Note on Terminology: "Rationality", not "Rationalism"
post by Vladimir_Nesov · 2011-01-14T21:21:55.020Z · LW · GW · Legacy · 51 commentsContents
51 comments
I feel that the term "rationalism", as opposed to "rationality", or "study of rationality", has undesirable connotations. My concerns are presented well by Eric Drexler in the article For Darwin’s sake, reject "Darwin-ism" (and other pernicious terms):
To call something an “ism” suggests that it is a matter ideology or faith, like Trotskyism or creationism. In the evolution wars, the term “evolutionism” is used to insinuate that the modern understanding of the principles, mechanisms, and pervasive consequences of evolution is no more than the dogma of a sect within science. It creates a false equivalence between a mountain of knowledge and the emptiness called “creationism”.
So, my suggestion is to use "rationality" consistently and to avoid using "rationalism". Via similarity to "scientist" and "physicist", "rationalist" doesn't seem to have the same problem. Discuss.
(Typical usage on Less Wrong is this way already, 3720 Google results for "rationality" and 1210 for "rationalist", against 251 for "rationalism". I've made this post as a reference for when someone uses "rationalism".)
51 comments
Comments sorted by top scores.
comment by Vladimir_M · 2011-01-14T22:44:50.324Z · LW(p) · GW(p)
As for the terms "rationalism" and "rationalist," they have a strong established historical meaning quite different from the way they're commonly used by many people here. The first thing that occurs to me when I hear them is the old philosophical notion of "rationalism" as opposed to "empiricism."
Also, it's important to note that historically these have never been terms of uncontroversial and unambiguous praise. In many contexts, they have been traditionally understood to convey criticism, not compliment. For example, when Michael Oakeshott titled his essay Rationalism in Politics, he definitely didn't aim to make the reader positively disposed towards the subject from the title. Whether and to what extent people on LW tend to commit the same errors and hold the same unsubstantiated beliefs that have traditionally been connoted by this term is certainly a complex and amusing question.
As the bottom line, the "rationalism" terminology is probably too deeply embedded in the LW folklore to ever be abandoned, but I would certainly advise against using it when talking to outsiders. Even if people understand the term precisely the way it's used here, describing oneself like that explicitly is a status-lowering way of qualifying oneself.
Replies from: Kaj_Sotala, Vladimir_Nesov, jfm↑ comment by Kaj_Sotala · 2011-01-15T11:07:18.711Z · LW(p) · GW(p)
The first thing that occurs to me when I hear them is the old philosophical notion of "rationalism" as opposed to "empiricism."
This has confused at least one newcomer to the site: I also remember correcting another person, but couldn't find the comment in question.
↑ comment by Vladimir_Nesov · 2011-01-14T23:05:28.138Z · LW(p) · GW(p)
As the bottom line, the "rationalism" terminology is probably too deeply embedded in the LW folklore to ever be abandoned
The term "rationalism" is not actually used on LW much, "rationality" is much more frequent. (I understand that you meant both in that phrase, but it's not clearly expressed.)
(3720 Google results for "rationality" against 251 for "rationalism"; added to the post.)
Replies from: Vladimir_M↑ comment by Vladimir_M · 2011-01-14T23:15:51.183Z · LW(p) · GW(p)
That's true, but "rationalist" is used extremely frequently as a noun or adjective. (Google claims about four thousand hits on LW for the plural "rationalists.") The word "rationality" indeed has a meaning separate from the traditional polemical sense of "rationalism," and it's not too far from what's commonly meant by it on LW. However, "rationalist" is not separable from "rationalism."
Replies from: Vladimir_Nesov↑ comment by Vladimir_Nesov · 2011-01-14T23:32:16.613Z · LW(p) · GW(p)
However, "rationalist" is not separable from "rationalism".
Word forms are very important when we are talking about connotations. "Rationalist" is a different story, see current version of the post.
(Added google results data on "rationalist" to the post.)
Replies from: Vladimir_M↑ comment by Vladimir_M · 2011-01-15T06:40:59.294Z · LW(p) · GW(p)
You say:
So, my suggestion is to use "rationality" consistently and to avoid using "rationalism". Via similarity to "scientist" and "physicist", "rationalist" doesn't seem to have the same problem.
I think this analogy is inaccurate. The suffix -ist has several distinct meanings, and my impression is that the general public will be apt to understand "rationalist" to mean "someone who subscribes to rationalism" (whatever exactly that connotes in the given context), not "a practitioner of/expert in rationality."
