Hacker News new | comments | show | ask | jobs | submit login
Lessons from the CIA’s classified guide to good writing (muckrock.com)
182 points by danso on July 20, 2017 | hide | past | web | favorite | 66 comments

My grad school teachers would kick my butt for writing like that. Look at this lede.

"The trouble with CIA writing is much the same trouble that there is with other governmental writing and with writing generally in the United States."

How about

"CIA writing baffles our readers. We share this problem with other parts of government and the United States at large."

Check out this sentence.

"CIA's problem is unique only in the relation of our profession to our writing but this relation is important, for it lends itself to bad writing and if we do not write well we risk losing valuable information and wasting dangerous effort."

WTF? I think this writer means

"At CIA we write about hard-won information. We lose that information and waste our agency's work when we write badly."

Look, it's easier to be an editor than a writer. Everybody knows that. It's always easy to cherrypick a sentence or two and make them better.

The point is this: good writing isn't a solitary job; it takes editors as well as writers. It's like software, which takes coders and testers.

Personally, I think these edits do not preserve the meaning of the original quotes: The changed sentences stress different things than the original quotes to the extent that one could call them distortions rather than edits.

From my own experience, I have seen extraordinary writers, but I have never seen an extraordinary editor: I think the reason for this is that exceptionally great editors work in such a way that their own edits most explicitly convey the original work both in its form and in its spirit, so that the editor becomes almost invisible. Invariably, the more bragging I have seen from editors in prefaces etc. about their own work on some books, the worse the result has looked to me in the sense that it had very little to do with the original, and typically felt worse to me.

I think good editing is at least as hard as good writing, probably even more so because as an editor, you are working subject to more constraints, in addition to facing the same challenges that writers also face.

On the one hand I agree with you about editing being harder. On the other hand, and having been an editor in the past, I can that overall it is much easier. Working within the writer's constraints is usually less difficult than trying to figure out what the constraints should be in the first place. As an editor it can actually feel pretty empowering to slash away any final barriers between the writer's intended message and the audience.

And, yes, good editing is invisible. It's a very good feeling, when you can discover the author's original mindspace, and hack away at it.

> The point is this: good writing isn't a solitary job; it takes editors as well as writers.

That's an empirical claim, and I'm wondering how often it is actually true.

I have no doubt that for many people, and probably most people, it is true. But how many people out there can (and do) write well without an editor?

They exist. They are called writers who know how to edit.

Most writers need editors because of the two d's: 1. Distance 2. Deadlines

Distance is necessary to read your work as a reader and not a writer. In order to polish your writing you have to have an insane amount of empathy for your reader. It's hard for people to edit their own work because it's hard to forget what you know. You know what you've meant by a phrase, of course, but your reader, coming from who knows what background, may not. Your diction, metaphors, etc. may be more idiosyncratic, arcane, limited, or just plain confusing than you realize. We're talking a whole tabula rasa situation which is not easy to pull off.

As to our second d, deadlines, they prevent writers from being able to address the first d alone. As with any art, critiquing your own work from the perspective of the audience takes 'fresh eyes', 'fresh ears'--when you're grinding on a piece you become so involved it's hard to look at it from the audiences perspective unless you have sufficient time (sometimes months) to step away and get your brain out of creative/obsessive mode. you're blessed if you actually have the time to simply take off from your writing like this. As such, an editor is your stand in--a set of fresh eyes that can vouch for the audience while you keep grinding away and ensure your hit that deadline.

Haha, yeah, I was thinking the same thing. Just give your operatives Strunk&White and be done with it.

Although I'd check out what actual grammarians and linguists think of S&W before doing that (hint: they're not fans)

Huh, what do they have against it? I could see a linguist considering it too prescriptive, but in this case the whole point is prescribing a particular register of the language for effective communication. But what about the grammarians?

Things like "Don't use the passive voice" whilst using the passive voice in the same paragraph and other egregious grammatical errors sprinkled wildly through the text.

Basically the same kind of dumb and wrong prescriptivism the linguists and dictionaryists hate about it.

Well, I'd check out how well grammarians and linguists write before I worry about their opinion of S&W...

