MAYNOOTH MEDIEVAL IRISH TEXTS IIL A First Old Irish | | Grammar and Reader including an Introduction to Middle Trish KIM McCONE DEPARTMENT OF OLD AND MIDDLE IRISH NATIONAL UNIVERSITY OF IRELAND, MAYNOOTH 2005© Kim McCone 2005 ISBN 0 901519 367 ISSN 1393-970X, Printed by ‘THE LEINSTER LEADER, NAAS FOREWORD Although (or perhaps because) it is intended primarily for students, including beginners, the present work has experienced a rather long gestation period of several years before finally making its way into print. Tn essence it has developed ‘ut of undergraduate (and some postgraduate) teaching over the last few years and iy thanks are due to those students who contributed to this process. In this regard special mention should be made of Louisa Moss and Ailis Ni Mhaoldomhnaigh, who are now engaged in postgraduate research towards a PhD. in NUI ‘Maynooth’s Department of Old and Middle Irish after successfully completing three years of study based upon the present course as it evolved. Both showed a useful aptitude for spotting mistakes and inconsistencies as this work underwent various different recensions until taking the form in which it has now been published, Needless to say, I alone am responsible for remaining errors and imperfections. Kim McCone Department of Old and Midale Irish NationalUniversty of Ireland, MaynoothTABLE OF CONTENTS (Chapter I: PRELIMINARIES 1 1A. Introduction 1 LA.L. The main phases of Irish 1 LA.2. Contemporary Old Irish sources 4 LA. Old and Middle Irish texts surviving in later manuscripts 3 LAA. The main aims of this book 9 LB. Spelling and pronunciation a LBL. Stops and corresponding frcatives 2 1B. Nasals and liquids 4 | LBS. The sibilant and f 4 | B.A. Stressed and final vowels 15 LBS. ‘Broad’ or non-palatal and ‘slender’ or palatal consonants 15 LB. Unstressed internal vowels 16 1LB.7. Diphthongs 16 LB. Initial mutations "7 (Chapter TI: Ta NOUN, ADJECTIVE AND ARTICLE 1 ILA, The noun 2 TLAA. Basic categories 21 (gender, number, cas; also word order, prepositions) ILA.2. The main patterns of noun declension 23 ILA. Full declensions 24 ILA. Further notes on noun declension 26 (the neuter, vowel changes, syncope and the dual) ILAS. Reading practice 7 ILB. Adjectives 28 ILB1. Basics 28 (gender, number, case and declensions) TLB2. Reading practice 29 (Gncluding notes on word order and syllabic metre) ILB3. Gradation 2 (Comparative, superlative and equative) TB. Reading practice 33 ILC. The definite article and mutations 3 TLCA. The article 33 (also the formation of adverbs) TL.C.2. Mutation of nouns and adjectives 35 ILC3. Reading practice(Chapter IU: THE VERB *T0 BE’ AND WORD ORDER TILA. ‘To be’ AIL-A.A. The main tenses and moods IILA.2. Copula, substantive verb and verbal noun IILA:3. Paradigms of the substantive verb TILA. Paradigms of the copula MILA Reading practice (beginning with anote on the augment) TILB, Word order pavieras TILBLL The basic order of constituents in the sentence or clause TILB.2. Patterns involving emphasis for topic or focus (the hanging nominative and the cleft sentence) I1L.B3. Reading practice (Gncluding a note on alliteration in verse) (Chapter IV: PRONOUNS, PREPOSITIONS AND NUMERALS IV-A, Pronominal elements TV.A.L. Independent personal pronouns TV.A2, Various adjuncts (emphatic particles, the anaphoric particle and sel?) TV.A3. Possessives IV.A.4 Pronouns and the cleft sentence IV.AS. Reading practice IV.AS. Suffixed and infixed pronouns with the verb ‘to be" IV.A. Reading practice IVB. Prepositions and thelr pronominal forms IV.B.L. The basic system IV.B.2. Paradigms IV.B3. Further adjuncts (cemphatic particles, the anaphoric particle or pronoun, ‘this’ and that’) IV.BA. Reading practice IVC Numerals: IVC, Cardinals 1V.C2. Ordinals IV.C3. Reading practice (Chapter V: THE PRESENT STEM, SIMPLE AND COMPOUND VERBS, (Onsecr PRONOUNS V.A. The present indicative of weak, strong end hiatus verbs V.Ad. Some basics or 7 6 (independent vs. dependent forms, conjunct particles and absolute vs. conjunet endings) V.A.2. Weak, strong and hiatus presents plus paradigms (plus farther notes on 1sg., 25g, deleition, 35g, and vowel alterations) V.A3. Reading practice (Gncluding a further note on metre) ¥.B. Compound verbs V.B.L. Verbal composition YV.B.2. Deuteroronie and proworone forms V.B3. Reading practice V.C.Sufftced and inftced pronouns V.Gi. The basi set YV.C2, Suffixing pronouns V.C3. Infixing pronouns (the panicle no infixing in general, class A and B pronouns) V.CA. Reading practice (including more details on metrics) VD. The imperative VDA. Basic inflection ViD.2. Compound forms and negation ViD3. Reading practice V.E. The imperfect indicative VEE&1. The forms V.E.2. Reading practice (Chapter VI: TH PRETERITE ACTIVE STEM "AND THE RELATIVE MARKERS. VLA. The preterte active VLALL. The s-preterite ‘VLA2. The e-preterite VLA. Reading practice VLA. The suffices preterite (‘eduplicated, d-, long-vowel and suppletive variants) VLA, Reduplicated s- and r-preterites VLA‘. Reading practice VIB. Relative