I suppose you have the latter meaning in mind when you make the analogy with "physicist," but I don't think that's how the term is likely to be perceived outside of LW. (Just like e.g. "historicist" doesn't mean "historian.")
comment by TheOtherDave · 2011-01-14T21:55:38.993Z · LW(p) · GW(p)
I agree that "rationalism" has an ideological implication that "the study of rationality" (or "rationology"?) does not.
I also think that for quite a lot of LW, "rationalism" is the correct word (although we may wish to use a different one for signaling purposes).
Replies from: Vladimir_Nesov↑ comment by Vladimir_Nesov · 2011-01-14T22:16:29.682Z · LW(p) · GW(p)
I agree that "rationalism" has an ideological implication that "the study of rationality" (or "rationology"?) does not.
"Rationology" is much, much worse, because it's an "-ology" and a new term that wasn't historically used much. (I invite you to flee in horror from the stuff that google search on "rationology" turns up.)
Replies from: NancyLebovitz↑ comment by NancyLebovitz · 2011-01-18T13:57:52.126Z · LW(p) · GW(p)
This at least raises an interesting question: Why are people capable of feeling worthless to the point that it compromises their ability to function?
More generally, I regret it if "rationology" isn't available for sensible use-- I'd like to see a formal study of how people reason.
comment by Jayson_Virissimo · 2011-01-15T18:44:49.514Z · LW(p) · GW(p)
I think people confusing rationalism with one side of the age old rationalism/empiricism debate in philosophy is a much more probable event than someone confusing it with a faith based ideology.
Replies from: Tuukka_Virtaperko↑ comment by Tuukka_Virtaperko · 2012-10-10T01:35:49.680Z · LW(p) · GW(p)
According to Rationality (1989) by Nicholas Rescher, who is for all intents and purposes a rationalist in the sense LW (not academic philosophy) uses the word, the LW rationality is a faith based ideology. See confirmation holism by Quine, outlined in "The Two Dogmas of Empiricism". Rationality is insufficient to justify rationality with rational means, because to do so would presuppose that all means of justification are rational, which already implicitly assumes rationality. Hence, it cannot be refuted that rationality is based on faith. Rescher urges people to accept rationality nevertheless.
comment by DSimon · 2011-01-14T21:48:02.547Z · LW(p) · GW(p)
I disagree mildly. As a practical matter, I see your point that we want to dissociate what we do at LW from the sorts of things that ideologies do badly (insularity from disagreement, cultishness, faith, etc.)
But the thing is: rationality is an ideology when it's used to the degree we do here. We might want to consciously not refer to it as such for PR purposes, but we still ought to be clear on the priority we place on being rational.
Replies from: Vladimir_Nesov↑ comment by Vladimir_Nesov · 2011-01-14T21:57:58.278Z · LW(p) · GW(p)
But the thing is: rationality is an ideology when it's used to the degree we do here. We might want to consciously not refer to it as such for PR purposes, but we still ought to be clear on the priority we place on being rational.
Taboo "ideology". According to the inferences you've borrowed from your usage of the word, mathematics is also an ideology (adhered to by mathematicians), which doesn't sound right.
Replies from: DSimon, rwallace↑ comment by DSimon · 2011-01-14T23:06:37.482Z · LW(p) · GW(p)
When words end in -ism, I tend to think of them as a framework or philosophy that is applicable in general terms, on a day-to-day basis. This has negative associations and examples (i.e. Darwinism, creationism), but on the whole I see it as something that applies strongly to what LWers mean by rationality, particularly instrumental rationality.
Mathematics might also be appropriately described that way, but not I think when it comes to mathematicians, who certainly use mathematics as a primary framework for their jobs, but may compartmentalize the rest of their daily activities and use non-mathematical approaches there. A lot of LW topics might appropriately be described as "mathematicism" in this sense, for example posts that give general day-to-day advice based on game theory or decision theory.
Edit: On further thought, my two examples actually don't match my stated criteria, since they're not about day-to-day activity. My revised explanation is that it has less to do with how often the ideas are used, and more to do with how much the person consciously identifies themselves with (or, in the case of "Darwinism", identifies other people with) the philosophy as a kind of group membership or point of pride.
I think rationality, and rationalists, fit under that -ism-ness as well.
