There is no doubt that the version here translated is a very old
one. The language in LU is almost uniformly Middle Irish, not more
than a century earlier than the date of the MS.; thus it shows the
post-thetic _he_, _iat_, etc. as object, the adverb with _co_, the
confusion of _ar_ and _for_, the extension of the _b_-future, etc.
But YBL preserves forms as old as the Glosses:--
(1) The correct use of the infixed relative, e.g. _rombith_, 'with
which he struck.' (LU, _robith_, 58a, 45.)
(2) The infixed accusative pronoun, e.g. _nachndiusced_, 'that he
should not wake him.' (LU, _nach diusced_, 62a, 30.)
(3) _no_ with a secondary tense, e.g. _nolinad_, 'he used to fill.'
(LU, _rolinad_, 60b, 6.)
(4) Very frequently YBL keeps the right aspirated or non-aspirated
consonant, where LU shows a general confusion, etc.
LL has no very archaic forms, though it cultivates a pseudo-archaic
style; and it is unlikely that the Leinster version goes back much
earlier than 1050. The latter part of the LU _Tain_ shows that a
version of the Leinster type was known to the compiler. The style
of this part, with its piling-up of epithets, is that of
eleventh-century narrative, as exemplified in texts like the _Cath
Ruis na Rig_ and the _Cogadh Gaidhil_; long strings of alliterative
epithets, introduced for sound rather than sense, are characteristic
of the period. The descriptions of chariots and horses in the Fer
Diad episode in YBL are similar, and evidently belong to the same
rescension.
The inferences from the facts noted in the foregoing sections may
be stated as follows: A version of the _Tain_ goes back to the
early eighth, or seventh century, and is preserved under the YBL
text; an opinion based on linguistic evidence, but coinciding
with the tradition which ascribes the 'Recovery of the _Tain_' to
Senchan Torpeist, a bard of the later seventh century. This version
continued to be copied down to the eleventh century, gradually
changing as the language changed. Meanwhile, varying accounts of
parts of the story came into existence, and some time in the
eleventh century a new redaction was made, the oldest representative
of which is the LL text. Parts of this were embodied in or added
to the older version; hence the interpolations in LU.