markers VLBA. The basic system VLB.2, Relative endings VLB3. Infixed relative lenition Gnchuding the use of the particle no) 68 70 n 7 na m4 m4 " 5 6VLBA. Reading practice 96 1X3. The a-future 18 VIB‘. Infixed relative nasalisation 98 1X4, Reading practice M0 VBS. Reading practice 99 IXS. The /-future ut (Gncluding a note on the main features of Early Old Irish) 1X6. The sfuture wa ‘VLB.7. Relative forms ofthe object pronouns 100 TXT. Reading practice 1 (class C ingixed pronouns) IX8. The conditional ry VIBS. Reading practice 101 1X9. Reading practice 8 (Chapter Vit: Ta: easstvi AND THE DEFUNENE 103 Chapter X: Tae AvoMENT VILA. The passive 18 XL Basic forms and functions 1st YVILALL. The third-person present indicative an imperative forms 1G X2.The augment ro 1s VILA.2. The first and second persons 105, (preverbal, prevocalic and proclitic) VILA, Reading practice 105 X3. Reading practice 15 VILA. The imperfect passive 106 XA. Other augmented forms ‘7 VILA. Reading practice 1 (ed, com, other preverbs and suppetion) VILA.6. The preterit passive 0 XS. Reading practice 19 VILA. The past passive priciple and verbal of necessity 19 VILA. Reading practice 18 Chapter XI: FonTHER READING VIER. The deponent in XLI, Some recommendations 16 VILB.L Present indicative, imperfect and imperative i X12. Reading practice 168 (aio normal active for deponent endings, including with suffxed pronoun) (Gided Cheltchair mai Uthechar) VILB2. The preterite us VILB3, Reading practice 16 (Chapter XII: A Basic IyrRoDUCTION To MIDDLE Tis (including a note on metes based on stress and alteration) XILA. Introduction B XILB. Key sound changes 14 Chapter VINE: Tar SuanONcTIVE SraM m XILBAL, Merger of prociic vowels and of unstressed final vowels 14 VIL. Preimivares ai XILB2. Shift in syllable cenre Ws (presenipas, function and main stems) XILB3. Loss of hits and changes inthe system of diphthongs 176 ‘VIUL2. The asubjuncive Py XILBA. Reduction of procities 1% VIS, The e-subjunctive 15 XILBS, Consonants in VINA, Reading practice 16 XILBS. Orthography 7 VANS. The s subjunctive ” XILG Significant morphological developments outside the verbal sytem 179 VLG. Reading practice a XILCAL. Loss ofthe neuter rAd VIILT. The past subjunctive 3 (anda noe on hypercorect aslisstion) ‘VIILS. Reading practice 134 XILC.2. Reading practice 179 XILC3, Confusion of nominative and acusative forms 1» Chapter 1X: Tur ForuRE STEM 1 XILC4. Nom ac. pl. -C9 versus gen. pl.-C 180 TA The ure 1 XILCS. Changes in deciension ist 1DC2. More on syncope an its effects 16 XILCS, Adjective and atte 8 (devoicing, assimilation of quality, deleition and epenthesis) (including the formation of adverbs)XILC. Reading practice XILC8. Numerals XILLC.9. Prepositions (including pronominal forms) XILLC.10, Reading practice XILD. Personal pronouns and relative marking XILD.L. Stressed independent personal pronons XILD.2. Suffixed and infixed pronouns (ncluding the petrification of neut. sg, forms and prosthetic) XILD.3, Relative marking XILDA. Reading practice (Gncluding farther notes on syllabic metres) XILE, Further major changes affecting the verbal system ‘XILE.L. Verbal stems and personal endings (@) The spread of weak atthe expense of strong stems (b) The endings ofthe passive, the depenent and the Ipl. () The present indicative (@) The present subjunctive (©) The future (The preterte active (g) The preterite passive (8) The imperative and imperfect endings XILE2. Compound verbs (Coubted inital preverbs and new simple verts for old compounds) ‘XILE.3. The augment ILE. The copula XILF. Reading practice ‘XILF.L. A poem illustrating different metres ‘XILF.2, Comthéth Léegairi “Loegaire’s Conversion” Vocaputary CHAPTER I PRELIMINARIES A. INTRODUCTION. AcL. The main phases of Irish, Trish (or Grelic, including the Scottish and Manx varieties; Ot. Goldale, Modlr, Gacilge te.) isa Celtic language whose closest relatives are the three members ofthe British Celtic family, namely Welsh, Comish and Breton, Irish and British constitute the two branches of Insular Celtic, which probably developed certain peculiarities (most notably the double ‘absolut’ and “conjunct’ verbal inflection discussed in IILA.3 and V.A. 1c) in Britain in the course of several centuries straddling the middle of the first millennium B.C. before being brought to Ireland some time later. Thereafter Trish (or Goidelic) and British began to diverge on their sepecate islands. The Insular Celtic languages are more distantly related to imperfectly documented and long dead Celtic languages on the Continent, notably Celtiberian in Spain (some inscriptions from the first and second centuries B.C.) and Gaulish (documented chiefly from various inscriptions ranging in date from the fifth century B.C. to about the thd century A.D) in France, Norther Italy and a swathe of teritory stretching eastwards along the Danube as far 1s what snow central Turkey. All ofthese languages together make up the Celtic family, touse ‘anow conventional linguistic term established by