↑ comment by rwallace · 2011-01-15T02:05:24.054Z · LW(p) · GW(p)
I don't think the accurate analogy is with mathematics itself.
Consider the belief that the universe must be fundamentally mathematical in nature, that if some aspect of it appears not to admit of mathematical description that's just because we don't understand it well enough yet, or our tools can't handle the complexity or somesuch, not because of any fundamental incompatibility.
That belief (to which I and probably most readers here subscribe) is an ideology as the term is being used here.
I will suggest that belief is to rationalism as mathematics is to rationality.
Replies from: Documentcomment by Student_UK · 2011-01-15T12:20:38.595Z · LW(p) · GW(p)
Does the same reasoning apply to all -isms? Empircism, materialism, internalism and externalism to name a few.
ism has a few different uses it can indicate a group of principles (empiricism), or a political movement (socialism), or a type of discrimination (sexism).
Your worry seems to be that "rationalism" looks like a political movement, but that sort of thing is more likely to be determined by how the people who use the term of themselves act. And that problem does not go away by refusing to use certain words. If people who call themselves rationalist try to spread their ideas through exerting political force rather than calm argumentation then it is going to look like a political movement, whether or not a ism is being used.
If you're worried about this then you need to deal with the underlying problems, not just worry about words (after all, I'm not aware that empiricists have this concern).
As an aside, I think the main reason for not using it is that it already taken: http://plato.stanford.edu/entries/rationalism-empiricism/
Replies from: timtylercomment by ata · 2011-01-14T21:49:14.939Z · LW(p) · GW(p)
Agreed.
"Rationality" can also have some bad or incorrect-from-our-perspective connotations — Spock, Ayn Rand, sociopathic CDT-like agents in economic/game-theoretic models, etc. — but at least without the "ism" it doesn't sound like we're pledging our allegiance to an ideology, and we can explain what we mean by it without people saying "But philosophers X, Y, and Z were 'rationalists' and they didn't believe that!"
Replies from: DSimon↑ comment by DSimon · 2011-01-19T20:11:33.757Z · LW(p) · GW(p)
Another first response I tend to get:
"So, if it were rational to kill yourself, would you do it?"
Replies from: TheOtherDave↑ comment by TheOtherDave · 2011-01-19T21:14:45.775Z · LW(p) · GW(p)
What do you answer?
My honest answer would be "I don't know; I'd like to think so" but I suspect that would fail to communicate usefully.
Perhaps "Well, for example, if by killing myself I could save the lives of ten other people whose lives were roughly as worthwhile as mine, then I'd like to think I would... though of course you never know until the moment for decision arrives. What would you do in that situation?"
comment by Vladimir_M · 2011-01-14T22:18:39.814Z · LW(p) · GW(p)
Drexler's article is, in my opinon, simply inaccurate. The suffix -ism is used not only for political, ideological, and religious opinions, but also for philosophical opinions and theories in exact sciences and mathematics whose accuracy is a matter of dispute, or has historically been a matter of dispute. He's also displaying what appear to be strange gaps in his vocabulary -- Mendelism (alternatively Mendelianism) and Newtonianism have been widely used terms for centuries, as a simple search of Google Books will show.
In the concrete case of Darwinism, the term is very useful to differentiate Darwinian evolution from other theories of biological evolution that have been proposed historically, like Lamarckism or intelligent design.
Replies from: Vladimir_Nesov↑ comment by Vladimir_Nesov · 2011-01-14T22:27:20.613Z · LW(p) · GW(p)
Connotations don't study history or philosophy, what matters in the context of Drexler's article is prevalence in popular perception, and resulting normative recommendation.
Replies from: Vladimir_M↑ comment by Vladimir_M · 2011-01-14T23:04:59.536Z · LW(p) · GW(p)
I don't see any evidence that the specific practice of naming scientific theories by the names of their founders, with or without the -ism suffix, actually has the effect that Drexler complains about, let alone that this effect is of such magnitude and importance that it would justify describing it as "pernicious." Do you?
Replies from: Vladimir_Nesov, Kaj_Sotala↑ comment by Vladimir_Nesov · 2011-01-14T23:13:50.335Z · LW(p) · GW(p)
I don't have specific references sufficient to convince someone who doesn't agree ("evidence"), but I have my reasons. I believe this probable, via feeling the negative connotation with "-ism" myself, knowing of other people who feel similarly, and understanding arguments for why such feelings could appear by association, so I don't need such references to hold the level of believe I have. The opposite effect (of "-ism" being beneficial) looks less probable for similar reasons. The effect is important because if it's real, it affects many people (or simply, it affects a big expected number of people).
