¹ E. Emerton, The Letters of Saint Boniface (“Records of Civilization, Sources and Studies, ”xxxi; New York, 1940), pp. 158–159. For the Latin, see Monumenta Germaniae Historica, Epistolarum (Berlin, 1892), iii, 368.

² A. Hauck, Kirchengeschichte Deutschlands, Sechstes Tausend (Leipzig, 1922), i, 540.

³ Emerton, p. 159. For the Latin, see Mon. Germ. Epist., iii, 368–369.

⁴ Zeitschrift für deutsches Altertum, xxxvi (1892), 135—141.

⁵ F. W. Hack, Untersuchungen über die Standesverhältnisse der Abteien Fulda und Hersfeld bis zum Ausgang des 13. Jahrhunderts (in “Quellen und Abhandlungen zur Geschichte der Abtei und der Diözese Fulda, ”vii; Fulda, 1911), p. 60. The origins of Baugulf (780–803) are uncertain; Ratgar (803–817) was an East Frank.

⁶ StD, pp. 1–8; = E. von Steinmeyer, Die kleineren althochdeutschen Sprachdenkmäler, Berlin, 1916. All OHG documents except the Tatian are quoted from this edition; cited forms are followed by a number indicating the line in which they occur. The Tatian is quoted from E. Sievers, Tatian, Lateinisch und altdeutsch,² Paderborn, 1892.

⁷ StD, pp. 39–40, 55–57.

⁸ On the dates and Fulda origins of these three documents, see MSD, i, xii; ii, 357; i, xxv (= K. Müllenhoff and W. Scherer, Denkmäler deutscher Poesie und Prosa, 3d ed. by E. Steinmeyer; 2 vols. [Berlin, 1892]); R. Kögel, Geschichte der deutschen Litteralur bis zum Ausgange des Mittelalters, 1 vol. in 2 parts (Strassburg, 1894–97), i, 226–227; ii, 497–498, 499–501; H. Pongs, Das Hildebrandslied (Marburg diss.; Marburg, 1913), pp. 194–207; StD, pp. 9 and 13, 41, 58; G. Ehrismann, Geschichte der deutschen Literatur bis zum Ausgang des Mittelalters, 1. Teil, Die althochdeutsche Literatur ² (“Handbuch des deutschen Unterrichts, ”vi, 1; Munich, 1932), pp. 122, 362, 352.

⁹ G. Kossinna, Über die ältesten hochfränkischen Sprachdenkmäler (“Quellen und Forschungen, ”xxvi; Strassburg and London, 1881), pp. 4–16.

¹⁰ For the following, cf. Pongs, pp. 196–207; G. Baesecke, ZfdA, lviii (1921), 276–277.

¹¹ AS influence is strongest in the BasRec. All 3 recipes (the 1st is in Latin) are written in insular letters, except that uu is used in place of the w-rune; see M. Enneccerus, Die ältesten deutschen Sprachdenkmäler in Lichtdrucken (Frankfurt a. M., 1897), plate 17. The 3d recipe is, in addition, such a mixture of AS and HG forms and spellings that it cannot be considered typical of the Fulda documents. As used in this paper, the abbreviation BasRec will refer only to the 2d recipe. The MonRec are written in AS letters; see Kossinna, p. 4. The SalLaw is written in Carolingian minuscule, but the insular w-rune is used; see Pongs, p. 200, StD, p. 55, note 4. The Hild is written in Carolingian minuscule, but many insular letters are used: the w-rune, round d, crossed d, the ligature œ, hooked , closed a, one insular f; see Enneccerus, plates 1–2, Pongs, pp. 44–45.

¹² See below, 3.1–2, and cf. W. Braune, Althochdeutsche Grammatik, 5th ed. by K. Helm (“Sammlung kurzer Grammatiken germanischer Dialekte, ”A, v; Halle, 1936), §§167, 135–136.

¹³ Cf. Braune, §163. Hild 98 t, no d, see C. Kraus, Zeitschrift für die österreichischen Gymnasien, xxlii (1896), 325–326; BasRec 23 t, 1 d (do 18), plus cullantres 15<Lat. coriandrum; SalLaw 78 t,1 d (indi 28). The extreme rareness of d is again perhaps evidence of Bavarian influence. On the more numerous d's in the MonRec, see Kossinna, pp. 36–44.

¹⁴ Cf. Braune, §76. Hildfurlaet 20, furnam 43; SaiLaw forlaazit 9, forstilit 23, 27, 29, 30, 34, 35, 36, foruz(z)an 24, 26, 28, 29, 31, 33, 34, 35, 37, forstolan 25. The prefix does not occur in the BasRec; Kögel, ii, 498, gives additional EFr. features of this document.

¹⁵ Cf. Braune, §131. Hild werpan 40, scarpen 64 (though these p's may represent part of the attempt to rewrite the document into Low German; cf. the treatment of p in sloptū 65, wabnū 68); BasRec peffur 7 (though an h has perhaps been erased before the p; see Enneccerus, plate 17, line 19; StD, p. 39, note 2); SalLaw pentinga 31; MonRec, see Kossinna, pp. 46–47.

¹⁶ Cf. Braune, §283.1a. Hild 9 her plus hera& 22 (=he raet or her raet?); BasRec he 12, 12, 13. The SalLaw has 6 er (lines 13, 15, 18, 19(2), 20); weo her 18 may be an error for hweo er; see StD, p. 59.

¹⁷ See below, 3.3, and cf. Braune, §144. The use of c before i and e is perhaps borrowed from AS spelling habits; cf. Braune, §142.1. The ch-spellings in the Hild are surely not evidence of affricate pronunciation; see Braune, §143.2. A clearly non-UG spelling of the Hild is hop 27, as against UG liup; cf. Braune, §47.

¹⁸ Cf. Braune, §71. Hild 17 gi-, see Pongs, p. 59; BasRec 9 gi-, 1 ge- (getanes 12); SalLaw 10 gi-. The prefix apparently does not occur in the names of the MonRec.

¹⁹ StD, p. 23. Only MS A will be discussed here. MS B is lost; we possess only a poor 17th century copy of it; see StD, pp. 25–26.

²⁰ See Enneccerus, plate 6.

