I've been interested in the general idea since reading 1984... where language shaping and limiting thought is a central idea. But I'd be curious to know how much scientific support this idea has.
The Wikipedia article you linked has some of the details, but they are obscured a bit if you aren't looking for them. The linked main article on Linguistic determinism has some more direct details in its Criticism section: https://en.wikipedia.org/wiki/Linguistic_determinism
Linguistic determinism is something more among the lines of Newspeak from 1984 or Apple's Emoji Gun Control: By making it harder to talk about something, you make it harder to think about it, and harder to put into action. Thus, you can force language into a certain direction and culture will follow.
"Harder" is actually the weak hypothesis. The strong hypothesis is much more "impossible".
The common critical dissent to the strong hypothesis is often the Piraha and their lack of numbers and counting words, but demonstrated ability to count and do simple arithmetic even if not name the results verbally in their language. In the strong hypothesis, they wouldn't be able to count at all. Certainly they exhibit some symptoms of the weak hypothesis, because they may have a harder time at it than those with words to describe the effort. (The common criticism to the Piraha criticism is that we've also taught horses, pigs, dolphins, primates, etc to count and do simple arithmetic with no mutually distinguishable language whatsoever, so maybe that's not an impressive example, because maybe counting/arithmetic are outside of the linguistic faculties of the brain.)
> You can always invent things first and name them later.
I find a more impressive example of a disproof of the strong hypothesis and implication for the weak hypothesis to be the Calculus. Though the example is still in the domain of math, I feel that particularly as math trends towards more abstracted thought its entanglement with language seems ever stronger (see computing and programming languages as further examples).
Newton and Liebniz were both polymaths that spoke multiple languages and referenced many of the same primary sources (in those primary sources' own languages), but they had distinct native languages and cultures (Newton's English and Liebniz's German), yet both co-evolved the Calculus independently (to most accounts; let us dismiss the accusations of theft in one direction or the other). They evolved very different notations/jargon/language for the same concepts, but you could still see the underlying concepts as isomorphic, and that those concepts didn't particularly have names or words until the pair invented names and words for them.
Liebniz's notation was superior, in that it made it easier to see similar concepts as similar and easier to work with some of the more complex concepts, which is why the notation we use today is closer to Liebniz's than Newton's (though still somewhat of a pidgin/fusion of the two). Yet, Newton, despite using an "inferior" notation to do so is still best known for discovering so many more of the Calculus' implications to other sciences, especially Physics. In that case a potential disproof of even the weak hypothesis, or rather one anecdotal outlier of the weak hypothesis that seems to indicate how particularly "weak" it may be.
http://www.slate.com/articles/podcasts/lexicon_valley/2014/0...
It is not necessary to have a grammatical future tense to talk or think about the future. In fact, English does not have a future tense in the technical grammatical sense of 'tense'.
More generally, there's no evidence for the existence of human languages with the sort of lack of expressive power that you're talking about. All languages appear to be more-or-less equivalent in their expressive power, although of course that does not imply that exact translation is possible in all instances.
I was thinking the other day about that language that has no relative left/right and instead everything is based on (effectively) compass direction. The speakers reportedly have excellent direction sense compared to the average non-speaker. Their language is much more efficient at certain tasks we struggle with (English has crutches like "stage left", "starboard", and "your left" to compensate, but unless expected and practiced those are always effort/error prone. I myself fail at simple right/left way too often.)
Obviously this is a linguistic (and arguably cultural) trait with some value, without denying the value of also having relative kedtýright.
This language is unique (? or at least very rare), which tells me it failed to spread. Logical, as it would be hard for non-native speakers to pick up.
But that means "losing the evolutionary race" has no particular relation to the VALUE to us.
Would be OK to forget how to make stone tools because we don't need them for now?
Traits are useful in some circumstances and not some useful when circumstances change, who say you would not miss them in the future?
Knowledge is knowledge, it adds to our always imperfect understanding of the world.
https://numberwarrior.wordpress.com/2010/07/30/is-one-two-ma...
Idioms. Metaphors. Reference to background material. Such things may not be expressible in as few words, but there is no idea that cannot be expressed eventually.