Cormac McCarthy's tips on how to write a science paper (2019) [pdf](gwern.net)
gwern.net
Cormac McCarthy's tips on how to write a science paper (2019) [pdf]
https://gwern.net/doc/science/2019-savage.pdf
86 comments
When I read papers I often think: if only the author had the space here to write two sentences instead of one, then perhaps I would immediately understand what they are trying to say.
Oh, that's the only way to write a good paper - first you write 1.5x pages to figure out what you want to say, and then, with this knowledge, you replace entite confusing paragraphs with short sentences focused on exactly that. When i don't know how to express an idea, i ask myself "-so what idea are you trying to communicate? -well, that XYZ holds -if so, just go ahead write literally that!"
FWIW I think this is exactly the opposite of what the OP was trying to say. I think the OP meant that they often wish authors had used more space to explain their ideas, because that would immediately make them more accessible.
And I must agree. I often feel papers are written with an extra 20% over the page budget, then condensed in the last minute to fit the constraints, which hurts the exposition.
And I must agree. I often feel papers are written with an extra 20% over the page budget, then condensed in the last minute to fit the constraints, which hurts the exposition.
It's a great way to write fiction too. But only if you're not marrying to what you've written.
Sometimes two sentences is the minimum necessary number of sentences, but everybody should be wary of that instinct
"Use minimalism to achieve clarity. While you are writing, ask yourself: is it possible to preserve my original message without that punctuation mark, that word, that sentence, that paragraph or that section?"
This is rich, coming from the guy that spends multiple paragraphs describing an empty ditch in the desert.
This is rich, coming from the guy that spends multiple paragraphs describing an empty ditch in the desert.
Apparently he thinks that is the minimum required. There is no metric included with the statement.
Reminds of this quote attributed to Dieter Rams:
> A designer knows that he has achieved perfection not when there is nothing to add, but when there is nothing to take away.
> A designer knows that he has achieved perfection not when there is nothing to add, but when there is nothing to take away.
That's from Saint-Exupéry.
https://english.stackexchange.com/questions/38837/where-does...
https://english.stackexchange.com/questions/38837/where-does...
Aside: I consider McCarthy's residency at SFI an ideal job
This is a testament to just how multifaceted he was.
This is a testament to just how multifaceted he was.
Sorry, but he takes it too far. McCarthy's omission of punctuation makes his books difficult to understand who is saying what, and a challenge to follow especially with dialogue. The Road and No Country for Old Men both do not contain quotation marks for speech, and he omits the common speech tags like "he said" or "she exclaimed" which makes it a challenge to know who is saying what. It is a choice and the art form he's choosing, but is far from writing for clarity.
I would assume that his suggestions for clarity in "scientific papers" and his literary style don't overlap all that much to infer the former from the later.
This is certainly the case, but it does make it all the more amusing that the myth
> Commas denote a pause in speaking.... Speak the sentence aloud to find pauses.
made its way into this article. Hard to imagine that this particular point, to which I might attribute many of the comma splices I see in scientific writing, actually came from a professional writer.
> Commas denote a pause in speaking.... Speak the sentence aloud to find pauses.
made its way into this article. Hard to imagine that this particular point, to which I might attribute many of the comma splices I see in scientific writing, actually came from a professional writer.
McCarthy's books involve unrelenting violence. If he viewed commas as pauses, it makes sense that he would never use them.
It goes without saying that you're a better writer than Cormac McCarthy.
Tell us something beyond that.
The implication was that likely not all of the advice in this article, which was written by biologists, is actually attributable to McCarthy.
> the former from the later
my writing advice:
never use the former and the latter
my writing advice:
never use the former and the latter
Does it make me sound pretentious? That's fine, we're debating literary styles after all. :D
no, it's just a stylistic pet peeve of mine. lacks specificity and always makes me have to think about which is the latter and which is the former, no matter how many times i look it up. scrambles my brain.
I recently read Blood Meridian, the only one of his books I've read. I agree this was a bit jarring and confusing at the start, but I got used to it by the end.
Though I haven't read any scientific papers, so can't comment on those.
Though I haven't read any scientific papers, so can't comment on those.
