Hebrew borrows its name Sapir from the Old Iranian name “valuable one of Saturn” Saniprigyam, Sani being the planet Saturn. (Klein.) Probably these Indo-Europeans view Saturn as the farthest planet, grinding across the celestial ceiling to guide the cycles of the seasons below. For Hebrew-speakers too, the gem manifests the heavens above.
Saniprigyam → Sapir
The Hebrew gem name Sapir comes from the language of the merchants who bring lazurite into the horizons of the ancient world.
(Vere, Stein. ©2011. “Sapir Caravan”.)
Evidently, one source supplies the known world lazurite, the ancient mining area of today Ladjuar Medam, in the remotest reaches of Afghanistan. Around there the tribes speak the Iranian languages.
Between their tribes in the east and the tribes of Yisrael (Israel) in the west, the mutual caravans trek the traderoutes meeting along the River of Prat (Euprates) in today Iraq.
(Vere. ©2011. “Sapir Routes”.)
Much of the lazurite heads from caravan to caravan to Egypt.
While the Tanakh (Hebrew Bible) keeps traditions about the Bronze Age, these mainly crystallize later within Hebrew dialects of the Iron Age, around the years −700s to −500s (namely the 8th to 6th centuries BCE).
Yisrelim already use lazurite a millennium before, but their royal scribes keep the language current, updating the ongoing Tanakhi traditions. Probably the name Sapir keeps current too, making Old Iranian during the Iron Age the contempory source.
Note the Hebrew language is actually one of several Canaanite dialects. Possibly, mutually intelligible Canaanite-speakers develop this Hebrew form of the gem name. For example, to the north in Lvanon (Lebanon), the Knaanim (Canaanites proper later called Phoenicians) boast an international gem trade, hosting merchants from three continents.
Even so, Hebrew-speakers themselves pursue the traderoutes, and probably know the Old Iranian-speakers firsthand.
At first, Hebrew-speakers attempt the foreign sound, Sanipriˈgyam.
The stress is on the final syllable -ˈgyam (the consonant -gy- is possibly a voiced palatal frigative /ʝ/). There is some hypothetical Hebrew realization, such as Sanipriˈyam סַנִפְרִיָּֽם . But it proves unstable. The unstressesd short -i- /ɪ/ is weak in Hebrew and often expands to long -i- /i/ or oppositely contracts to a shwa /ə/ even going silent, thus Sanpriˈyam. But then the resulting consonant cluster -npr- needs a vowel (this r is probably a dental tap /ɾ/, sounding like a quick gentle -d-). So the Hebrew adds a shwa -ə- to alleviate the consonants, thus Sannpəriˈyam. But then two squeezed vowels before the stress on last syllable -pəriˈya- proves difficult to maintain. Dramatically, the upcoming vowel -i- swaps around with the -r- (metathesizing -riy- to -iyr-), thereby replacing the shwa -ə- to alleviate the consonantal cluster, trying for Sanpiyˈram. But this -iy- vowel with palatizing consonant produces a strong long -i-, and the main stress of the syllables shifts to it. Thus, Sanˈpirˈam. Thus the final vowel -a- loses emphasis and contracts, with -ˈpiram- becoming -ˈpirm-. But in the potential Sanˈpirm, the final consonant sequence -rm is awkward (again this -r- sounds somewhat like d). So the generally weak final -m fails to happen. Thus Sanˈpir. Ultimately, the generally weak -n- assimilates into the oncoming -p- (geminating the -pp-). So finally, the gem name stabilizes within the rhythm of the Hebrew language as: Sapˈpir [sap:ˈʰiɾ] סַפִּיר . Whence Sapir.
Compare how Aramaic-speakers render this Old Iranian name. They add another syllable, the suffix “-on”, for some hypothetical realization, such as Sanipriˈyamon סַנִפְרִיָֽמוֹן (with the stress on -ˈyam-), possibly meaning something like “a little (gem) of the Saniprigyam (mineral)”. They too realize the -gy- sound as -y-. They too drop the weak i vowel, thus Sanpriˈyamon. But then they too must alleviate the consonantal cluster -npr-. They too add a shwa vowel -ə-, thus Sanpəriˈyamon. But (instead of metathesizing the -riy- like the Hebrew-speakers do), this shwa assimilates into the oncoming -i- and palatizes and becomes a strong -i- like it, because the weak i is unstable, thus Sanpiriˈyamon. Like in Hebrew, this -iy- vowel with a palatizing consonant is strong, and the syllable stress shifts to it. Then the unstressed -a- contracts, thus Sanpiˈrimon. Possibly, the nasal sound -m- assimilates into the nasal sound of the final -n. In any case, the gem name stabilizes in Aramaic as: Sanpiˈrinon סַנְפִּירִֽינוֹן .
By the way, at that time among the Indo-Iranian languages, Old Iranian and Sanskrit Indian relate closely. The cognate in Sanskrit is Shyani-priya [ɕɐ.nɪ.pɽɪ.yɐ] श नि प्रि य. You can see how easily the name can drift: Saniprigyam → Shyanipriya → Sanpir → Sapir. Originally linguists suggest Hebrew gets the name from Sanskrit.
But probably the nearer Iranian-speakers are the point of contact. So.
Saniprigyam → Sanipriyam → Sanpir → Sapir
Something like that.
• (Vere, Stein. ©2011. “Sapir Cravan”. brakha.blogspot.com. Derivation. McDowell, J. 2008 share alike. Wikipedia. http://en.wikipedia.org/wiki/File:Kuchi_caravan_in_Badakhshan,_Afghanistan.jpg, 2011.)
• (Vere. ©2011. “Sapir Routes”. brakha.blogspot.com. Derivation. Yahoo. 2011. http://maps.yahoo.com/#mvt=h&lat=35.499354&lon=51.105545&zoom=5, 2011. Gross, Ross. “Silk Roads”. Travels and Writings. 2007. http://rolfgross.dreamhosters.com/HistoryAtlas-GE/Post/SilkRoads.jpg, 2001.)
• (Klein, Earnest. ©1987. “סַפִּיר”. A Comprehensive Etymological Dictionary of the Hebrew Language for Readers of English. New York: Macmillan Publishing Company.)
Note Klein transliterates the Old Iranian name as Sani-prijám with stress on the final syllable, and derives both Hebrew and Aramaic from it. This -j- possibly represents a voiced palatal /ʝ/, as I suspect it is, paralleling the unvoiced plosive /c/ but weakening, whence my transliteration as -gy-. Hebrew, Aramaic, and Sanskrit all render it as an approximate -y-, tho it relates -dzh- in later Avestan.