and Luick, we come across terrible verse structures in which arses and theses are distributed in a most irregular manner; we meet verses, which have enormous theses of two, three and even four syllables and arses undivided by theses. Anacrusis of three or four syllables is not rare. Words otherwise strongly stressed must often come in the thesis, unless one assumes with Luick a 'rhythmical subsidiary stress' in these cases."

NOTE. When Luick (Anglia Beibl. 12, 40 ff.), quoting many examples, shows that in NE. in anapaestic verses words, otherwise strongly stressed, may come in the thesis, he forgets that this anapaestic rhythm is something quite modern. What is more important, however, he forgets that in alliterative verse it is the alliterating sound which is to mark the rhythm of the verse and to make the most strongly stressed words prominent, so that it would be absurd to put an alliterating noun in the thesis.

§ 158. b) Rosenthal, Trautmann etc., Kuhnke, Bunzen.

Rosenthal (Die alliterierende englische Langzeile im XIV. Jahrhundert, Halle 1877 and Anglia 1, 414-459) was the first to apply Lachmann's four-beat theory to the ME. alliterative verse and to assume "for the half-verse two haupthebungen and two nebenhebungen." And, as we shall find below (§ 159), four (strong or weak) beats can be easily seen in the first half-lines of ME. alliterative verse. In the second half-lines, which are shorter, this scansion would be possible only if

we were to give the disyllabic words with a long root-syllable (×), which are generally found at the end of the verse, two beats. But in the regular short rimed couplet of Havelok and the later romances (§ 122 f.) and in the second half-line of the septenary in Passion and Josephslied (§ 132 f.) the feminine ending (-x) has one beat only. We can, therefore, give only one beat to such words in the alliterative verse which arose in the middle of the fourteenth century.

Thus Trautmann seems to be right, when he assumes four beats for the first half-line and three for the second: "The ME. alliterative line is one of seven bars and has four in the first part and three in the second", (Zur Kenntnis der mittelenglischen Stabzeile, Anglia 18, 83--100). Here (p. 85 ff.) Trautmann has shown the possibility of scanning the long lines with seven beats by giving some short passages from various poems, whilst his pupils, Mennicken (Versbau und Sprache in Huchowns Morte Arthure, Bonner Beitr. 5, 33—144), Steffens (Versbau und Sprache des mittelenglischen stabreimenden Gedichtes 'The Wars of Alexander', B. Btr. 9, 1-104) and J. Fischer (Die stabende Langzeile in den Werken des Gawaindichters, B. Btr. 11, 1-64) have examined in detail the verse structure of some longer alliterative poems.

Trautmann surely is wrong when he looks on all the beats of the verse as of equal strength, just as he does in the case of OE. verse (§ 57 ff.).

For some of the beats, which do not alliterate, are weaker than others, and do not in themselves form a bar, but form a foot of two members in conjunction with a preceding stronger beat.

This mistake is avoided by Kuhnke, who in other respects holds Trautmann's point of view. In his examination of Die alliterierende Langzeile in der mittelenglischen Romanze Sir Gawayn and the Green Knight (Studien zum germ. Alliterationsvers 4, Berlin 1900) Kuhnke arranges the verses of Sir Gawayn according to Sievers' types. scanning with four beats, however, so that the different strength of the four or three beats is clearly shown. His results are that in the first half-line the commonest types are D2, B and A2 b, which end with an independent hebung or a strong nebenhebung, and in the second half-line A, C and D1 only occur, which have 'feminine' endings. The last member of these types is no longer to be counted as a beat. The individual types, however, cannot be so easily recognized and distinguished as in the OE. period.

Now since both the ME. alliterative verse and the septenary have four beats in the first halfline and three in the second, Kuhnke sees in the early ME. septenary the long sought link between the OE. and ME. alliterative verse and holds the latter as a "septenary without rime and therefore adorned with alliteration".

Asmus Bunzen in his Ein Beitrag zur Kritik

der Wakefielder Mysterien. (Kiel 1903) opposes Trautmann's scansion of the ME. alliterative verse and the application of the two-beat theory more decidedly than Kuhnke, whose work he evidently has not read. He says (p. 26): "In spite of all attempts to prove that the ME. half-line has two beats, it has not two beats, but always three or four beats. These beats are, however, not of equal value, as Trautmann Trautmann assumes, but we must distinguish between haupthebungen and nebenhebungen, i.e. the structure of the ME. verse is dipodic.... If Trautmann makes the mistake of not distinguishing between haupthebungen and nebenhebungen, giving the latter the same value as the haupthebungen, Luick on the other hand makes the mistake of neglecting the nebenhebungen and of giving his whole attention to the haupthebungen".

Bunzen is certainly right when he distinguishes between strong and weak beats in ME. alliterative verse, but we must notice that even in the early ME. rimed verse, e.g. in Lagamon's Brut etc., the old types have become very loose, that the first member of B and C verses, the second member of A verses, the third member of B verses, have become more independent than they were in the OE. period, whenever these are independent words and not inflexional syllables. We cannot indeed scan with Trautmann: Máistúr in mágesté, mákér of álle, but with Bunzen: Máistùr in mágestè, mákèr of álle, but not with Bunzen: Ich hèrde

mén upo molde máke much món, but rather with Trautmann: Ich hérde mén úpo molde máke múch món, so too as téllen óure bókes, þére fást byside etc. The combination of two beats in ME. alliterative verse is limited to those cases, in which the second beat falls on an inflexional or derivative syllable, e.g. Hit bi|týddè sum | tyme in þe | térmès of Jude (Patience 61), whilst the beats which fall on pronouns, prepositions etc. are weaker than those on nouns, adjectives and verbs, but are independent as in the rimed verse of equal bars, e.g. þat ín þat pláce át þe póynt i pút ín þi hérte (Patience 68).

NOTE. When Bunzen says (p. 26): "The structure of the ME. verse is dipodic, but that of the Anglo-Saxon verse was not dipodic", I must refer him to § 63 ff. where the dipodic and tripodic structure of the OE. alliterative verse is shown in detail.

$ 159. The Rhythmical Structure of the Middle English Alliterative Verse.

From what has been said we may draw the following conclusion: The ME. alliterative verse has no direct connection with the OE. alliterative verse, but rather with the early ME. septenary, not with the septenary with equal bars of Poema Morale or Orrmulum, but with the verse of God Ureisun (§ 130), Passion (§ 132), Josephslied (§ 133) and of the rimed chronicle of Robert of Gloucester (§ 153), which is freer in its structure and was influenced by the OE. alliterative verse. Like the