I agree about clarity, so this is just an aside but that's what makes it a fun experience for me. It's unlike reading anyone else (although I haven't read many authors). I'd say no country for old men was still pretty straightforward, but I had to re-read sentences and whole paragraphs with blood meridian.
The work makes it worth it, makes it that much more rewarding to me personally. It's like choosing to play a difficult videogame, because you know once you overcome it, it'll be great.
The work makes it worth it, makes it that much more rewarding to me personally. It's like choosing to play a difficult videogame, because you know once you overcome it, it'll be great.
I agree, his literary work is unique, and does take a bit more work to read, and with that it includes additional meaning behind it. For example, in The Road often times it doesn't even matter if its the boy or the man saying it.
However, I wouldn't take his advice on how to write for clarity. I too often found myself rereading paragraph, "wait is this description or dialogue", "who said that" - this is not what you want in scientific papers
However, I wouldn't take his advice on how to write for clarity. I too often found myself rereading paragraph, "wait is this description or dialogue", "who said that" - this is not what you want in scientific papers
> The Road and No Country for Old Men both do not contain quotation marks for speech, and he omits the common speech tags like "he said" or "she exclaimed" which makes it a challenge to know who is saying what.
I am reading NC4OM right now and this is not, technically, the case. He does use those “speech tags”.
I am reading NC4OM right now and this is not, technically, the case. He does use those “speech tags”.
technical writing and fiction writing are two totally different forms of writing. the ability to modulate between those disciplines is the sign of a good writer.
I routinely reduce my Readmes by 30% or more in the third draft.
"Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away."
- Antoine de St Exupery
- Antoine de St Exupery
"If I had more time, I would have written a shorter letter."
Yes, I came here to say this sounds a lot like Strunk & White
"A foolish consistency is the hobgoblin of little minds."
Yet behaving consistently according to well-chosen rules is a path to probable prosperity and, often, greatness.
I guess that means that machines will be greater than man could ever hope for.
What kind of greatness? Or prosperity? Well-chosen according to whom?
You’re using an absolute attachment to relativism to critique objectivism. This is a category error.
Strawman word salad.
Big brain socialist “everything is a special case” quote.
Would love to read any of the scientific papers that McCarthy supposedly edited diligently, should anyone happen to know of some. Very curious to see what they read like.
Not exactly a scientific paper, but he wrote this essay about language: https://nautil.us/the-kekul-problem-236574/
Here’s an economics article at least: https://hbr.org/1996/07/increasing-returns-and-the-new-world...
And here’s a short article about his contribution to that article: https://andrewbatson.com/2016/12/13/cormac-mccarthys-contrib...
And here’s a short article about his contribution to that article: https://andrewbatson.com/2016/12/13/cormac-mccarthys-contrib...
Cormac McCarthy was deeply interested in physics and mathematics and was a trustee at the Santa Fe Institute, which has a heritage connected with Los Alamos National Laboratory. I don't know a lot about this side of him, only reading about it after reading his last two novels which do show a mastery of physics that really seemed to mirror his master of bridles and guns and culture in the old west. I don't remember reading that he had actually published any of this himself but he was spoken of as intensely curious about physics.
one thing i've noticed about scientific writing:
many early-career folks are afraid to make things too simple and easy to understand because they (subconsciously?) fear that it makes their work seem simplistic or trivial.
when you're an academic that has built a great deal of your self identity around being perceived as 'the smart one', it takes a fair amount self-confidence to start presenting yourself in a way that is easy to understand
many early-career folks are afraid to make things too simple and easy to understand because they (subconsciously?) fear that it makes their work seem simplistic or trivial.
when you're an academic that has built a great deal of your self identity around being perceived as 'the smart one', it takes a fair amount self-confidence to start presenting yourself in a way that is easy to understand
I keep hearing this exact same idea and it puzzles me a great deal. Is it a computer science thing? I'm doing a PhD in signal processing / engineering and people seem to care a lot about giving simple and clear explanations so I don't really relate!
In my experience in neuroscience it even differs widely across programs/universities. Some good professors care about giving good talks, and if you're lucky it becomes contagious in the program. Others think less of you if it's clear, some are too naive to realize obscurity is not a virtue.
That's professional deformation, because signals are supposed to be clear and easily identifiable among noise.
I think there was a study at some point which showed that the worse the university, the more jargon the researchers use in their papers.
