origin unclear

O’Donovan was mistaken in identifying this placename with Ceann Fuaid, the location of a Viking battle in 917CE (cf. FSÁG 4 s.n. Ceann Fuaid). In his edition of the Annals of the Four Masters (ii 588–591) he prints the name-forms from the relevant annalistic entry as follows: ‘oc Cind fúaith [-th sic] in airer Laighean’ … ‘cath Cinnfuait’ (588), ‘do chath Cinnfuaid’ … ‘i cCind Fuaitt’ … ‘ó Gallaibh Chinn Fuait’ (590), which he translated as ‘at Ceann-fuait [-t sic], in the east of Leinster’ [recte on the borders of Leinster], ‘the battle of Ceannfuait’ (589), ‘concerning the battle of Ceann-Fuaid’, ‘at Ceann-Fuaid’ and ‘the foreigners of Ceann-Fuaid’ (591). O’Donovan, who was working from Charles O’Conor’s edition of this part of the Annals (Rerum Hibernicarum scriptores veteres iii: Quatuor Magistrorum Annales Hibernici usque ad annum M.CLXXII 1826), was correct in spelling the placename ‘Ceann-Fuaid’ in Modern Irish orthography in his English translation: O’Conor used the autograph copy RIA C iii 3, which clearly reads oc cind fúait in airear Laighean ... cath cinn fúaitt (f. 350r), i ccind fuaitt and o gallaibh chinn fuait (f. 350v), and the corresponding entry in the Annals of Ulster under the year 917 reads oc cínn fhuait in aíríur laígin … cath cínn fhuait in Bodleian MS Rawl. B. fol. 29r. This Middle Irish Cenn Fúait cannot possibly be the same name as seen in the earliest references to Confey, Co. Kildare: ‘(Let)confi’ (1179), ‘Cunefi’ (1202), ‘Conephy’ (c.1228), etc. (For a full discussion which also rules out Confey on other grounds, see Colmán Etchingham, ‘The battle of Cenn Fúait, 917’, Peritia 21 (2010) 208–32.)

When O’Donovan had originally assessed the evidence for the Ordnance Survey in 1837, he had concluded that its precursor must have been a close compound which he spelled Con-magh (i.e., Conmhaigh [< cú+maigh] ‘hound-plain’). Although this proposed form is itself problematic and is unlikely to be correct, the analysis was far more reasonable than his later attempt to shoehorn the open compound Ceann Fuaid *[ˌkʹanˈfuədʹ], with unlenited final /d/ and stressed second element, into Confey [ˈkɑnˌfi].

The Middle Irish precursor to Confey is obscure. An Coiste Logainmneacha / Placenames Committee have not yet examined the administrative names of this part of Co. Kildare; the Irish form in official use since the 1980s is Confaí, a non-etymological phonetic approximation to the majority of the surviving forms. However, the very earliest spelling, ‘Letconfi’ on the Glendalough charter of 1179, when taken in context, seems to indicate that the second syllable was not originally long. Compare in the same charter ‘Tehcheli’, ‘Cellusailli’, ‘Techcumni’, ‘Cellbicsigi’, ‘Domnachrignaigi’, ‘Cellfiunnaegi’ (< Middle Irish Tech Chéle, Cell Usaille, Tech Cuimne, Cell Bicsige, Domnach Rignaige, Cell Finnmaige), in which final -i in the charter regularly stands for a final unstressed central vowel after a palatalized consonant in the Irish original. In this context ‘Letconfi’ seems to represent (leth) Coin(n)fe (‘the moiety of Coin(n)fe’) or similar. (It is unlikely to represent a longer form of the placename beginning lec ‘flagstone’.) For another example of the mistaken transcription of Middle Irish leth ‘moiety of’ in the same charter, even showing the anticipated Irish declension of the following placename, see ‘Lachcluanamor Moedoc’ = leth Clúana Móir Maedóc ‘the moiety of Clúain Mór Maedóc’. If this spelling ‘-confi’ was intended to represent the genitive form of the placename, as seems likely, it would indicate that *Coin(n)fe, or similar, was understood as an io- or iā-stem noun . A substantivized usage of Middle Irish coinnfe ‘fitting, proper’ (eDIL s.v. coindfe) would be very unusual but not impossible; compare in the glossaries cuinnisem .i. coindfe airisma .i. dún co trebaib.

