Location and history:
The stone was found by R. Hitchcock (thus according to Brash, OIM, 234 and PRIA 15, 1871, 27) in the ruined church of Kilbonane, a townland located half way between Killarney and Killorglin and about 2 M. west of Beaufort Bridge. According to Ferguson, OI 108 (179.), it was at that time covering a tomb "at the north side of the altar", although being "broken across in the middle"; according to Brash (OIM, 234 / PRIA 15, 1871, 29) who visited the site in Aug. 1869, the vault in question was "immediately under the east window". It was Brash who proposed to the Academy to fix the stone at the wall in order to prevent it from further damage. Afterwards, the stone was visited by Windele who did not publish anything about it, though.
The stone has meanwhile (when? at least before MCIIC was published) been moved to the entrance of the Gap of Dunloe where it has been arranged with the Coolmagort stones {197 ff.} in a small enclosure. It was lying prostrate when the site was visited in 1978 and 1981, but had been erected leaning against a concrete pillar before 1998.
The stone is peculiar in both its edges and the surface being covered with Ogham characters. The ones on the surface are arranged on three (virtual) stem lines that are obviously "independent of those on the two arrises" as Ferguson stated; the same author regarded the inscription as "the longest inscription in legible Ogham letters yet found".
Size according to Brash, OIM 234: 6'3" x 1'4" x 7"
No size is given by Macalister, CIIC.
Published drafts:
- Brash, OIM pl. XXXIII
- Macalister, Epig. 2, 65; CIIC 2, 236.
- Macalister, CIIC I, 236.
1. inscription:
ᚌ...ᚅ...ᚋᚐ...ᚆᚔᚐᚇᚇᚔᚂᚑᚅᚐ
G N MA IADDILONA
Between G and N there is room for one vowel mark, between N and M a "space of seven inches"; between the "A and the score following is also a space of six inches". This can be restituted as:
ᚌᚐᚅᚅᚋᚐᚊᚔᚐᚇᚇᚔᚂᚑᚅᚐ
GANNMAQIADDILONA
The last vowel is restored as I in PRIA. - For Gann cp. the same name appearing on an inscription found at Tinnehally, about six M. from Kilbonane {256}; these are "certainly the initials of the name" (a "pre-historic name" according to OIM, 235). ADDILON is a "historic" person: Adhlan, son of Eighneach, died AD 950 (Ann. 4 Mas.); cp. also Adamair, son of Fearcorb, A.M. 4783; Adamnan, A.D. 684 (ib.).
2. insription:
ᚅᚔᚏ<<<<ᚋᚅ...:ᚊᚐᚌᚅᚔᚓᚄᚄᚔᚉᚑᚅᚔᚇᚇᚐᚂᚐ
NIR MN AGNIESSICONIDDALA
[In PRIA, the second I is misprinted as a T.] - For the writing on the surface, cp. the stones at Kilcoleman {249: Brackhill?}, Lomanagh {227} and Callan mountain {52a}. The fourth and fifth groups of characters bear a "resemblance to the scale known as the Ogham Consoine"; possibly they were written by a "later hand". "I have no doubt that an incised stem-line originally existed".
3. Inscription:
ᚂᚐᚋᚔᚈᚐᚔᚇᚐᚌᚅᚔ
LAMITAIDAGNI
This is also arranged on the surface. Most probably, it was intended as a "continuation" of the second one. It is "perfectly preserved and legible".
4. Inscription (right hand angle):
ᚅᚌᚑᚅᚔ
NGONI
The edge containing this (fragmental) inscription was obviously not accessible to Brash; at least he was not able to find any more letters. - Possibly the inscriptions on the two edges contain a name and a patronym, whereas the ones on the surface may denote "some particulars as to the person commemorated".
B[ON]AGNiMA[Q]IHADDILO/ENA
NIReMNAQAGNIESSICONIDDALA
NGO/UNIMUC/TOI LAMITA/OIDAGNI
"b(o)nagni maqi haddilona niremnaqagni essiconi ddalangoni mucoi lawithaidagni"
The beginning of the inscription is "a little lapidary rebus of b `o' (from) n, which, with the agni following, make up the name of Bonagnus, Bonán, the founder and giver of its name to the original church". All in all, the inscription names "Bonan, son of Adlon, son of Nireman, son of Esscu, son of Lamidan, son of Dangon"; the "intermediate maqi's" are omitted and the inscription ends with the elements of the formula Maqi Mucoi which "as in previous examples, are not part of the pedigree, but something extrinsic and formal". Should these formulas "relate to our Lord"?
