^{Note 1 in page 453} Julius Zupitza, ed. Ælfrics Grammalik und Glossar, Erste Abteilung: Text und Varianten, Sammlung englischer Denk-mâler, Vol. I (Berlin, 1880). References in the text are to this edition, by page and line: I have in citing certain verbal forms altered Zupitza's spelling -j- to -i-, and omitted his marks of quantity throughout. The complete Grammar has been published once before: William Somner, Diclionarium Saxonico-Latino-Anglicum (Oxford, 1659). Fragments of it have been printed: Thomas Philipps, A Fragment of Ælfric's Anglo-Saxon Grammar (London, 1838); A. Birlinger, Bruch-stûck aus Älfrics Angelsachsischer Grammatik, in Pfeiffer's Germania, xv. The Grammar, in whole or in part, is extant in 15 known manuscripts: see Zupitza's Vorwort.

^{Note 2 in page 453} Aelius Donatus (4th cent.) wrote a grammatical treatise in 2 sections: the first is known under the title Donati De Parlibus Orationis Ars Minor, or more briefly the Ars Minor; the second, much longer, is the Ars Maior or Donali Grammalici Urbis Romae Ars Grammatica. (See W. J. Chase, The “Ars Minor” of Donatus, Univ. of Wis. Studies in the Social Sciences and History, No. 11, Madison, 1926.) Pris-cian's work, the Institutiones Grammaticae (6th cent.), is a more inclusive and advanced treatment than the Ars Minor. (See editions by Augustus Krehl, Prisciani Caesariensis Grammalici Opera, 2 vols., Leipzig, 1819-20, and by H. Keil, Prisciani Grammalici Caesariensis Institutionum Gramma-ticarum Libri [Grammalici Lalin-i, n, iii], Leipzig, 1855-59.)

The question of Æfric's indebtedness to Donatus and Priscian has not been worked out in full and exact detail. L. K. Shook (“Ælfric's Latin Grammar,” unpub. Harvard diss., 1940) deals with it in an illuminating way in connection with specific passages.

^{Note 3 in page 454} See, e.g., Alfred's Preface, King Alfred's West-Saxon Version of Gregory's Pastoral Care, ed. H. Sweet, EETS, O.S. 45, 50 [London, 1871-72], pp. 2 ff.

^{Note 4 in page 454} Note, e.g., the 29 categories of nouns listed in the Grammar, some overlapping others (pp. 11 ff.); the similar series of classes of adverbs; the long listing of nouns of the third declension according to their endings and of verbs according to the form of their perfect tense.

^{Note 5 in page 454} The present article was completed before Shook's unpublished dissertation on Ælfric's grammatical terminology was known to me (see n. 2 above). His work is a valuable contribution to studies of Ælfric's methods: its major part consists of “an investigation of the etymology, semantic development and grammatical use of the Latin as well as the corresponding Old English terms,” the stated purpose of the whole being “to investigate this Old English grammatical nomenclature and to determine on what principles these new technical terms are formed.” (See the abstract of the dissertation in Harvard Univ. Summaries of Theses, 1940 [Cambridge, Mass., 1942], pp. 360-363.) Our discussions, differing widely in purpose and scope, have necessarily touched on a few of the same points.

^{Note 6 in page 454} We read, e.g., “The special grammatical terms of his original are not borrowed wholesale but are translated into their Old English equivalents… On the whole it seems probable that this grammatical terminology died out, not from any lack in itself, but because it depended for its existence on constant literary use. Had it not been for the Conquest we should probably still be using Ælfric's grammatical vocabulary” (R. M. Wilson, ed. Sawles Warde, Texts and Monographs: Leeds School of Eng. Lang., No. iii, 1938, p. vi).

^{Note 7 in page 455} The tabulation, so far as it goes, is complete, all instances of the appearance of the terms within the given pages being counted. A few of the grammatical terms used are not listed.

^{Note 8 in page 456} There is to be noted in the later sections of the Grammar an increasing use of the English terms without the corresponding Latin: andwerd, fordgewiten, toiverd (tid).

^{Note 9 in page 457} Such an instance occurs in connection with GERUNDIA: “Das fif word synd… gecwedene GERUNDIA of am worde gero ic bere, foran e hi bera manega andgytu” (135:16-136:1).

^{Note 10 in page 457} In his work on Old English terminology, O. Vocadlo (“Anglo-Saxon Terminology,” Prague Studies in English, iv [1933], 59-85, esp. pp. 75-76) takes full account of these translations-by-phrase, noting in particular the translation of GENTILIA, COIXECTIVA, etc. Tentative in his general conclusions regarding them, he states his opinion clearly that it was not necessarily lack of skill on Ælfric's part that gave rise to some of his more involved, less succinct translations, and implies in the course of his discussion the opinion that it was not always Ælfric's intention to find simple equivalents for the Latin terms and the Greek terms adopted by the Latin grammarians.

^{Note 11 in page 457} “Anglo-SaxonTerminology”; “A Technical Construction in Old English,” Mediaeval Studies, ii (1940), 253-257, and “Ælfric's Latin Grammar” (n. 2 above); Ælfric, Sermon-naire, Docteur et Grammairien (Paris, 1943), pp. 259 ff.

^{Note 12 in page 458} Such terms, of which leorningcniht for Lat. discipulus is a classic example, together with the compounds, etc., mentioned in the next section, represent the additions to the vocabulary known as “loan-translations.” On this topic see in particular Shook, “A Technical Construction in Old English,” p. 253, and H. W. G. Gneuss, Lehnbildungen und Lehn-bedeutung im AUenglischen (Berlin, 1955).

^{Note 13 in page 458} See p. 456 of this article.

^{Note 14 in page 458} See Shook, “A Technical Construction in Old English.”

^{Note 15 in page 458} Note also the names of all other moods and cases.

^{Note 16 in page 458} John C. Pope has called to my attention, in this connection, unasecgendlic and other examples. See Bosworth-Toller, Anglo-Saxon Dictionary, the un- combinations in particular.

^{Note 17 in page 459} See R. M. Wilson, Introduction to Sawles Warde (n. 6 above).

^{Note 18 in page 459} It also translates Lat. FORMA at least once: “SIT HOC SATIS DE SEX CASUALIBUS FORMis beo is us genoh be am syx gebigendlicum hiwum gesaed” (91:7-8).

^{Note 19 in page 459} Zupitza, 108:5-6; 112:10; 146:10-12 (“miratus sum ic wundrode to werlicum hade, mirata sum to wiflicum hade, miratum to naSrum cynne”).

^{Note 20 in page 459} Zupitza, 108:16; 217:11; 105:21.

^{Note 21 in page 459} Zupitza, 130:6, 136:14, 94:15. Note also 78:15, where declinung renders Hie CASUS.

^{Note 22 in page 460} The term homonym here must be understood to include reference to one word with multiple meanings, as well as to etymologically different words.

^{Note 23 in page 460} Latin offered a choice of two words in this instance, NOMEN and VOCABULUM, the latter occurring frequently.

^{Note 24 in page 460} Nama ‘noun’ is frequent. In the following passages nama is to be interpreted ‘name’: 5:4-6; 5:11-13; 25:19; 29:16; and possibly in 8:13-9:2. Occasionally there is ambiguity: 5:15; 14:22; 33:1; 30:11; etc.

^{Note 26 in page 460} Cf., for example, the following passages: 96:3-5; 11:1-2; 125:9-12; 6:12-13; 5:11-13; 18:19: 135:18.

^{Note 26 in page 460} The general meanings of the other English words that Ælfric used in preference to their Lat. counterparts, notably cynn, getel, tid, dœl, had (Lat. GENUS, NUMERUS, TEMPUS, PARS, PERSONA, respectively) did not interfere with the grammatical senses to any extent. A reader might hesitate at times over tid and the occasionally introduced lima: Are both used to mean ‘tense’ in a technical sense or are they used to differentiate between ‘time’ and 'tense'? The difficulty is slight. Getel may have shared in the historical development of its related verb tellan, which lost its meaning 'to count' in the extension of the meaning ‘to relate, to narrate.‘

^{Note 27 in page 460} Both the OED (s.v. Part, sb.2.a, Pars, sb.) and Bosworth-Toller note the occurrence in Old English of the forms cited here, with reference in each dictionary to Æfric's Grammar only. The word appears in Byrhlfer's Handboc (c. 1050) in the sense of 'portion.' Its next appearance recorded in the OED is that in the 13th century as an adoption of French part, pi. pars.

^{Note 28 in page 460} In Bosworth-Toller this word appears in the forms casus, casu, casum, case, with reference to Ælfric's Grammar; it appears, in the form casus only, in Clark Hall (4 Concise Anglo-Saxon Dictionary, Cambridge Univ. Press, 1931). The OED does not make note of its appearance in Old English, giving 1225 as the date of its first appearance in the language (in the sense 'event,' 'occurrence') and 1393 as the time of its first appearance in a grammatical sense.

^{Note 29 in page 461} The initials referring to individual MSS are those used by Zupitza.

^{Note 30 in page 461} Clark Hall gives declinian, un/declinigendlic (with reference to Ættnc), and declinung. Bosworth-Toller also gives the word. The OED gives no citation of the verb used in the grammatical sense before 1387. M. Braunschweiger {Flexion des Verbums in Älfrics Grammatik, Marburg diss., 1890) notes Ælfric's use of various forms of the verb.