For many years, I have used "Words of Estimative Probability" - a similar CIA "on writing" piece: https://www.cia.gov/library/center-for-the-study-of-intellig...

Also written in the 60s, it reflects upon their experience with how to convey uncertainty in prose, in a meaningful and consistent manner. Imagine an analyst writing a two pager for the President, who is going to use it to decide go/no go on a covert action. It's a big, ugly and complex situation - with a mixture of highly confident assessments, educated guesses and total speculation. Synthesizing it into something simpler is The Art of Writing; doing so while not losing the anecdotal quantification of uncertainty of the discrete issues is a more unique skill.

If you are in a position where you write about uncertain situations, it's worth the read. I'm a security guy and frequently find myself in that role (think annual risk assessments and estimating the likelihood of a successful attack). It's been a go-to reference for many years to refresh my thinking.

One of my backburner projects is to check out the declassified CIA documents, OCR and clean up them, and make these available on my website [1].

The idea was to look for stuff that was interesting as a resoure for role playing games, mostly, but I also included one about "how to produce estimates"[2] that is close in spirit to the one about writing.

[1] http://www.pa-mar.net/Main/Games/Games/Espionage%20Files.htm...

[2] http://www.pa-mar.net/resources/The-evolution-of-some-techni...

USG documents are automatically declassified after 25 years by default. CIA makes their declassified documents available via the CREST database[1]. Data.World has published the metadata about these documents in a nice format[2]. Might be worth getting in touch with Noah Ripper to coordinate your efforts [3]

[1] https://www.cia.gov/library/readingroom/collection/crest-25-...

[2] https://data.world/cia-crest-files/cia-crest-archive-metadat...

[3] https://data.world/nrippner


by @reedes (Reed Esau) is my favorite tool to automatically help me with many of the mentioned points.

Spell-corrects/warns you about:

  * Weak and lazy words
  * Redundancies
  * Puffy & Jargon
  * To be and passive voice
  * Colloquialisms, idioms, and similies
  * Lots more..

Is there an emacs version?

Or better yet, if you notice that your workplace is starting to resemble the old, bloated, wildly checkered bureaucracy of the CIA run like hell. This guide seems less like a style guide in general, and more like an attempt to flush some nonsense out of the CIA in particular.

> it says that CIA writing is “full of jargon, of would-be professional language, of clichés; it is even opaque”

That's what I'd have expected frankly, which is what you'd expect from people trying to build towers out of sand. If your ideas are lacking, bullshit, and if bullshit isn't working, obfuscate... that's the alphabet-agency way.


and spelling cliche with an apostrophe isn't ;-)

That is an accent, not an apostrophe

oops my bad

Apostrophe? :-?

I love your username, and assume it's from 'The Point'?

There is a lot in here that reminds me of the Air Force's Tongue & Quill (public). For example, the CIA's "avoid jargon" section looks eerily similar to page 82 of T&Q.


I think they're both proper but I feel like it'd be better titled "The CIA's guide to writing well.".

Just being pedantic though maybe.

If I were wearing my editor hat, I'd cross out 'well' and suggest that in that case it's redundant. "The CIA's Guide to Writing" is just as strong.

It's an internal memo, so "The CIA's" is redundant.

Which leaves us with "Guide to Writing". Can we do better?

Stephen King's book on writing is called:

"On Writing".


As I recall "On Writing" does not have a great deal to say about writing, it's much more about the life of a writer. To get a pretty good book that is actually about writing you'd want to upgrade to William Zinsser's "On Writing Well." So...

I read the first two sections of On Writing and that was enough for me. The first is a memoir from Stephen King in which he goes through major events in his life. The second is a writing style guide which centers around one theme: simplicity in writing allows for greater imagination in the reader. The second section goes through sentence structure, syntax, verb/adverb choice, pronoun placement, ect.

I would agree that the first section does not say a great deal about writing, but the second section reminded me of sitting in my High School english class with my favorite teacher writing a couple sentences on the board as an example and passionately (oops) dissecting the structure with arrows and margin notes to explain how and why a certain picture was playing in my mind while reading them.

Placing King's personal anecdotes and thoughts about the "life of a writer" before the actual lesson felt to me, when I began the second section, like I was sitting in the classroom of an artist who has far more passion for his work than I have ever felt for anything in my life before then. It made the writing lesson more impactful, if anything.

