There is a demand right now for companies and industries to be seen addressing problems of systemic racism. You can look deep into your internal processes, identify how unconscious and conscious biases are undermining the acquisition or advancement of underprivileged minorities, and modify them to create more equitable outcomes. Or you can tweak some jargon to be more linguistically inclusive, a problem that (as far as I'm aware) no one has actually complained about.
In other words, it's an attempt to be inclusive by claiming very loudly that you're inclusive while doing absolutely nothing to actually help change that.
https://www.reddit.com/r/programming/comments/m78lk2/github_...
- The "strong" version, that says effectively language controls thought and you cannot think things you cannot express in words.
- The "weak" version that says language influences thought.
It was realised that the strong version was wrong before World War 2. The weak version is probably also wrong, because viewed through modern eyes a lot of this research is exactly the sort of stuff that's unlikely to replicate. The claims that seem like they might replicate are about trivial and arbitrary differences e.g. Swedish speakers using "a long time" vs Spanish speakers saying "a lot of time".
For example: one major problem with Sapir-Whorf research is that a lot of it is based on anthropologist's claims about obscure languages, often the languages of isolated tribes. These are inherently difficult to replicate because often these tribes disappear or their language changes (supposedly) after contact with western civilization. Additionally the field is plagued with bad research that should have immediately raised suspicion, for example, the 1981 claim that Chinese people have difficulty reasoning about counterfactuals because of grammatical differences in their language, which turned out to be simply due to bad translations [1]. Many other studies are very weak conflations of correlation/causation, for example, this field is littered with claims of the form: speakers of languages with feature X go to college more often/are richer/are more or less confrontational. This sort of thing is ripe for all the classic ills of social science research, like P-hacking.
Outside of academia the most famous example of belief in strong Sapir-Whorf is surely communism and their insistence on calling each other "comrade". This ubiquitous term didn't make them behave in comradely ways as assassinations, executions and widespread corruption were all common in Soviet societies.
It's probably not a surprise that the left has difficulty letting go of Sapir-Whorf. This is partly because it's easy to mix up cause and effect with language control. Being in a position to severely punish people for something trivial, like use of a previously acceptable phrase, indicates to all observers that you have total power, that you can do whatever you like regardless of how unreasonable or extreme it is. Those observers may incorrectly conclude that language control is the actual source of those people's power, when in reality, it is the public punishment for arbitrary and harmless acts of disobedience that is the real source of power. Language is just a highly visible way to achieve that.
[1] https://www.sciencedirect.com/science/article/abs/pii/001002...
https://www.realclearpolitics.com/video/2019/08/23/tucker_ca...
I'm tired of thinking and thinking and thinking about these ideas that always terminate in foundational disagreement. I expect to encounter disagreement, and I'm fine coexisting with it, but I need the ability to isolate myself and those I care about from the effects of it.