-reading it after 35 years. It's pretty interesting to see which bits Bruce Sterling accurately predicted and which were a bit off the mark.
- If anyone here hasn't read Borges, I'd strongly recommend him. Pretty much everything he wrote was short, <20 pages, and so it's really easy to sit down and read one of his stories over a lunch break. The common recommendation would be to try out Tlön, Uqbar, Orbis Tertius and see if you like it. If so, it's part of Labyrinths, which is (in my opinion) his best collection of short stories. The best edition in English is probably Penguin's Collected Fictions.
Regarding the content of this interview:
>If you compiled an enormous dataset of everything Borges read, and combined it with an exquisitely sensitive record of every sensory experience he ever had, could you create a Borges LLM?
This is my Kantian way of thinking about epistemology, but I don't think that LLMs can create synthetic a priori knowledge. Such knowledge would be necessary to create Borges out of a world without Borges.
In this interview, Simon's view feels much more like the way Hume viewed people as mechanical "bundles of sensations" rather than possessing a transcendent "self". This led to his philosophical skepticism, which was (and still is I guess) a philosophical dead end for a lot of people. I think such epistemological skepticism is accurate when applied to machines, at least until some way of creating synthetic a priori knowledge is established (Kant did so with categories for humans, what would the LLM version of this be?)
