The Relevance of Advanced Vocabulary to Rationality

by aletheianink1 min read28th Nov 201341 comments


Personal Blog

Edit: I realise that I foolishly over-complicated and worded my question in a way that obscured what I actually meant. In essence, my question was: if we didn't have specialised vocabulary for things - say, in the area of rationality - would our rationality be hampered by our inability to be specific without long-windedness? Often words are created to bridge this gap when new concepts are created, so if we didn't have those words, would it take longer for us to understand or communicate and idea (to others or ourselves) and make it more difficult to be rational?

From the direction of the comments the general answer to my initial question is coming across as: "words are useful for communicating explicitly, and so an extensive or highly specialised vocabulary can be useful, if and only if the person/people with whom you are communicating understands those words". The internal understanding of concepts does not need words and thus a vocabulary.

I am curious about the relevance of vocabulary to rationality. I'm not talking about a basic vocabulary, but a vocabulary beyond that of the average, English-as-a-first-language adult. I believe there are a few correlations between intelligence as measured by IQ and vocabulary, as well as vocabulary and income(via IQ), but anecdotally I think it's fair to say that there are certainly people who are highly intelligent, but often irrational.

In reading through LW, I've come across a lot of new terms specific to certain areas of study, and I've had to look them up to fully understand that discussion of rationality - I assume this is probably true of most people new to the field, and applies to most specialised fields. Jargon is obviously useful within given fields where there is a need for detailed discussion of highly specialised topics, and helps one to discuss that area, but is it necessary to understand that jargon in order to practice in the field?

For example, I would think that a general practitioner would have trouble within his field if he did not hold the language to be able to specify what, in particular, was wrong with a patient, even if he knew what it was. Or could he not even be able to understand, say, that a patient was having a heart attack if he did not have the words for it? I suppose history might be a good indicator of this, or new scientific phenomena.

The field of rationality is one of both practice and theory - but if we didn't have an advanced vocabulary, could we still be highly rational? For example, my stepfather didn't finish high school, and makes up words like "obstropolous" (which I think kind of means stubborn and difficult to deal with on purpose) to say what he means, but he's also the type of person who, in a emergency, takes the most logical, rational course of action without panicking or doing something silly. On the flip side of this, he makes grand generalisations about races, religions and people while refusing to discuss the possibilities of individuality, or conceding any part of his argument to, well, evidence.

So do you have an argument for or against the need for an advanced or specialised vocabulary to be rational? Is it a question that's too vague, or with too variable an answer? I couldn't find any scientific papers on rationality and vocabulary, so I don't know if there's any data for or against, but I think it's an interesting question.

(This is my first LW article, so please be gentle but thorough with any criticisms you may have - I'm happy to improve or clarify!)


Personal Blog


41 comments, sorted by Highlighting new comments since Today at 5:39 PM
New Comment


So, obstreperous is an actual word, that means something very similar.

So do you have an argument for or against the need for an advanced or specialised vocabulary to be rational?

Have you read the A Human's Guide to Words sequence? A lot of irrationality in arguments centers around misunderstanding what words are and how they work, but I don't think size of vocabulary has a huge impact on that understanding (once you've corrected for IQ).

He does actually say obstropolous, but he must have read obsteperous somewhere and mispronounced it. Thank you!

I have bookmarked it because I want to read pretty much every link but don't currently have the time to do so.

Are you saying then, that if we fully understood what other people were saying, there would be less irrationality?

Are you saying then, that if we fully understood what other people were saying, there would be less irrationality?

I think I'm pointing the arrow in the other direction. A common mistake people make is to misunderstand each other and themselves; if they did not make that mistake, they wouldn't necessarily be better at understanding each other or themselves, but they would at least be clearer about the boundaries of their knowledge. (This could lead them to behave in ways that do actually make them understand themselves and others better.)

Consider the double illusion of transparency.

I see what you mean; that makes sense. I think that's something LW has certainly pointed out for me - by knowing one's own boundaries of understanding, one can try to further one's knowledge of the unknown.

