v1.0
Published 01/10/24
tabh, an m. [əᵰ̪ ˈt̪ʰav], gen. taibh or indeclinable, used with the article, ‘the open sea, the Atlantic Ocean’. This word, along with its variants (below), is derived from ON haf nt. ‘sea, ocean’.
Cameron (in MacBain 1894a, 645); MacBain (1896; 1911); Goodrich-Freer (1897, 67); Mackay (1897, 94: Ice. haf); Henderson (1910, 137, 148); Mackenzie (1910, 385); MacLennan (1925); W. J. Watson (1932, 375); Sommerfelt (1952a, 230); Stewart (2004, 415); McDonald (2009, 360); Ó Muirithe (2010); Cox 2022, 853–56.
ON haf *[hav] would formally be expected to yield SG [hav]. Two main strategies (A–B) have evolved to deal with initial h-, which otherwise would not normally occur in radical position in Early Gaelic. What appears at first sight to be a third strategy (C) most probably constitutes a back-formation from (b). Variation in the final consonant (-bh, -f, -mh) occurs in each strategy.
A. SG na h-abh, na habh [ᵰ̪ə ˈhav], na haf [ᵰ̪ə ˈhav̥]
The plural article na prefixes h- to a following vowel, but in this instance accommodates and therefore preserves the original aspirate in ON haf. This strategy is found in the southern Outer Hebrides.
A 1.
(i) na h-abh (Borgstrøm 1936, 295: Na h-Abh, Barra; Faclan bhon t-Sluagh˄: South Uist).
(ii) na hābh (Henderson 1910, 148: Barra, Eriskay) and, omitting the article, àbh (McAlpine 1832, s.v. tabh: Lewis). Henderson’s contention that the word is ‘sometimes long’ (C 2(i)) does not seem to be supported by the evidence: although McAlpine writes àbh in line with his entry tabh [tâv] (indicating a long vowel
McAlpine does not use lengthmarks in headwords.
), the latter may have arisen through conflation with tàbh ‘hand-net’ (B 1). For the spelling na habh, as opposed to na h-abh, cf. SG Na Hearadh (Eng. Harris) from ON *Herað or (pl.) *Heruð ‘(the) district(s)’ (e.g. Cox 2022, 817–20).
(iii) Rona na nav (1561, Collectanea 1847, 4) may be a typesetting error for *Rona na hav, for SG Rònaigh na Habh ‘Rònaigh of the ocean’.
The island Rònaigh (Eng. Rona) HW811324 lies about 70 km (44 miles) north-north-east of Lewis (Cox 2022, 487–56). For other forms of this alias, see B 2(ii) and B 3(iii).
A 2.
The final fricative -bh is devoiced,
Before pausa in Gaelic, but perhaps also more generally under the influence of Scots haaf [hɑ(:)f], also from ON haf (SND˄).
which is consequently reflected in the orthography.
(i) na haf (Henderson 1910, 137: Barra, Eriskay; Borgstrøm 1937, 229: na Haf [nə haf], Barra (so also Sommerfelt 1952a, 230); McDonald 1972: South Uist), and
(ii) haf (Goodrich-Freer 1897, 67; McDonald 2009, 360), omitting the article.
A 3.
The final fricative -bh is replaced by -mh, cf. tabh ~ tamh (B 3) and sabh ~ samh (C 2).
For spontaneous nasalisation in stressed syllables in Scottish Gaelic, see Ó Maolalaigh 2003, 109–17.
(i) amh (Shaw 1780; Armstrong 1825: †amh
Armstrong conflates amh ‘ocean’ with †amh ‘water’, the latter for †abh ‘river’ (EG aḃ; see W. J. Watson 1926, 477).
), and
(ii) àmh (HSS 1828, citing MacFarlane 1815, s.v. tabh, who in fact gives short amh), both omitting the article.
B. SG an tabh [əᵰ̪ ˈt̪ʰav], an taf [əᵰ̪ ˈt̪ʰav̥]
Initial h- is taken to represent the lenited form of SG t- and preceded by the masculine singular article an. This strategy is found in the northern Outer Hebrides and Argyllshire.
B 1.
(i) tabh (Shaw 1780
Shaw gives the senses ‘a sort of fishing-net, the ocean’, conflating tabh with SG tàbh ‘hand-net’, s.v.; so also Armstrong 1825.