the great Welsh scholar Edward Lhuyd inthe carly eighteenth century. This i then related ata further remove or removes to various other language families, notably Germanic (eg, English, German and the Scandinavian languages), Italic (notably Latin and ts modem Romance descendants such as Italian, Spanish end French), Baltic (notably Lithuanian and Latvian), Slavic (e.g. Polish, Serbo-Croat and Russian), Albanian, Greek, Armenian, Indo-Iranian (further subdivided into Iranian and Indic branches) as well asthe long extinct Anatolian (notably Hitite with records from the second millennium B.C. in Asia Minor, present-day Turkey) and Tocharian (documented from the sixth to the cighth centuries A.D. in the Karim Basin, now in eastern China). Ultimately all of these are ‘branches of the vast Indo-European family of languages assumed to descend from @ common ancestor known as Proto-Indo-European tha has left no records (since its speakers could not, ‘write) but must have been spoken around the fourth and third millennia B.C. somewhere in aster Europe or Western Asia, an area ust north ofthe Black Sea currently beng the best bet Irish seems to have been steadily going its own way for the best part of a thousand years in Ireland before it began to be written down, frst on very brief Ogam inscriptions around the early fifth century A.D. and then in continuous texts in manuscript in the course ofthe seventh century A.D. Prior to thisits precursor shared various demonstrable nodes of development with other languages, namely an Insular Celtic stage (with British Celtic) around the middle ofthe first millennium B.C., a Proto-Celtc stage (with all known Celtic languages, deed or living) around the later part of the second and the early fits sullleanium B.C, and finally & Proto-Indo-European stage that probably drew to an end in the course of the third millenniamn2 B.C. Inthe absence of written documentation these stages are only accessible by means of scientific historical linguistic comparison and econstriction. On this basis the development of Irish can be traced beck, at least in outline, some five thousand years, although its directly recorded history begins no earlier than some sixteen hundred years ago with the scanty Ogam. evidence and the language has only been adequately documented forthe last thirteen centuries or so since the emergence of a manuscript tradition. The long undocumented period can be regarded asthe language's prehistory, andthe shorter period (albeit a longer one than can be ‘boasted by eny other Wester European language except Latin) since more or less continuous records began as its history. The latter, of course, iste main concer here, ‘The linguistic phase from the fifth tothe sixth centuries A.D. is best designated Primitive Irish and is marginally historical by virtue ofthe exiguous Jocumentation provided by a couple of hhundred short and formulaic Ogam inscriptions These and comparative linguistic considerations combine to show that it was a watershed during which Irish underwent major changes that transformed an essentially ‘old Indo-European’ profile similar to that of the Continental Celtic languages or Latin, say, into a sgnificantly different type, certain Key features of which have characterised Irish ever since (CE. the remarks on Ogam INIGENA versus ‘Old Irish ingen ‘daughter’ in B.8 below), ‘The reliable manuscript records ofthe language ofthe seventh century are few and far between but they do suffice to show that the Karly Old Irish of this period was not unduly different from the ‘classical” Old Irish of the following two centuries, the main divergences being of a ‘relatively trivial phonetic nature (see VIB 62). The tenth century saw the emergence ofa stage ofthe literary language Imown as Middle Irish (see XID that was characterised by considerable innovations and fluctuations in morphology above all,especially where the verbal system was concemed. Around the end ofthe twelfth century this in tum gave way to the appreciably more streamlined grammatical system known as Early Modern Irish, which continued in use until about the end of the sixteenth century. Thereafter texts increasingly show various dialeet and ‘other features sen in the Modern Irish of the present éay Irish, then, has a continuous textual tradition in the Roman alphabet going back at least as far asthe seventh century A.D. This ‘grammar and reader is chiefly concerned with the early or ‘Old Irish’ part ofthat evolution, a period extending roughly from the late seventh to the early tenth century A.D., but the final chapter offers a basic introduction to the subsequent ‘Middle Irish’ stage extending down to approximately the end ofthe twelfth century. ‘The long period of development culminating in Old Irish as known tous from various sources didnot, ofcourse, take place in complet isolation. As inthe case of other recorded languages, certain external influences can also be detected. It has, for instance, sometimes been claimed that, whereas the basic