[A?]b[.....]agni maqi Addilona Niraumnaqagni Essiconi [?for Essic-goni] Ddalangoni mucoi Lawithaidagni
"Patronymic values must be assumed for -gni to construe such inscriptions as that at Castletimon {47} .. (doubtful ..), Alatto Celibattigni at Whitefield, Co. Kerry {215}, and apparently" in the present inscription. "This formula seems to be translated into Latin in the non-Oghmic inscription at Colbinstown {19}, IVVENE DRVVIDES; where apparently -IDES represents *-gnos." - [Additional remarks in Epig. 2,, 63 ff. (76.):] "The interpretation .. given .. at p.9 of part I .. was based on the copies of previous transcribers. Even a brief inspection of the original shows it to be untenable, and it must be withdrawn. - The stone bears the largest number of Ogham letters on any known monument. But it becomes immediately obvious that this long inscription is really two independent legends, the one cut on the angles of the slab, and the other cut at some subsequent time in two lines on its face. The face-inscription interferes with the angle inscription, and the technique of the scores is quite different; in the face-inscription, the interspaces between the digits of the letters are narrower, the scores are finer, and the strokes of the letters of the m-group are vertical to the line."
"Left angle":
ᚐᚌᚅᚔᚋᚐᚊᚔᚐᚇᚇᚔᚂᚑᚅᚐ
]AGNIMAQIADDILONA
"About a foot in front of the a is a b; but this has been apparently erased by short transverse scores cut across, and is probably an engraver's mistake which has been scratched out. At a space 7 3/4 inches before the a there is a very faint trace like an m, and it is just possible that the name may have been Modagni, also found at Windgap in Waterford {307}. .. The name Addilona is not infrequent in MSS. as Aedhlon, but is generally masculine; here .. it must be feminine."
"Right angle":
ᚅᚐᚌᚓᚅᚋᚒᚉᚑᚁᚔᚇᚅᚆ
iNAGENeMUCOb/?IDAaNoh?/i
"The first word is, I take it, inagene, for an earlier inagenes, the genitive of inagena, `daughter' .. there is plenty of room for the i, and .. one faint notch is visible. It must be noted .. that the stone was always intended to lie prostrate, and that there is no blank butt for standing in the ground. .. The d appears as an l, transferred to the face-inscription, in previously published copies; and the n has not previously been recorded in print. Restoring the vowels, we are left to choose between muco Bidanoh, which is the inscription actually on the stone; mucoi Dani which is an inadmissible emendation; and the possible compromise muco Bidani. The last name would be comparable with the Baethin of the MSp."
"Face" inscription:
ᚅᚔᚏ<«<<ᚋᚅᚔᚇᚐᚌᚅᚔᚓᚄᚄᚔᚉᚑᚅᚔᚇᚇᚐᚂᚐ | ᚐᚋᚔᚈᚐᚔᚇᚐᚌᚅᚔ
NIR ? MNIDAGNIESSICONIDDALA | AMITAIDAGNI
".. doubtful .. is the letter following the second n. Certain chipped and weathered marks above its digits look like overscores, and in consequence it has been read, with the following d, as aq .. That we have really id is, however, beyond question. An initial difficulty meets us in the angled characters following the letter r .. The Book of Ballymote tract .. shows nothing comparable with them. Such angled characters occur singly on the Ballyspellan brooch, and in MS scribbles, as word-separators or as indications of the commencement of lines; other examples are reported from Grange in Waterford, and from Clonmacnoise. Here, however, they must have three letters, respectively of one, one, and two digits. As these cross the line, it is obvious that only two solutions are possible, if we are to regard these as variant forms of ordinary Ogham lettters; these are amo and mag. .. The scribe used these eccentric forms as a means of word-differentiation. If so, Nir must be one word and amo or mag another. There is no such word, apparently, as amo, but there is a word mag, mog, mogu-s, `a servant,' the dative of which occurs on a stone at Drumloghan, in Waterford {272}. .. I .. suggest that the inscription commences Nir mag Mnidagni - `Nir the servant of Mnidán.'" - "Essicon suggests a name Esconn, known to us from the Martyrologies; and the final Taidagni is a feasible name, Taedán; but amitaidagni seems most reasonbly to resolve into amita Idagni, where amita is an intermediate form between proto-Celtic ammanti-s and Middle Irish ammait, a grandmother. .. The score already read h in the alternative Bidanoh above .. may after all be a b, and belong to the face inscription, giving us a quite unintelligible amitbaidagni." - [New reading proposed in Epig. 3, 11 ff.:]
ᚅᚐᚌᚓᚅᚓᚋᚒᚉᚑᚁᚔᚇᚐᚅᚔ
iNAGENEMUCOBIDANi
"Delete .. any suggestions involving the suppression of the b, or the substitution of oh for the final i."
