ONlwSG

rubha 

Historically, hiatus in SG rubha has been represented orthographically by gh (MacDomhnuill 1741, 7; Shaw 1780; Mac Farlan 1795; MacFarlane 1815: rughadh; and Armstrong 1825: rugha, rughadh; while Dwelly (1911) cross-references rugha to rudha) dh (HSS 1828; M^cAlpine 1832; Dwelly 1911; MacLennan 1925; and Dieckhoff 1932; while Armstrong (1825) cross-references rudha and rudhadh to rugha) and even th (MacEachen 1842: rutha); modern usage employs bh, e.g. Wentworth 2003, s.v. point: Gairloch, and AFB^˄.

m. [ˈɍu-ə] (sometimes with a closing fricative -[ɣ]) or [ˈɍu-u] (with vowel harmony), 

Cf. /Ru-ə/, /Ru-u/ (Oftedal 1956, 79–80: Lewis). For vowel harmony, see McKenna 2011.

occasionally [ɍuː], 

Cf. Dieckhoff 1932: Glengarry.

gen. idem, ‘a point, promontory or headland projecting into the sea or loch’ is derived by Borgstrøm (1940, 223) from ON hrúga (also hrúfa) f. ‘heap, pile’; Cox (1991, 492) follows Borgstrøm, and McDonald (2009, 365) considers the loan likely. Oftedal earlier (1956, 79–80) merely notes Borgstrøm’s thesis, 

Pace Cox (1987 II, 7.61; 2002a, 151.85 (so also Murray 2006, 134); and 2022, 5–6), who incorrectly states that Oftedal here concurs with Borgstrøm’s thesis.

but later (1980, 173) writes that ‘the etymology from Norse is not convincing but better than other guesses I have heard’.

Medial ON g [ɣ] could yield either SG [ɣ] (e.g. ON stig acc. > SG stiogha [ˈʃtʲiɣə] ‘path on a slope’, s.v.) or hiatus (e.g. ON *Smuga ‘(the) cleft’ > SG Smudha *[ˈs̪mũ-ə] NC419675 (Eng. Smoo) (Cox 2022, 891–92)), while medial ON f [v] could yield either SG [v] (e.g. ON *Grafir ‘(the) hollows’ > SG Grabhair [ˈɡ̊ɾavəɾʲ] (ibid., 753–54) or hiatus (e.g. ON klofa obl. > SG clobha [ˈkʰɫ̪ o-ə] ‘tongs’, s.v.). The development of ON ú here is more problematic. Long ON ú might be reduced to a short vowel before hiatus in Gaelic – at least in Lewis (Oftedal 1956, 40, 68) – yet SG rubha has a short stressed vowel except in Bernera (Lewis), where Borgstrøm (1940, 223) records [ʀuː əɣ], although this might be due to analogy with the homophonic rùghadh ‘the arranging of peats to dry in small stacks’, s.v. rùghan.

?Cf. also the occasional variation found in the dialect between long monophthongs and disyllables, e.g. in the village name SG Bradhagair [ˈb̥ɾa-aˌɡ̊əɾʲ] > [ˈb̥ɾaːˌɡ̊əɾʲ] (Cox 2022, 127–28).

Stokes (1893a, 447), however, already equates SG rubha with EG ruba (i.e. ruḃa) ‘a point of land’, so also MacBain (1896; 1911, s.v. rudha: EG rube [sic]) and Watson (1904, lxi: EG ruba ‘a promontory’; 1926, 287–89: 288: ruba ‘a cape, point, also copse wood’); cf. Meyer 1911, 126: EG ruba ‘a brake, clearing’. MacBain’s reference to Ir. rubha is probably to O’Reilly’s Irish-English dictionary’s rugha, 

O’Reilly 1817: (in Gaelic script) ruġa, (in Roman script) rugha ‘a promontory, cape, headland’; (in Gaelic script) rúġaḋ [sic], (in Roman script) rugadh [sic] ‘a cape, headland, promontory’; O’Reilly 1864: (in Gaelic script) ruġa, (in Roman script) rugha ‘a promontory, cape, headland’; (in Gaelic script) rúġaḋ [sic], (in Roman script) rughadh ‘a cape, headland, promontory’.

but rugha here is probably a ghost word adopted from Scottish Gaelic.

Cf. Shaw 1780: rugha; Mac Farlan 1795: rugha; MacFarlane 1815: rughadh.

EG ruḃa, however, survives in place-names in Ireland, e.g. Rubane (Ir. [An] Rubha Bán ‘(the) white clearing or thicket’) in Down (Hughes and Hannan 1992, 110; McKay 2007, 127), and the tautological Rue Point (Ir. [An] Rubha ‘(the) headland’) in Antrim (McKay ibid.).

Further, see Hogan’s (1910) Onomasticon Goedelicum, p. 589, s.vv. rubha, ruba. Joyce’s (1912 II, 342–44) analysis of Ir. rubha in a number of place-names as meaning Eng. rue (the herb) seems onomastically unlikely. See also Murray (2006, 134) for the possibility that O’Davoren’s Glossary (Stokes 1904, 438.1366) contains an instance of ruḃa in the sense ‘clearing’.

The senses ‘brake, clearing’ are extensions of ruḃa as the verbal noun of *ro-ben- ‘strike, cut, hew etc.’ (cf. Meyer ibid.; Vendryes 1996, s.v. rubae; eDIL^˄, s.v. ruba(e)).

MacBain (1896; 1911, s.v. rudha) derives EG ruḃa from *pro-bio- ‘being before’, from the root bu of the verb ‘to be’. Darton (n.d., 235, s.v. rubha) compares SG rubha with W rhiw ‘hill’, but no connection has otherwise been shown. (For W rhiw, cf. the element -rew in Cornish place-names (GPC^˄).)

For ruḃa’s range of senses, cf. EG ros ‘a wood, frequently of a wooded height or of a promontory on the shore of a lake or river’ (eDIL^˄; Watson 1926, 116, 496–97).

Cox (1987 II, 7.61) suggests that EG ruḃa in the sense ‘clearing’ might be due to influence from ON ruð nt. ‘clearing’, and that the sense ‘mound’ might be due to influence from ON rófa f. ‘tail’, but neither suggestion is a desideratum.

For ON hrúga, hrúfa, s.v. rùghan.