sounds and forms (or phonology and morphology) of Old Irish can usually be accounted for as the outcome of a continuum of internal change stretching hack through Proto-Celtic to Proto-Indo-European itself, certain key features of the language's syntax, particularly where word order is concemed, are due to the so-called ‘substatun" {influence of a non-Indo-European language o languages encountered by early Celtic speaking settlers in Britain and Ireland, However, this can be no more than unproven and unprovable speculation in the absence of any record of the pre-Celtic speech of these islands and this approach has been rejected by most scholars inthe field forthe good reason that plausible {intemal derivations ofthe features in question are available. Various studies of other languages have revealed that external influences are most likely 0 rake themselves felt in the field of vocabulary, and Celtic, Insular Celtic and Irish are ro exception to this tendency. Although the bull of their vocabulary can be shown to have widar Indo-European connections, Irish and British Celtic nevertheless contain anumber of obviousiy related words suchas Olt. muceand MW moch-en ‘pig’ (< *mukk-), Olt. brat and MW bretk-yn ‘cloak’ (< *bratt.), Olt. fannall and MW gwennaw 'swallow" (< *we/aNl-) that nt only lack ‘obvious cognates elsewhere in Indo-European but also have a distinctly non-Indo-Europeen Took. That being so, they are likely to have been taken over from an otherwise unknown non-Indo-European language or languages at prehistoric Irish (or British), Insular Celtic or Proto-Celtic stage. Given the already mentioned inadequacy ofthe documentation of Gaulish and Celtiberin, lack of attestation of similar forms thee so far hardly suffices to prove that, ‘these extemal elements made their way into the language(s) in question at an Insular rather than ‘an earlier Proto-Celtc phase. On the other hand, attestation in one or both of these does constitute solid grounds for positing borrowing as early as Proto-Celtic, a case in point being the PC but apparently non-IE *kurmi ‘ale’ underlying Olt. cuirm, OW curum (MW cwrw), ‘Gaul, curmi, Celi, kurmi-. However, words ike these afford no more than occasional glimpses into early preistorie borrowings from unidentified sources. ‘Unsurprisingly, the frst securely identifiable extemal source of borrowings into Irish (and, fir ‘hat mater, British) is Latin. As the official language ofthe Roman Empire and ofthe Catholic (Church it ot only survived the former's downfall but also sueceeded in extending its reach ‘beyond the old imperial boundaries. Apar romahandful of arguably pre-Christian borrowings relatingto trade and seafaring such as Olt. ingor ‘anchor’ (Lat. ancora), the bulk of early Latin Toanwords sms, predictably enough, to have entered Old Irish in the wake ofthe introduction ‘of Christianity into reland, which occurred a least as early asthe first half ofthe fifth century AD. Detailed studies ofthe various ways in which ¢ considerable number of early borrowings from Latin were adapted into Irish have shown that this was a virtually uninterrupted proceis etween a more retracted or non-palatal (oroad’) and a more fronted or palatal (‘slender’) articulation of each and every consonant. This contrast between non-palatal (C) and palatal consonants (C’ as in /in’y’@n/ in B.1,/sub'e/ and ‘ke:l'y in B.4 above) applies equally to Middle and Modem Irish. A consonant before stressed a, 0 and u (short or long) was ‘non-palatal and a consonant before stressed e and (short or long) was palatal as a rule in Old Irish. Being quite automatic, the distinction between palatal and non-palatal initial consonarts ‘was non-phonemic and hence is not marked in eases such as penn (phonemically ‘pen! but phonetically [p’en), tech tex! (or (tex), cen ken (or [ke], ben foen/ (or (ben), gel gel’ (or (g’ell). The distinction between palatal and non-palatal quality was phonemic after a siressed vowel. No orthographic change took place when the following consonant was ‘on-palatal: e.g, cath ‘ka, bds foal, secht Ise, fe fer, th AOI, fir If, rom) tro,6 sl6g slo: crwth her, drigh /éru:8/ in B.4 above, However, when the following consonant was palatal a glide # was usually but by no means invariably inserted between stressed a, 0, u or (less regularly) ¢ and the palatal consonant, especially where this was word-final : 6, ‘maith ‘good’ /ma®’ (mod. maith, dil ‘desr” [a Berd or betrid ‘bears’ fhet’88'/ (mod. betreonn), céle orcéile'fellow” Feil el ithid eats’ f0"80'/(mod.tkeann),fir of teat fi (nod, fir), -oirg slays’ for’ s1dig‘of host’ slo mod. slua),sude or sue siting, eat ‘Isu8'e! (rod, sul), sil ‘eye’ /su:l'/ (mod, sil) and sil! or sul ‘eyes" (sui (emo. sie). [Note thatin Ula trish, exceptin the ease ofa iphthong on whicl see B.7 Velow), only the fist ina sequence of two vowel leters ina sessed slate is to be pronounced as norm, the second (in practice, being a mere glide to indicate a following palatal consonant. However, in the case ofan