Hacker News Reader

Comment by f5ve | Hacker News Reader

[flagged]

In addition to this being pretty mean-spirited, I'm also fairly certain it's wrong. I have no trouble coming up with examples of the adjective "cogent" modifying a noun indicating a person; you can quickly find one such example ("one of the most cogent writers") in the archives of The Atlantic, and all over Google Scholar. I dug through every usage of the word "cogent" in the New Yorker and, while it took awhile, I did find a couple. Alas, there was no section in Garner's Usage Dictionary covering the use of the word, presumably because there's nothing controversial about it.

I think the jury is pretty well in: obviously, the word "cogent" describes the expression of an idea or an argument, not a person, directly, but idiomatically, a "cogent writer" is simply one who produces cogent output.

Wholeheartedly agree with this reply. Attacking grammar is only very, very rarely the right move. It’s most useful as a distraction akin to an ad hominem attack. Everyone knows what is meant by the author even if they don’t know the dictionary definition of cogent.

Add to that that dictionaries are lagging indicators of meaning. If people start using cogent differently, dictionaries will catch up later, but we will all understand each other ahead of that point.

It's been decades since some of the examples I pulled.

Boo. Sure it’s probably wrong to call someone “cogent”. But is there such a huge difference between saying “Harry has cogent writing” and “Harry is a cogent writer”? Is the reader confused about the meaning?

I might use “prosaic”, “flowery”, or “confusing” in the same way, even though strictly speaking those words are describing the writing and not the writer.

Even if the meaning gets across, the latter is a less cogent sentence.

To clear this up once and for all, we can refer to the much-respected Oxford English Dictionary (OED) [1].

The current definitions are: "1.a. Constraining, impelling; powerful, forcible."; "1.b. esp. Having power to compel assent or belief; argumentatively forcible, convincing."; and "1.c. [as part of certain] dependent [phrases]."

A historical definition of "cogent" was: "Of persons: Employing force or compulsion, peremptory. Obsolete. rare," which includes an example phrase from 1672 that demonstrates this usage as a descriptor for people.

This confirms that "cogent" is no longer a strictly valid descriptor for people according to the OED, though this usage was once valid in the past. However, I agree with other readers that this doesn't take away from the article's substance, which draws from the author's experience in software.

[1] (Paywalled) https://www.oed.com/dictionary/cogent_adj?tl=true

The OED can say whatever it likes, but if it invalidates "cogent" as a description for an actor rather than their actions, it disagrees with modern usage, and is thus itself, in a word, wrong.

It turns out that your point is part of a fair criticism of the OED, as critics such as Oxford linguist Roy Harris [1] claim that dictionaries should be descriptive instead of prescriptive. The OED itself has also varied on its presentation of itself across its history as a resource that describes versus prescribes proper English usage.

I still believe that OED definitions are reliable for checking usage in formal writing, though its definitions are less important for personal articles such as the author's. But in any case, the author still makes a good point that clear writing is useful when solving engineering problems in a team.

[1] https://en.wikipedia.org/wiki/Oxford_English_Dictionary#Rece...

Reply on news.ycombinator.com