Suggestion: site-wide taboos
Every so often, someone on Less Wrong uses a word wrong.
What does it mean to use a word wrong? Can’t we use language however we want, as long as we manage to successfully communicate? Well, yes, we can, but we shouldn’t. Jargon terms, in particular, are used by professionals in a certain field in order to communicate concepts that are applicable chiefly in that field. They often have very precise definitions—”incunable”, for example, means “book printed in Europe before the year 1501”, and “sweet crude oil” means “petroleum with a sulfur content less than 0.42%”.
The thing about precisely-defined terms like these is that if you use one of them in a way that’s at odds with its official definition, you can cause people to have more misunderstandings later on. I admit I can’t think of a great example, but “obsessive–compulsive disorder” seems like a decent one: people often say “I’m so OCD” to mean that messy things annoy them, which seems like it could lead people to misunderstand when people actually have obsessive–compulsive disorder.
There are just two words I don’t really like LW’s usage of:
“Signaling”. I’m not actually sure exactly what “signaling” means—which is arguably reason enough for us not to use it. I get the impression that it’s usually used to mean exactly the same thing as “indicating”. If that’s the case, we should stop using it (or else only use it when everyone knows exactly what we mean by), and just say “indicating” instead. Or perhaps we don’t use “signaling” to mean exactly the same thing as “indicating”, but if that’s the case, I don’t know what the difference is, and I don’t know whether or not it matches the “real” meaning of the word.
“Affect” (the noun). Wiktionary defines it as “a subjective feeling experienced in response to a thought or other stimulus; mood, emotion, especially as demonstrated in external physical signs”. LW seems to use it as an exact synonym of “emotion”.
One person or even a few people not understanding some term isn’t a good enough reason for the rest of us to stop using it. It might be a good reason for someone to write a tutorial about it though… if Yvain hadn’t already written one.
Alternative suggestion, then: let’s update the wiki pages so that their definitions are actually correct.
Please do!
I’ve changed the page “Signaling” to contain Yvain’s definition, and changed its definition on “Jargon” to an abbreviated version of the same. I also gave the jargon page the following definition of “affect”, which I have doubts about: mood or emotion as demonstrated in external physical signs. “Affect” currently has no page.
Thank you. :-)
If these really are synonyms, then doesn’t signaling win by having fewer syllables?
Using it as a synonym of emotion would indeed be wrong, but I haven’t noticed people using it that way. Got a concrete example? I also disagree with the definition you gave, though; it fails to capture what I see as the defining quality of affect that distinguishes it from emotions in general, which is that it has only one axis which runs from good to bad.
Whoops, I was unclear. When I said it seems like “it’s usually used to mean exactly the same thing”, I meant that some people on Less Wrong use it that way, not that that’s the correct usage.
I suppose the only example that comes to mind is the term “affective death spiral”, which doesn’t seem (to me) to have anything in particular to do with affect rather than emotion. I’m tempted to declare “affect heuristic” an abuse of the term as well, except that “affect heuristic” is a term that’s actually used by experts.
Is that your best guess based on seeing the term used many times, or do you have some other type of evidence? That “defining quality” doesn’t seem to agree with what Wikipedia says at all.
I think the name is derived from the affect heuristic.
Googling for the phrase, the first hit is the LW wiki article about it.
I decline and/or oppose as appropriate.
I know what ‘signalling’ means far more clearly than I know what other words of equivalent ambiguity referring to subjects equivalent complexity. It is standard terminology) for a rather important concept and a and it would be hard to discuss most of the interesting subjects on lesswrong without it.
I don’t believe you. If there was a site-wide substitution :s/affect/emotion/ then many usages would either outright stop making sense or at very least mean something different to what they once did. Including whole swathes of the sequences.
I’ll concede that point; I haven’t been on Less Wrong much at all in the past year or so, so I wouldn’t relaly know.
You’re probably right, but I can’t actually think of any examples of this (apart from the phrase “affect heuristic”). I would be interested in learning of one.
Trivial point: Do you mean “Site-wide Taboos?” The current title is “side-wide taboos.”
Whoops.
There was recently a discussion of Lesswrong’s use of the word “Signaling,” and it seemed to me that upon consideration it was shown that we’re pretty much using it to mean what it means in a broader academic context. See this comment in particular.
With respect to the use of “affect,” I again disagree but there aren’t really any examples I can point to. I think its use in many cases is very similar to “emotion,” but I also think that its use fits pretty perfectly with your stated definition. For example, someone might say, “I have a really strong positive affect towards ponies, so my assessment of this fanfiction’s quality might be biased” (I have almost no idea why that is the example I thought of), and this fits pretty well with ” a subjective feeling experienced in response to a thought or other stimulus.”
With “affect,” I think we might have “good reasons” for using the word, but I concede that its use isn’t really necessary and that its use probably is because people want “to seem Less Wrong-y.”
This example supports Warrigal’s claim ‘affect’ as a psychology term is used incorrectly on the site. Vide beoShaffer’s link to the psychology wiki, particularly:
The statement with correct usage would then become:
Hah, thanks. So one cannot use the word to reference their own “subjective feeling” but can use it to reference others’?
(Sidenote: If you’re right, I guess most of its usage here is incorrect, and perhaps misleading, but it seems like we’d be wrong in an silly, pedantic, “what silly rules for word” sort of way. We’d still be wrong though.)
I think the idea is that affect is the outward appearance of a feeling or emotion or whatever. You could tell what your own affects are, but you’d have to look in a mirror or something.
You can feel sad, but affect happiness*, and appear to others as displaying a highly positive affect**.
* non-scientific usage
** scientific usage
You can feel sad, but signal happiness, and appear to others as displaying a highly positive affect.
Ok, so how would one signal happiness?
Signalling has an academic definition in economics, for sure. It’s used both in an intentional sense (“workers signal their conscientiousness to employers by making their way through a 4-year college degree”) and an unintentional sense (“being a high school dropout signals to the employer that a worker is in the bottom 5th percentile”)
However, I do think LW uses it in a intellectual hipster sense as well - “Do you really think that, or are you just signalling?”. The difference seems to me that instead of jockeying for economic advantage, we are accusing someone of jockeying for social status. Of course, such social jockeying is widespread, simply by dint of human nature. But I suppose we could replace this use of the word with “posturing” or something of the sort.
Signaling is a very commonly used word in economics, especially game theory.
The definition psychology wiki gives is notably different from wiktionary’s http://psychology.wikia.com/wiki/Affect specficly it defines affect as a persons apparent mood regardless of their internal emotional state. Signaling means something similar to what I think you mean by indicating, but I would consider ‘indicating’ to be awkward if not outright wrong in most places were I would use ‘signaling’. It may be a regional difference something. However, I would like a taboo on using rationalist as a synonym for rationality because rationalism already refers to a rather different concept in philosophy.
For some reason my comment isn’t displaying properly. It should end with a link to plato.stanford.edu/entries/rationalism-empiricism .
IIRC that weird disappearing act happens when a link doesn’t have proper formatting, such as missing the “http://” in the URL.
[Test Link](http://plato.stanford.edu/entries/rationalism-empiricism/) gives:
Test Link
(or just drop down the URL with the “http://” directly in the comment, as you can see above it automatically becomes a link)
You do realize you can edit your comments?
Yes, it stayed broken after multiple edits.
I had trouble with a link in my comment as well.