Maybe I'm overstepping, but I think 1. you're right 2. that it's driven by insecurity. I've experienced many instances of people trying to "protect their knowledge" so to speak by hiding behind jargon.
I like the advice about avoiding footnotes. Citing sources is fine. Almost all other footnotes and information from links should either be omitted or incorporated into the main text. They are too disruptive to the flow of reading.
> Find a good editor you can trust and who will spend real time and thought on your work.
Haha, sure, I will send it to my LLM -- ... I mean "editor." :)
> Find a good editor you can trust and who will spend real time and thought on your work.
Haha, sure, I will send it to my LLM -- ... I mean "editor." :)
Footnotes are very tempting in a methods section where there’s no end to additional details to add in. However, it’s probably better to put all of that into an appendix/supplemental material section (if the journal allows it). Having a section like that can be a bit freeing — it’s an info dump meant to be read by those looking for a specific implementation detail so it doesn’t need to “read well”
I think it depends on what the situation and format is. When I was writing research notes and a few books, I found bottom of the page footnotes really useful as, not only references, but also as parentheticals and other purposes that some readers might find useful or interesting but didn’t really justify breaking the flow of the text.
Pretty sure footnotes disrupt the flow of reading. The improvement of the "flow of the text" is an illusion which would only exist if the text wasn't sprinkled with footnotes.
Footnotes make scientific work more accessible and enrich the conversation. In Laurie Garrett’s “The Coming Plague”, the footnotes alone are worth the price. Even in monographs, when reviewing literature in an area where I’m not a specialist, I find footnotes valuable.
> Footnotes make scientific work more accessible and enrich the conversation.
I disagree. Readers have no way of knowing whether they can be safely skipped or not. Relevant information should go into the main text. If a paragraph starts out uninteresting enough, readers will skip it anyway.
I disagree. Readers have no way of knowing whether they can be safely skipped or not. Relevant information should go into the main text. If a paragraph starts out uninteresting enough, readers will skip it anyway.
Footnotes effectively have no place in CS, engineering or natural sciences. Other disciplines treat footnotes very differently, I think.
This is certainly too strongly worded to be correct. I use footnotes quite a bit in my papers (physics) as a strategy to handle the "two audiences" problem - that many or most readers just skim for main ideas, but some (and those whom I might argue are more important) try to follow the details closely. I presently use footnotes for the latter audience for certain supplementary details or technical qualifications that would break the reading flow or add unnecessary length for the former.
I do appreciate the arguments that footnotes can be distracting, or that one doesn't know whether to skip them, but at present I see them as the best option for keeping the main body streamlined/as short as possible without sacrificing points that I'd like to make that wouldn't make for or fit into an appendix.
I do appreciate the arguments that footnotes can be distracting, or that one doesn't know whether to skip them, but at present I see them as the best option for keeping the main body streamlined/as short as possible without sacrificing points that I'd like to make that wouldn't make for or fit into an appendix.
I disagree completely. There are many different sorts of works you can publish relating to a given field, and some of them benefit from the asides and additional context that footnotes can provide, particularly pedagogical works targeting an audience of a wide breadth of experience.
I recently wrote a paper for a conference that ended up being rejected, with a split review (2 for, 1 against).
As a non-academic that wrote a paper for the first time, I'll say that writing a good science paper takes an absurd amount of time, even on a topic you are very well versed in. It is also way different than other forms of writing, like blogs or technical documentation.
Frankly, I'm astounded I've even managed to get that result, working on it just one month from the submission deadline (a ridiculously massive time crunch) and only in my spare time, not having touched LaTeX in almost a decade. But if a proper heretic like me managed to get that far on their first try, then everyone who's considering writing a paper for the first time ought to have hope.
I might've somehow got an invitation to present a poster out of this, but that's a story for another day (still wrestling with Inkscape on that one).
As a non-academic that wrote a paper for the first time, I'll say that writing a good science paper takes an absurd amount of time, even on a topic you are very well versed in. It is also way different than other forms of writing, like blogs or technical documentation.
Frankly, I'm astounded I've even managed to get that result, working on it just one month from the submission deadline (a ridiculously massive time crunch) and only in my spare time, not having touched LaTeX in almost a decade. But if a proper heretic like me managed to get that far on their first try, then everyone who's considering writing a paper for the first time ought to have hope.