unstressed final vowel the glide was inevitably inserted between it andthe preceding consonant. Anon-paatal consonant was dirctly followed by final ~a,-o and -u in spelling (¢, cosa, silo and fru in 8.4 but after e palatal consonant a glide was usually inserted to give -ea, -e0 and -iu asin -léicea ‘may leve’ Neg’ al, Iictheo ‘of leaving’ Teg8 ol, guidiu I pray” gu0"w). Conversely -e and were writen directly afer a palatal ‘consonant (eg. sue, cfile and si above) but aglide a was optionally inserted before them after anon-palatalo giv luge or lugae “oats Atel, lug or lgai ‘fan oth’ Auyil. As ‘ comparison of su()de with Iug(a)e indicates, ambiguity as to a consonant’s quality could result fom failure to writ the appropriate glide before or after it B.6. Unstressed internal vowels. (Old Jong vowels were shortened in unstressed syllables (c.g, tt ‘goes’ but con:é(i)-et “goes ‘in with, accompanies") but some secondary long vowels remained (e.g. cenél ‘kin’. Unless in absolute final position (see B.4 above), all short unstressed vowels (including shortened long, ‘vowels except a! (eg. lé(efud ‘leaving’ /e:g' ub, romus ‘measurement’ Novus) had fallen together as an ‘obscure’ mid-central phoneme /6/ or ‘Schwa’ (pronounced like the ein Eng. ‘mother ee). This was then automatically coloured by the surrounding consonants, being spelt ‘a between two non-palatals (CaC; e.g. marbad ‘ili’ /marved), e between a palatal and a ‘non-palatal (C" eC; e.g. aicned ‘nature’ /ag’n’90), between two palatals(C'sC’ seg, be(rid ‘bears’ /ber’98’/) and (a)i between a non-palatal anda palatal (C(a)iC'; e.g. tabir or tabair “give!” aver’), B.7. Diphthongs. In this case two vowels are combined to produce a sound sliding from one to the other. Old Irish had the following i-diphthongs: /ai spelt af or de, fi! spelt of or de, and Ai spelt ui (all > mod. ao()) asin colm or eéem ‘fir"(nod. caomh), moidem "boasting® (tod. maiomh), drut “druid (mod. draoi). The first two ofthese were already beginning to fall together in Old Irish, as in maidem alongside commoner moidem in Wp. “he mark of length is often omitted in ‘manuscripts and, where its writen, may be placed quite arbitrarily on the frst or the second clement ofthe diphthong. Modern editors prefer consistency, placing the length mark over the ‘rstelement (e.g. cdem or cdem) except where this would cause confusion with long Vowel plus ” # (e.g. colm or caim as opposed to di or di for long f:/ ot /a:/ plus palatal glide), Short and long 4,¢, could forma diphthong with following u (e.g. dat. sg, new ‘strength fur man’, cenéul ‘Kin’ or Ise. -biu ‘I am wont tobe’) but au had mostly become u (eg. acc. pl Bally or Bulle ‘limbs, members") and duu mostly 6 (Isg. a:tdu or a:16 “I am’) in Old Trish, Two further diphthongs were fa! fa (e.g. cial ‘sense’ /kial/, mod, call) and /ual tia (e.g. slag ‘hos? ‘sluxy/, mod. sua, alternating with slg /so:Y). BB. Initial mutations. twas remarked inB.1 above that conconanto were Hable tobe lenited after vowel within word and thatthe intemal vowel responsible for this change was then quite often lost by syncope 80 that a lenited consonant could stand after another consonant in Old Irish. This, Jenition coutd also occur between two closely connected words such as article plus noun ce noun plus adjective, the final vowel responsible then being prone to be lost by so-called apocope well before the Old Irish period (compare apocopated and syncopated Olt. ingen ‘daughter’ with Ogam IN1GENA in B.1(v) above). As a resulta lenited initial consonat could stand efter a word ending in aconsonant in Old Irish, asin ingen chéem ‘fair daughter’ /inya ‘oii! (mod. inion chaomh), where ch- for c- is due othe leniting effect ofthe -2 with which Ingen once ended, Since it was no longer obvious that a lost final vowel had caused this change, alterations such as that between ingen chdem with lenited and fer cdem ‘fair man’ /ferkoi¥/ (mod. fear caomh) with unlenited following adjective were seen as depending on grammatical factors such asthe ‘gender of the noun. Old Irish, then, had a system of grammatically conditioned initial ‘mutations thats still well preserved in Modern Iris. Consequently even in intial position fi, ‘nhl and Po, fl, /a/ could be lenited by certain preceding elements to), /8/, and iv, ‘ry! withthe same spelling conventions applying as in B.1 above: lenitedp--,c- were writen Ph, th, ch- ut lenited b-, d-, g- were not distinguished in spelling (until the Modem Irish invention of dh, gh-) from the unlenited initials. Likewise initial m-, -,r-, -were always \wrtten single when Ienited but were sometimes written double when unlenited asin B.2 above In initial position /s/ became fl and f disappeared completely through lenition but, apart om. sporadic omission of lenited in spelling, these changes were not indicated in writing until the ninth century, when a dot came tobe optionally placed over leniteds and fto give and f. In certain circumstances words ended in anasal that affected a closely connected following