Read on Twitter
Lot of dunking on this tweet, all of it good.
When I studied linguistics, I was trained to look to how language is functional and to avoid judgment. https://twitter.com/KordingLab/status/1296459722313392131
When I studied linguistics, I was trained to look to how language is functional and to avoid judgment. https://twitter.com/KordingLab/status/1296459722313392131
Many of the problematic words are problematic because they are vague. For example, "mediates". Or "crucial role", or "pivotal".
Vague is good though, when your point is "these things seem related, but, huh, can't really say much more than that right now."
Vague is good though, when your point is "these things seem related, but, huh, can't really say much more than that right now."
In that case, criticizing the word is akin to saying we shouldn't stop and publish before solving the problem. That is, we should write books but never papers.
So, sure, it's a weasel word, but weasel words are a good thing, and it's good that the English language has them.
So, sure, it's a weasel word, but weasel words are a good thing, and it's good that the English language has them.
"Represent" is different - it used, innocently, to mean "brain activity correlates with" by one segment of systems neuroscience, but has a more restrictive meaning in another. Both uses are perfectly find; it's just potentially confusing when they talk to each other.
Anyone who's been around for a while knows to be careful (I usually use "encode" on the assumption that it's less loaded).
So I guess pointing out the possibility of confusing is useful for the newbies, but probably better to explain rather than to criticize.
So I guess pointing out the possibility of confusing is useful for the newbies, but probably better to explain rather than to criticize.
Of course, lots of people hate "encodes" or "codes" for the same reason they hate "represent", but, often this is precisely what the author means - what people dislike is the implication that correlation is enough to conclude something stronger.
In other words, some of the language policing is empirical disagreement (which is good) disguised as linguistic policing (which is misdirected).
Science is building a boat while on a leaky raft in the middle of the ocean during a storm. Meaning, we don't know all the answers yet. 90% of all description is guessing what it will look like when we know more. So it's going to have to be problematic.
So scientific language for newly discovered results is unavoidably problematic if we want it to have rigorous meaning.
We don't live in that world. All we can do is touch our little part of the elephant.
We don't live in that world. All we can do is touch our little part of the elephant.
And language, as it is actually used, not as it is imagined by prescriptivists, is flexible enough to handle that admirably.
(Note this is true for everyday language too, where ambiguity is often used strategically, but that's a different - and fascinating - topic)
(Note this is true for everyday language too, where ambiguity is often used strategically, but that's a different - and fascinating - topic)
In conclusion, weasel words are (1) functional, (2) typically used because they have usefully weaselly properties, and (3) highly recommended for writers and speakers of all levels.
