The language of the Dream of the Rood presents few problems for the student in its spelling and grammatical forms since it is written in very regular West Saxon and thus is close to the forms of the paradigms given in the standard grammar books. There is also considerable consistency about the way individual words or word-roots are spelled.The following description will pick out the most obvious of the features which characterise it as West Saxon as well as those places where it does show evidence of dialectal or late linguistic features.
Old English sound changes are not explained here. They have been given their traditional names and can be looked up in standard Old English grammars e.g. Campbell, Hogg.
Characteristic West Saxon spellings and/or sound change results:
The text regularly spells the result of OE "a" before a "breaking group" of /-lC/ or /-rC/ as "ea" e.g. "eall" (6), "gealga" (10), "healfe" (20), "wealdend" (67), "bearn" (83). There are no examples of the Anglian /-alC/ or Northumbrian /-arC/ forms.
It also shows /ea/ spellings in forms like "geseah" (14, 21 etc) and "meahte" (18) rather than the Anglian forms with "smoothing".
Characteristic West Saxon "back-mutation" is found e.g. in "heofon", "weorod", and combinative back-mutation in e.g. "worulde" (133). Curiously "weruda" (51) doesn't show either combinative back-mutation (which would be unusual) or ordinary back-mutation which one would expect and which is found in this word elsewhere in the text; "werede" (124) may also lack back-mutation or may be a form with an alternative suffix that did not contain a back vowel. In such a regular text it is quite unusual to find a word with three different spellings. "feala" occurs 3x (50, 125, 135) although it is unusual to find back-mutation of /e/ by /a/ in West Saxon.
The majority of the spellings of the inflections in this poem represent the unstressed vowels by i,e,a,o,u as they are used in standard grammars. However, there is some evidence of the later West Saxon spelling of terminations which can use some of these vowels interchangeably. While this may in part indicate reduction in distinctiveness of unstressed vowels, the variation seems also to be limited to certain inflections e.g. -en (the plural subjunctive) may vary in spelling, -en (the past participle of a strong verb) does not seem to do so.
Preservation of earlier /u/ - except in the /-um/ and /-u/ terminations (e.g. wynnum, sunu) West Saxon tends to represent earlier unstressed / u / by /-o-/. Occasionally in this text -u- remains e.g. "worulde", "weruda" "ærur"
Places where an original back vowel (o in standard spelling) is represented by (e):
wunedon (3, 155), forwunded (14), bysmeredon (48), heofenes (64), heofenum (85, 134), heofenas (103) (cf. heofonum 140, 154). With the exception of /forwunded/ all these concern an unstressed vowel which is followed by a further unstressed syllable and the use of /e/ here is very common in Old English spelling. It is quite surprising how frequently the back vowel appears in the spelling in this text. The /eo/ spelling in the stressed syllable of /heofen-/ is assumed to be the result of the sound change "back-mutation" which would require an original back vowel in the /-on-/ syllable. werede (124) (cf. weorode 69, 152) may be similar but see above.
geniwad (148) ; the /-ad/ in place of more regular /-od/ might be simply late variation of spelling but non-West Saxon varieties of Old English tend to have -a- as the medial vowel of weak II verbs rather than -o- which is regular in West Saxon. This spelling, therefore, might reflect a dialect form from an earlier manuscript version.
genaman (30), þurhdrifan (46), cwùman (57), þolodan (149). In every case the regular /-on/ is the pl.past indicative inflection. Again this is a very common later spelling variation and it is perhaps surprising how infrequently it occurs in the poem. The majority of forms have /-on/ while the past participle has /-en/, the infinitive /-an/ and the plural subjunctive /-en/ (onginnen 116).
This seems restricted to adjectives and demonstratives and it is difficult to be certain in some cases whether one has a poetic feature of syntax or late spelling confusion: "of beorhtan stane" (66), "on deopan seaþe" (75) both could be explained as poetic use of the weak adjective without a preceding demonstrative; however, "mid deorcan næglum" (46) because the plural is needed must have /-an/ for /-um/. The variation in the spelling of "on þyssum lænum life"(109) and "on þysson lænan life" 138) would suggest confusion of /-um/ for /-an/ in "lænum" in the former and /-on/ for /-um/ in "þysson" in the latter. This tendency to confuse /-n/ and /-m/ possibly accounts for "þan" (121) in place of "þam". The existence of the instrumental "þan/þon" may have helped the spelling confusion.
In the majority of instances this is regular but "manegum" (99) seems to show the tendency of unstressed vowels to reduce in distinctiveness. "hefian"(61), "manigeo" (151) show the spelling variation that can be used to represent this syllable with /-i/ and -ige/ repectively. "frineð (112) shows that /i/ can equal /ig/ in stressed syllables as well.
Places where scansion suggests that the written form does not match the spoken : 120b æghwylc sawl - this appears only to have 3 syllables rather than the required 4. When said sawl must be given the full form of its unstressed syllable, sawol/sawel. 121b wunian þenceð , on the other hand, needs to uncontracted form of þenceð as written.
The number of present indicative verb forms is limited in the poem but whereas "cwyð" (111) shows the i-mutated and contracted form of 3sg.pres.indic. of "cweðan" characteristic of West Saxon, "frineð" (112) and "þenceð" (121) do not show contraction and "bereð" (118) shows neither i-mutation nor contraction. Such reformation is more characteristic of Anglian texts. Since frineð is unstressed and "bereð" as two resolved short syllables is the equivalent of "birð" as one long one, the scansion does not tell us whether these are scribal or possibly original. As mentioned in the previous point, "þenceð" requires two syllables since "wunian" must resolve the first two syllables.
Individual points of grammar or syntax which raise problems are indicated in the notes.
Extracts taken from Ann Squires essay "The Language of the Poem"
June 1994