My original version of this list was prepared for Academy of Saint Gabriel report 1738. In further research to prepare this article, I ran into a few typos and errors of grammatical classification in the previous version. To correct them and to avoid having to re-do the research, I have put together an annotated reference version of the data. One of the reasons for setting out the reasoning in such detail is that I think it helps me avoid the sorts of interpretation errors that got included in the previous version. This appendix presents a detailed discussion of how and why I reconstructed the forms the way I did.
All references are listed on another page.
1. Masculine o-stem, with deuterotheme <-agnas>, genitive <-agni>. Both declensional forms are mentioned in McManus p.107, although the nominative appears to be reconstructed rather than found in any inscriptions. Very straightforward group.
|316||*Brocagnas||Brocagni||#Broccán||Corrected from previous typo <Crocagni>; see also Latin forms in inscriptions 372, 478.|
|71, 166||*Coimagnas||Coimagni||Cóemán, #Cáemán|
|41||*Covagnas||Covagni||Cuan [from Ziegler], #Cúán||The name does not appear in the index to McManus, but occurs in the text; as quoted in CISP, Ziegler (1994) 156 argues that this represents OIr <Cuan>.|
|441||*Curcagnas||Curcagni [Lat.]||Corcán, #Corccán|
|432||*Dovagnas||Dovagni||#Dubán||Previously listed as <Dobagni> but that was from the Latin inscription.|
|60, 160, 258||*Mailagnas||Mailagni||#Máelán|
|135||*Minnaccagnas||Minnaccanni||[gen.] Mincháinn, presumably nom. Minchán|
|75||*Rodagnas||Rodagni||Ródán > #Rúadán|
2. Masculine o-stems with adjectival suffix. Suffix is discussed and identified as o-stem in McManus p.108, but no nominative form is given there. Nominative <-acas> is postulated by me in parallel.
|89||*Biracas||Biraci||#Berach, #Berrach||The inscription actually reads <BIRACO> but this is generally interpreted as an error for <BIRACI> and I have gone with the "corrected" reading instead; OC&M lists <Berach> as masculine and <Berrach> as feminine -- it's unclear how significant the spelling difference is.|
|37||*Dovatucas||Dovatuci||#Dubthach||This name also occurs in an á-stem form, q.v.|
|170||*Qenilocas||Qeniloci||#Cellach||This name is not specifically identified as having the adjectival suffix, however whether or not it has this suffix, the genitive form suggests an o-stem anyway.|
3. Other masculine o-stem nouns. The source for the identification of the declensional class is given when available, otherwise the conclusion is based on the form of the genitive.
|3a. Deuterotheme = <-maglas>, gen. <-magli>.|
|425||*Catomaglas||Catomagli [Lat.]||Cathmál, #Cathmáel|
|3b. Deuterotheme = <-genas>, gen. <-geni>|
|126||*Soginas||Sogini||[gen.]Sogain [read: SOGENI]||Presumably OIr nominative <Sogan>.|
|3c. Deuterotheme = <*-valas>, gen. <-vali>. The Gaulish cognate of the deutherotheme is presented as o-stem in Evans p.421 and Schmidt p.284.|
|504||*Cunavalas||Cunavali||#Conall||Previously listed as <Cunovali> from the Latin inscription.|
|375||*Totavalas||Totavali [Lat.]||Túathal, #Tuathal|
|368||*Barrivendas||Barrivendi||#Barr(fh)ind||The deutherotheme is an o/á-stem adjective in OIr, see DIL <finn>; Schmidt p.296 shows the cognate Gaulish element behaving as o-stem.|
|154, 162||*Corbbas||Corbbi||#Corb, #Corbb|
|128||*Culidovas||Culidovi||Cúldub||In OIr, the adjective from which the deuterotheme is derived is u-stem (but like all u & i-stems at that period borrows the o-stem genitive form), see Thurneysen p.227, DIL <dub>; I can find no examples of the cognate in Gaulish names.|
|342||*Cunacennas||Cunacenni||Conchenn||The deutherotheme is an o-stem neuter noun in OIr, DIL <cenn>.|
|3||*Cunalegas||Cunalegi||Conlang?||This may be the same name as the late á-stem genitive <Cunalegea> in (275), which would represent nom. <Cunalega>.|
|154||*Cunamaqqas||Cunamaqqi||Conmac, #Conmacc||The deutherotheme is an o-stem noun in OIr, see DIL <mac>.|
|449||*Cunatamas||Cunatami||?||Listed previously as <Cunotami> from the Latin form; as referenced in CISP, several authors identify this with OW <Cunutam>, MW <Condaf>, but do not offer an Irish equivalent.|
|63||*Dovalescas||Dovalesci||Duiblesc||If the deutherotheme is identified with OIr <lesc> (DIL), it is an o/á-stem adjective in OIr.|
|368||*Vendubaras||Vendubari||Findbarr, #Finnbarr||The deutherotheme is identified as o-stem in McManus p.102.|
|185||*Vlatiamas||Vlatiami||#Flaithem||The deutherotheme (actually a suffix) is identified as o-stem in McManus p.108.|
|3e. Uncertain. McManus gives no clue to what declensional class this name falls in, and the form doesn't help much. The deuterotheme is an adjective that, in Old Irish, takes an o/ á-stem declension, so the best guess is to treat this as a variant of an expected o-stem genitive <Bodibevi> and reconstruct nominative <Bodibevas>.|
|378||*Bodibevas||Bodibeve [Lat.]||Búaidbéo||The Ogham inscription is only partial, reading <BODDIB...>.|
4. yo-stems. Discussed in McManus p.115 where he notes that the genitive ending of these is indistinguishable in the written Ogham form from that of o-stems, but the declensional class can be retrieved from the appearance of a final vowel in Old Irish nominatives. No examples of an Ogham nominative are mentioned. None of my other sources speak directly to the question either, but my best guess would be a nominative <-ias>. The genitive is <-i> representing an underlying <-iyi>, but there wouldn't be a parallel reason for the stem <i> to disappear in the nominative.
