I spend so much time complaining about the idiotic things non-linguists say about language that I like to give public kudos when they say sensible things, and this footnote on page 727 of How to Read the Bible (see this post) is so full of interesting details it’s worth quoting in extenso (the topic is the passage in Deuteronomy in which the temple is said to be the place where God “caused His name to dwell”):
On the Akkadian roots of this expression see Richter (2002). Richter’s thesis is that the “name theology” attributed to Deuteronomy by modern scholars is the result of a great misunderstanding: the biblical phrase “to cause My name to dwell” is essentially a cognate translation of the Akkadian expression šuma šakānu, which refers to the erection of a “display monument” marking a victory and a claim to the land where the monument is erected. Despite this erudite element [...], her critics have rightly countered that Deuteronomy itself offers ample evidence of its far more abstract concept of deity than that of earlier writers. As Richter herself notes of this phrase, Deuteronomy and the Deuteronomistic history contain, beside the Hebrew cognate of this phrase, other noncognate expressions, “to build a house for the name” of God, to “offer praise to the name,” and so forth. These would suggest that, whatever the origin of “to cause My name to dwell,” the idea of God’s “name” as a kind of a divine hypostasis is reflected in these other uses; “name” had been freed from its specific meaning in the original Akkadian idiom. Indeed, her overall argument appears to be based on a misconception, that because a word or phrase meant X in its original language, it will also mean X when borrowed by another language. Reality is full of examples of precisely the opposite. Thus, the French loan-word outrage suggests to most speakers of English an element of anger that is quite lacking in French. The reason is that English speakers unconsciously analyze the word as a combination of out + rage, whereas French speakers, having no morphological out, do not isolate the element rage (indeed, most native speakers will correctly perceive -age as the nominalizing suffix of outre, “beyond” [Latin ultra]). The legal phrase corpus delicti originally meant “the body of the offense,” that is, “the actual facts that prove that a crime or offense against the law has been committed.” But many people (including some lawyers) with a poor grasp of Latin understand corpus in the specific sense of a “(dead) body,” corpse. It is true that in a murder trial, the corpse does constitute the corpus delicti, but the phrase of course has much wider applicability — it can mean the stolen bicycle or the broken storefront window as well. Nevertheless, corpus delicti has actually developed in English the secondary meaning of a dead body — even in some dictionaries. Other examples could be given. Thus, the fact that šuma šakānu had the meaning it had in Akkadian does not guarantee that it ever had the same meaning in Hebrew. [...] Indeed, it is not hard to imagine the learned Deuteronomist borrowing this foreign idiom with the specific intention of creating an authoritative-sounding equivalent that would support his new theology. That is, he consciously took over šuma šakānu to help legitimate the idea that God had merely caused His “name,” but not Himself, to dwell in the earthly temple devoted to Him.
Extra points for talking about corpus delicti developing “the secondary meaning of a dead body” rather than calling it an error!