(I welcome any references that would convince me of this claim more strongly or disprove it.)
Replies from: Kaj_Sotala, Vladimir_M↑ comment by Kaj_Sotala · 2011-01-15T10:57:57.660Z · LW(p) · GW(p)
I don't have specific references sufficient to convince someone who doesn't agree ("evidence"), but I have my reasons. I believe this probable, via feeling the negative connotation with "-ism" myself, knowing of other people who feel similarly, and understanding arguments for why such feelings could appear by association, so I don't need such references to hold the level of believe I have. The opposite effect (of "-ism" being beneficial) looks less probable for similar reasons.
My experiences agree. I have a memory of having talked to several people who've expressed notions of the type "anything ending in -ism is something to avoid", though these have been off-line conversations so I can't give direct references to them.
↑ comment by Vladimir_M · 2011-01-14T23:40:47.227Z · LW(p) · GW(p)
To me, this sounds like a typical example of a theory that sounds neat and plausible when stated vaguely, but which turns out to be unsubstantiated on a closer examination. You can certainly tell a neat story to make it sound convincing, but when I consider in more detail the concrete mechanisms by which opinions of different sorts of people are formed in the modern society, I don't see any grounds to conclude that this concrete issue has much relevance.
(On a related note, the actual dynamics of the contemporary public opinion with regards to evolution and Darwinism are, in my opinion, very different from the standard story within which, among others, Drexler also frames his argument.)
Replies from: Vladimir_Nesov↑ comment by Vladimir_Nesov · 2011-01-14T23:48:40.638Z · LW(p) · GW(p)
To me, this sounds like a typical example of a theory that sounds neat and plausible when stated vaguely, but which turns out to be unsubstantiated on a closer examination.
Could be. But most beliefs (even most correct beliefs) are like that, you can't set some magical default level of certainty for everything that was not formally studied. The power of scientific evidence is in ability to conclusively destroy incorrect beliefs upon closer examination, but it doesn't hold monopoly on construction of correct beliefs.
Replies from: Vladimir_M↑ comment by Vladimir_M · 2011-01-15T06:57:10.628Z · LW(p) · GW(p)
There are many different grades of understanding and evidence between the extremes of absolute ignorance/confusion and a true no-nonsense scientific approach. Based on what Drexler writes in this article, I do think that my understanding of the issues at hand (i.e. the state of the current public opinion about Darwinism, the factors influencing it, and various linguistic claims he's made) is significantly more accurate than his, and my understanding suggests that his hypothesis is false. Of course, to substantiate this claim I would have to explain my position at length, for which I could hardly find the necessary time and space in the context of this discussion.
Nevertheless, if some evidence could be found for his position, it would provide an interesting challenge to my own ideas about these issues, and it could lead to an interesting discussion which wouldn't require me to first write a lengthy explanation for why I believe his ideas are mistaken. That's why I asked.
↑ comment by Kaj_Sotala · 2011-01-15T10:55:29.738Z · LW(p) · GW(p)
Of course, even if we presume that naming ideologies as "isms" causes a bad reaction but naming scientific theories doesn't, then I suspect that a randomly chosen outsider is at least as likely to put the LW usage into the "ideology" camp as he is to put it in the "scientific theory" camp.
comment by steven0461 · 2011-01-14T22:14:54.541Z · LW(p) · GW(p)
"Rationalism" has a meaning that "rationality" does not: namely, the attitude that rationality is a good thing. You can't just throw out a word without offering a working replacement.
Replies from: Vladimir_Nesov↑ comment by Vladimir_Nesov · 2011-01-14T22:20:00.006Z · LW(p) · GW(p)
You can, actually, it's just going to take longer to communicate (and you get a less-ambiguity bonus).
Replies from: DSimoncomment by jsalvatier · 2011-01-14T21:37:33.346Z · LW(p) · GW(p)
I agree that 'rationalism' has undesirable connotations.
'rationalist' also has these connotations, but I am not sure what other name I should call someone who devotes themselves to 'rationality'. Ideas?
Replies from: Zack_M_Davis, TheOtherDave, Vladimir_Nesov↑ comment by Zack_M_Davis · 2011-01-15T00:17:58.760Z · LW(p) · GW(p)
I use aspiring rationalist a lot.