; Armstrong 1825
For Armstrong’s comment that ‘[t]o the Celtic root ta may be traced the names of all waters which begin with ta or tam’, see Nicolaisen 2001, 244–45.
; HSS 1828; Cameron, in MacBain 1894a, 645; MacBain 1896; 1911; Henderson 1910, 137, 148: Harris; MacLennan 1925; Stewart 2004, 415; McDonald 2009, 360; and Faclan bhon t-Sluagh˄: Scalpay), omitting the article. McAlpine’s form (1832, s.v. tabh: [tâv], with a long vowel) may have been influenced by SG tàbh ‘hand-net’, s.v. (cf. Shaw and Armstrong, above).
(ii) taibh (MacDomhnuill 1741, 6; Shaw 1780), omitting the article, is perhaps in origin a genitive form (see below).
(iii) Genitive an taibh occurs in the place-names Sgeir an Taibh ‘the skerry of the ocean’ NA686464, the outermost rock of the Flannan Isles, and Camus cùil an Taibh (sic) NM267236 (J. C. Watson 1965, 44, 154). According to the orthography, the latter name means ‘the bay of the nook of the ocean’, but the derivation is problematic (for a discussion, see MacArthur 2022˄).
B 2.
The final fricative -bh is devoiced, which is consequently reflected in the orthography.
(i) taif (Armstrong 1825), omitting the article, is perhaps in origin a genitive form.
(ii) Genitive an taif (MacDomhnuill 1776, xv: a ’n taif, 27 a ’n Taif, an Taif
Not an taibh as Henderson (1910, 137, 148) claims, but so written in J. C. Watson’s edition (1965, 44–45): Ri fuaim an taibh | Is uaigneach mo ghean ‘At the ocean’s sound, my mood is forlorn’.
) occurs in the first line of a song by Màiri nighean Alasdair Ruaidh.
The first edition of the songs of the Harris poetess Màiri nighean Alasdair Ruaidh (c.1615–c.1707) was published by MacDomhnuill in 1776; the second by Mackenzie in 1841; the most recent by J. C. Watson in 1965.
Similarly, Rona-a-taiff (1700, Morrison MS, in Robson 1991, x) is for SG *Rònaigh an Taif ‘Rònaigh of the ocean’, with genitive of the specific.
(iii) The radical form *an taf is represented in an unconventional orthography in an Haf-a-Siar from Lochs in Lewis (Christiansen 1938, 4, but 15 (without article and erroneously) haf-a-shiar), which doubtless represents SG (Lewis) *an taf a-siar *[ə ˈᵰ̪ʰav̥ ə ˈʃiəɾ], with the adverb (a-)siar used adjectively in the sense ‘in the west, western’ (< EG s-iar ‘to the west’).
B 3.
The final fricative -bh is replaced by -mh, cf. A 3 and C 2.
(i) tamh (Shaw 1780; Armstrong 1825), omitting the article.
(ii) an t-saimh (Faclan bhon t-Sluagh˄, Harris) is for an taimh and is perhaps in origin a genitive form: *[ə ˈᵰ̪ʰɛv] (or *[ə ˈᵰ̪ʰɛ̃v]) implies either orthographic taimh or t-saimh (the latter with mutation of initial s- + vowel after the genitive masculine article). However, given the Harris connection, we should probably read taimh (taibh); for saimh, see under C 2(i).
(iii) Genitive an daimh occurs as specific in SG Rònaigh an Daimh, ‘which appears to be “Rona of the stag” but may in fact be Rònaigh an Taibh “Rona of the Atlantic” ’ (Taylor 2011, 124, s.v. North Rona): an daimh for an taibh is likely to be the result of folk etymology.
C. SG an sabh
While initial h- could theoretically represent the lenited form of SG s- and, preceded by the masculine singular article an, yield an sabh, the form an sabh most likely arises from an t-sabh, a back-formation from an tabh, cf. (B 3(ii)).
C 1.
(i) sabh (Mackay 1897, 94: Sutherland
Dwelly (1911) gives samh.
), omitting the article. Mackay extrapolates sabh from Mackenzie’s 1841 edition of a song by Màiri nighean Alasdair Ruaidh: (p. xliii) an t-sáimh, (p. 22) an t-shăimh), (p. 376) sámh, but MacDomhnuill’s earlier 1776 edition gives an taif (B 2(i)); cf. C 2(i), below.
