By David Stifter, OG(H)AM’s Irish Principal Investigator
This is an abridged excerpt from the article ‘The Donaghmore ogam stone (I-KID-008)’, which will appear in the 2025 volume of the Journal of the Kildare Archaeological Society.
The formula word MUCOI occurs around 80 times on ogam stones; mostly on stones from Ireland, with only three exceptions (one each in Wales, Cornwall and Man). The meaning of MUCOI as a word that indicates membership of a kin-group is well established, but its etymology has not been clarified so far (cf. Ziegler 1994: 211–212). Previous explanations are unsatisfactory: Holder (1904: 649) reconstructs *muk-o(ṷ)o- from a root *meṷk- ‘to untie, strip off’, which is semantically unconvincing and does not account for geminate *kk. Detschew’s (1957: 312) comparisons with Thracian μουκα-, μοκ(κ)α-, muca-, Lycian muka, muki, and Ossetian mukkag ‘seed, race, offspring’ are too distant to be convincing. Vendryes (1960: M-3) connects maccu etymologically with OIr. mac ‘son’, without discussing the involved problems; he is content with stating that “la formation en est obscure”. De Bernardo Stempel’s explanation of early OIr. maccu as a petrified ablative makṷkṷūd < *-ōd ‘originating from the son of’ suffers from weaknesses in diachrony and chronology: she explains some of the oldest attested forms via much younger sound changes, while she treats some of the youngest forms as most archaic. McManus (1991: 119–120, 180 n. 69) refrains from etymological speculations. Ziegler’s (1994: 211–212) proposals *mukkos-ī̆ or *mukkoi̭-ī̆ are phonologically possible, but operate with suffixal morphology that is rare at best within Irish; furthermore, she is silent about the semantic side of the derivation.
I will argue here that through a close morphological analysis, taking seriously the chronology of the attested forms, we can arrive at a compelling, albeit surprising, solution. The word is evidently related to Early OIr. moccu, maccu ‘belonging to the gens or family of’ (eDIL dil.ie/31187), but the precise relationship between the two words has remained obscure. This comparison requires that MUCOI represents a form with geminate *kk, which, however, like in the case of MAQI for *makṷkṷī, is rarely graphically expressed. The most common spelling is MUCOI, which occurs approximately 57 times (54 times in Ireland and three times outside Ireland). MUCCOI with two CC is found 7 times in Ireland. Other spellings (MOCOI, MOCO, MUCO, MOQOI, MU) account for approximately 21 cases. Given the difficulty in reading some stones, all of these figures are only approximations.
The position taken here is that MUCOI and moccu, even though they are clearly connected, are not morphologically the same form, but represent different inflectional forms of the same etymon. The point has been made repeatedly in the past that MUCOI appears to be trisyllabic (e.g., Holder 1904: 649; McManus 1991: 120; De Bernardo Stempel 1991: 218; Ziegler 1994: 212). Moccu likewise presupposes a trisyllabic preform, namely *mokkVCū, where C represents a consonantal segment that was lost during the Primitive Irish period (*s, *i̭, or *ṷ).
It has not been sufficiently appreciated in the past that MUCOI preponderantly occurs in the syntagma MAQI MUCOI (and its orthographic variants), which accounts for c. 40 examples. The sequence MAQI … MUCOI, with an intervening name, occurs 16 times; MUCOI on its own is found 21 times. Several of the latter are evidently late; and there seems to be a predilection for this construction in the Waterford region. Other examples are too fragmentary for a decision. The naïve interpretation of the syntagma MAQI MUCOI is that of a noun (MAQI) followed by an adjective (MUCOI), according to the fundamental word order of Irish where the head is followed by its determiner. Those cases where MUCOI is not immediately preceded by MAQI can be analysed as cases where the adjective was substantivised according to a natural tendency. Furthermore, some late examples of MUCOI on its own, especially in the evidently late spellings MOCO, MOCOI and MOQOI, may be influenced by early OIr. moccu, which, as will be argued below, represents a syntactically different construction. At that late stage, the original adjectival role of MUCOI was probably no longer transparent.