↑ comment by TheOtherDave · 2011-01-14T21:56:42.829Z · LW(p) · GW(p)
As below, I think it's the correct word. That said, if we're looking for a better word for signaling purposes, I propose "clear-thinker" (modeled on "free-thinker").
Replies from: Vladimir_Nesov, endoself↑ comment by Vladimir_Nesov · 2011-01-14T22:07:21.008Z · LW(p) · GW(p)
"clear-thinker"
A rationalist works on improving their thinking, but of course, being merely human, never actually thinks clearly.
↑ comment by endoself · 2011-01-16T02:27:31.971Z · LW(p) · GW(p)
What is wrong with free-thinker? Does it connote more emphasis on atheism as opposed to rationality in general?
Replies from: TheOtherDave↑ comment by TheOtherDave · 2011-01-16T07:06:13.176Z · LW(p) · GW(p)
Huh. Now that you mention it, I don't know... it mostly just feels outdated, I guess, which is not in and of itself a reason to discard it.
↑ comment by Vladimir_Nesov · 2011-01-14T21:55:09.460Z · LW(p) · GW(p)
"Rationalist" looks OK, with less undesirable connotational impact than "rationalism". (Compare: scientist, physicist.)
(Added to the post.)
comment by shokwave · 2011-01-15T02:04:29.157Z · LW(p) · GW(p)
Agreed. Related: -isms give me a prickly feeling of "self-trusting system".
Replies from: Eugine_Nier↑ comment by Eugine_Nier · 2011-01-15T05:21:48.963Z · LW(p) · GW(p)
So? Changing the name doesn't change whether "rationalism" is a "self-trusting system", it only changes your impression of it.
Replies from: shokwave, Kaj_Sotala↑ comment by shokwave · 2011-01-15T16:32:29.783Z · LW(p) · GW(p)
Of course not, but 'rationality' feels like a toolbox to me, and 'rationalism' feels like a collection of beliefs. I don't want to give people the impression that I believe something because it's part of rationalism, and I think it's right because rationalism said so.
It was mostly just a feeling, though, that strongly agreed with the concerns Vladimir_Nesov expressed.
↑ comment by Kaj_Sotala · 2011-01-15T11:11:06.414Z · LW(p) · GW(p)
If you have the impression that your selected identity implies some way of thinking, then you will tend to adopt that kind of thinking. Identities are dangerous things.
comment by kilobug · 2012-07-25T09:21:55.472Z · LW(p) · GW(p)
Well, my own understanding is that "rationality" is the art of being rational (of using reason and its related tools to improve your map of reality, and therefore be able to steer the future in a direction you desire) while "rationalism" is the philosophy that says using "rationality" is a good thing.
For example, a post that explains what Bayes' theorem is and how to use it, what is the "fundamental attribution error" and how to avoid it, or explaining the basis of game theory are "rationality". A post that explains the importance of "wanting to be stronger", how you shouldn't accept blissful ignorance (aka "doublethink") or why we need to "raise the sanity waterline" are "rationalism".
That's the difference between saying "how you do it" and "why you should do it".
Of course in many cases, the limit isn't clear. Words are fuzzy boundaries, after all.
comment by JamesCole · 2011-01-19T13:21:29.878Z · LW(p) · GW(p)
So, my suggestion is to use "rationality" consistently and to avoid using "rationalism". Via similarity to "scientist" and "physicist", "rationalist" doesn't seem to have the same problem. Discus
A while back I argued against using the term "rationalist".
comment by EchoingHorror · 2011-01-15T13:05:54.361Z · LW(p) · GW(p)
Ernst Mayr's What Evolution Is uses "evolutionism" for "belief in evolution" and isn't even slightly worse because of it. I'm not sure we should let people who are wrong whenever they use a certain word prevent us from using that word. I feel the connotation, but I know where it's coming from.
As for "rationalism", I don't see a good reason someone would prefer it over "rationality" and it isn't understood with the same separate meaning for everyone. So until a need to universalize it comes along, I'll stick with "rationality", and would recommend the same, for those less-ambiguity points. Clear communication matters, and such.
comment by David_Gerard · 2011-01-15T12:13:45.983Z · LW(p) · GW(p)
Yes. The implied claim ownership of the word "rationalism" is more than a little annoying.