Comment by simias | Hacker News Reader

Comment by simias | Hacker News Reader

Parent Full thread

I'm not a linguist but I do enjoy linguistics and language learning as a hobby so I suppose I've gained a certain perspective that I feel most monolinguals lack.

Basically here's my rule of thumb: any linguistic claim coming from a mainstream source which seems to imply the Sapir–Whorf hypothesis[1] or other kind of objective criteria to compare languages is a huge red flag and extremely likely to be entirely bullshit. Out of those claims the "language X is more precise/able to express more nuance than Y" is probably the reddest flag of them all.

I think generalities about languages ("French like to say no") fall in the same category as horoscopes: if they're vague and general enough you'll probably recognize yourself in them to some extent. People like to feel special, they like to feel special for speaking a given language so they'll gladly accept any claim that sets them apart.

>If true (I suspect a methodology shenanigan there) I really doubt this is a factor. Very few French or English speaker will have a mastery of 70,000 words, to say nothing of 500,000.

Yeah, this claim alone shows that the author really stayed at surface level. I suspect the main reason for these numbers is that English, traditionally less prescriptive than French, considers basically any word once used in an English text to be an English word while French curates more heavily. The average English speaker definitely doesn't draw from a reservoir of 500k words, probably closer to a tenth of that. Maybe somebody could link this website to the author: http://testyourvocab.com/

[1] https://en.wikipedia.org/wiki/Linguistic_relativity

>Out of those claims the "language X is more precise/able to express more nuance than Y" is probably the reddest flag of them all.

This may be nonsense as an overall comparison between languages but this is definitely the case for specific concepts. Here is a paper where the reactions of 30 Dutch people and 30 Jahai tribesman to a set of odors were compared:

https://royalsocietypublishing.org/doi/full/10.1098/rstb.201...

First two paragraphs of the conclusion:

Olfactory abstraction varies across cultures: while Dutch participants confirmed the often-touted claim that ‘olfactory abstraction is impossible’ [19] by providing mostly concrete language in response to odours, Jahai speakers overwhelmingly described odours with dedicated, abstract language. In addition, their responses were faster and shorter, providing converging evidence that the Jahai are communicatively adept in talking about odours.

Even for the monomolecular odours used in this study, which do not have a single object entity associated with them, Dutch participants predominantly tried to identify a source (e.g. flowers), or situation (e.g. house that isn't aired), corresponding to that aroma. Their grappling to identify concrete sources was in sharp contrast to the fluent abstract Jahai responses. Previous studies have shown that Standard Average Europeans struggle to identify odours [5,13–16], as also illustrated by the Dutch here. The greater ease of linguistic expression demonstrated by the Jahai is not unique, however. It appears that hunter-gatherer communities in particular find odours easier to talk about [23].

It's not really an objectionable concept. It's essentially the same as how any kind of discipline with a reasonable amount of information to learn comes up with their own terms. It makes it easier to navigate relevant information.

This paper shows that Jahai are capable of "olfactory abstraction" but I don't see how that says anything about the respective languages. I mean the last sentence:

>The greater ease of linguistic expression demonstrated by the Jahai is not unique, however. It appears that hunter-gatherer communities in particular find odours easier to talk about.

If that's true that probably means that they have a greater need to express these concepts and as such have become better at it and developed a specific jargon. Maybe if they tried with tulips or windmills they'd find that the Dutch are much better able to describe them succinctly.

Besides the paper says that the Dutch subjects are urban dwellers while the "All Jahai still pursue traditional foraging, although they reside in a resettlement village much of the time, and so are exposed to modernity". Attributing the results solely to linguistic differences would be rather bold (and the paper doesn't in fact do that as far as I can tell>

So if you're saying "culture which have a lot of exposition to X and often need to precisely talk about X develop linguistic tools to do it efficiently" then yeah, that seems fairly uncontroversial. Just look at any HN technical discussion about "bugs", "segfaults", "DDOS" or "transpilers". I still don't think it makes a good case for comparing entire languages as a whole.

>I still don't think it makes a good case for comparing entire languages as a whole.

