Meaning: '(wild) goat, gazelle'
Akkadian: armû (arw/biu, arwû, fem. arw/bītu, armītu) 'gazelle' OAkk. on [CAD a2 294], [AHw. 73].
Eblaitic: a-wi-um = Sum. [DÀR]A?.DÀ [MEE IV 1251']. // To be normalized as /ʔarwiyum/ (on the orthographic conventions underlying this interpretation v. [Conti Ebla 19, 32ff.]). Cf., however, Sem. *ʔiwVy- No. 95. Hardly to be identified with *ʔayyal- No. ... as proposed in [Bonechi-Conti 10].
Arabic: ʔurwiyyat- 'chèvre de montagne' [BK 1 959], [Fr. II 214], [Lane 1196-7], [LA XIV 350] (ʔal-ʔunt_ā̀ mina-l-wuʕūli; also ʔirwiyyat-). V. further [Hommel 281].
Epigraphic South Arabian: (?) Sab. ʔrwy-n (pl.) '(female) mountain goat, ibex' [SD 7]. // A Hapax in the dificult passage RES 4176/5: ḥt_̣r tʔlb s1ʔr ʔrwyn bn ns3g bn mṣrn ('the god [Tʔlb] has prohibited the rest of the mountain goats from being prevented from feeding' according to [Biella 26]). The traditional interpretation of this term as an animal name was rightly put to doubt in [Sima 25] ('Tʔlb hat ein Verbot erlassen bezüglich der ʔRWY, dass man (sie) nicht wegtreibe? von dem MṢR').
Mehri: ʔarī́t 'goat' [Nakano 116].
Notes: The Akk. form is to be treated, at least synchronically, as a separate lexeme different from armu (on which see below); only forms with noncontracted or plene-written Auslaut are thought to belong here. Aboundantly attested from OAkk. on, they are found almost exclusively in proper names and lexical lists (ar-mu-u = ṣabītum in Malku V 42, ar-wi-um = MAS̆.DÀ in CBS 8538:17) which made doubtful the genuine Akk. origin of the term (cf. the explicit statement in [CAD]: "the WSem. loan armû"). Now, however, a. is attested not only in OB Mari (šētētum <ša> ar-wi-i 'nets <for> gazelles' ARM 14 38:6), but also in the OB Baby-Beschwörung OECT XI 19 22 (liṣbassu-mi ša iṣbatu ṣabītam likassīšu-mi ša ukassû ar-wi-[am] 'may that who caught the gazelle catch him, may that who tied the deer tie him!', cf. [Far ber...]) so that its good Akk. origin is beyond doubt. // According to the commonly accepted point of view, the above forms should be united with Hbr. ʔaryē 'lion', Gez. ʔarwe 'wild animal' and similar terms (cf. No. ...) into one PS root meaning 'wild beast' (v., among many others, [Hommel 281-2], [Landsberger 100], [DRS 32], [LGz. 40], [Lipiński Lion 213], with reservations also [OS No. 56]). At present, we prefer to treat the two groups of terms as belonging to separate PS roots (doubts about the correctness of the traditional approach cf. already in [Nöldeke MG 167]). Note in particular that if the traditional approach is correct, one has either to assume that the rather specific meaning shift 'wild beast' > 'wild goat' occurred independently in both Akkadian and Arabic (languages quite distant both geographically and genealogically) or to posit it already for the earliest stages of PS (which practically amounts to postulating two separate roots as it is done here). On the other hand, the obvious advantage of the traditional approach is that it takes into account the complementary distribution of the two sets of meanings: in no Sem. language both *ʔarwVy- 'wild goat' and *ʔarwVy- 'wild beast, lion' are reflected. // Definitely inconvincing is the comparison of both groups of terms to Akk. erû 'eagle' (for which v. No. ...). // Likely related to the present root are semantically similar forms with the base *ʔVr- augmented with suffixed -m- or -n- (cf. *ʔVrn/m-, No.). // For bibliographic references v. next item.