Reminiscent of Austroasiatic synonyms like Old Khmersi(“male”) or MK words for "man, male" like *ʔŋsiil, *ensir, *kəsəy on the Malay Peninsula; Schuessler (2007) noted that foreign *-r sometiems left traces in OC initial complex. These relations, if, valid, would keep 士1 "bachelor, man, male" distinct from 士2 "servant, retainer, officer, scholar".
"take or give an office, serve", "servant", "retainer", "officer", "scholar"
Schuessler (2007) noted that one could naturally assume the semantic devlopment "male > man > servant > to serve" in order to posit that 士1 "bachelor, man, male" is the same word as 士2 "servant, retainer, officer, scholar". Yet, the exopassive derivation 事 (OC *ʔsrɯs, *zrɯs) "assignment, affair, thing" and Tibeto-Burman counterparts demonstrated no association with "man, maleness"; & "male" hardly derives from "to serve".
Therefore, Schuessler derived these forms from 理 (OC *rɯʔ) "envoy, jail official, matchmaker" & therefore ultimately Austroasiatic. In terms of phonology, MC *dʐ- normally does not occur with *l- and *ʂ in an ST word-family, apparently confirming a non-ST provenance.
However, MC *dʐ- here could go back to OC *s-r- (unlike MC *ʂ- from OC *sr) (related to AA) or an OC word loaned into Tibeto-Burman as Tibeto-Burman (Matisoff, 2003), > Burmese [script needed] (ca, “thing”) & Tibetan [script needed] (rdzas, “thing, matter, object”) (Gong, 1999). Even so, Tibeto-Burman cognates of this etymon and 所 (suǒ) are difficult to distinguish.