ᚅᚔᚏ<«<<ᚋᚅᚔᚇᚐᚌᚅᚔᚓᚄᚄᚔᚉᚑᚅᚔᚇᚇᚐᚂᚐ | ᚐᚋᚔᚈᚁᚐᚔᚇᚐᚌᚅᚔ
NIR ....MNIDAGNIESSICONIDDALA | AMITBAIDAGNI
"This inscription sounds like an ordinary Ogham legend, and it has in its course certain well-established Ogham letter-sequences, notably -gni. But it is incapable of being analysed into Irish of any period .. We cannot fail to be struck with certain coincidences. Idagni and .. agni, oniddala and Addilona, baidagni and muco-Bidani, all correspond to another in position, and to some extent in sound. The residue, amit and essic .. do not resemble words known in any language likely to be found in an inscribed stone in Kerry. That is to say, any natural language ..". Macalister then mentions the Irish vocabularies Dúil Láithne and Béarlagair na Saor, further Sim, "a distortion of Arabic spoken for secrecy by Egyptian nomads", and Shelta, "a tinker dialect, which consists of Old Irish words pronounced backward or otherwise artificially modified", as well as Volapük, Esperanto, "and other monstruous inventions". If oniddala corresponds to Addilona and -agni to -agni, we should expect essic to equal maqi. This "might be a compound of aes (ess), which is a Bearlagair word for `a man'". - The first name mentioned could have to be restored to Bidagni, and its correspondent on the surface could be read as sBidagni instead of Nidagni. - Lastly, amit could stand for muco, having emerged by an "interchange" of letters starting from maqv; and in the same way, nirmagms could be an anagram of inagenes. The substitution of ns by rs is a feature of Shelta. - Like this we arrive at:
Bidagni maqi Addilona inagene muco Bidani
Bidaggni essic Oniddala nirmagms amit Baidagni
The word for `daughter' is "misplaced" in this arrangement. In order to explain this, Macalister adds a theory about the usage of "wooden loops" which the inscriptions were copied from. In the present case, the four "angled characters" cannot correspond to the "cor fá chasán symbols" "which in the great vellum manuscripts mark the filling up of the blank at the end of one line of writing by matter belonging to another .." (20) because they have to be taken as letters (21). Macalister ends his presentation about the Kilbonane stone with lengthy remarks about the Ogham script having emerged from a "gesture alphabet, like the deaf-mute finger-signs" (thus following D.H. Haigh, JKilkAS, Ser. 2/2, 175.
.........AGNI MAQI ADDILONA
NIR <«<<MNiDAGNIESSICONIDDALA
AMITBAIDAGNI
.........NAGENi/eMUCOB(A)IDAN
As against Macalister's interpretation, McNeill underlines that O.Ir. ingen "daughter" goes back to inigena as represented on the stone at Eglwys Cymmum {362} (INIGENA CUNIGNI AVITTORIGES) and that neither in Ogham nor in ms. tradition, there is any example of ingen being used in connection with a moccu-formula; furthermore, there are no examples at all of matronyms. Therefore the inscriptions on the edges have to be separated. - ADDILONA might represent a name *Aidlenn, but only Saidlenn is known so far. This could be represented here if the lenited initial to be expected after MAQI (*H for S) was not written. Saidlenn is attested several times in the genealogies of the Ui Ferb: BB 157a, 18 ff. [CGH 301 u.] in several graphic appearances (cp. Saigleand BB 146, 3x; Saidleann 159 b9 a.o.: n. 15). From the Onomasticon Goidelicum, we can further compare the family of Mo Chutu, died 637, who must have lived in the same area. The primary form to start with would be a nom. *Saidliu (or *Sadiul, < *Sadilus, *Sadillus: n. 17); the bearer would have lived in the 6th century (a pedigree is given on p. 50). - NAGENi must represent a name with -gen as its final compound member. BIDAN must be the same as BAIDAN because the name reappears like this in the third inscription; cp. the stone at Aghacarrible {139}, where BAIDANN could be read instead of LAIDANN. The corresponding ms. form, Báitán / Báetán, was frequent in the 6th and 7th centuries. The same name appears in the inscription on the surface which reads:
NI Raba aMNe DAGNI (DAGNEA?).
ESSI CONIDD ALA AMIT BAIDAGNI.
"Let it not be so he makes it.
but (?) let him compose it thus `Baidagni.'"
raba = O.Ir. rop. com-la plus "perfective ad" yields .ad-la > .ála; cp. the corresponding abstract comol, comul, comall as given by Pedersen: conid.ála `he ought to construct it, put it together, compose it.' Cp. ZCP 8, 306: Conailla Medb míchura `Medb had contrived injurious contracts'; "read Conála?" Further cp. conlat noted in O'Mulconry's Glossary under the heading étged which might be read as *con.láat (?) `they join, combine, form a group'. - AMIT is `so, thus' in accordance with Thurneysen's Grammar, 846; cp. Amne "thus" = am-ne, from (s)am- < samo-, which is also present in the Ogham name VLATIAMI as a second compound member as well as in sam-il-dánach as a first member. - ESSI is "an adversative conjunction, meaning `but'" in the present context; it belongs to eks like acht `but'. - In NI RABA = ní rop, the first A represents an "unstressed neutral vowel", the second one was "unpronounced" like the I in late MAQI or both A's in CUNACENA. The reading ABAA proposed for the "cipher symbols" is "based on the context". "The use of these symbols here suggests that the author of the inscription, having begun with the ordinary symbols as far as NIR, changed to a cipher, as though to make his criticism legible only for the expert few; that then, finding the cipher symbols to require too much space and too much labour, he reverted to the ordinary symbols." -
McNeill's view was argued against by Bergin, Ériu 11, 1932, 107 ff. according to whom the normal form of the patronym in question was not Saidlenn but Saiglenn in the M.Ir. genealogies as can be shown by a comparison of the ms. tradition (ms. Rawl. 502, 12th cent. as against the Book of Ballymote). Furthermore, the pedigree as established by McNeill cannot be relied upon because it too rests on the abbreviated genealogy as quoted in the BB. Like this, a name Saidliu as a 6th century equivalent of ADDILONA does not exist. The latter must keep unexplained; possibly it is a "compound beginning with Ad-, containing the suffix -ilo which appears in Gaulish names". - McNeill's restoration of the "sentence" appearing on the surface is equally doubtful, because after ní rop we should expect a subjunctive expressing a wish, viz. dagné; this, however, is not an a-subj. so that DAGNEA must be excluded. conid.ála, if a verbal form, would have to contain the marker of a "rel.subject" as well as a "pron.obj."; the "non-rel." form would have been cot-ála. If AMIT `thus' represented samit = Cymr. hefyd `also' we should have to cope with the missing S-; besides, a meaning `also' would hardly be justified in the present context. What remains is BAIDAGNI and the initial NI, "probably the negative".
"Dexter angle":
B[AID]AGNI MAQI ADDILONA
After the B, there is a gap of 1' which "must at one time have contained vowels". Of AGNI "the last letter (is) broken". MAQI ADDILONA can be taken for granted "though Q1-4 are only just visible and the distal ends of the L are slightly broken".
"Sinister angle":
NAGUNI MUCO BAIDANI
Much less than this is readable: NAG N , the gaps measuring 4" each. Comparing the single N with its width of 3,5", we can assume the second missing vowel to have been an I. The same holds true for the first missing vowel "but the appearance of the broken surface does not encourage us to fit so many into it: it is more likely to have been an E, or even an U". - "I cannot press my former restoration inagene." The following words are "evidently MUCO (not MUCOI) BIDANI (or BAIDANI: there is sufficient space for the two vowels between the B and the D). The vowels have practically all disappeared, and C1 is broken. .. The use of MUCO instead of MAQ or MAQI seems surprising .. It discourages us from interpreting the dexter inscription as a later addition, commemorating a son of the owner of the first epitaph."