I'm about to put child to bed so I haven't time to read the link right now, but I'll certainly be on it first thing in the morning!


So, obstreperous is an actual word, that means something very similar.

The OED has an entry on "obstropolous" adj.:

Etymology: Humorous alteration of "obstreperous" adj.

Chiefly humorous and regional. Now rare.

There's also an entry on "obstropolos" n., which was a nonce word coined by James Joyce for Ulysses:

Etymology: Alteration of "obstropolous" adj., perhaps after ancient Greek -ος, ending of masculine singular nouns, probably also with punning reference to classical Latin ōs mouth (see "oral" adj.).

An obstreperous mouth.

I think that having words for things is very useful - both for communicating about things and thinking about them. I do think that having a larger and more specialized vocabulary about a topic actually makes you better at thinking about a topic.

For an example of the benefits of verbal labels spilling over into non-verbal domains, check out this experiment, which showed that giving verbal labels to things facilitates learning and raises peak performance on a visual discrimination task. Original paper.

Even taking into account purely personal experience, I feel extremely confident of this opinion. I am actually rather surprised at the number of users who believe that possessing a large vocabulary isn't extremely important for learning and thinking and general purpose / domain specific intelligence...for my part, I think it's actually one of the most important things, even more important than quantitative skills.

Could you list something like the ten most important words for rationality that the average person doesn't know and where they would benefit from learning them?

It's hard, because "rationality" is really general. This would be a lot easier with something more domain specific (epistemology? emotional regulation? efficiency?). Here's an attempt to start a list though:

Necessary and Sufficient - helps verbalize many real world logic problems

Opportunity Cost -" Your current action might be beneficial, but what could you be doing instead?"

Diminishing Returns-"Doing what worked before isn't always the best strategy

Rate-Limiting-Step- To speed up any process, Identify it

Pleiotropic constraint - (try applying it to non-genetic things)

implicit / explicit

I missed one of those words "pleiotropic", I add it to Anki. How wwould the 10 or 20 word list look like for epistemology/emotional regulation/efficiency?

Thank you for the link - this was essentially what I was looking for! I have yet to read the article, but it's an interesting conclusion - perhaps other commenters were simply going by their intuition or what they felt, instead of looking for evidence?

I think that in general people have a tendency to create too much notation rather than too little ("advanced vocabulary" is really just a type of notation). The LessWrong community seems particularly susceptible to this: we often create new words when existing concepts already exist, or when it wouldn't be that hard to explain what we mean in the first place. This is problematic because it makes idea transfer with outsiders far more difficult: not just because they are less willing to engage, but because one now needs to translate between notations, which is often much harder than it would naievely seem. So I would actually be in favor of less vocabulary, more effort to tie existing concepts to concepts that exist in other communities, and more effort on simple explanations of ideas that don't require someone to have read the Sequences.

we often create new words when existing concepts already exist

Every time I read this, it comes without specific examples. And it keeps returning for years.

If some people think this is a real problem, the best way to solve it would be to publish a dictionary of LW-specific words and their equivalents used by the outside world. I believe that such article would be upvoted, and if there is an agreement that those words really are equivalent, many of them would be replaced. At least you could give a hyperlink to that article every time this topic appears in the debate.

If writing an article is too much work, writing a short list or just a single example in the Open Thread would be a good start.

we often create new words when existing concepts already exist

Every time I read this, it comes without specific examples.

It doesn't come up much (Google only gave me two examples on LW), but "funge against" winds me up. I don't see what the phrase does that "displace", "crowd out", and "substitute for" don't do already. It's a needless, opaque neologism.

Edit, 5 days later: oh no!

Here are some. One reason that translation between jargons is hard is because there are often not words in one jargon that can be used exactly in place of words in the other jargon. I've therefore included substitute words as long as they point to the same concepts. The words below were taken from the LW wiki.

Belief as attire: jumping on the bandwagon, groupthink Dark arts: hard sell, manipulation Fully general counterargument: reductio ad absurdum Fuzzy: warm fuzzy feeling, self gratification Luminosity: self awareness, emotional awareness Ugh field: aversion

Thank you for the specific examples!

I like: jumping on the bandwagon, manipulation, self awareness. On the other hand, I disagree with the reductio ad absurdum. Reduction as absurdum is like this: "if X, then Y, but Y is obviously silly, therefore not X." The Y is somehow derived from X. Fully general counterargument is something that actually does not depend on X (this is what makes it fully general); it is a chain of words where you can substitute any value X and get the result "therefore not X". Being a fully general counterargument is a semantic property of some arguments; and probably most of them are syntactically reduction ad absurdum.

More meta: that's the point. If we have specific examples, we can discuss them specifically, and perhaps accept some and refuse others. (And even the act of refusing is helpful for communication, because it makes more clear what exactly we mean by saying something.)

Thanks for the feedback. I agree that reductio ad absurdum is the weakest of the examples I gave, but let me try to justify it anyways: if X is a fully general counterargument, then we can use it to argue against true statements as well as false ones. So applying X without any additional justification would lead to patently false conclusions, and therefore (by reductio ad absurdum) X is not a valid form of reasoning. Perhaps this is not the best word for it, but it is similar to a very pervasive idea in mathematics, where when formulating possible approaches to prove a theorem, a key criterion is whether those approaches can distinguish between the theorem and similar statements that are known or suspected to be false.

ETA: And yes, I agree that specific examples are good!

Yes, that's the usual application, but it's the wrong level of generality to make them synonyms. "Fully general counterargument" is one particular absurdity that you can reduce things to. Even after you've specified that you're performing a reductio ad absurdum against the proposition "argument X is sound", you still need to say what the absurd conclusion is, so you still need a term for "fully general counterargument".

It often happens that you invent new words precisely because you aren't familiar with existing words for a given task.

For example, I would think that a general practitioner would have trouble within his field if he did not hold the language to be able to specify what, in particular, was wrong with a patient, even if he knew what it was. Or could he not even be able to understand, say, that a patient was having a heart attack if he did not have the words for it?

The second idea is called strong Sapir Worpf hypothesis. It doesn't seem to be true.

Take a word like aunt. It includes the sister of your mother but it also includes the wife of her brother. Obviously you can separate the two mentally even if you don't have separate words. is a nice podcast that describes it's state.

When you talk about rationality it's still highly useful to distinguish between aliefs and beliefs. There are probably plenty of people out there who don't make that mental distinction. Having different words makes it easier to learn to make the mental distinction.

It also makes it easier to remember thoughts about the mental distinction. Both natural remembering and constructed remembering via Anki.

Also, I strongly suspect there are typical mind fallacy effects at work here.

Some people can think clearly without having words in their mind, and tend to assume that of course thought is possible without language. Other people can't think at all without words, and tend to assume that of course language is required for thought.

There's also a philosophical literature on 'thought without language' that I've never got to grips with, and the associated pop-philosophy stuff that's even harder to make sense of.

Also, I strongly suspect there are typical mind fallacy effects at work here.

Whether or not you need words is an empirical question. That question was investigated pretty thoroughly in linguistics.

I don't have to extrapolate from what I think about my own mind but can extrapolate from empirical research.

Great post. Right now, I think that you do need an advanced vocabulary to understand what's on LessWrong. In general, most people that have been involved with LW-style explicit rationality as it exists today are unusually intelligent and this are likely to use such terms.

However, I think that that is a flaw with our own methods rather than a flaw with rationality-in-general. A more accessible form of rationality would be very useful.

One post that you might want to read is lukeprog's "Explicit and tacit rationality."

Thank you - I think the article was actually rather weak, on review, but thank you!

I'm going to read the article now.

Your initial point was what prompted my thoughts on the issue - essentially, as I read through LW, learn new words, new ways of thinking, new approaches, will I become more rational? I suppose that's not solely vocabulary - it includes the ideas that spawned that vocabulary - but looking up definitions has something I've definitely been spending a lot of time doing!


I'm not talking about a basic vocabulary, but a vocabulary beyond that of the average, white, English-as-a-first-language adult.

Why white?

I have no idea why I put that. I was trying to just be very specific, so people wouldn't ask "well, what if they hadn't heard of x" or whatever ... it may be because I'm used to reading about the entitlement of average, white, English-speaking people (specifically men), and just linked that in without thinking. It's irrelevant, so I'll go fix it - thanks.

Would Magnus Carlsen be unable to play chess if he had no words for it? This old post suggests some people don't have mental imagery at all. If that's true, these people would probably make some interesting claims about thinking.

Or could he not even be able to understand, say, that a patient was having a heart attack if he did not have the words for it?

I can visualize a "heart attack" in my mind with all the relevant steps involved, no words needed. People (and animals) understand other kinds of phenomenoms they don't have words for, so I would deduce that it applies to this situation also. Symbols are needed for understanding phenomenoms you're not directly experiencing, but they don't have to be words.

So do you have an argument for or against the need for an advanced or specialised vocabulary to be rational? Is it a question that's too vague, or with too variable an answer?

Specialized vocabulary usually makes thinking and communication more effective by allowing shorthand for complex concepts. This applies to many kinds of thinking, most of it not necessarily rational. You'd probably find correlation, but good luck with the causation part.

I had a complicated point to make about the interplay of vocabulary and simplifying ideas in order to make thinking more clear (and thus perhaps rationality?) but I think I kind of lost that in the post and have made it sound more like "can people think if they don't have words?".

I agree with what you've written, and I'd say that your last sentence rather answers my (intended) question: correlation may exist, but causation is a lot trickier to pin.

I think a doctor would do just fine inventing her own word for heart attack, if she didn't have any colleagues to communicate with. A rationalist would only get extra value out of using "real" words for concepts rather than ones they made up if they could speak with other people who knew of the same concepts.

Now that I think about it, some of my friends have done something sort of like inventing words when we needed to refer to concepts that didn't have an existing label. One of my programmer friends tends to take a common word that's sorta close and write it like a PHP variable ($protect, etc.). I use "ouroboros" to mean a being that can't exist without hurting itself, and I'll tell people that I use the word that way if the concept comes up in discussion so that I can refer to it later without having to re-explain.

What sorts of things can't exist without hurting themselves?

Or is it pointing out the effect of entropy on all living things?

The concept came up in the first place when a friend and I were arguing over whether it would be ethically good to instantiate all possible minds, if you had a large enough simulation for them to live in that would allow them to self-modify and modify their circumstances however they wanted. (There are technical difficulties with that, of course - like how to keep all these minds from hurting each other without the ones that want to hurt people being unsatisfied - but this is assuming there's some way to resolve that.) My objection to doing this was that some parts of mindspace will have extremely unhappy lives, but will not change or kill themselves if given the opportunity. For example, they might believe it is immoral to self-modify.

In the real world, mental illnesses like severe depression seem similar, in that it's painful because of the way one's mind is. Depressed people do want to stop being depressed, but it's very difficult to do that.

As I said above, I think I kind of misphrased my question while trying to make it clear in my head - almost ironically, my inability to find the right words hampered my ability to communicate what I meant.

I agree completely about people making up words for new ideas - I suppose that's what I meant to bypass: we make up words as shorthand for longer concepts, because if we didn't, it would take a lot longer to say or explain what we meant. My question was meant to be along the lines of, if we didn't have those new words, would our rationality be hampered by the lack of specific words (even if we knew what we meant in our minds)? (You don't have to answer that, I was just trying to clarify!)

You seem to differentiate between words and concepts. And with words you mean words with a specific interpretation in a specific context. Not words everyone 'knows'.

Take e.g. 'correlation' (the first candidate from your post). The laymans reading is roughly 'some kind of relationship'. Being used in a LW context you may assume that the reader knows about the mathematical definition and its application to everyday use. Now as long as you use 'correlation' as in your post and do not attach weight on the consequences of that use there is no risk. But once you want to use consequences of that definition to support or weaken a point then you must consider what concepts your reader really knows.

As the other commenters noted the Sapir Whorf Hypothesis is mostly wrong. You can have concepts without words. But words clearly help.

The question also is: When do you really have a 'word' for a concept. 'correlation' alone is no concept. There are stochasitic correlation between stochastic variables or correlations between signals. Those that are discrete or continuous or multi dimensional. Pick your meaning. And arguing that the 'concept' is pure; the abstraction or generalization of these specific instances doesn't really work because nobody really knows what the pure concept is. Thus you cannot communicate it but only appeal to it by reference to particular cases.

Stripping a word of its context (who and when and where it is uttered, onto which other words it is applied) leaves not a pure concept but only a sequence of letters.

Words give the hearer an indication of your indicated meaning but they are not the meaning. That is what EY means with

How words [...] are secretly a disguised form of Bayesian inference.

Words make a certain reading more plausible and (hopefully) let your mental representaticonverge locally toward that of the speaker (your brain will update its structures mostly bayesian, but it may fail given noise).

A large vocabulary then is a large set of 'signals' for which both the speaker and the heaer know the concepts signalled.

Because words are also used internally by conscious thought they can also be used to signal to yourself and thus a large vocabulary can be conducive to efficient reasoning.

Because private and public use of the vocabulary are often mixed one should reflect on what the other really knows.

For more on this see and .

It's not particularly useful to have a big vocabulary - it's actually often counterproductive for getting your point understood. That said, it's sometimes useful to substitute words to prevent using a single word for multiple distinct meanings.

It's counterproductive if you have a big vocabulary but use it to show off (which is unfortunately how most people use them). A big vocabulary used correctly can be invaluable.

There are two ways to develop your vocabulary. You can make it more florid, by adding lots of long obscure synonyms. This is the image most people have of a large vocabulary. Or you can make it more precise, by taking words you already know, that are more specific or more vivid than their everyday equivalents, and focus on being able to actually use them in conversation.

For instance, consider the word "burp". If I were being more florid, I might use "eructation" instead. You're right that this would be counterproductive. However, if I were focusing on precision, I might use "belch". Most poeple know what a belch is, and it sounds more vivid and exciting than "burp", but few people use it in conversation.

For an example of a large vocabulary that is precise rather than florid, Walt Whitman's Leaves of Grass has one of the densest vocabularies of anything on Project Gutenberg, yet it's still quite readable.

It's counterproductive if you have a big vocabulary but use it to show off (which is unfortunately how most people use them). A big vocabulary used correctly can be invaluable.

It's not clear to me what "correctly" means here. An illustrative example: I recently wanted to use "exacting" in a paper because, well, it was exactly the word to use in that sentence. I know about twelve similar words that each didn't quite fit for one reason or another, and I am not at all trying to show off my vocabulary by using it.

My advisor looks at the draft and says "look at this typo, did you mean to type exciting?" (He is not a native English speaker.) "No, I meant exacting; it means that it has precise requirements." "Oh. Well, that's confusing, you should change it."

Most of the people in my field are not native English speakers; if he has trouble understanding it, likely they will as well. Is it correct to use it?

Most of the people in my field are not native English speakers; if he has trouble understanding it, likely they will as well. Is it correct to use it?

No. The purpose of writing is to be understood, if your writing will not be understood then it should be changed. Write to the audience you have, not the audience you want to have.

Yes. That reaction of your advisor is so annoying! If he doesn't understand a word, he should just look it up, instead of assuming you made a typo. It makes me think of that translation I had to revise once, where the translator had translated 'quartile' as if it said 'quarter'. I didn't know that word either, so I looked it up, and there was a perfectly good Dutch translation in the dictionary.

Are the people in your field all/mostly the same nationality, BTW? If not, it's impossible to take everyone's deficiencies into account anyway. A Frenchman, a Dutchman and a Spaniard would each misunderstand different words, and they're all from the same language family.

Another thing is that words that native speakers think of as 'difficult' may not be so difficult for non-native speakers. The 'difficult' words are often imported from Latin or Greek, and have been imported in many other languages as well. Whereas words that are perfectly ordinary for you may be very difficult for us, like the names for kitchen implements, foodstuffs, articles of clothing, and the like (what do you mean a pocketbook is a woman's handbag? that one confused me so much, the first time I encountered it).

It's not useful if other people don't understand it. If you're communicating with other experts, it's very useful.

Med students for example have to learn thousands of new words to practice as doctors and without this communication would get ridiculously long-winded.

I agree that sometimes using unusual, uncommon or long words when a shorter one will do can be counterproductive, but what about topic-specific vocabulary - words which are common in given circles (for example LW) but have complex ideas or meanings behind them? Or would you consider that to fall under your latter sentence?

Topic-specific vocabulary and technical jargon has to get finessed to be used properly. In a perfect world, you can use such words in a way that the context makes their meaning clear. Or, at the very least, gives an obvious way to research what the word means. Acronyms and abbreviations are an excellent anti-example for this - I'd never use an abbreviation unless it's absolutely necessary, and use the full phrase nearby before it. Garbage-in garbage-out is a useful way to talk about algorithms failing by seeding them with bad input, but randomly dropping GIGO in a conversation is a quick way to lose someone.

Jargon and inferential distance are closely related, and you need to keep track of both together. People who are already thinking much like you are likely using the same words for the same concepts, and vice versa.

Also, even if you can talk specifically in technical contexts, learning to ease people into jargon and introduce lay members to engineering disciplines is a tremendously valuable skill to learn. As in a 4-year engineering degree and this skill is a quick way to get promoted to whatever position is in between management-types and engineering-types. It's really straightforward - just have enough technical chops to not completely embarrass yourself in a technical context, and learn how to explain engineering like your audience has an MBA, and management is going to want to talk to you about engineering projects rather than other people.

tl;dr - go ahead and use technical words in technical contexts, but try to do so in a way that those unfamiliar with the context can at least figure out what questions they need to ask.

[-][anonymous]6y -2

Your vocabulary literally changes the universe.

''Perhaps the most different counting system from that of modern Western civilisation is the “one-two-many” system used by the Pirahã people. In this system, quantities larger than two are referred to simply as “many”. In larger quantities, “one” can also mean a small amount and “many” a larger amount. Research was conducted in the Pirahã culture using various matching tasks. These are non-linguistic tasks that were analyzed to see if their counting system or more importantly their language affected their cognitive abilities. The results showed that they perform quite differently from, for example, an English speaking person who has a language with words for numbers more than two. For example, they were able to represent numbers 1 and 2 accurately using their fingers but as the quantities grew larger (up to 10), their accuracy diminished. This phenomenon is also called the “analog estimation”, as numbers get bigger the estimation grows [1] Their declined performance is an example of how a language can affect thought and great evidence to support the Sapir-Whorf Hypothesis.


Language also seems to shape how people from different cultures orient themselves in space. For instance, people from an Aboriginal community called Pormpuraaw define space relative to the observer. Instead of referring to location in terms like “left”, “right”, “back” and “forward”, most Aboriginal groups, such as the Kuuk Thaayorre, use cardinal-direction terms – north, south, east and west. For example, speakers from such cultures would say “There is a spider on your northeast leg” or “Pass the ball to the south southwest”. In fact, instead of “hello”, the greeting in such cultures is “Where are you going?” and sometimes even “Where are you coming from?” Such greeting would be followed by a directional answer “To the northeast in the middle distance”. The consequence of using such language is that the speakers need to be constantly oriented in space, or they would not be able to express themselves properly, or even get past a greeting. Speakers of such languages that rely on absolute reference frames have a much greater navigational ability and spatial knowledge compared to speakers of languages that use relative reference frames (such as English). In comparison with English users, speakers of languages such as Kuuk Thaayorre are also much better at staying oriented even in unfamiliar spaces – and it is in fact their language that enables them to do this.[2]''