I wasn't making such a case. I was pointing out that in limited contexts, different languages may have more or less effective means for accurately describing things within that context.

Attempting to place a valuation on languages and comparing them wholesale is stupid, but that doesn't imply that a language can't be lacking in vocabulary in some space compared to another. That also doesn't mean that there is something fundamentally keeping it from developing vocabulary for such spaces. Which I'm sure they do in niche spaces, such as the academy.

"Out of those claims the "language X is more precise/able to express more nuance than Y" is probably the reddest flag of them all."

It seems like there are plenty of cases where some language is able to express some nuance another is not. It's just that there will be some tradeoff in return because there isn't a language that has all the details. The real problem is the implication that because some language is nominally superior in some particular dimension ("more precise", "more nuanced", "more deeply integrated into the grammar") in the vast set of dimensions languages have then it must be somehow "better", or that we can assume this relationship will be shown in all possible dimensions of comparison.

But I agree it gets oversold. English can't properly express the concept of schadenfreude... except I just did, because we imported it because we liked it, and it's an English word now too.

Right, that's what I had in mind. Just because some language doesn't have a "built-in" word or grammatical feature doesn't mean that it's impossible to express the concept. Spanish has two words for "to be", one for temporary states (I'm tired) and one for more permanent attributes (I'm tall). Because English only has "to be" doesn't mean that it's incapable of communicating this nuance when necessary.

I can't disagree more with "language X is more precise/able to express more nuance than Y"

Japanese has 4 words for "If" each one is used in a different set of circumstances and has a slightly different meaning and flavor.

http://www.guidetojapanese.org/learn/grammar/conditionals#Ho...

While all of those circumstances are covered by the English "If", the flavor and connotation is lost.

Similarly with Japanese and personal pronouns.

There is one word for "I" in English, but many many words for "I" in Japanese, each with it's own flavor and implications.

http://nihonshock.com/2009/11/the-many-ways-to-say-i/

These other "I"s express additional information about the person using them. It's not that this information cannot be expressed alternatively, but it's very efficient to simply use the right "I" to convey the right meaning.

>It's not that this information cannot be expressed alternatively

So we agree then.

>but it's very efficient to simply use the right "I" to convey the right meaning.

How do you quantify efficiency here? Having more generic words can arguably be more efficient in situations where you don't actually care about which "I" you want to communicate: "I want an ice cream Sunday please". Besides some languages (and I believe Japanese to be one of them) are pro-drop and let you drop the pronouns in most cases, isn't that even more efficienter?

My native language is French, which unlike English kept the T/V (thou/you) nuance. You have familiar you and polite you. Do you consider that more efficient? Because as somebody who's usually pretty shy I've often found myself in awkward social situations where I simply want to say "how are you?" and "vous" feels too formal but at the same time I worry that "tu" might be a bit too brash and quite frankly I'd gladly trade them for a generic, neutral English-style "you" in situations like these.

Some other romance languages, like Brazilian Portuguese for instance, almost completely dropped the T/V distinction. Are they becoming less efficient or more?

What you describe comes across more like versatility.

It could be argued either way.

You could say its more efficient not to have the T/V distinction because then you don't have to think about how you want to express yourself.

Conversely, if you wanted to be very polite or very rude, people would have to rely more on context or vocal cues. Thus less efficient because now you have to express the politeness or rudeness in other ways.

But in Japanese, you are not only convening politeness information, which can be in the form of polite verbs (Keigo), polite verb endings, and polite pronouns, and honorifics, you are also partially conveying your own perception of your self importance and sense of masculinity or femininity.

Surface level. Sums up the book pretty well if you ask me. Started not too bad with some anecdotes, yet didn't build up on that on any depth. Maybe helpful, or with some bad luck quite the opposite, for people without any intercultural experience themselves. For people with some experience it is just common sense.

One other point the book misses is the complexity of all of that. Only hinted at are the effect personality has. Let alone the overlaying patterns of culture, personality and organizational culture.

And yes, the theoretical basis is almost totally absent.

Reply on news.ycombinator.com