© COPYRIGHT, Dr Nora White 2009 2015-07-22
Early OIr. moccu is in several respects different from MUCOI. eDIL (dil.ie/31187) defines it as “a word belonging solely to the O.Ir. period, always used in apposition to a preceding [proper name] (occasionally a place-name) and folld. by another [proper name] in gen[itive] case.” The word in the genitive is the name of the eponymous ancestor. Its latest contemporary use in the Annals of Ulster is in 690, the last instance of its retrospective use is found in 789. The early 9th-century Félire Óengusso ‘Martyrology of Óengus’ still makes frequent use of it. The common spelling maccu in Old Irish sources instead of earlier moccu is evidently influenced by macc ‘son’. See further Mac Neill (1907); O’Brien (1973: 218–2019); Byrne (1995: 44, 50–51).
What is most relevant for present purposes is that in the Early OIr. formula X moccu Y ‘X belonging to the kin-group of Y’, moccu is never preceded by the word for ‘son’. It appears to be a petrified adverbial case form with ablatival meaning, i.e. literally ‘X (who is) from the kin-group of Y’. It is similar to corcu ‘belonging to the population group of’ (eDIL dil.ie/12444), itself synchronically the dative case of corcae ‘oats’ in a petrified adverbial, appositional usage, continuing earlier *korkii̭ū < ablative *korkii̭ūd. For ‘oats’ with reference to a population group, Mac Neill (1911: 64) compares síl ‘seed’ as a word for ‘race, descendants, offspring’ in Old Irish.
Starting from the hypothesis that PrIr. MUCOI and OIr. moccu continue two morphologically and syntactically separate instants of a common preform, the two can be combined under the adjectival reconstruction *mukkou̯i̯o-. MUCOI is the masculine genitive singular *mukkou̯i̯ī, agreeing with its head noun *makṷkṷī, while moccu is the petrified appositional substantivised ablative *mukkoṷi̯ūd. Moccu derives regularly from its preform. For MUCOI, we have to assume that suffixal *-oṷi̭ī is graphically expressed by -OI. Ziegler (1994: 212) pointed out that intervocalic *ṷ is well preserved in the ogam corpus, so its loss in such a preform would be unexpected. However, there is a wide-spread phonetic tendency for *ṷ to ‘disappear’ (or rather to be interpreted as a sub-phonemic glide) in the sequence *oṷi (cf. Stifter forthc. with numerous examples in Gaulish). The ogam corpus contains only one genuine example of *oṷi, i.e. the bilingual inscription W-DEN-001 (= CIIC 399) in Clocaenog, Wales, which has TOVISACI in the Roman script and [TO]VISACI in ogam. However, in this word OVI stands in stressed position, which may have favoured its retention, whereas the unstressed, suffixal position in MUCOI may have favoured the early loss of the graphic representation of *ṷ.
Having established the phonology of the word, we can now turn to its morphology and semantics. There is only one sufficiently similar noun in the lexicon of Irish from which it can be derived as a secondary adjective, i.e. OIr. mucc ‘pig, sow’. While that word is synchronically a feminine ā-stem in Old Irish, the u in the first syllable, which did not undergo the expected lowering to **mocc < **mukkā-, proves that the word historically belonged to the small and extinct class of earlier feminine ū-stems. In Primitive Irish, the word must therefore have been *mukkū- ‘pig’. In adjectival derivation, the feminine suffixal stem vowel *-ū- < PIE *-u-h₂- probably behaved parallel to its more common masculine and neuter equivalent *-u-, i.e. it could form a full-grade thematic suffix *-eṷo- (maybe via *-eṷh₂o-) which regularly gave *-oṷo- in Celtic. This was in turn extended by the additional suffix *-i̭o- to give *-oṷi̭o-. The original meaning of the resulting adjective *mukkoṷi̭o- must have been ‘belonging to a pig or sow’, which, applied to the offspring of a sow, could then mean more generally ‘being part of the litter’.