But it certainly had a lot to say about writing.

The document itself is titled "Essays on CIA Writing". And, it's largely on trends observed within CIA writing, so I'd argue that it's well titled.

Or, complying with the CIA's guidelines regarding capitalization:

"On writing".

Well once you've distilled it down that far, might as well drop 'On' because it's not adding much.



Isn't the "on" implicit if the title were just "Writing".

If the book was just writing and not concerning the subject it would be "writings" so there should be no confusion?

This is quickly turning into a new comic idea for Randall Munroe (XKCD)

Writing That Works

Pedantry would also require that you drop the "the." It's not the TLA, or the IBM or the GE.

It is "the" Central Intelligence Agency though. I think "the" is appropriate in this case.

They prefer “CIA” without an article, oddly enough. At CIA, CIA concludes, etc. When speaking with other Washington types the A is often dropped, too. “(So and so) at CI” is something one hears a lot.

I'm guessing they also have a coding manual.

Tabs or spaces?

Spyces, of course.


Apart from this being very elementary instructions, I think the bigger story is that the CIA is abusing its power to classify information so extensively that simple writing advice was classified for 55 years.

Haha, I think you are missing the much more obvious explanation. In any organization dealing with classified information, there is a very well known management trick: if you want your employees to read a document, make it classified and give them clearance to read it. People will be much more interested in it.

That case sounds like a textbook example of a document where you want to use that trick.

>obvious >trick

It's not as contradictory as it sounds. Many such practices are fairly well know to general audiences from causal reading of articles just like this, but the practices often aren't identified as such by those who are subjected to it, or if it is identified, the practice works regardless. Take for example here on HN the large number of people who dismiss many of the practices of social media likes but get upset when their post is downvoted.

Such practices have a major effect on people regardless of how aware they are of what's actually going on.

"In order to popularize potatoes in France, Antoine-Augustin Parmentier placed armed guards around his potato fields, instructing the guards to accept all bribes and allow people to "steal" the crop."


Advertisement is an obvious trick that works.

While I agree with the idea that the CIA abuses power, techniques that enable effective communication are a HUGE advantage. Knowing how to write in a way that other people can actually understand (and take action on) is absolutely advantageous (as is any other technology that reduces miscommunication). Of course at this point in time you can take a course like Little Red Schoolhouse and come away with many, many practical actions you can take to improve the clarity of your written communication, so probably in effect is has already been declassified.

I can't wait to see the CIA guide to shitposting once it's declassified.

It just says: "Look at the top post, disagree with one point, bring up a distracting point"


- United States Marine Corps, School of Advanced Warfighting


I submitted that a few months ago! In general people don't realize the extent that information warfare is taken seriously in some circles.

I know Little Red Schoolhouse as the informal name for a particular writing class at the University of Chicago. Is it more general than that?

There are similar courses at other universities, but I don't know them by name, the general idea being that the courses are based on the data and theories that have come out of linguistics over the past 60 years.

It's easier to classify than de-classify. It's also easier to get dinged for under-classifying than for over-classifying.

Styleometry is a thing. You can somewhat identify an author by analysis of word choice, grammar (or lack thereof) and other features. With computers this can be done surprisingly accurately.

So yes there is a very good reason to keep the CIA writing advice classified. Even though it seems banal.

Are there ways to de-Styleometrize writing?

I destylometrize online sometimes: Use USA spelling, verbize nouns, use words like faucet and miss out Oxford commas.

There is adversarial stylometry software like JStylo-Anonymouth.

Translate to a non-Latinate language and back.

Follow a guidebook/protocol religiously

This doesn't surprise me at all. Consider what else you can learn about the CIA from the discussion. They could have released a redacted document but it would only have any information that was already publicly available. There's no value in that.

Anyone else noticed 2 spaces after a period and 1 space after a comma in the writings? I've always done the same, but for some reason 1 space after a period is correct.

That is how we were trained in typing class.

So 2 spaces only on a typewriter?

"...these tips will make that next report on regime change into your magnum opus"

Obligatory "you won't believe #5!"

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact