INTRODUCTION
13
somewhat earlier than Wurzburg: we suggest that it was composed
about the middle of the 7th century. A terminus post quern is
provided by the reference in §§31-2 to Congal Caech, whose death
is recorded in 637 A.D. This passage can hardly be dismissed as a
later interpolation, because it contains the archaic connective -eh.
For a discussion of Congal Caech’ s career, see notes to §31.
1. -eh. The archaic connective -eh occurs in §§32 ba-ch, 44
ro-ch-lamethar , 45 na-ch laimethar, 54 ro-ch-fintar ( Grammar §880).
In his article on this connective in Celtica v 77-94 Dr. Binchy has
concluded that it became obsolete before about 700 A.D. except in
the special combinations noch is and seek is.
2. Unstressed There are some possible cases of the
preservation of archaic unstressed e after non-palatal consonants,
cf. ached 'Thes. ii 278.2, toceth 47.26 for Classical (i.e. 8th-9th century)
O. Ir. achad , tocad.
We follow Thurneysen in taking the verb do-etet to be from
*to-ad-uss-tet (see notes to §37) which in later Irish became do-etat
(do-autat). If this identification is correct, archaic unstressed e is
preserved in the 3 sg. pres. ind. do-etet (occurring at §§37, 41, 43), in
the passive do-etegar (§42, spelt thus in both A and B, but updated
to do-etagar in O’Dav.), and in the plural past participle tetechtai
(§§36, 39, 41, 54). In the case of §§44, 45 in-otat [* in-oss-tet ], on
the other hand, archaic e has become a. It is probable that the
archaic e of do-etet survived only because successive scribes — familiar
with con-e(i)tet — took it to be do-e(i)tet. Marks of length are almost
always omitted in legal manuscripts.
Another possible case of archaic unstressed e is in §12 where the
B commentary (perhaps retaining the spelling of the original text)
has cro ime thor- ‘ an enclosure about its fruit ’ against cro ima torad
A, cro imma torud C. With ime of B one can compare the spelling
of the 3 sg. possessive pronoun as -e in ine (ince) ‘ in his ’, faire ‘ on
his ’ in the archaic Cambray Homily (Thes. ii 245.36-7, 246.8,
246.22-3); see Grammar §§441. However, the issue is clouded
because im(m)e- occurs beside im(m)a- in Classical Old Irish in words
such as immenetar Ml. 26b20, immenetor Sg. 28a 10 beside immanetar
Wb. 13c26, 27b21,etc. and immelei Wb. 10a6, immelle Ml. 68d9 beside
immalle Wb. 9b 1 7, Ml. 61b23, etc. Similarly, both im(m)a- and im(m)eoccur as relative forms of the preverb imm in Class. O. Ir. (Grammar
§493.4). So, while ime B may be the original spelling, it could also
be an orthographic variant of ima introduced by a later scribe.
14
INTRODUCTION
3. -th. Archaic -th (Grammar §130.2) is preserved in §§4
deolaith A, 10 alith B, ailit (= ailith] A, 24 uathath A, 40 tintuth A.
Much more commonly, however, it has become -d.
4. to-. The only case of archaic to- (Grammar §855 B) in the
text is §33 tobert; elsewhere this preverb appears as do-. The gloss
./. toich aslaiet (in the hand of the third A glossator) on §39 docoislet
suggests that the text originally had tochoislet here. Toich is the
regular ‘ etymological ’ gloss on toch -, e.g. §27 i tochumlat which is
glossed i comimluaidid . . . gu toich , cf. Celtica v 83. This glossator
may have had access to another copy of BB in which some of the
original spellings were better preserved than in A — see p. 6. For
the preverb di- see notes to §14 di-renar.
5. -tubart. We take the form §32 conid-tubart (comidubart A)
as an indication that the text is earlier than the Wb. Glosses. Already
in Wb. ta- is three times as common as to- (tu-) in prototonic position
in this cpd. (see DIL s. vv. do-beir and tabart).