wore before the final syllable was lost. The result was a further mutation known as ‘eclipsis' ot nasalisation that changed intial /p, /, to ba, // (til written p- -,c- as in the case of internal fy, /4, /g after a vowel in B.1),/ ov (sil written) and prefixed a nasal to iit Pol [8 gf ora vowel (whence mb-,nd-,ng-, nV), although original mb, /nd! and /yg/ were probably assimilated in pronunciation to /mi,‘N/ and /y/ during the Old Irish period, Certain elements that neither Tenited nor nasalised could prefix /b/ to a following vowel but the Presence or absence off in spelling is no sure guide, since it was a ‘silent leterin Old Irish asin later Latin (and still in French words such as honeur ‘honour’; e.g. Olt. ised ors hed “it ie” eB (mod, is ea) verous nied oF ned ‘its not fais eB! (tod. mt hed).18 owel) ‘These mutations are tabulated below (e stands for any V. Unmodified — Lenited ‘Nasalised Prefixed h-(V-only) op ‘9 ph to! Pp wot 18) the ia Moe If che gf ° hob Nob ‘mb! > eal b= ia & wo & Ind! > IN nde Woe oe fap >I 6 Mmm) oe aw (mn)- INI nfo i nin) ikl 1 Aloe ed 1) WW wv wy 10 wf WW Wl £ hi i) WI s Jel (ie Fe ie ine ne Ie! (ae- “The above table makes it clear that Old Irish orthogrepy only accorded limited recognition to initial mutations. Lenition ofp, ¢, ¢ was clearly shown but nasalisation was not, The reverse applied to d, dg, nasalisation of which was regularly indicated whereas lenition was not, Double spellings of m, n, r, 1 definitely denoted non-lenition but single spellings were ambiguous, At first no mutation of for s was expressed in writing but optional use ofa dot later emerged as a clear indicator of lenition, Owing tothe inconsistent use off only nasalisation cof an initial vowel was clearly indicated in Old (as opposed to Modern) Irish, As in noted B.1 above, Modem Irishs often ahelp in resolving earlier orthographic ambiguities. The examples below will use the thid person possessives. Comparison with related languages shows that 4 “his, is’ (eniting) had an old final vowel, that f. a *he, its’ (non-mutating except for prefixing h- to a vowel) once ended in -h <-s and that a ‘their’ (nasalising) once ended in -n ‘<-m, The same mutations survive after these elemerts in Modern Irish. _penn ‘pen’ fpeN! (04, peann): aphenn ‘is pen’ /a fe! (mod. a pheann), a penn ‘her pen! Ja ‘peN! (mod. a peann), a penn ‘thei pen’ Ja beN! (mod. a bpeann). tech ‘house’ ite (tod. teach): a theck ‘his house fe Ge (mod. a theach), afech ‘her house” (a tex/ (tod. a teach), a tech ‘their house’ /a dey! (mod. a dteach). ‘cenn “bead! ike (food. ceann): a chenn ‘his hea’ fa en! (mod, a cheann), a cen ‘her head” Jaken! (mod, a ceann), a cenn “their head /a gen! (mod. a geeann). 16 ‘cow" (mod. bean): a bé ‘his cow’ /a vor! (mod. abhé), a é “her cow /a bo: (Mod. abé), 12 mb ‘their cow" fa mb)or! (mod. a mb) 9 deug ‘drink’ (mod. deoch: a deug ‘his drink’ a Beuy! (mod. a dheoch), a deug “her drink’ /a eu! (mod, a deoct), a ndeug “their drink’ Jandeuy/> fa Neu! (mod. a ndeock). ‘galar ‘illness’ (mod. galar:« galar ‘hs illness’ /a Yale (mod. a ghalar),a galar‘heillness ‘is galee! (mod. a gaiar), a ngalar “thet ilness a ygal6r/ > fa yalOr! (mod. a ngalar, ‘mac ‘son’ /tmak/ (tod, mac): a mac *his son’ /a Yak (mod. a mhac), a m{m)ac “her son’ /a ‘mak’ (mod. a mac), a m{m)ae “their son’ Ja mak/ (rod. a mac). inert ‘strength’ (Ner! (mod. neart: a nert ‘his stength’ /a ner (mod. a neart), a n(njert ‘ter -smengih”/a nerd (mod. a nears), a niet ‘their sength’ fa Nerv (mod. a nearn), riking’ /Ri:) (tod. i): ari his king’ Jari: (mod. a7), a (7) ‘her king” (a Ris! (mod. a), ‘a r(r)i “heir king’ /a ni! (mod. ar). lam “hand! /.a:t (mod. lémh): a lm “his band’ fa a:0/ (mod. a Lim), a (0)ém “her hand? a asi! (mod. a Kémh), a i()ém “their hand? fa Lact! (mod. ald), fer ‘twan” lel (mod. fear): afer (later also a fer) ‘his man’ /aer! (mod. a fhear), a fer ‘ber rman’ fa fer! (mod. a fear), a fer ‘their man’ /a ves! (mod. a bhfear). sil ‘seed Js: (mod. sof: asi ater also asi) ‘his seed” Jail (mod. a sol, a sil‘her seed asi (mod. a siol), a sil “their seed’ (a si (mod. a si). ‘enech “face, honour’ fen’ (mod. eineach): a enech ot a henech ‘his honour" fa. en’ @x/ (mod. a eineack), a henech ora eneck ‘her honour’ /a hen’ 9x (mod, a heineach), a n-enech "their honour /a nen’@{f/ (mod. a n-eineach), Itisto be noted that, since p-only occurred in loanwords (normally from Latin; e.g, penn above from Lat, penna), there was no pre-existing pattern of mutation to which it could be directly adapted. Consequently itremains unlenited more often than not in Old Irish, whence frequently «penn ‘his pen’ /a peN/ etc, and was presumably equally resistant o nasalisation (a penn ‘their pen’ /a peN/ etc.) despite the ambiguity of Old Irish spelling on this point. Only gradually did extrapolation ffom the patterns ¢/U, len. sh /0/, nas. // or Mi len. ch //,mas.