The presence of medial /f/ in the Middle Irish form of any placename is problematic; similar examples include the nearby Life / Liffey and the historical Aífe (cf. FSÁG 2 s.n. Baoi Aoife; Gearóid Mac Eoin, The Celtic languages in contact (2007) 122). As no attested topographical elements seem suitable for lenition of *sw- > /f/ (e.g., in close compound with *konno- ‘hound’, *kondo- ‘head’) the Middle Irish form *Coin(n)fe may also represent a pre-Celtic placename.

An inquisition held c.1240 into historical appointments to the church of Confey shows that Adam de Hereford (†1216), owner of Leixlip and Confey, installed an Irishman as patron, apparently in the late 12th century. This was ‘Tatheig’ father of ‘Gillecondi Maclother’. As noted by Kenneth Nicholls (Peritia 5: 414–415), this surname seems to correspond to that of ‘Machenlodher’, the Gaelic lord mentioned on an earlier charter as the builder of a castle in a nearby part of modern Co. Kildare. The name Cú Lothair (gen. Con Lothair) is attested in this part of Leinster in the 11th and 12th centuries (see M.A. O’Brien, Corpus genealogiarum Hiberniae, where he provides a macron on the -o- in the index .i. Cú-lōthair; cf. Baile Chon Lóthair / Ballyconlore, Co. Wexford). On Tadhg’s death, Adam granted both churches to the Abbey of Saint Thomas (probably c.1205), but arranged that Tadhg’s son ‘Gillecondi’ would be permitted to hold the church of Confey, subject to an annual pension. The inquisition records that ‘Gillecondi’ was still paying this pension at some point between 1216, by which time Adam had died, and the year 1223; we also know that ‘Gillecondi’ himself was deceased when the inquisition was held c.1240. We have no evidence to throw light on the relationship between Tadhg Mac Con Lothair and the Gaelic lord who built the castle in the 1170s. However, Adam’s direct appointment of a member of this Gaelic family to the church of Confey seems to hint at an accommodation with native interests in the early years of the colony, and the arrangement of hereditary succession to church office is very noteworthy.

The younger Mac Con Lothair’s given name ‘Gillecondy’/‘Gillecondi’ is also worthy of consideration. Nicholls (ibid.) suggested, very tentatively, that the name might stand for Middle Irish Gilla in Choimded. Another possibility to consider, equally as tentatively, is that the -nd- in these latinized spellings represents Middle Irish -nn-, and that Gillecondi could be read as Middle Irish Gilla Coinne, an unattested personal name indicating devotion to Coinne, a pet form of Colm (see Pádraig Ó Riain, A Dictionary of Irish saints, s.n. Mochonna). Whatever the original Irish name of this ‘Gillecondi’, we know that the church at Confey was dedicated to Colm Cille by at least the later medieval period: it is attested as the church of ‘S. Colme de Comfye’ (1464), and the local Irish name of the blessed well which appears on the maps as St. Columb’s Well OS was ‘tobar choluim cille’ (1837). Indirect evidence that the cult of Colm Cille may have predated the Anglo-Normans in the area of Leixlip and Confey is found in the Middle Irish saints’ lists and martyrologies, where we find an otherwise unidentified ecclesiastical foundation named Middle Irish Ernaide (Modern Irish (An) Urnaí ‘the oratory’) in connection with Columb, and again, in his hypocoristic guise of Mochonna (Columb i ndErnedi CGSH 708.69; Mochonnae Ernaide MartT. 25 Jan. = MoChonna Ernaide MartG. 25 Jan). The pre-Anglo-Norman name of the church in modern Léim an Bhradáin / Leixlip (#1205), situated on the borders of Meath and Leinster, seems to have been Middle Irish Ernaide, but the identification with the places just mentioned is very uncertain. As to the dedication of Confey itself, note that the cult of Colm Cille was promulgated in the 9th century by Clann Cholmáin, the historical rulers of the vast area of Meath which adjoins Confey to the north. (See Máire Herbert, Iona, Kells, and Derry (1996); cf. Pádraig Ó Riain, ‘Boundary Association in Early Irish Society’, Celtica 7 (1972).)