I might've somehow got an invitation to present a poster out of this, but that's a story for another day (still wrestling with Inkscape on that one).
Congratulations on getting that far! There's a story there – how did you get into writing this paper? What is it about?
Well, that's a long story. I'll try and keep it short.
So I've worked in my spare time for the past three years on an extremely esoteric and mind-bending reverse engineering technique I call delinking [1] and my tool for it [2] developed a small user-base. At some point I saw in a Discord server a call for papers for the SURE workshop, shitposted that it'd be funny to fry academic brains for a change, then got baited into writing it.
What started out as a long paper (12 pages) with quantitative case studies quickly got cut down to the bone and then some into a short paper (6 pages) that merely introduced my take on it and two qualitative case reports, because I realized it would take an amount of work on the scale of a master's thesis to do the long one. I barely managed to get that out of the door as my usual writing style is extremely unsuited for scientific papers (it took all the might of Gemini and Copilot to even wring out something that vaguely resembles academic vernacular from my first draft). I've submitted it, exhausted and deeply unsatisfied by the compromises I've had to do under that severe time crunch.
Then one month later, the reviews came in. The feedback was that the topic and my take on it were interesting; the criticisms can be summarized by saying that I've missed existing related work and it would've taken the long paper to address all the deficiencies in my paper. Fair I guess, but I definitely don't have that kind of spare time to spare on writing papers.
But alongside the notification decision, something unexpected was also sent: an invitation by the program committee to present a poster at the ACM CCS 2025 conference in Taiwan. This is the kind of world-class conference with a $1800 entry ticket, happening in a convention center whose area is measured in hectares, attended by professional members of an academic, industrial or state organization, all expenses paid in a five star hotel room with a king bed.
Somehow, a hobbyist who can charitably be described as a Doctor Frankenstein but with bits instead of gibs, who creates unholy chimeras of programs in his free time in spite of ABIs and common sense, received an invitation from the Hogwarts School of Witchcraft and Wizardry equivalent of the cyber-security academic world to present a poster. I can't even begin to express how utterly impossible this is.
So of course I said yes and now I'm in a mad dash to get everything ready in time for the conference, three weeks from now. Bonus point: it's on the other side of the planet and in my entire adult life I've basically never got farther away than the next county (and groaning while doing so). There's even more to that story, but I'll leave it for a blog post once it's done.
[1] Others call it unlinking, binary splitting or binary reassembly.
[2] https://github.com/boricj/ghidra-delinker-extension
So I've worked in my spare time for the past three years on an extremely esoteric and mind-bending reverse engineering technique I call delinking [1] and my tool for it [2] developed a small user-base. At some point I saw in a Discord server a call for papers for the SURE workshop, shitposted that it'd be funny to fry academic brains for a change, then got baited into writing it.
What started out as a long paper (12 pages) with quantitative case studies quickly got cut down to the bone and then some into a short paper (6 pages) that merely introduced my take on it and two qualitative case reports, because I realized it would take an amount of work on the scale of a master's thesis to do the long one. I barely managed to get that out of the door as my usual writing style is extremely unsuited for scientific papers (it took all the might of Gemini and Copilot to even wring out something that vaguely resembles academic vernacular from my first draft). I've submitted it, exhausted and deeply unsatisfied by the compromises I've had to do under that severe time crunch.
Then one month later, the reviews came in. The feedback was that the topic and my take on it were interesting; the criticisms can be summarized by saying that I've missed existing related work and it would've taken the long paper to address all the deficiencies in my paper. Fair I guess, but I definitely don't have that kind of spare time to spare on writing papers.
But alongside the notification decision, something unexpected was also sent: an invitation by the program committee to present a poster at the ACM CCS 2025 conference in Taiwan. This is the kind of world-class conference with a $1800 entry ticket, happening in a convention center whose area is measured in hectares, attended by professional members of an academic, industrial or state organization, all expenses paid in a five star hotel room with a king bed.
Somehow, a hobbyist who can charitably be described as a Doctor Frankenstein but with bits instead of gibs, who creates unholy chimeras of programs in his free time in spite of ABIs and common sense, received an invitation from the Hogwarts School of Witchcraft and Wizardry equivalent of the cyber-security academic world to present a poster. I can't even begin to express how utterly impossible this is.