|Specifically identified in McManus as yo-stem:|
5. á-stems. See McManus p.115, where he discusses how this class appears to have borrowed forms frim the ya-stem declension. As a nominal declension, this class is normally feminine, with <-á> in the nominative and <-ias> in the genitive, however there is no doubt that these are the names of male individuals. (There are even clearer examples of Gaulish men's names in the á stem declension, see Lambert "La Langue Galois" p.54.) Male names in this class tended to shift to the o-stem declension, witness the two examples here which apparently occur in both forms. To the best of my knowledge, the length of the <a> in the nominative ending would not affect the Ogham spelling.
|275||*Cunalega||Cunalegea||Conlang?||This may be the same name as o-stem <*Cunalegas>.|
|431||*Dovatuca||D[o]v[a]tuceas||#Dubthach||This name also occurs in an o-stem form, q.v.|
6. i-stems and u-stems. McManus (p.116) indicates that these behave essentially identically in the genitive. The earliest forms have <-os> in the genitive (the u-stem form generalized to both declensions) which later becomes <-o>, and presumably the <-u> is a variant on this stage. (He notes that Old Irish forms begin with the <-o> and later become <-a>.) Again, no nominative forms are available. Lambert (p.59) gives some Gaulish examples of i-stem nominatives in <-is>, and Gallo-Greek examples of u-stem nominatives with <-omicron-upsilon-sigma> which corresponds to <-us> in Roman letters. When u-stem and i-stem elements occur as protothemes in Ogham inscriptions, the composition vowel appears as <u> and <i> respectively (although u stems may also appear with <o>). All this leads me to suggest <-us> and <-is> as plausible for the Ogham nominatives, unless better arguments are offered. MacBain p.xxxiv drew the same conclusion.
|6a. u-stem, deuterotheme <-catus>, gen. <-catos>, <-cato>|
|500||*Ambicatus||Ambicatos||#Imchad||The form <Ambicatos> was previously mis-identified as nominative.|
|19||*Ivacattus||Ivacattos||Éochad||The form <Ivacattos> was previously mis-identified as nominative.|
|353||*Trenaccatus||Trenaccato||Trenchad?||My guess; the Ogham actually reads <TRENACCATLO> and various authors, as reported in CISP, venture interpretations of this form, rather than assuming an error for <TRENACCATO> as McManus does, relying on comparison with the Latin form <TRENACATUS>.|
|6b. u-stem, deuterotheme <-gus(s)us>, gen. <-gus(s)os>, <-gusu>, <-goso>|
|107||*Cunagussus||Cunagussos||Congus||The form <Cunagussos> was previously mis-identified as nominative.|
|428||*Trenagusus||Trenagusu||Trengus||CISP cites Westwood 1879 as referring to a later form <Trengus>.|
|168||*Ircittis||Irccitos||Ercaid, [gen.] Irchada||The form <Ircittos> was previously mis-identified as nominative.|
7-9. Consonantal Stems. See McManus p.116. The genitive ending in the Oghamic forms is first <-as> and in later examples <-a> and finally null. The nominative form depends on the nature of the stem.
7. n-stems. Only two deuterothemes occur here, genitive <-aidonas> (OIr <aed> "fire") and genitive <-conas> (OIr <cú> "dog"). It's interesting to note that in Old Irish, the latter retains the nasal declension while the former loses it and borrows forms willy-nilly from other declensions. (The OIr genitive appears to be borrowed from the u-stems.) As usual, no Ogham nominatives are available. Gaulish examples of the nominative for this declension have <-ú> (Lambert p.61, also Evans "Gaulish Personal Names" p.427) and this is the vowel of the OIr nominative (e.g., <cú>). Thurneysen p.58 mentions Celtic */u:/ from IE */o:/ and notes that it survives in the nominative singular of n-stems. When it doesn't appear, either the final consonant has u-quality, or it's from a source that he thinks resisted u-quality. Overall it seems clear that he's thinking of a nominative singular in <-u>. I agree, and suggest nominative <-u> to go with genitive <-onas> (variant: <-unas>).