¢/ produce corresponding labial series p/p, len. ph nas. po.20 CHAPTER I ‘THE NOUN, ADJECTIVE AND ARTICLE A. THE NOUN. AA. Basie categor (@) Like Moder Irish, French or German (but unlike Modera English) Old Irish ascribed grammatical GENDER fo its nouns. Whereas Moder Irish and French have a two-gencer opposition between masculine and feminine nouns, a three-gender opposition between ‘masculine, feminine and neuter applies o the nouns of Old Irish as to those of German sd Latin (reduced to two in descendants of Latin such as French; cf. the development from O1dt0 Modern Irish). Although this grammatical distinction correlates to some extent with natural gender (eg. Olt. ben ‘woman’ and fer ‘man’ are grammatically feminine and masculine respectively, this is very often no the case: eg. flaith (“lord clock (£) ‘stone’ but lie (m.) ‘stone’, briathar (£.) ‘word’ but focull(n.) ‘word’. As in Modem Irish, French, German or the like, the gender of a noun in Old irish was liable to have various grammatical consequences aid these will be encountered at various points below. (Old Irish nouns make a familiar distinction between singular and plural NUMBER (e.g. ben woman’, mn ‘women') but also have a special dual form used exclusively after the numeral “two! (eg. a mnai ‘two women’) ‘They also undergo various formal changes in CASE depending upon their grammatical function inthe sentence, Some languages (uch as Moder French and English) make next to no formal case distinctions preferring to use other devices such as word order and prepositions in oréer to indicate the grammatical fnctions in question: e.g ‘the man!" Killed "0 the dog* ‘with?©om the stone" versus “the dog"™= killed "=" the man" with!™™="™ (its) teeth (oF “the stone!™™#7 killed 1 the dog!™*°", Other languages rely more heavily upon changes inthe formof the nouns themselves in such circumstances. For instance, formal changes ofcase ‘would be the means of indicating these basic grammatical relations in the following Latin equivalents ofthe above three sentences,namely homo! canem! lapidd™%0")pecist*0 ‘versus canis! hominem dentibus™=™%0% geciif™\ or lapis*™) canemi™ occisit™=# ‘Whereas a change in their order i prone to alter the basic grammatical relations between words in Modern English, a in ‘the man killed the dog’ versus the dog killed the man’, in Latin sul changes can be made for different types of emphasis without affecting the basic grammatc.al, structure and hence meaning of the sentence. Thus canem lapide ocesit homo or lapide kono ‘canem oceisit still mean ‘the man killed the dog with the stone’ (outwith emphasis upon “dog” and ‘stone’ respectively) and hominem canis occist dentibus or dentibus hominem occisitcaris still mean ‘the dog killed the man with its) teeth’ (but with emphasis upon ‘man’ and ‘teeth respectively),2 (Oe Irish occupies a oughly intermediate positon insofar as it has quite rigid word-order pater end males extensive we of peposone iy, Moder Engh but do omaly fisinguithes five separate grammatical cases rather likeLatin, which has six. Whereas Modem English word orderis typically Subject-Verb-Object (SVO) as inthe above examples, the basic order of these constituents in an Old (Middle or Modern) Irish sentence is Verb-Subject-Object (VSO) followed by other elemeats such as prepositional phrases and adverbial expressions: ¢g., béoigiir™ in spina) in corp in feckt so™™*) ‘the spirit Vivities™ the body! (at) this time!" (from the Warzburg Glosses already discussed in A.2). The five Old Trish eases and their basi uses may be stmarised as follows: (©) The nominative denotes the SUBJECT ofthe sentence (typically its main point of reference and very often the performer(s) of verbal action) as well asthe predicate of the verb ‘to be” Janown asthe copula, Thus rt in fr ‘the man (nom. sing subj.) goes" iagait ind fir ‘the men (nom, plur subj.) go" and és sacart infor ‘the man (nom. s8. Subj.) is priest (nom. sg. pred.) (©) The vocative denotes someone (or rarely something) addressed and is regularly preceded by the particle a (leniting), Except in masculines of groups I and IVa below (eg, « fir “O tan’), the voc. sg, is identical with the nom. sg. (eg. a ben ‘O woman’). The vor. pl. is ‘identical with the ace pl in all noun classes (e.g a fr "O men’). (@) The accusative designates the DIRECT OBJECT (‘ypically the experiencer(s) of the action) (of an active transitive verb. Thus cavaid in fer in mnei ‘the man (nom. sg. subj.) loves the ‘woman (acc sg. 0b)", carait ind fir inna mnd ‘the men (nom. pl. subj.) love the Women (ace. pi, ob.) and carat ina mnd inna fru ‘the women (rom. pl. subj.) love the men (ace. pl. 0b}. ‘The accusative is also used with certain prepositions such as la ‘with, by’ or imm ‘around, bout" and itis to be noted that, when a verb shifts fromactive to passive, the accusative object ‘becomes nominative subject, as in carthair in ben la for ‘the woman (nom, sg. subj.) is loved bby aman (ace. sg. after prep. la)’ or cartair inna mn Ixsna fra ‘the women (nom. pl. sub.) are loved by the men’ (ace pl. after prep a). Itis also frequently used in expressions of time such as in fecht so “(at) this