So of course I said yes and now I'm in a mad dash to get everything ready in time for the conference, three weeks from now. Bonus point: it's on the other side of the planet and in my entire adult life I've basically never got farther away than the next county (and groaning while doing so). There's even more to that story, but I'll leave it for a blog post once it's done.
[1] Others call it unlinking, binary splitting or binary reassembly.
[2] https://github.com/boricj/ghidra-delinker-extension
That’s wonderful :) it sounds like you have an adventure ahead of you! Thank you for the story
You sound like exactly the sort of wizard they want to see!
You sound like exactly the sort of wizard they want to see!
Care to post the paper here? Would like to read it.
I think it's too weak as a paper for me to put out there. I also haven't even applied the minor corrections from the reviews.
If you want to take a peek at the case studies, I've blogged about my butchering of aln across C standard libraries and operating systems [1]; there is also widberg's outstanding write-up of their FUEL decompilation project [2], which uses ghidra-delinker-extension as part of the magic.
If you want to read a paper on the technique, there is the one for Ramblr [3] which I became aware of after the reviews came back.
[1] https://boricj.net/atari-jaguar-sdk/2023/11/27/introduction....
[2] https://github.com/widberg/fmtk/wiki/Decompilation
[3] https://www.ndss-symposium.org/wp-content/uploads/2017/09/nd...
If you want to take a peek at the case studies, I've blogged about my butchering of aln across C standard libraries and operating systems [1]; there is also widberg's outstanding write-up of their FUEL decompilation project [2], which uses ghidra-delinker-extension as part of the magic.
If you want to read a paper on the technique, there is the one for Ramblr [3] which I became aware of after the reviews came back.
[1] https://boricj.net/atari-jaguar-sdk/2023/11/27/introduction....
[2] https://github.com/widberg/fmtk/wiki/Decompilation
[3] https://www.ndss-symposium.org/wp-content/uploads/2017/09/nd...
Semi-related, someone made a hilarious spoof of McCarthy's writing in a blog, Yelping with Cormac[1].
[1] https://yelpingwithcormac.tumblr.com/post/11950502897/taco-b...
[1] https://yelpingwithcormac.tumblr.com/post/11950502897/taco-b...
Cormac McCarthy had a level of emotional competence that is somewhat transferable from his works. This list of tips becomes easier to follow as emotional insecurities and desperation are better managed.
I agree with most of the advice given here; but the footnotes thing, I feel like there are many authors (Freud and Kant come to mind immediately) where losing the footnotes would take away from the content. There is definitely some value in having non-linear writing, even for work that isn't creative.
Would love to get David Foster Wallace's advice on scientific papers
I like a lot of the advice here, except
> Keep sentences short, simply constructed and direct. Concise, clear sentences work well for scientific explanations. Minimize clauses, compound sentences and transition words — such as ‘however’ or ‘thus’ — so that the reader can focus on the main message.
Repetitive sentence structure is not engaging and lulls a reader to sleep, no matter the context. Clauses and transition words and nontrivial sentence structure allow for qualification and clarification, juxtaposition and contrast, and emphasis, often with many fewer words than if written as a series of single independent clauses. A short sentence following longer ones punctuates its point and can effectively lead into subsequent sentences that express more complex ideas/explanations.
In my own scientific writing I also frequently use compound sentences to indicate that the ideas are related (causally or otherwise). It's also unclear to me how one could more efficiently communicate logical or causal flow between ideas than with transition words like "thus" or "therefore."
> Keep sentences short, simply constructed and direct. Concise, clear sentences work well for scientific explanations. Minimize clauses, compound sentences and transition words — such as ‘however’ or ‘thus’ — so that the reader can focus on the main message.
Repetitive sentence structure is not engaging and lulls a reader to sleep, no matter the context. Clauses and transition words and nontrivial sentence structure allow for qualification and clarification, juxtaposition and contrast, and emphasis, often with many fewer words than if written as a series of single independent clauses. A short sentence following longer ones punctuates its point and can effectively lead into subsequent sentences that express more complex ideas/explanations.