²¹ Gmc. b appears only as b (16 times). Gmc. appears 4 times as th- (them 3, thie 4, thrinisse 9, thuruh 12); it appears as d only in medial or unstressed initial position (den 3[2], unholdun 1, 2, heidene 4, iode 13); enclitic pu after s appears as tu (forsahhistu 1, 2, 3, gilaubistu 6, 7, 8, 9, 11, 12, 13).

²² Cf. Kögel, ii, 449; J. Franck, Altfränkische Grammatik (“Grammatiken der ahd. Dialekte, ”ii; Göttingen, 1909), p. 8; StD, p. 26; Baesecke, ZfdA, lviii, 275–276. Ehrismann, p. 299, suggests that it is a RhFr. copy of a Fulda original; similarly Pongs, p. 173. RhFr. is d for WGmc. d in indi 2, 3(2), and geldom 4, as against 2 t-, 10 -t-, 3 -t. EFr. are the spellings for- (forsahhistu 1, 2, 3, forlaznessi 12) and fur- (fursahu 1, fursahhu 2, 5; these fur-'s are all in red letters).

²³ StD, pp. 62 and 327–328.

²⁴ Kossinna, p. 94: “nach a. 841, aber nicht zu lange danach, wohl schon a. 850 ”(this is probably too late an estimate); MSD, i, xv: “im neunten jh.; ”Kögel, ii, 502: “Mitte des 9. Jahrhunderts ”(based on Kossinna); E. Schröder in W. Arndt, Schrifttafeln zur Erlernung der lateinischen Palaeographie, ed. by M. Tangl (Berlin, 1903), iii, 37, places it between 812 and 822, and Tangl, loc. cit., considers that this fits the paleographic features of the document better than Kossinna's late date; StD, p. 63: “frühestens im 2. Dezennium des 9. Jh.; ”Baesecke, ZfdA, lviii, 275: “gegen 830; ”Ehrismann, p. 350: “Kopie des 9. Jh.s. ”

²⁵ So Kossinna, p. 95; Baesecke, Beiträge zur Geschichte der deutschen Sprache und Literatur, xlix (1925), 349; Ehrismann, pp. 316 and 459. Kögel, ii, 542–543, agrees on 830 for the original of MSS A, B, C, but assumes an even earlier prototype. That the original was written later than our copy of the HamBD is perhaps indicated by HamBD teofun 13, 18, 21, as against FuldConf liogannes 5 in all three MSS.

²⁶ For the text of A, see also F. Pfeiffer, Akademie der Wissenschaften in Wien, phil.-hist. Kl., Sitzungsberichte, lii (1866), 41–42; for C, F. Pfeiffer, Germania, xiii (1868), 386–387. In C, blank spaces are left for the capital letters, which were presumably to be added in color; I shall reconstruct them in square brackets.

²⁷ M. Tangl in Arndt, Schrifttafeln, iii, 37. In i, iv (4th ed.; Berlin, 1904), Tangl calls it “fränkische Cursivminuskel. ”

²⁸ HamBD 1 p- as against 22 b-, 5 -b- (perenfirst 14, but -burg 3, 20, -bot 5, -beraht 5, 7(2), bruning 6, -bah 10, 11, 12(2), 14, 18, -brunnon 15, 17, 19, burg- 16, buochun 17, blenchi- 17, -berg 19, brunnen 21, plus hamalunburc on the cover [see StD, p. 62, line 11 from bottom]; gruoba 13, obanentig 15 (2), houbit 16, ubar 17). FuldConf A 1 -p as against 9 b-, 5 -b-, 4 -b (gop 14, but bigihtig 1, 21, giburili 7, 20, Bisprachido 9, nibiheilt 10, bi 17, bin 21, buozziu 22; Ubilero 5, 9, ubartruncani, 7, selbes 8, forgibig 23; Abunstes 8–9, urloub 14(2), forgib 24). FuldConf C 8b-, 5 -b-, 6 -b (lacking initially is [B]isprachido 9; added finally are gab 14, nigab 11).

²⁹ HamBD 12 th-, 3 -d-, 8 -d, 1 -t (thie 13, 16, 18, then 13, 16, 18, 20, 22, thaz 17, 20, theo 21, themo 21; adal- 7, nendichen- 13, -frides- 20; nid- 3, hruod- 5, suuid- 5, 6, leid 7, -nand 7, -frid 7, -ueld 13, exception is gunt- 4; do attumar 6 and malten 18 contain geminated ?). FuldConf A 16 th-, 11 -d-, 1 -th-, 1 -d, and 5 -d- < (githanco 2, Thes 3, 16(2), 17, thuruh 6, 7(2), 8, Thaz 9, themo 12, then 13, githahti 18–19, thir 22, thinan 24, thin 25; UU1RDU 1, andran 4, uuidar 5, 16, eido 5, anderes 8, 16, Girida 8, Nides 9, Bisprachido 9, sculdig 17, exception is kinthisgi 7; uuizzod 13; Odo 4, 6, 7(2), 8). FuldConf C 15 th-, 12 -d-, 1 -d, and 5 -d- < (lacking initially is then 13; added medially is kindisgi 7).

³⁰ Hauck, ii, 629, note 4, and 630.

³¹ Cf. Baesecke, ZfdA, lviii, 276.

³² Sievers, §118.

³³ Baesecke, ZfdA, lviii, 274.

³⁴ For the following see Sievers, p. xii; Baesecke, ZfdA, lviii, 251–252. A clear distinction should be made between the original translators of the first rough draft, and the scribes who from it copied our present manuscript. The question of how many authors made the original translation, and who translated which parts, is still unsettled, and will not be discussed in this paper; see 1.5 below, and note 42. The number of scribes who made the more finished copy, however, and the parts which each one copied out, can easily be determined by noting differences in handwriting. Sievers does this on p. xn. Scribe γ's handwriting extends from manuscript page 124 line 7 through page 164 (chapter 82.11 thér thar through 103.5 in then).

³⁵ Cf. the statistics on the spellings of the various scribes in Sievers' introduction, §§5–117.

³⁶ ZfdA, xvii (1874), 82.

³⁷ Zeitschrift für deutsche Philologie, vii (1876), 421, 429.

³⁸ ZfdA, lviii, 259.