(ii) an sabh (K. Mackenzie 1910, 385). Mackenzie (ibid.) records an sabh a-siar ‘the western sea’ for Lochs in south-east Lewis, but this may also be a back-formation, with an tabh a-siar *[ə ˈᵰ̪ʰav ə ˈʃiəɾ] interpreted as an t-sabh a-siar; cf. the alternative interpretation an Haf-a-siar, also from Lochs (B 2(iii)).
C 2.
The final fricative -bh is replaced by -mh, cf. A 3 and B 3.
(i) sāmh (Henderson 1910, 137, 148). In contrast to Mackay’s extrapolation of sabh from Mackenzie’s 1841 forms (C 1 (i)), Henderson takes sāmh directly from Mackenzie’s contents’ page and glosssary (p. lxiii: (gen.) an t-sáimh, p. 376: sámh). However, the first line of the song in Mackenzie’s edition actually reads (p. 22) Ri fuaim an t-shăimh, with the breve indicating short [ɛ̃], with assonance between t-shăimh and ghean. It seems likely that Mackenzie used a breve to indicate the short vowel length of a relatively unfamiliar word, but that acute accents instead of breves were typeset in error in the contents’ page and glossary. Of course, Mackenzie’s form was itself a reinterpretion of earlier an taif (B 2(i)), and it follows that Henderson’s can be dismissed.
(ii) samh (Stewart 2004, 412). While Henderson cites long sāmh in the sense ‘surge’ (C 2(i)), Stewart cites short samh in the sense ‘the crest of a breaking wave’, following MacLennan (1925: ‘surge, white crest of wave, waves as they break on shore’). Mackenzie’s 1841 edition of Màiri nighean Alasdair Ruaidh’s song not only changes former an taif to an t-shăimh, it redefines the word from ‘ocean’ to ‘surge, the agitation of waves on the sea-beach, the crest of whitened billows’ (p. 376).
In his own edition of Màiri nighean Alasdair Ruaidh’s song, W. J. Watson (1932, 198, 375) gives short an t-saimh but the sense ‘the ocean’.
This new sense is noted in Dwelly’s dictionary (1911: ‘surge, agitation of waves on the sea-beach; crest of whitened billows’, citing Mackenzie), repeated by MacLennan and Stewart (as above), and more recently incorporated into the online Gaelic dictionary Am Faclair Beag. In addition to the sense ‘surge’, however, AFB˄ gives ‘gust’, the latter from a reading of the form samhan in Le gaoth bho thuath ’s i bagarrach le samhan uisge mìn, | Is chan fhaiceadh tu le dorchadas am fear a bha ri d’ thaobh (Comann Eachdraidh Tholastaidh bho Thuath 2005, 83) ‘with a northerly wind and with it threatening samhan uisge mìn, because of the darkness you couldn’t see the man beside you’. However, samhan uisge mìn seems better read as meaning ‘fine mists of rain’, with the plural of samh ‘haze, mist’
Or with a diminutive form samhan: ‘a fine mist of rain’.
(cf. McDonald 1972, s.v. samh: South Uist; Faclan bhon t-Sluagh˄, s.v. idem: Tiree). In consequence, the form samh in the sense ‘surge; gust’ can be considered a ghost word.
According to Bugge (1912, 295), ON haf also gives Ir. tabh, but the Irish form is taken from O’Reilly’s Irish dictionary (1817; 1864
The word is not listed in O’Brien’s Irish dictionary (1768).
), where it was no doubt adopted from Shaw’s 1780 Scottish Gaelic dictionary. De Vries (1962, 201) follows Bugge, so also MacLennan (1925) and McDonald (2009, 360), but, as Marstrander (1915a, 126) points out, tabh occurs only in Scottish Gaelic.
The Latin text of the Treaty of Perth (1266) contains the word haffne: ... Manniam cum cæteris insulis Sodorensibus, et omnibus aliis insulis ex parte occidentali et australi magni Haffne (Munch 1874˄, App., 27) ‘Man and all the islands of the Western Isles, and all the other islands from the western and southern part of the great ocean’. Magni is the genitive (masc./nt.) of Lat. magnus ‘great’, and haffne is presumably also in the genitive case or at least in genitive position. Haffne does not appear to be Gaelic; neither could it represent an Old Norse genitive hafs or (with suffixed article) hafsins ‘the ocean’. It might, however, represent a petrified form of the Old Norse dative, e.g. (í ) hafinu ‘(in) the ocean’ (Cox 2022, 853–56).