In my own scientific writing I also frequently use compound sentences to indicate that the ideas are related (causally or otherwise). It's also unclear to me how one could more efficiently communicate logical or causal flow between ideas than with transition words like "thus" or "therefore."
I second that.
> short, simply constructed and direct [sentences]
require an honed writing skill; a finely tuned feeling for language, reading flow and contiguous thought; and while time and effort can culminate in such abilities, they shouldn't be prioritized.
It's better to have adept editors who like fiddling--sorry, tinkering, with syntax and semantics.
> short, simply constructed and direct [sentences]
require an honed writing skill; a finely tuned feeling for language, reading flow and contiguous thought; and while time and effort can culminate in such abilities, they shouldn't be prioritized.
It's better to have adept editors who like fiddling--sorry, tinkering, with syntax and semantics.
Cormac McCarthy was an amazing writer. For anyone with a strong stomach I would recommend "Blood Meridian". It's one of the most extraordinary books I have ever read. At times achingly beautiful, at times utterly horrifying. It's a really remarkable journey and the memory of it still troubles me.
David Foster Wallace wrote a memorable review of it once in a piece for Salon, which read (in its entirety)
"Don't even ask."[1]
If you've read the book, you know.
[1] The article is "Five direly underappreciated U.S. novels > 1960." https://www.salon.com/1999/04/12/wallace/
David Foster Wallace wrote a memorable review of it once in a piece for Salon, which read (in its entirety)
"Don't even ask."[1]
If you've read the book, you know.
[1] The article is "Five direly underappreciated U.S. novels > 1960." https://www.salon.com/1999/04/12/wallace/
This would ideally be paired with a Pynchon guide to publishing a work of mathematics.
These are good tips for any non-fiction writing, not only limited to technical papers
> Dashes should emphasize the clauses you
consider most important — without using
bold or italics —and not only for defining
terms.
So this is why the LLMs do it
So this is why the LLMs do it
"Gemini write me a paper in the style of Blood Meridian about Ribosomal DNA's function in cell respiration."
I havto eA uh a lot ayaarrrr you by
Most of the advice is good, though not particularly different from an advice you would give about writing any essays. This one though:
> Avoid placing equations in the middle of sentences. Mathematics is not the same as English, and we shouldn’t pretend it is.
I don't know what to make of it. Equations are supposed to be part of sentences, and mathematical equations are compact expressions of relations. For example, the sentence,
> Avoid placing equations in the middle of sentences. Mathematics is not the same as English, and we shouldn’t pretend it is.
I don't know what to make of it. Equations are supposed to be part of sentences, and mathematical equations are compact expressions of relations. For example, the sentence,
Newton taught us that force is equal to mass times acceleration, where both mass and accelerations are inertial quantities.
Newton taught us that $F=ma$, where both the mass $m$ and acceleration $a$ are inertial quantities.
I think it's just wrong. Related, linked recently here: https://dwest.web.illinois.edu/grammar.html
The authors are biologists, so I suspect they're not particularly versed in mathematical writing (and that McCarthy was not likely providing them much advice on it).
The authors are biologists, so I suspect they're not particularly versed in mathematical writing (and that McCarthy was not likely providing them much advice on it).
The link is a pretty great. A bit long though. Not sure if I can get my charges to read it all and apply it to their writing.
Yes, this is just bad advice if your audience is fluent in mathematical notation.
There's another piece of advice that is bad for formal mathematical writing:
> And don’t use the same word repeatedly — it’s boring.
If the word has a formal definition, this is bad advice. For example, in a game theory setting, we might say there are n agents. If we start calling the agents players, people, etc., it can get confusing, especially if there are other entities involved besides the agents (such as a mechanism designer or so on).
There's another piece of advice that is bad for formal mathematical writing:
> And don’t use the same word repeatedly — it’s boring.
If the word has a formal definition, this is bad advice. For example, in a game theory setting, we might say there are n agents. If we start calling the agents players, people, etc., it can get confusing, especially if there are other entities involved besides the agents (such as a mechanism designer or so on).
dhfbshfbu4u3(2)
This echoes advice I first read in Strunk & White. It remains the most actionable tip for better writing I'm aware of, technical or otherwise.
Aside: I consider McCarthy's residency at SFI an ideal job