³⁹ Walahfrids deutsche Glossierung . . . und der ahd. Tatian (“Hermaea, ”xvi; Halle, 1926), esp. pp. 140–143. But see the objections raised by H. Brinkmann, Sprachwandel und Sprachbewegungen in ahd. Zeit (“Jenaer Germanistische Forschungen, ”xviii; Jena, 1931), pp. 102–103.

⁴⁰ G. Baesecke, Einführung in das Althochdeutsche (“Handbuch des deutschen Unterrichts, ”ii, 1, 2; Munich, 1918).

⁴¹ J. Schatz, Althochdeutsche Grammatik (in “Göttinger Sammlung idg. Grammatiken und Wörterbücher,“ Göttingen, 1927).

⁴² For a review of research on the authorship of the Tatian translation, see the most recent work on the subject: H. C. Meyer, The Authorship of the OHG Tatian: Addition and Non-addition of Pronoun Subjects (Chicago diss.; Chicago, 1936), pp. 25–27.

⁴³ Sievers, §10.2.

⁴⁴ Hild non-relative der 34, 58, relative der 59, de 60; BasRec non-relative de 17, relative de 17; SaiXaw relative de 3, der 1, 4, 6, 8, 9, 10, 14, 15, 27. There are no examples in the MonRec, FrBapOath, HamBD, FuldConf.

⁴⁵ Sievers, §10.2.

⁴⁶ Following Sievers' practice, references are usually to chapter and verse; but where the form in question is not cited, reference is made to the page and line of Sievers' 1st edition as given in the center of each page of the 2d edition. Abbreviations (cf. Sievers, pp. viii–ix): s = scribe, c = the corrector , r = Rasur, n = Nachtrag, z = Zeile. The letter or letters to which such an abbreviation refers are in bold-faced type. Ellipsis (…) before or after an erasure indicates that the erasure includes a preceding or following letter or letters.

Sievers' statistics on γ in the second table of §10.2 are reversed and contain two omissions. They should read (including doubtful forms, as Sievers does): thie [i.e. the] 2 (10), ther 79 (13). In §69.2, for 87.8(5) read 87.8(6).

⁴⁷ Sievers, p. 350^b. Line 7 from bottom, for 103.3(3), 4 read 104.3(3), 4.

⁴⁸ If the dot under the H of no. 20 was intended as a sign of deletion, this is another example.

⁴⁹ See above, note 16. There are no examples in the MonRec, FrBapOath, HamBD, FuldConf.

⁵⁰ Sievers, §69.2. Sievers omits my no. 18 among the the-spellings, my nos. 9, 15 among the erasures.

⁵¹ “Das vereinzelte sia 136.1 wird blosser Fehler sein”; Sievers, §69.2,

⁵² In the footnote on this form, Sievers writes se: r (e), but seems to mean se: r (e), because in the text he writes see.

⁵³ Sievers, pp. 455–170.

⁵⁴ in thie leofun gruoba 13; in thie michilun buochun 16–17; in thie teofun clingun 18. Only the spelling of the article indicates that these words are all acc. sg. fem.; the noun and adjective endings can also be acc. pl. fem. However, the acc. pl. fem. of the article is spelled theo in one sure example: in theo teofun clingun 21 ; hence thie is probably singular.

⁵⁵ Sievers, p. 420. Under sia = eam, for 127.2(2) read 127.2.

⁵⁶ Schatz, §§21, 421.

⁵⁷ Sievers, §69.1.

⁵⁸ hiez 90.1 (not in Sievers, §69.1), fie:lun (l) 91.3, zuogiengun 93.2 (not in Sievers, §69.1), intfiengun 93.2, ingieng 93.2, giengun 94.2, ::fiel 97.4, intfieng 97.6.

⁵⁹ e is certain in ingeng 84.8, zuogeng rc in ie 92.4; very probable in ingiengunt rc (g:n) 87.8, ubargeng rc 97.7 (i written over e), gⁱeng (c?) 99.3 (dot under i), zuogiengun rc (Raum für nur 2 Buchstab.) 100.2, gieng rc (für nur 2 Buchstaben) 101.2. Uncertain are giengun rc 82.12, ingieng rc 83.1, giengun rc 85.3, giengun rc 87.2, giengun rc 87.7, gieng rc 88.3, gieng rc 88.5, zuogieng rc 91.3, zuogiengun r 92.1, gieng rc 92.2, uzgieng rc 92.6, giengun rs, das ie rc 92.8, gieng rc 97.2, gieng rc 98.4, gieng rc 101.1. The 1st of these uncertain examples is quite obscure, since the corrector erased the whole word. In 10 of the remaining 14 he was apparently forced to erase several letters in order to fit his ie into the word. If γ had written ea or any other two-letter combination, this would have been unnecessary ; probably, then, y wrote only one letter, e.

⁶⁰ γ probably wrote ea in uorliez r (ea?) 87.7, mieta. r (ea::?) 87.8, gispien rc (ean?) 100.3. Uncertain are uorliez rc 89.4, uorliez … rc 99.2, forliez … rc 99.2, uorliez rc 99.4, uorliez u r in f, iez rc 100.4, gifiel r 102.1.

⁶¹ … hir … rs 91.2, hir 91.2, hier r (für 1 Buchst.) 97.3. Uncertain is hier r 90.6.

⁶² For a possible explanation of this i, see Franck, §42.2, Anm.

⁶³ The only examples are Hild urhettun 2 (or<Gmc. ai?), hœtti 17, furlaet 20, 63; SalLaw hwe 5 (but weo her 18), heer 20. The later Fulda documents show ie: MonRec Uuielant (from the year 841; Kossinna, p. 28); FuldConf C nibihielt 10, nifirliez 12, intphieng 14, gihiezi 17, forliezi 18. FuldConf A seems to have been copied from a document where i was written over e, since it shows both ei and ⁱe: nibiheilt 10, nifurleiz 12, intpheing 14, gikⁱezi 17, forlⁱezi 18.

⁶⁴ Sievers, §74.