Prototonic 3 sg. preterites with tu- are found in two other early
sources. The spelling contubart is attested in the Additamenta to
the Lives of St. Patrick (Thes. ii 242.20 — Patric. Texts 178.11).
The spelling condetubert (with archaic unstressed e in final syllable)
is attested in the Laud Genealogies (ZCP viii 308.34). For a suggested
explanation of the development of to- (tu-) to ta- in this verb, see
Grammar §82.
In the Notes we discuss other linguistic features (some possibly
archaic) which are not fully covered in Thurneysen’s Grammar.
They include the omission of the possessive in bes tech torad (§20),
fer batar beich (§33), the collective suffix -id in lestrae (§30), the use
of nad- rather than nach- before an infixed pronoun in nadidlamethar (§34), the intransitive use of suidigithir (§37), and the
occurrence of pres, indie, pass, -fogbaither (§§46, 48) beside -fogabar
(§47). For the Latin loan-words in the text see Introd. p. 41.
THE VALUE OF THE GLOSSES AND COMMENTARY
The glosses and commentary range in date from about the 9th
to the 16th century. In the case of B, there is no clear division
between gloss and commentary, but in A the glosses are written above
the lines of text whereas the commentary is written in the margins.
Though the first A glossator (Lucas) is thought to have worked in
INTRODUCTION
15
Ihc first half of the 14th century (see p. 4) it is likely that most of
his glosses and commentary are earlier, probably transcribed from
the same copy of BB as was used by the main scribe. Many features
of Lucas’s glosses show them to have been copied. Thus, he does not
always place his glosses above the word or phrase in the text to
which they refer. In two cases he provides glosses for words which
have been omitted by the main scribe. In §8 (20b20) he gives a gloss
(./'. roindsaigidh) corresponding to the word rosaig, omitted by the
main scribe. This word was later added in the margin by the third
glossator, Cairbre. Likewise in §43 (24b3) his gloss trian don fir bis
ina coimideact indicates that the original had do fiur , though this is
absent in the main scribe’s text. It was added later, also by Cairbre, to
the left of Lucas’s gloss.
Lucas’s glosses frequently contain the correct reading where the
main scribe (or a predecessor) has made a mistake. For example in
§7 the text of A has rind , but the gloss has the correct reading ind ,
as in B. In §31 the text has cetaraced , but an original -rucad is
indicated by the gloss rugadh, and confirmed by E and F. In §34
the text has a acht , but the context requires a ucht , as in the gloss
(again confirmed by E and F). Other cases where Lucas’s glosses are
more accurate than the A text include §§17 toraid { torad text,
toraidh gloss), 29 sciith {said text, saith gloss), 30 uili (aile text, idle
gloss), lestar ( lestra text, leastar gloss), 43 oci n-elat {oatelat text, o
nelan siad gloss).
Many of Lucas’s glosses are the conventional glosses of the
law-schools, and are found in other legal MSS. Thus, his glosses on
mud, ecmacht and dirainn in §§48-9 are also used in CIH 59. 7-8 and
2066.40-1 (Heptads). His glosses on fo theol 7 taidi (54c>d) are
similar to the glosses on i teol 7 taithe in Airaiccept na Riar {CIH
2336.14).
Glossators II- V worked between the 14th and 16th centuries.
In some cases a word or phrase of the text has been glossed a number
of times. For example, §52 direnar landtre is glossed icaidh se
laneneclann by Lucas. Further glosses were added in turn by
glossators II, III and IV. Later glossators generally expand the
preceding gloss (e.g. 2b, 4a, 10b, 14d, 15a), but sometimes contradict
it (e.g. 34b). An additional gloss usually follows the syntax of the
preceding gloss, but sometimes the connection is broken e.g. 50d
./. ime uile I (na cesa III) ‘ i.e. in the matter of all of it, the hives ’ cf.
45 d, 48b.