"Face of the stone":
NIR*** MN[I]DAGNIESSICONIDDALA AMIT BAIDAGNI
"The asterisks represent three characters in the formula .|.. the short strokes being V-like zigzags turned sideways, the long stroke a similar W-like zigzag. These I should be inclined a priori to read as AMO, but there is nothing to indicate what the intention of the lapidary may have been, why he wrote these three pseudo-cryptical letters, and why, having begun thus, he wrote only three. The only reasonable interpretation for this inscription so far offered is that by Prof. MacNeill .."
Reading Gippert (1981 / 1998):
Dexter angle ("up"):
B[AID]AGNI MAQI ADDILONA
ᚁ[ᚐᚔᚇ]ᚐᚌᚅᚔᚋᚐᚊᚔᚐᚇᚇᚔᚂᚑᚅᚐ
ᚁ[ᚐᚐᚐᚐᚐᚐᚆᚆ]ᚐᚋᚋᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐᚋᚐᚆᚆᚆᚆᚆᚐᚐᚐᚐᚐᚐᚆᚆᚆᚆᚐᚐᚐᚐᚐᚁᚁᚐᚐᚁᚁᚁᚁᚁᚐ
Sinister angle ("up"):
N(A)G[E]NI (MUCO BAI)D[A]N[
ᚅ(ᚐ)ᚌ[ᚓ]ᚅᚔ(ᚋᚒᚉᚑᚁᚐᚔ)ᚇ[ᚐ]ᚅ[
ᚁᚁᚁᚁᚁ(ᚐ)ᚋᚋ[ᚐ]ᚐ(ᚐ)ᚐᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐ(ᚋᚐᚐᚐᚆ)[ᚆ](ᚆᚆᚐᚐᚁᚐᚐᚐᚐᚐᚐ)ᚆᚆ[ᚐ]ᚁᚁᚁᚁᚁ[
The angle has suffered very much, probably not only from natural erosion (and from the fracture) but also from the removal of the stone and its replacement at Coolmagort. The vowel notches at the end as assumed by others cannot be ascertained in any way. The M is broken in the middle (i.e., at the angle); on the B-surface, there seems to be a second stroke, slanted in the opposite direction so that a ᚕ-forfed might be suggested; there is no continuation of the stroke on the H-surface, however. Another -stroke seems to be recognizable near the end of the angle, but this is rather likely to be a scratch added by a visitor.
Surface: Long axis ("up"):
NIR<«<<M(N)[ID]AGNIESSICONIDDAL[
ᚅᚔᚏ<«<<ᚋᚅ[ᚔ]ᚇᚐᚌᚅᚔᚓᚄᚄᚔᚉᚑᚅᚔᚇᚇᚐᚂ[
ᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐᚋᚋᚋᚋᚋ<«<<ᚋᚁᚁᚁᚁ(ᚁ)[ᚐ](ᚐ)ᚐᚐ[ᚐᚆ]ᚆᚐᚋᚋᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐᚐᚐᚐᚐᚁᚁᚁᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐᚆᚆᚆᚆᚐᚐᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐᚆᚆᚆᚆᚐᚁᚁ[
The amount of character elements that have been lost at the fracture can hardly be esteemed; it seems that the space was enlarged when it was cemented. The characters following after the fracture are badly damaged; a sequence ID is just one possible reading, even a C or Q could be assumed. The final A as read by earlier investigators cannot be ascertained at all, but the end of the stone may have been broken later.
Short axis ("up"):
AMIT BAIDAGNI
ᚐᚋᚔᚈᚁᚐᚔᚇᚐᚌᚅᚔ
ᚐᚋᚐᚐᚐᚐᚐᚆᚆᚆᚁᚐᚐᚐᚐᚐᚐᚆᚆᚐᚋᚋᚁᚁᚁᚁᚁᚐᚐᚐᚐᚐ
It is not clear at all whether this is an independent inscription or a continuation of the one on the long axis.
Additional literature:
- J.Kilk.Arch.Soc. 2, ser. 2, 175: D.H. Haigh.
Dexter angle ("up"):
Sinister angle ("up"):
Surface: Long axis ("up"):