|7a. deuterotheme <*-aidu>, gen. <-aidonas> (McManus p.103):|
|504||*Bivaidu||Bivaidonas||#Béoáed||This is a duplicate here, correcting the previous typo <Vibaidonas>.|
|93||*Ercaidu||Ercaidana||Ercáed?||My guess at the OIr equivalent -- O'Brien lists an Ercaid, Erchaid, but McManus derives it from <Ircitt-> and it doesn't match the pattern for the <-aidonas> group.|
|7b. deuterotheme <-cu>, gen. <-conas>, variant <-cunas> (McManus p.102):|
|134||*Assicu||Assicona?||Assiucc||This is missing from the index in McManus, but appears in the text.|
|86||*Cliucu||Cliucoanas||?||This is generally interpreted as a variant form of Cliucunas.|
|191||*Gamicu||Gamicunas||?||CISP indicates that Ziegler has a discussion of the name, but does not reproduce any of it.|
|159, 252||*Glasicu||Glasiconas||Glaisiuc, Glaschu|
|446||*Maglicu||Maglicunas||?||CISP indicates that Nash-Williams 1950 suggests a nominative <Maglicu>, however the only known parallels in later naming are Welsh <Maelgwn> from the oblique stem.|
|126||*Vedacu||Vedacuna[s]||[gen.]Fíadchon, nom. Fíadchú||My guess.|
8. Dental stems. For the dental and velar stems, the stem consonant is theoretically present in the nominative, followed by <s>. Although, as usual, no nominatives are present in the data, McManus (p.108) confirms the forms as nominative <-ts> genitive <-tos>. The question is whether this represents only the underlying phonology or whether it also represents the expected written Ogham form. I'm thinking, for example, of Latin dental-stems, where sound-rules assimilate the <t> to the nominative <s> ending and it isn't written. At least as late as Gaulish, there was some distinction in sound made between final <-ts> and final <-s>, with t-stem nominatives ending in a "barred d" (Lambert p.60) which appears to have been used for a variety of non-basic dental sounds, including affricates and fricatives (Evans p.419). So the possibility seems still open for some sort of final cluster in this situation in Ogham. The Old Irish forms don't give us strong evidence, since this declensional group seems to regularly give us duplicate forms, one apparently derived from the Oghamic nominative and one from the non-nominative forms. Thurneysen (p.110, Sect.177) writes that in absolute auslaut all consonant groups containing <s> except <-rs> and <-ls> have been lost, specifically, groups 'such as <-ks>, <-ts>, <-ns>, <-st>, which had presumably fallen together with single <-s> at an earlier date'. He adds that in Ogam inscriptions final <-s> is sometimes preserved and sometimes not. It isn't entirely clear, but I get the impression that he's putting the assimilation to /s/ early enough not to show up in Ogam. This theory requires that all the parallel Latin inscriptions that use final <x> in the nominative are either purely archaic holdovers, or are consistently analogizing to the Latin velar-stem declension. Either is possible -- especially the latter, I would think, but it would be nice if he addressed that specifically. The lack of nominative forms in the Ogham data makes it really difficult to argue the point. So for the moment I'll go with McManus's nominative as representing a written form, simply because it gives a direction to jump.
|244||*Coillabbots||Coillabbotas||Cóelbad, Coílboth, #Cóelub|
|300||*Cunanets||Cunanetas||Conne, [gen.] Connath|
9. Velar-stems. See McManus p.116. Called guttural stems by Thurneysen. These present a similar problem to the dental stems. When names in this group appear in Roman letters at a similar period, we see nominative <-x>, representing <-cs>. (Similarly in Gaulish.) With no Oghamic nominatives, we're left to guess at nominative <-cs> genitive <-cas>. The stem-consonant may be either <c> or <g> in the non-nominative forms, but I have assumed that the latter devoices in the nominative (as appears to be the case in Gaulish examples). Note that, as with the dental-stem names, the Old Irish forms may derive from either the nominative or non-nominative forms.
|9a. Deuterotheme = gen. <-deccas>|
|9b. Deuterotheme = gen. <-veccas>|
|250||*Rittavvecs||Rittavvecas||[gen.]Rethech > Rethach > Ráthach||CISP gives the reading as <Rittuvvecas> compared to McManus' <Rittaveccas>; O'Brien lists both <Rethach> and <Rathach> as nominative forms, with genitive <Rethach> and <Retha>. It's tempting to consider O'Brien's nominative <Rethu>, but I'm not solid enough on the sound changes to have any confidence.|
|9c. Deuterotheme = gen. <-rigas>|
|(xxi)||Cunorix (n) [Lat.]||*Cunorigas||Conri, #Conrí|
Layout, editting, and publishing Arval Benicoeur.