⁶⁵ eo is sure in teof 87.3, seohhoro r in c 88.1, seocho 88.2, leoht 91.3, leobar 91.3, manodseoh 92.2, leohtfaz r in i 96.5, the^onon 97.7 (o written over e), reof 85.2, reof r zu i 92.2, ueor 89.3, 5, eo 86. 2, 88.12, eogilih 84.7, éogilih r in i 95.5, eogilicheru r in i 98.3, neo 97.7(2) (Sievers, §74.1, omits 1 of these 2), neoman 82.11a, neomannen rc in a 91.3, neomanne 91.4, uueo 87.2, theo 85.4, 97.2; eo is probable in lioht… rs, das erste i für e 88.11, liohte r 88.11, uorliose rc (-e.::a) 90.5,… íouuiht rc? 88.7, … íouuiht rc 88.10, íogilih r 98.2 (not in Sievers, §74.1), niomanne rc 90.3; e is sure in neman 87.7, neuuiht 92.8, probable in nioman r (dem Räume nach nur für 2 Buchstaben [ne?]) 97.2. Uncertain are íouuiht r 82.11a,… nioman rc 95.2.

⁶⁶ io appears in MonRec Liobhilt, Liobheri from the year 779; the other 6 examples of io are from the years 823 and 825; see Kossinna, p. 33. eo appears in Hild irmindeot 13, theotrihhe 19, deotrichhe 26, eo 27(2), 51, leop 27, neo 31, sceatantero 51; BasRec deo 9, eogiuuelihha 9, geoze 10, feorzuc 12, neouuiht(es) 13, 17; SalLaw eowih: 4, weo 18, deonoste 18–19, feorzug 31; MonRec 79 examples, see Kossinna, p. 33 (Kossinna counts 78 but cites 79). e appears in Hild detrihhe 23; eu in Hild hevwun 66 (v written over e), MonRec Theutoni (Kossinna, p. 34). Uncertain is MonRec Tienen-, Teinenheim (Kossinna, pp. 33–34). There are no examples in the FrBapOath.

⁶⁷ Braune, §§306.5, 371–376, 379.1; Brinkmann, pp. 150–152.

⁶⁸ Sievers, §62.

⁶⁹ -st in uueist 90.4 (plus uueislu 84.7) and in bist 87.5, 88.5, 90.2(2), 3, 4 (plus bistu 87.3). Otherwise only -s, 17 times: habes 82.12, 87.3(2), 5(2), bitis 87.2, uuessis 87.3, batis 87.3, scefes 87.3, habetos 87.5, suochis 87.7, sprichis 87.7, gibiniis 90.3, zilosis 90.3, mugis 92.4, fintis 93.3, gistrⁱunis 98.1 (dot under i). (Sievers, §62, omits the examples in 82.12, 92.4.) γ also writes -s in bis 82.12, 97.8, 103.2, bist n 87.2; cf. Braune, §379.1.

⁷⁰ Hild bist 39, pist 41, but sis 11, sages 12, nigileitos 32, spenis 40, f^vortos 41 (v written over o), habes 47, 57. FrBapOath Forsahhistu 1, 2, 3, Gilaubistu 6, 7, 8, 9, 11, 12, 13, with enclitic -tu, are not comparable forms. The other Fulda documents contain no 2d sg. forms.

⁷¹ Sievers, §7.

⁷² Sievers, §7a; aruueccu 82.11 is rs, not rc.

⁷³ Sievers, §7b; plus 2 enti 85.1, 100.1.

⁷⁴ Sievers, §7d; not pertinent are fon samariu 87.2, in uuostinna 96.2.

⁷⁵ The page and line references are: γ wrote -iu 168.3(2), 169.4, 7, 8, 170.15, 19, 171.7, 172.7, 174.24, 36, 175.17, 34, 177.33, 178.12, 179.25, 181.25, 183.33, 185.12–13, 186.16, 188.13(2), 195.2, 14, 196.36, 197.4, 15, 23(2), 201.35; γ wrote -u 166.22, 169.4, 4–5, 8, 171.33, 174.37, 178.9, 181.8, 11, 12, 185.11(2), 12, 13, 14, 196.2, 4, 198.4, 200.9; uncertain are minu, thimxu corr. in iu 195.14, gibuntanu rc 196.3, uuelichu rc 196.10, uuizu … r(s?) 175.18, arlóstu r 196.6. Sievers, §7e, counts 29 -iu, 25 -u. His figures for δ should apparently be 26 -iu, 0 -u, not 10 -u; see the material that follows his table.

⁷⁶ Dat. pl. -un where we would expect -on is common; see Sievers, §111.3.

⁷⁷ “Das zweite i später unten angehängt”; StD, p. 1, note 13.

⁷⁸ From the years 821, 824; Kossinna, p. 61.

⁷⁹ Cf. Braune, §118.3.

⁸⁰ Sievers, §7.

⁸¹ This last example omitted in Sievers, §7.

⁸² uuillon 87.8, uuillo 96.4, lebento 97.1, burgliuto 97.2, suntota 97.3, 4, sunto 98.1, ur: cundono 98.2, gisuntol 98.4.

⁸³ Braune, §58.1.

⁸⁴ Braune, §§198.4, 209.3, 226, 205.1–2.

⁸⁵ Sievers, §7; cf. also §§103–104, 106.

⁸⁶ -e is sure betere 87.5, asnere 97.3, scribere 101.2, each with -e rc in -a; -e is probable in lichazera rc (e für a?) 84.5, buockera: rc (e?) 91.4, buochera rc 91.6; uncertain is lichezera rc 103.4.

⁸⁷ -en is sure in üfstiganten rc in a 82.11a, uzuuerfenten 95.1, habenten 95.4, einougen 95.5, habenten 95.5; -en is probable in quementan rc 90.6; -an is sure in ligentan 88.2; uncertain are tuontan rc 88.7, tuontan rc 96.6. γ also writes 1, perhaps 4 -en in the corresponding a-adj. form: thesen 90.3, ebanlichan rc 88.6, sinan rc 92.4, halzan rc (e?) 95.4. There are 34 -an in other a-adjectives, plus 64 acc. sg. masc. inan; uncertain is uuenan … r(s?) 90.1.

⁸⁸ -e in groube rc in a 84.7, fuzze 87.3; -ea in redea. rc in a 99.1(2); -a in minna 88.13. An uncertain example, if it is a fem. jō-stem, is tura rc 102.1. Sievers, §103b, and Braune, §207.1, consider groube an ō-stem; but see Franck, §§47, 139.

⁸⁹ thiuba 84.9, friuntinna … zrc 96.5. γ perhaps wrote -e in ace. pl. fem. scultra rc (e?) 96.2, an ō-stem. A possible example of -e<-ja is acc. pl. fem. mitte rc in a 86.1, where we would expect -o (or perhaps -eo). However, γ and most of the other scribes often write nom. acc. pl. fem. blinta etc. on the analogy of nom. acc. pl. fem, geba; see Sievers, §110.1. In mitte (a ja-, jō-adj.) γ may have written -e on the analogy of nom. acc. pl. fem. ∗sunte.

⁹⁰ The full phrase is inti mittiu der heilant gisah zuolouffante thie menigi ‘et cum videret Ihesus concurrentem turbam.‘ Sievers, p. 380, considers zuolouffante acc. pl. masc; cf. also §110.1, Anm. γ also writes-e in the acc. sg. fem. of an ō-adj. (alle rc in a 90.5), where we would expect -a. This is a very common spelling, however; cf. Sievers, §103c (to which add alle ζ 196.2). On p. 303^a Sievers lists for acc. sg. fem. 5 alle, 11 alla.

⁹¹ e<ja also in Hild eilen 55<∗aljan, though this -jan is a suffix, not a case ending.

⁹² Kossinna, p. 61.

⁹³ Braune, §§314–316.

⁹⁴ Sievers, §§89–90.

⁹⁵ 13 -an (incl. 2 -a), 16 -en (incl. 1 -e), 11 -an rc; see Sievers, §89a, for references. In 2 of these last 11, several letters are written on the erasure: uorgeltan rc 99.2; uuesan r (s?) 97.4; in the remaining 9, however, only -a- or -an is written on the erasure, so that γ probably wrote -en in 9 more cases.

The figures in Sievers, §§89–90, are largely correct. In §89b, forlázzanne rc (e:::) 100.4 should not be listed as uncertain; it surely belongs in the -enne column, giving 8 -anne, 5 -enne. In the 5th line from the end of §89, for spinnan read spiuuan; in the 2d line from the end add 3 gangenti 84.7, 86.1, 97.6. To the type sizzen (§90a), add as a probable example richoson r in i, das zweite o r (e?) 95.5; here γ seems first to have written richosen instead of richison. In the type sizzenti (§90c) there are 16, not 15 -enti; in the 3 examples which Sievers lists as uncertain, γ probably wrote -anti: antuurtenti r 97.7, nahenti r 101.2, antuurtenti rc 102.2.

⁹⁶ ja-verb forms: Hild seggen 1 (also repeated above line 1; see StD, p. 1, note 1), sitten 20, bihrahanen 57, hrumen 61; SalLaw gimenen 19, menen 20. a-verb forms: Hild infahan 37, wer pan 40, 42, sceotantero 51, hauwan 53, werdan 54, giwinnan 56, uualtan 62, scritan 63; SalLaw suganti 23.

⁹⁷ Braune, §118.

⁹⁸ ja-verbs: A, 2 e, 2 a: sizzenti 20, liganti 20–21, zigiuuircanne 24–25, zigifremenne 25; C, 1 e, 1 i, 2 a: sizzenti 20, liganti 20–21, cigiuuircanne 24–25, cigi freminne 25. a-verbs: A, 6 e, 4 a: Liogannes 5, Stelannes 6, ezzenti 15, trinchanti 15, slafenti 15, uuahchanti 15–16, slaffenti 19, uuahhenti 20, gangenti 20, stantenti 20; C, 7 e, 3 a: [L]iogannes 5, [S]telannes 6, ezzenti 15, trincanti 15, slafenti 15, uuachenti 15–16, slafenti 19, uuachenti 20, gangenti 20, stantenti 20.

⁹⁹ The only exceptions in γ are abasnit 95.4, heithin 98.2; see Sievers, §23.

¹⁰⁰ Sievers, §§18–19.

¹⁰¹ Sievers, §20.

¹⁰² Sievers, §§21, 62.

¹⁰³ γ uses miltiu to translate Lat. dum (88.2, 91.1) and nam (92.2), and 19 times to translate cum ‘when, because.’ But he writes mil thiu twice (82.12, 87.2) where cum describes accompanying circumstances, and once (87.2) to translate a relative pronoun. Sievers, p. 465^a, under 20, lists these mit thiu separately, under the meanings ‘während, da doch, obgleich.’

¹⁰⁴ Not 62 times, as Sievers, §20.2, states.

¹⁰⁵ 2 of the 3 th-spellings (83.1, 88.2) are also the only non-r-forms of the article; see ve, 2.1.

¹⁰⁶ The crosses on the d's in lines 1 and 3 are paler than the other two, and seem to be additions; see StD, p. 1, note 1.

¹⁰⁷ Kossinna, pp. 44–45.

¹⁰⁸ For non-initial p, the BasRec shows 3 -d-, 2 -t-; cf. also, in the 1st (Lat.) recipe, the German word antor 3. The Hild shows, after line 5, 13 -d-, 10 -d, and in gemination eddo 11, 13 (StD, p. 2, note 6), 54, erdo 62. In the MonRec, d is the usual spelling; t is occasionally used, especially in final position; see Kossinna, pp. 45–46. The SalLaw shows 21 -d-, and in gemination erdo 17(2), 23, &tes kwelihemo 17; the one d-spelling, alode 2, was probably so written to conform to the Lat. spelling alode; cf. Braune, §167.3.

¹⁰⁹ Sievers, §27 and p. 370.

¹¹⁰ “[Dieses Wort] verdankt sein p dem Einfluss des vorhergehenden t”; Sievers, §26.

¹¹¹ Sievers, §28.

¹¹² Spellings with -b: in β, 18.4; in a‘, 127.3; in ζ, 134.8, 156.5, 178.4, 203.4, 208.1, 211.4(2); in δ’, 220.5, 232.1; not counted is giscribit a’ 129.5 (error for giscrib). Spellings with -p: in a’, 129.7; in ζ, 185.5, 9; in d’, 229.1 (plus 1 in γ, 88.13). See Sievers, p. 425.

¹¹³ pipoz 8, petti 12, protes 13, nenpizt 14, nipado 15, niinpiize 15, piuuartan 18; nipuz ( = ni ipu iz) 17, simplum 17–18, ipu 19; but uuegabreita 8; suebal 8, ribanne 9, gebe 17, habe 18. Cf. also, in the 1st (Lat.) recipe, the German word pipaoz 2 and the Lat. word putdiglas 1 (= ∗butticulās ‘bottles‘).

¹¹⁴ Kossinna, pp. 47–50.

¹¹⁵ Pongs, p. 63.

¹¹⁶ biliugit 7, bifillit 8, gibanni 16, biuzan 30, bistoozit 32; ibu 13, 14, 18, 19, 25, nihab& 14, 15, obana 20, giscriban 21, iubiu 22, 32, haubitgelt 24–25, 29, 32, leben 31, haubit 33, 34, 35, 37; wiib 10.

¹¹⁷ Braune, §143. Gmc. k probably remained unshifted also after s; see Braune, §146. The documentary spellings of k in this position, however, are often different from those of unshifted k in other positions, and would have to be considered separately. They need not be discussed here, since γ's treatment of sk does not differ much from that of some of the other scribes; cf. Sievers, §§50–52.

¹¹⁸ See Franck, §115.

¹¹⁹ BasRec trincen 11, cullantres 15 (<Lat. conundrum), niuuirce 17, tranc 17, gitruncan 19 (plus flasgun 9, 18, 20). SalLaw nicuimit 13, 15, urcundeom 16, cueme 16, cuenun 17, gicunde 18, cuninges 18 (plus scazloos 1, scillinga 14, giscriban 21).

¹²⁰ Kossinna, pp. 50–53.

¹²¹ In A, k- in kinthisgi 7, and in krislanheiti 17, but ch- in chirichun 9; c in githanco 2, ubartruncani 7, giuuircanne 25, but ch in trinchanti 15; qu in niquam 9–10. In C, k- in kindisgi 7, kirichun 9, and in kristanheiti 17; c in githanco 2, uuerco 3, ubatruncani 7, trincanti 15, giuuircanne 25; qu in niquam 9–10. Gmc. sk appears as follows: A kinthisgi, C kindisgi 7, scolta 10, 11 (twice in C), 12, 14, sculdig 17.

¹²² In A, ci 9, Uncitin 14–15, 15(3); in C, ci 9, 24, 25, [I]nuncitin 14–15, 15(3).

¹²³ Initially chind 13, 53, chunincriche 13 (ch- on an erasure), chud 13, 28, chonnem 28, cheisuringu 34, chuning 34, chludun 65; cnuosles 11; quad 30, 49, 58. Medially folche 10, folches 27. Finally folc 51. In gemination otachres 18 (chrcs on an erasure), otachre 25, dechisto 26; reccheo 48. cc is written for HG hh in harmlicco 66.

¹²⁴ The other spellings of Gmc. sk are sciltim 64, scilti 66; scerita 51, sceotantero 51; scal 37, 53, scarpen 64; scurim 64; scritan 63.

¹²⁵ Pp. 186–188, 201.

¹²⁶ Sievers, §46.

¹²⁷ keliches 83.2, kelicho 84.4, kind 85.4, 93.3, 97.8, 99.2, kirichun 90.3; cind 89.3, cindheiti 92.4, ckindo 85.4; uncertain is kind r 94.2. Not counted is urceolo rc 84.4 ‘urceorum.‘

¹²⁸ trinkit 82.11(2), 87.4(2), uuirkit 88.6, uuerke 90.6, scalkes 99.2, karkeri 99.3, uuirkenne 103.3. In trinkan rc 87.2 he probably first wrote -ken. Uncertain is thenket rc 89.5.

¹²⁹ uuirche 87.8, uuecchit 88.7, folche 89.2, uorsenchit 94.4. γ perhaps wrote ch in eben-scalkes rc 99.4 and zikin rc 97.7. In arqueketa rc (uic?) 97.5 and arqueketa rc (u::) 97.8 he seems first to have written arquuicta; cf. Sievers, p. 406.

¹³⁰ §§47-48.

¹³¹ Sievers, §63.2.

¹³² γ surely wrote a double vowel in gicoos 82.12, eer 83.1, 85.4, 88.2, :ér (e) 90.6 (Sievers' text reads eér, his footnote :er [r]), :ér (e) 91.4, se: r (e) 84.1, see 84.4, eere 84.2, eeret 84.3, 5, flanzoota re in ta 84.7, huus 84.8, méer 88.6, gibóot r in t 89.4, giboot 90.3, 91.4, geet 92.1, 94.2, tool 92.6, tót: r (ot) 97.5. γ probably wrote a double vowel in gi:hortemo rc (hoo?) 84.7, ::áhtitun (ha?) 88.6, gibót: r (ot?) 89.2, brotes r (ootes?) 97.3. γ wrote aha = in gito:an (h) 100.1. Further possible examples are ér rc? 82.11a, arlóstu r 98.3, iár rc (nur der Accent ist alt) 102.2. γ possibly wrote see in sie rc 84.1, sie r 89.5, sie rc 91.1. Sievers, §63.2, is probably correct in believing that the ii's in driio 91.2, friiu 93.3, thriio 98.2 are to be interpreted as drījo etc., rather than as drīo etc.

¹³³ SalLaw scazloos 1, losii 7,forlaazit 9 (corrected from forlazzit; see StD, p. 55, note 6), wiib 10, huuse 16, heer 20, daar 29, mooter 30, bistoozzit 32. In the Hild there are no doublings, in the BasRec only niinpiize 15 (=ni inpiize), in the MonRec Luutra and, where we would expect a short vowel, Bootbarta, Miilrat; see Kossinna, pp. 32, 29.

¹³⁴ Sievers, §63.1.

¹³⁵ Sure are hérro rc 87.3, hérro rc 87.5(2),… úfscouuon zrc, n nc 103.1,… iár rc 103.5 (so Sievers' footnote; his text reads íar); probable are ér rc? 82.11a, arlóstu r 98.3.

¹³⁶ ::áhtitun (ha?) 88.6, gibót: r (ot?) 89.2, gibóot r in t 89.4, :ér (e) 90.6 (see above, note 132), :ér (e) 91.4, tót: r (ot) 97.5.

¹³⁷ Page and line references. In a corrected word 166.32, 172.5, 176.10, 193.5, 17, 18, 22, 27, 195.16, 196.34, 197.22–23 (circumflex), 198.27–28, 201.8; between corrected words 200.23; after a corrected word 175.17, 195.24–25 (circumflex), 200.29. Another example in a corrected word is 191.32–33, if this was corrected to íogilíh.

¹³⁸ Page and line references. 1 surely by the corrector 179.13; 2 probably by the corrector 178.7, 195.32; 8 in a corrected word 166.2, 8, 13, 18, 33–34, 35, 177.22, 195.20; 23 apparently by γ (the last 12 are in chap. 101–103) 167.7, 19, 178.6, 180.4, 14, 184.4, 190.29, 191.32–33, 192.11, 22, 198.28, 199.29(2), 31, 35, 200.5, 7, 33, 35, 201.17, 23, 28, 202.1.

¹³⁹ Cf. Sievers, §107.1.

¹⁴⁰ Braune, §288.3a.

¹⁴¹ γ wrote -e in theser nc 101.2; γ probably wrote -a in theset re (a?) 87.9, 94.3, 97.5, and in theser r (a?) 97.7; γ perhaps wrote -a in theser re 94.2,97.8.

¹⁴² γ wrote -e in thanne 91.4, 93.3(2), 96.2, 5, 6, 97.4, 98.4, 100.5, donne 90.6; γ wrote -a in thanna 87.6, 91.3, 95.5, 96.3, danna 87.5, 89.5, 90.6; thanna. re in e 88.8, 95.4, 98.2, danna rc in e 87.5; γ probably wrote -a in thanne rc 98.1, 2, 102.2, danne rc (a?) 88.2.

¹⁴³ γ wrote -e in arsterbe 84.2, queme 87.5, neme 90.5, arstante 91.4, uoruuerde 96.4, and analogically in tuo: (e) 102.2; γ wrote -a in uuerda 82.11a, uuesa 85.4, giueha 87.8, uorliose rc (-e::a) 90.5, and (with a rc in e) in uuolla 90.5(2), uorsacha 90.5, gilimpha 91.4, bisuuicha 95.4, 5, zisceida 100.3, and analogically in tuo: (a) 95.2, 98.1; γ probably wrote -a in bifahe rc 100.6.

¹⁴⁴ γ wrote -e in thurste 87.5, uuirche 87.8, thuruhfreme 87.8, gibure 88.5, 97.1, ruoge 88.13, hore 98.1,2(2); γ perhaps wrote -a in sente rc 88.2, hore r 98.2.

¹⁴⁵ Spellings with -e (page and line references): 166.24, 167.4, 35, 168.13, 34, 169.36, 175.27, 30, 178.16, 20, 179.5, 181.13, 20, 31, 182.15, 20, 36, 183.2, 3(2), 186.35, 187.26, 188.27, 190.15–16, 191.21, 193.10, 14, 38, 196.13–14, zuuei:: (es stand zuuene) 198.20, 38, 199.5, 16, 17, 27, 200.15, 16, 201.36; cf. also manege menigi 100.1 (Sievers, §110.1, Anm.; same phrase and ending twice in β, 22.3, 46.1) and manege ziti r in o 88.2. The sure and probable spellings with -a are listed in Sievers, §107.1aγ, except that there are 8, not 6 sure sina: 82.11a, 84.7, 85.3, 86.1, 87.3, 7, 8, 91.4; cf. also sinaorun rc in u 86.1. Of those which Sievers lists as rc, 4 are uncertain: aruiurte rc 100.6(3), giborane rc 100.6. Of the probable 13 -a (add sine rc 96.2, 103.5), all have only the final -e on an erasure, all are rc except heile r 88.11. γ wrote -ae in beidae 84.7.

¹⁴⁶ -a is possible in mitte rc 94.2. The spellings with -e are all present participles; page and Une references: 166.24, 168.7, 13, 169.1, 17, 18, 28, 170.36, 172.7, 8–9, 9, 175.4, 181.4, 182.2, 185.17–18, 33, 186.2, 11–12, 27, 29–30, 187.1, 190.10, 198.10. Another possible example is zuolouffante 92.6; see above, note 90, and Sievers, p. 380 and §110.1, Anm. Not included are gisehante rc in e, das Schluss-e aus i corr. 99.4, artonte: (n) 102.1.

¹⁴⁷ Included as e-spellings are Iosebe 87.1, Iacobe 87.3, Iohanne 88.11, 12, 91.5, Petre 93.2; not included are himile: r 82.11, uuazzare rc 87.4, uuazzare rc 87.4, gote … rc 88.13, cuninge rc 99.1. γ surely wrote -a in sambaztag: (a) 88.6, salza r in e 95.5, reua corr. in 100.6 (cf. Sievers, §64.1), and (with a rc in e) in uuara 87.9, morgana 92.1, toga 93.1, uuega 94.1(2); γ probably wrote -a in morgane rc (a?) 92.1. Cf. also zisamena rc in a, e 100.3.

¹⁴⁸ cunne 85.2, giruornisse 88.1, riche 90.6, 94.2, 3, touffare 91.5, bigangere 102.2, gibente 103.5.

¹⁴⁹ See Braune, §315.1.

¹⁵⁰ γ wrote -e in haltanne 84.4, be:tonne (t) 87.5, betonne 87.5, arnonne 87.8, infahanne 89.4, sellenne 93.1, uorkoufanne 99.2, miltenne 99.4, gebanne 100.4, gihiuuenne aus i? corr. 100.5, uuirkenne 103.3, losenne 103.5; 7 wrote -a in nemenna 85.4, uuerfenna 85.4, ezzenna rs 87.8, ezzanna 87.8, nemmenna 88.4, uuartenna rc in e 89.6, uorlazzanna 100.2, arslahanna 101.2; γ probably wrote -a in uuesanne rc 91.2, perhaps also in arslahanne rc 88.6, forlázzanne rc (e:::) 100.4; cf. Sievers, §§89b, 90b. It may be seen that in the a-verbs there are 3 spellings with -e, 7, perhaps 10 with -a; in the ja-verbs only 6 with -e; in the ōn-verbs only 3 with -e; in the ēn-verbs 1 with -a.

¹⁵¹ See above, notes 89 and 90.

¹⁵² Cf. Sievers, §103a.

¹⁵³ Add to §107ba sinen rc 88.13; to β, arslahet r 97.5, uuollent r? 88.13; to γ, sagant rc in e 88.13, sagetun r 102.1. There are 5, not 4 spellings of sagata etc. (86.2, 87.7, 88.5, 11, 12). The 9 -ant in the 3d pl. pres. of a-verbs might also be interpreted as the older, inherited ending; the spelling -ent, which appears always in the other scribes and 17 times in γ, has been borrowed from the ja-verbs; see Braune, §309. In the 3d pl. pres, of ja-verbs, γ writes only -ent, 8 times. There are no 3d pl. pres, forms of either a- or ja-verbs in the other Fulda documents.

¹⁵⁴ P. 60.

¹⁵⁵ Another possible agreement between y and the earlier Fulda documents is the treatment of medial p after long vowels. The other scribes write only f, but γ writes 9 f, 4 ff; see Sievers, §35 (to which add f in 99.2). There is only one comparable form in the older Fulda documents: BasRec nislaffe 16. The later Fulda documents show f: FrBapOath taufunga 12, FuldConf A slafenti 15, slaf-fenti 19 (slaf at the end of a line), C slafenti 15, 19.

¹⁵⁶ See Sievers, §§32–34.

¹⁵⁷ phuzi 87.3, phorzicha 88.1, hellephorta 90.3 (not in Sievers, §32); flanzunga 84.7, flanzoota rc in ta 84.7, fuzze 87.3, giflanzotan 102.2. Uncertain is … pfendingo rc 99,3. From Sievers, §32, delete phigboum 102.2(2), with ph- surely from Lat. f- (fīcus).

¹⁵⁸ sceffen 87.2, scefes 87.3; uncertain is … scephen rc 87.5.

¹⁵⁹ gilimphit etc. 87.5, 91.4, 103.3, gilampf 97.8, gilimfit 87.5, 90.4, 92.1, gilamf 87.1, 103.5, gilamf r 99.4, thamfta 99.3; hilf 85.4, 92.4, 5, uuelfa 85.4; uuerfenna etc. 85.4, 92.2, 8(2), 93.3, 95.1, 4, 5, 97.1, thorf 97.2. γ probably wrote f in uzuuirphu rc (fu?) 92.1.

¹⁶⁰ Cf. Sievers, §7e; Baesecke, §56.1; Baesecke, ZfdA, iviii, 259; Schröter, passim; Ehrismann, p. 289; Braune, §§58.3, 131.4, 135.3, 143.2, 167.3, 308.3 (Alemannic), 306.4 (Upper German).

¹⁶¹ Braune, §167.2.

¹⁶² Braune, §136.3.

¹⁶³ Braune, §144.

¹⁶⁴ Braune, §306.4.

¹⁶⁵ Braune, §248.6.

¹⁶⁶ Braune, §308.3. The 12 spellings with -nt are listed in Sievers, §13.5 (uuollent is 88.13, not 88.11). In the present of strong verbs there are 8 -nt, 18 -t through chap. 92, 13 -t after that. In the present of preterit-presents there are 2 -nt, 1 -t, all before chap. 92. In the preterit of all verbs the first 2 occurrences are tatun 84.3, 4; then through chap. 92 there are 2 -nt, 9 -t, after that 2 -t. There are no spellings with -nt in the present of weak verbs (44 -t), or in the irregular verbs (birud 84.8, sit 88.11, geet 92.1, 94.2, tuot 102.1).

¹⁶⁷ See Sievers, §13.5.

¹⁶⁸ Braune, §131.4.

¹⁶⁹ Braune, §131.5

¹⁷⁰ Braune, §149.

¹⁷¹ But for evidence that γ unvoiced g in final position, cf. trang 82.11 (= tranc). Spellings with -c for -g are common in the other scribes, esp. ζ; see Sievers, §28.

¹⁷² Braune, §39a.

¹⁷³ Included as an uo-spelling is ruorta rs? 86.1. Uncertain are zuogangente rc 84.7, huor rc 84.9, guot rc 85.4,… riorta… rc 88.1 (=ruorta), ár uuor rc 88.4, in tuom rc 88.8, tho uorstuontun rc 91.5, arsluogi rc 97.7. γ wrote uo for short o in voba 82.11a, vvuolaga 92.3, and perhaps for ō in tho: das o scheint aus u gemacht zu sein 85.4 (cf. tho … rc 91.5, tho r 94.1), as against 121 tho.

¹⁷⁴ Franck, §211.

¹⁷⁵ Braune, §380b.

¹⁷⁶ Braune, §§380c, 40.4.

¹⁷⁷ Braune, §320.

¹⁷⁸ γ wrote o in giloubtomes rs 82.12, gihortomes 87.9, uueritomes rc in u 95.1; but u in mohtumes 92.8. γ perhaps wrote o in giofnotun r 86.1, uuntrotun rc 87.7. Uncertain is ar-uiurtun rc 100.6.

¹⁷⁹ γ wrote o in gisahomes o durchstrichen und vn darüber geschrieben 95.1, but u in uorstuontumes 82.12. Included as a u-spelling is arquant:: (un) 91.6. Not included are … quadun zrc 82.11a, giengun rc 82.12, giengun rc 87.2,… uorstuontun rc 91.5. γ also wrote 8 u in the present of preterit-presents (page and line: 174.9, 10, 175.7, 176.3, 180.20, 183.19–20, 192.27, 193.10–11), and in birut 84.8.

¹⁸⁰ See Sievers, §112d.

¹⁸¹ Braune, §320.2.

¹⁸² Braune, §382.

¹⁸³ Braune, §§221.2, 255.

¹⁸⁴ Not included is dat. sg. théismon r in e 89.4, in which γ seems first to have written an acc. ending, and the uncertain spellings uuirseren r(s?) 88.5, selben r 92.1, hênen rc 99.4, uuingarten rc 102.2.

¹⁸⁵ Braune, §§221.3, 255.

¹⁸⁶ Cf. Sievers, §111.2. Included is dat. sg. theismon r in e 89.4. Not included are acc. pl. masc. nahiston rc 96.2, and … nahiston … zrc 96.5. (In this last word we would expect a fem. ending; cf. Sievers, §112f.)

¹⁸⁷ Where UG documents show her, the h is prothetic; see Braune, §§283.1a, 152a.

¹⁸⁸ Braune, §47a.

¹⁸⁹ See above, 4.1–2, 2.6.

¹⁹⁰ See above, 2.1–4, 3.1.