The final question relates to the original vocalism of ‘pig’ itself. The British Celtic cognates (Welsh moch, Late Cornish môh, Cornish mogh, Breton moc’h ‘pigs, swine’), as well as the Gaulish theonym Moccus of unknown meaning, but which formally possibly belongs here, presuppose *mokku-, not *mukku-. Testen (1999) proposed to etymologise these words as *mo(H)-pk̑-u- ‘big (?) stock’. This explanation works for British and Gaulish, but does not for OIr. mucc, since the voiceless geminate *kk should have blocked the raising of *o of *mokkū- to *u in the prehistory of Irish. The consistent spelling MUCOI in the oldest ogams attests to the presence of U in the early period. Either we assume a sporadic, very early vowel raising of *o > *u in Irish (e.g., caused by final *ū), or the British Celtic forms are due to ā-affection, for instance in the collective neuter plural *mukk(ṷ)ā. In that case, Gaulish Moccus has to be removed from the dossier of words for ‘pig’. Whatever the case may be, OIr. mucc ‘pig’ inevitably presupposes a *u for its immediate preform, i.e. *mukkū-. Reconstructing *mukkū- for the immediate precursor of the Irish word receives some support from the fragmentary kin-group name MUCCỊ[ (I-CAR-004), which looks as if it were derived from ‘swine’.
3D model by Nick Parish on Sketchfab (CC-BY-NC).
To summarise: Parallel to the cereal metaphor for a kin-group as the ‘seed’ or the ‘oats’ of an eponymous ancestor, a similar livestock metaphor was in use in which the offspring was thought of as being part of the ancestral ‘litter’. While porcine imagery is commonly used with a pejorative intent (e.g., Dmitriy Medvedev, 28 February 2025, speaking of democratic countries), the etymology proposed here presupposes that the early medieval Irish agrarian society emphasised the positive connotations of swine, for instance their fertility. This imagery would be particularly apt if MUCOI/moccu displayed a significant predilection for female ancestors. However, while this is true for some kin-groups (MAQI MUCOI DOVVINIAS, maybe Mac Cuill maccu Greccae, Lugaid moccu Óchae), it is not a general principle.
The original collocation was that of a noun + adjective, i.e. PrIr. *makṷkṷī mukkoṷi̭ī. Literally, it can be translated as ‘a litter-son’, i.e. ‘a son from the litter’, i.e., somebody related by blood-relation and not brought in from outside through marriage or other types of social bonds. This collocation is the basic formula MAQI MUCOI on ogam inscriptions. Occasionally, MUCOI on its own could be used to convey the identical sense. MUCOI is the adjective in the genitive case, agreeing with MAQI or a personal name. The etymological meaning of the Early Old Irish formula X moccu Y was ‘X from the litter of Y’.
Functionally, these formulae indicated the genealogical affiliation of persons and would have been notionally removed from the sphere of animal husbandry, even though, at least in Primitive Irish, the etymological connection with ‘pig’ must have been transparent to speakers of the language.
References:
Byrne, Francis J. 1995. ‘Dercu: the feminine of mocu’, Éigse 28, 42–70.
De Bernardo Stempel, Patrizia 1991. ‘Archaisch Irisch maccu als morphologisches Relikt’. Historische Sprachforschung 104, 205–223.
Detschew, Dimiter 1957. Die thrakischen Sprachreste. Wien.
Holder, Alfred 1904. Alt-celtischer Sprachschatz. Band 2. Leipzig: Teubner.
Mac Neill, Eoin 1907. ‘mocu, maccu’. Ériu 3, 42–49.
Mac Neill, John 1911–2. ‘Early Irish population groups: their nomenclature, classification and chronology’. Proceedings of the Royal Irish Academy 29 C, 59–114.
McManus, Damian 1991. A Guide to Ogam. Maynooth: Department of Old and Middle Irish.
O’Brien Michael 1973. ‘Old Irish personal names’. Celtica 10, 211–236.
Stifter, David forthc. ‘La phonologie du gaulois’, in: Coline Ruiz Darasse (ed.), L’épigraphie monumentale en gaulois, Bordeaux.
Testen, David 1999. ‘Stem-final *-kk- in Celtic terms for ‘pig’’. Ériu 50, 161–164.
Vendryes, Joseph 1960. Lexique étymologique de l’irlandais ancien. M N O P. Paris: CNRS – Dublin: Dublin Institute for Advanced Studies.
Ziegler, Sabine 1994. Die Sprache der altirischen Ogam-Inschriften. Göttingen: Vandenhoeck & Ruprecht.