¹ The finest of these studies of Old English case-syntax is that of J. E. Wülfing, Syntax Alfreds des grossen (Bonn, 1894); somewhat later is the monograph of H. Winkler, Germanische Casussyntax (Berlin, 1896). There are many other works on individual problems. The most complete bibliography is included in Morgan Callaway's “Studies in the Syntax of the Lindisfame Gospels,” Hesperia no. 5 (Baltimore, 1918). Note also G. K. Anderson, A Study of Case-Syntax in some Old Northumbrian Texts, summarized in Summaries of Harvard University Ph., D. Theses for 1925 (Cambridge, Mass., 1928).

² Of the two glosses, the Lindisfarne gloss is the more reliable for the student of Old Northumbrian. The Rushworth gloss is mixed—that portion of the gospels from Matthew i, 1 to Mark ii, 15 (Rushworth i), the work of the scribe Farman, is Mercian with Northumbrian admixture; from gehlionade in Mark ii, 15 to the end of John, the work of the scribe Owun, is Northumbrian with Mercian admixture. John xviii 1–3, however, is the work of Farman. R. J. Menner's recent article in Anglia for January, 1934 (lviii 1 ff), not only affords a fine review of the whole vexed matter, but establishes clearly the individuality of the glossator Farman. The most cautious statement, in a mass of rather fine-spun inferences, is that of K. Bülbring, Altenglisches Elementarbuch (Heidelberg, 1902), §25: “Spätmercisch (aus der zweiten Hälfte des 10. Jahrhunderts)—jedoch mit nordhumbrischen und sächsischen Formen gemischt—ist die Interlinear-glosse zum Matthäus-Evangelium im Rushworth MS” [Rushworth i]. In §24 he affirms the Northumbrian nature of Rushworth ii. But Owun's portion (Rushworth ii) needs more investigation. Other scholars have detected even West Saxon peculiarities in Rushworth i (cf. Menner's bibliography in his article mentioned above), but such details do not come into the province of my present study.

³ The Holy Gospels: Anglo-Saxon and Northumbrian Versions (Cambridge, 1871–1887).

⁴ In the afore-mentioned Hesperia, no. 5 (Baltimore, 1918).—I might add here that the only subsequent contributions to the field are the thesis of Anderson (already noted), a collation by Lindelof of the Durham Ritual Book in MLR for June, 1923, and a complete discussion of Caedmon's Hymn by M. G. Frampton in Mod. Phil. for August, 1924, and Menner's article in Anglia, lviii, mentioned in note 2 above.

⁵ in EETS 83.

⁶ pro rather than coram.

⁷ All references to the Ritual Book and to the Introduction to the gospels are to page and line; for the gospels proper, to chapter and verse; for the archæological fragments, to line.

⁸ As in note 7.

⁹ That is, accepting the statement of Sievers' Grammar of Old English, tr. of A. S. Cook (Boston, 1899), §237, n. 2—and of J. Wright, Old English Grammar (Oxford Univ. Press, 1908), §334, note—that the instrumental ending of the masculine a-stems (I.E. o-stems) was originally -i. Accepting also the fact that most masculine proper names in Old English, when inflected at all, are usually declined as a- stems.

¹⁰ Wülfing's Syntax Alfreds des Grossen, ii, 658, §1067—(“von Hauptwörtern gebildeten Präpositionen”).

¹¹ Other examples of this same idiom: Int. Luke 10, 16; Int. Luke 10, 18; Int. Luke 11, 6; Int. John 6, 16; Int. John 3, 8; Int. John 4, 11; Ritual 17, 7; Ritual 18, 1; Ritual 94, 5; Ritual 94, 8.

¹² Note the treatment, customary in these glosses, of the Latin ablative absolute, and incidentally the treatment of the Latin deponent verb in the example immediately preceding from Int. Luke 4, 6. Callaway in Hesperia, no. 5, devotes considerable time to a discussion of absolute participial constructions in the Lindisfarne texts.

¹³ This affords us a good example of the slavish imitation of Latin word order on the part of the Northumbrian scribe.

¹⁴ There are four other examples of this construction: Int. Luke 3, 19; Luke 7, 45; 13, 7; 24, 21.

