Language, Thought and Reality
Selected Writings of Benjamin Lee Whorf
The pioneering linguist Benjamin Whorf (1897–1941) grasped the relationship between human language and human thinking: how language can shape our innermost thoughts. His basic thesis is that our perception of the world and our ways of thinking about it are deeply influenced by the structure of the languages we speak.
@appassionato @bookstodon @linguistics Well, that's a bit too much of hype, because there is very little evidence that Whorf's basic thesis is relevant in the real world. The only artefact I am aware of is a study on colour perception that is different for speakers of languages like Russian that have two words for blue (denoting azure and indigo as basic colours).
@jk @appassionato @bookstodon @linguistics What this evidence for linguistic relativity actually shows is how speakers name things, not how they percieve things: two related objects may have different names in one language, but only one in another; but if you present both things to the speakers at the same time, they both will agree that the things are different.
This wouldn't be the case if language (that is, names) were to influence your perception of the world.
@LupinoArts @jk @appassionato @bookstodon @linguistics there's research (ref Lea Boroditsky) that people speaking languages that have distinct words for shades of blue (Russian, Italian) are better at distinguishing the actual shades of blue
@LupinoArts @mapto @appassionato @linguistics It is not as easy as that. A proper experiment on this (like the one by Maier and Abdel Rahman I am alluding to) does not require the probands to name the colour. Instead, shapes in different shades of blue are presented and the task is to recognise the SHAPE of the objects. Due to the short reaction time some objects are missed by the test persons. 1/2
@LupinoArts @mapto @appassionato @linguistics The missing rate is lower for native speakers of Greek or Russian: They recognise some more of the shapes if the shades of blue cross the azure-indigo line in their native languages.
As a bonus: The difference vanishes when the probands have to memorise a sequence of number over the experiment, indicating that language processing is involved in the effect. 2/2
@mapto @jk @LupinoArts @linguistics
"We conclude that linguistic enhancement of color contrasts provides targets with a head start in accessing visual consciousness. Our native language is thus one of the forces that determine what we consciously perceive."
@appassionato @mapto @jk @linguistics It's the same as with the other paper: All they show is that a broader lexicon improves a subject's ability to put things into categories, but I don't quite get how that should be an argument for linguistic relativism (let alone determinism, as the wording in your quotation suggests): all those experiments say something about linguistic processing, but not (necessarily) about general world perception.
@LupinoArts @appassionato @mapto @linguistics First, it is not just a broader lexicon but a lexical difference in the native language (L1) that is significant here, and the putting things into categories is quite automated here. "World perception" sounds very big here (a bit like German *Weltanschauung*), but the authors of the study see a measurable effect on a little bit of perception (here: shades of "blue").
@LupinoArts @mapto @jk @appassionato @linguistics
But wouldn't the fact that language categorisations might tend to 'prime' perceptions and interactions be the point.?
I don't 'buy' strong linguistic determinism, but the priming effect of language would surely be exactly what you'd expect if a weaker linguistic influence was a reality. The question then becomes what kind of effects and influence ...
@pneumaculturist @LupinoArts @mapto @jk @appassionato @linguistics As far as I am aware, it is well-accepted that language can promote or produce mental proclivities, even new abilities, but through forcing the speaker community to perform certain tasks a lot, thereby promoting the skills associated with that task.
This is ultimately contrary to any sort of linguistic relativity: It's an alternative, competing model.
@androcat @pneumaculturist @mapto @jk @appassionato @linguistics "forcing the speaker community to perform certain tasks a lot": It's less about "being forced", like, from the outside; but by having a name for a category in your native language, you automatically engage with that name and the items that belong into that category and therefore "train" yourself and become quicker at perfoming naming tasks. That's basicly what i meant when i said that those results could be explained by priming.
@LupinoArts @androcat @mapto @jk @appassionato @linguistics
Thanks for those responses. I'd not really understood linguistic relativity in the way that you outlined and I'm still not sure if I 'get' the distinction you seem to be making. Are you able to point to something that explores it more?
@pneumaculturist @LupinoArts @mapto @jk @appassionato @linguistics There's probably some people who still hold to it. I mean some people probably still believe in Fodor or Chomsky as well.
You can start by reading Sapir's account of it, but with a grain of salt.
Overall, the "hard" version of it was the "speakers can't think things that their language doesn't provide categories for". That's been strongly rejected for 30 years at least.
The weaker form is just "language influences how we perceive the world", but there are models that better explain how, as outlined, and then it's more correct to apply those models than to hang on to the weakened form of relativism.
