The author claims that real, measurable differences in how people think based on how they speak is too insignificant, because it doesn't constitute a "worldview

The author claims that real, measurable differences in how people think based on how they speak is too insignificant, because it doesn't constitute a "worldview." I rather think he sells short the importance of seeing actual empirical ways in which language does indeed "speak us". Say there are thousands, maybe millions, of such minute alterations in metaphor or thought pattern – is it really so hard to imagine that the aggregate of these alterations form unique, interesting ways to interpret and respond to the world? Do we not see evidence of this when we read, say, the poetry of different cultures? (It's a rhetorical question – I certainly think we do!) I got a book last year about learning Old English. I haven't gotten very far into it yet, but just the fact that I can go back a thousand years in human history and see where the roots of my language came from is an incredibly cool thing. I am also inspired by the alliterative schemes of old Anglo-Saxon poetry, and have it on the back-burner to see how or how not that could work in modern language (e.g. in Rebsamen's "Beowulf" translation). Similarly, whether or not I actually "speak" programming languages like Lisp, Forth, and APL in my daily work, they are a mine of cool ideas and experiments that I can still draw inspiration from, decades after the machines that they were originally built on crumble back into sand. Which probably emphasizes the "cold storage" value of language, not necessarily the "active speaking" value of it. I think the cold storage value of retaining language is indisputably massive, and projects like the Rosetta Project are really interesting to me for that reason. What's the value of actively speaking it, even when the community of speakers dwindles? Perhaps it is in helping to cement our understanding of the language before it fades away.