Germanic vs Latinate Words in Writing

with many forebīsena for your quick teachings in wordcraft

You’ll often see mixed in with other writing advice an instruction to prefer Germanic words to Romance/Latinate ones:

Bad writers, and especially scientific, political and sociological writers, are nearly always haunted by the notion that Latin or Greek words are grander than Saxon ones, and unnecessary words like expedite, ameliorate, predict, extraneous, deracinated, clandestine, sub-aqueous and hundreds of others constantly gain ground from their Anglo-Saxon opposite numbers.

• George Orwell, “Politics and the English Language”

. . . publish stuff online, because an audience makes you write more, and thus generate more ideas; print out drafts instead of just looking at them on the screen; use simple, germanic words; learn to distinguish surprises from digressions; learn to recognize the approach of an ending, and when one appears, grab it.

• Paul Graham, “Writing, Briefly” (emphasis mine)

Avoid the elaborate, the pretentious, the coy, and the cute. Do not be tempted by a twenty-dollar word when there is a ten-center handy, ready and able. Anglo-Saxon is a livelier tongue than Latin, so use Anglo-Saxon words.

• Strunk & White, The Elements of Style (emphasis mine again)

Notice “the livelier tongue” (from the Proto-Germanic *tungō through Old English tunge) and not “language” (from the Latin lingua through Old French langage).

But I’m skeptical that these writers practice what they preach. More generally I doubt that skilled writers typically have an explicit grasp of what makes their own writing compelling. Strunk and White famously misunderstand the passive voice in their section criticizing it.

I checked whether Orwell and Graham in their essays actually use more Germanic language than peer writers. A script that checks for the words “French,” “Latin,” and “Greek” in a word’s Wiktionary entry does pretty well (see the matrix below, where a positive result for a word means that it’s Latinate)—better than GPT, which thinks most words are Germanic and makes up funny etymologies when asked to show reasoning.

I compare Orwell to Chesterton and Paul Graham to Sam Altman. Here’s the percentage of words that the script marks as Latinate, based on ~50,000 words each:

• Altman: 69.1%

• Graham: 68.6%

• Chesterton: 67.9%

• Orwell: 66.3%

So I don’t see meaningful differences.

There’s definitely something to this advice though. Consider Anglish, a version of English with “many fewer words borrowed from other tongues.” I can try to write like this myself using only Germanic words and you can hear a likable ring from time to time. It can become silly when no fitting word can be found without falling back on Latin or French. Take this bit of the page (uh, leaf?) on Julius Caesar:

Bewhile a trip to Greece, Caesar was kidnapped by seareavers and held for eddeeming. When the seareavers told him that they plotted to ask for twenty silvers for his befreeing, he wrathly onheld that he was worth fifty silvers at least. He upholded friendly bonds with his kidnappers, at one time laughingly tiding them that upon his freedom he meant to hunt them down and roodfasten them. Meal their awe when, after they leased him, he did swith that.

“Eddeeming” is ransom and “roodfasten” is crucify. At some point it becomes silly—I rewrote the sentence before the last quote to avoid the word “example” since the only alternative in the Anglish wordbook is the bizarre Old English “forebisen.”

The Fowler brothers warn about overdoing it in their first comment on the rules they list in the first chapter of The King’s English:

Prefer the familiar word to the far-fetched.
Prefer the concrete word to the abstract.
Prefer the single word to the circumlocution.
Prefer the short word to the long.
Prefer the Saxon word to the Romance.

These rules are given roughly in order of merit; the last is also the least. It is true that it is often given alone, as a sort of compendium of all the others. In some sense it is that: the writer whose percentage of Saxon words is high will generally be found to have fewer words that are out of the way, long, or abstract, and fewer periphrases, than another; ​and conversely. But if, instead of his Saxon percentage's being the natural and undesigned consequence of his brevity (and the rest), those other qualities have been attained by his consciously restricting himself to Saxon, his pains will have been worse than wasted; the taint of preciosity will be over all he has written.

He points out that words like “brave” and “battle” come from French but maintain the light touch of Saxon. This plain, rustic sound that comes from short words, hard consonants, and general old-timeyness is what’s worth having intuitions about. The etymology is just a good predictor of this sound and the related feelings of formal vs informal and concrete vs abstract.