The wikipedia page probably has some useful literature, but as it is missing a criticism section, grain of salt : https://en.wikipedia.org/wiki/Linguistic_relativity
@LupinoArts @pneumaculturist @mapto @jk @appassionato @linguistics The extreme form of what I am talking about can be found in Space in Language and Cognition by Stephen Levinson.
Mandatory grammatical categories do "force" language users to perform the categorizations required for applying the linguistic properties correctly.
A much less extreme form is stuff like color terms, where there will be a communal expectation for certain distinctions to be made, and where users are "forced" to use that sort of distinctions if they want to be understood or just avoid causing needless discussion.
@androcat @LupinoArts @mapto @jk @appassionato @linguistics
Is there a good source to compare and contrast the two models -so I don't take up your time with too-basic enquiries? (I'm returning to this field of enquiry after many years without engaging)
@pneumaculturist @LupinoArts @mapto @jk @appassionato @linguistics RE: compare and contrast, I am suggesting that any model that takes as input a distribution pattern in the language domain and compares it to a distribution pattern in the cognitive domain will be competing against any other models doing the same thing.
In this, the relativity model is "language users can think (only or easily) in the ways that language categorizes reality", whereas a (to make up an umbrella term) "learning by doing" model suggests "users will become relatively adept at categorizations that are required by their language, even outside of linguistic domains".
Of these two, I would say the latter has a far better prior, since we already do accept "learning by doing" as a pattern in general cognition: People do tend to get better at tasks that they perform more - if the task is at all learnable, that is.
@pneumaculturist @androcat @mapto @jk @appassionato @linguistics not out of the top of my head, sorry. Linguistic relativism isn't really taken seriously in linguistics, maybe apart from it being a part of linguistic research history.
If it came up beside that, then mostly because some non-linguist put out a paper and we made it an exercise to debunk it...
@LupinoArts @androcat @mapto @jk @appassionato @linguistics
I'm a lingust by background (read Sapir) but been 'out of it' for while. I share from that background the distrust of determinism and spent some time a few years ago disabusing non-linguists of it myself. But the priming and activating frames seemed to be suggesting a legitimacy of a form of 'soft determinism' -maybe that should be further softened to 'influence'?
@pneumaculturist @androcat @mapto @jk @appassionato @linguistics "influence" is what "linguistic relativism" is about; "linguistic determinism" is the "hard" version of that: that our language *determines* the way we perceive the world and you cannot see the world outside of the boundaries of your language.
@pneumaculturist @LupinoArts @mapto @jk @appassionato @linguistics I think it makes sense to look at what is being said in a more detailed way. "influence" is too nebulous - one might be tempted to say "not even false".
If Relativism is a corral, limiting ability, then priming and other such models are more about boosting than limitation.
It's the difference between saying "general cognition is X, and after linguistic influence it is X-n" and "general cognition is X, and after linguistic influence it is X+n"
@androcat @pneumaculturist @mapto @jk @appassionato @linguistics I think the crucial point is whether or not there is a "X" vs. "X±n". And also, i'm sorry, but I blurred the line a bit between Language processing as a cognitive function, and the interaction between language as a system and general cognitive abilities. Linguistic determinism/realtivism (in my understanding) claim a more or less strong connection between the latter, while the aforementioned experiments can only test the former.
@pneumaculturist @mapto @jk @appassionato @linguistics well, priming influences how quickly information is accessed during processing. It has not per se something to do with *how* we percieve the world around us as linguistic relativism would claim.
@LupinoArts @pneumaculturist @mapto @appassionato @linguistics But "priming" is, as I understand that term from psycholinguistics, something different, it is the effect of a stimulus that happens relatively shortly before the experiment. Like evoking a semantic frame in linguistics.
@jk @pneumaculturist @mapto @appassionato @linguistics It's been a while and i might be wrong. My line of thinking was that if you see a color patch, the name for that category activates and naming the category goes faster. If the colour borders two different categories for one speaker, activation for both is weaker, while for a speaker that has another category in that area of the spectrum, the activation for that category is stronger and that leads to quicker reaction times.
@jk @pneumaculturist @mapto @appassionato @linguistics But yeah, mid way through the debate i wondered if it isn't actually "frequency effects" or some other term i forgotten since i left university ;)
@mapto @jk @appassionato @linguistics Just read this paper: https://www.researchgate.net/publication/245631970_Effects_of_language_on_color_discriminability Interesting findings, but i assume the effects could also be explained by priming: If you have a more differentiated lexicon for categories, you are faster at sorting elements into those categories, but this does not necessarily mean that you percieve those elements differently. I wonder, how the results would look if they compared, say, artists against non-artists.