¹⁵ Wülfing has an excellent discussion of both prepositions in his aforementioned work, vol. ii.

¹⁶ Further examples in Int. John 3, 6 and Luke 14, 34. Manner, Attendant Circumstance or even a Comitative idea are other possible interpretations, to be sure. Winkler, in his Germanische Casussyntax, actually remarks (p. 453): “Der Instrumental mit Präpositionen kommt so gut wie gar nicht in Betracht; ich kenne ihn fast nur bei dem comitativ-instrumentalen mid.” He notes another case with for, but the Old Northumbrian seems to fall outside the confines of his rather flat statement.

¹⁷ Both Sievers and Wright (see note 9 on p. 948) in determining the -i to be the instrumental ending for a -stem masculine nouns, speak of this ending as an original locative (I. E. -ei). An interesting fact that this is an actual locative construction!

¹⁸ Of course, rod (and dun below) are generally feminine o-stems in Germanic, and it is natural so to regard them here, in which case they could be regarded as variants in -i of the normal dative feminine -e. But cf. Sievers Grammar of Old English (tr. A. S. Cook), §251, note: “In Northumbrian many of these feminines are also employed as neuters and masculines, and then conform to the inflections of these genders (§236, note)”, and again (§252, note 1) “… sporadically does the dative-instrumental [sic] exhibit -i, as in rodi … in which it is probably borrowed from the o-declension.” This may be rationalizing. Rushworth's on dune (Matt. 26, 30) makes me feel that duni may be an ignorant misspelling, and that both examples 1 and 2 on page 951 are inadmissible, yet a suspicion of their validity in these locative constructions persists.

¹⁹ Since this is a normally accusative construction, this may be even more surely a careless misspelling on the part of the scribe. By the way, the Old Saxon illustrates the locative with instrumental construction by an thiu; there is the well-known case in the Heliand of an with the instrumental to indicate Limit of Motion (an thiu holmcli$bSu, Hel. 4734). Or so I read it. Both these Old Saxon idioms parallel closely these Old English idioms under discussion.

²⁰ Again, note the Old Saxon an thiu phrase to introduce purpose-clauses. Cf. Holthausen, Altsächsisches Elementarbuch (Heidelberg, 1900), §539 note.

²¹ In addition to Ritual 10, 1 cited in the text—Mark 14, 2; John 7, 37; Ritual 9, 20; 173, 8; 11, 13; 20, 15; 115, 2.

²² The resemblance of this phrase to the Old Saxon an dag is obvious. Holthausen, Allsächsisches Elementarbuch (Heidelberg, 1900), §265, note 4), mentions this as an old locative (Urgermanisch-i; I.E.-ei). It would help us much, however, if the O.S. *dagi were not altered—according to Holthausen, by analogy—to its uninflected form.—An examination of A. Noreen's Altisländische u. Altnorwegische Grammatik (Halle, 1903) reveals (§138) the fact that an e/i interchange in unaccented syllables in Old Norse is not uncommon, although not, as it happens, in such specific cases as in the dative of the nouns dagr or dúnn. The possibility of Scandinavian influence upon Northumbrian cannot be ignored, but there is little direct evidence of any such influence here. For that matter, the same alternation of e and i in final unaccented syllables is in Old English throughout (cf. Sievers Angelsächsische Grammatik, §45, 2 and particularly note 1). The preservation of this tendency toward the interchange of final -e and -i in Northern English dialects is amply illustrated in J. Wright's English Dialect Dictionary (Oxford, 1905) vi, Grammar Supplement, §229. What I am calling here instrumentals in -i (dægi, duni, rodi) may be northern variations of datives in -e (dæge, dune, rode). On the other hand, perhaps not!

²³ O. Behaghel—Die Syntax des Heliand (Vienna, 1897).

²⁴ May we not consider the instrumental dægi, with such a form, such a background, and in such a construction, as a true locative? The Auslautsgesetze will not be unduly violated, I believe, and the whole explanation clarified. So Brugmann, Comparative Grammar of the Indo-Germanic Languages, III, §263.

²⁵ Some excellent examples of this disintegration are brought forward in W. F. Bryan's “The Midland Present Plural Indicative Ending -e(n),” Mod. Phil., xviii, 124.