time" above or fechtn-and ‘on(c)e (upon a) time” (lt. (at) time there"), (6) The genitive typically indicates the dependence of one noun upon another and as such Almost invariably follows another noua a Ol sh requently denotes possession andcan almost always be translated into Englishas“of or seg tech indir ‘the man's house’, benn ibe ‘mountain's peak’or ‘a peak ofamountait, tg fer ‘men's houses’ benna inna sléibe “he mountain(s’ peaks" or ‘the peaks of the mountains’ The so-called “objective” genitive is seen most clearly afer verbal noun (see IILA20) in Old Irish: eg oc precept soiscél “preaching the Gospel” (it. “at preaching of the Gospel’; nom. sg, soisedle, class IVa in A2 ‘elow). It is sometimes used on its own in expressions of time such as cecka biadnae ‘every year or as predicate ofthe copula in phrases suchas arom Dé ‘we are God's or ni tobi itis dot of prot, fie no ue" 2 (0 Apar froma couple of vestigial independent uses, eas an altemative tothe accusative in expressions of time such as ind fecht so (at) this time" or fecht and ‘on(c)e (upon a time” (see also TLB.3 on the dative of comparison and IV.C.1b on a usage with personal numerals, the dative is only used in Old Irish prose with those prepositions that do not take the accusetive (and sometimes with certain prepositions that tke ether acc. or dat). Thus 6/do fiur ‘frome ‘aman (dat 3)" (6/do * dat.) ori ig “ina house (dat. sg)" vs. tech ‘nto a house (ace. sg.) ( + dat. or ace.). The dat. pl. ending of all nouns is -(aib, asin 6/do feraib “fromto men’ or i tigib “in houses vs. tlge “into houses" Itcan thus be seen that some Old Irish prepositions govern the aceusative, others the dative ‘and still others the accusative or dative case. ‘A.2. The main patterns of noun declension: Like other early Indo-European languages such as Latin, Old Irish as a numberof differen! declension classes for nouns. This feature is still seen in Modern Irish and nurabers I-V below refer to the pattems underlying the corresponding declensions I-V in traditional grammars of Modern Irish. Where necessary, these ae futher subdivided by means ofan additional a, or «. Inthe case of V a mnemonic referring tothe type of consonant involved (lenden. = lenited dental, unlenden. = unlenited dental, gut, = guttural, nas, = nasal) seemed a more helpful subclassification, Category V1 is residual and does not correspond directly to a Modem Irish declension. There remain, of course, a number of individual imegular nouns (eg, ben £, ge. sg. mnd ‘woman’; dia m., gen. sg. dé ‘god’; bé f, gen. sg. bé ‘cow’; siur , gen. sp. sethar ‘sister’ that do no fit properly into any ofthe six pattems below. The designations ‘o-stems’ ‘etc. in brackets after the Roman numeral relate to the historical classification used in such standard descriptions as Strachan’ Old-Irisk Paradigms and Selections from Old-Irish Glasses ‘and Thumeysen’s Grammar of Old Irish so that these can be easily referred to when desired. Knowledge ofthe form ofthe nom. sg. (the basic citation or ‘dictionary’ form) ane the gen, sg. ofanoun is usually sufficient in order to assign ito the correct declension Is also necessary to know the gender (or another case form such asthe ace, sg. inorder to distinguish Il from Vib. Even in the absence of information as to gender confusion between IVe and VIb is impossible because the former are always disyllabic whereas the latter's stem is basically ‘monosyllabic and ends in -m(m). The list below is intended as a ready reference for the ‘identification of given noun's declension from the information supplied inthe vocabularies ofsuch essential introductory works a Strachan’s Stories from the Tain snd the aforementioned (Old-trish Paradigms... and Glosees. 1(o-stems) nom, C, gen. -C’, mJ/n. e.g, mac, maic (.) ‘son’; scél, scéu (n.) ‘tle’. 11 @-stems) nom, -C, gen. Ce (rarely -Ce), f: eg, tach, nlalthe (€) ‘petty kingdom Mila (-stems) nom. -C’, gen, -Cola, m/f. eg, Dsachuil,llachallora(m.) ‘eowherd’s» iio ee bla, iad) cor fa, m.fn.: €.g,, cath, catho/a (m.) ‘battle’, rind, rendo (8.) tm sens nom 4 ge. Cala, A 6 mi ata (moti ride) Va (io-stems) nom. -e, gen. -, m/n.: eg., dalt(a)e, dalt(a)i (m.) ‘fo a € iste) nom. go: math, mathe) ‘godnes We desens nom pen. {se laden Blade () Se ‘V (consonant stems) nom. -, gen. ~C, mf, i ‘V lenden. (j-stems) nom. -V (oft. -u), gen.-V4, m/f: ¢.g.,arae, seam i zut stems): nom. -, gen. -(V)ehip/e, mE: e.g. 7, rig (m.) ‘king's yee a ih crac (asec : talam, talman (m.) ‘earth’, cti, con (m.) “hound”; : (s-stems) nom. -C, gen. -C’e, n.: nem, nime (n.) ‘heaven’. : ‘rad (m.) ‘chariotees’ Fakes Han pt bende gop ena mane cn ahah aa acsearle ops apunt cn Soeventa iy mndam poo soupinlsenve Seopa uly I Osc cn nt ea lr pea ve Ca TV a henner na weve leorendr ale see ut uC echogenic aa ced en cara silat betors a (= ia een an ee spss Vb, en. cng es try ene een vores a 10 ili SEinranh hvac o-c ert 2 Omen whe aaa eta ih som op scape of cs Ve A ee et ane mt rosie ms nfs. Se NS mite Ve Ie aie orton ee dente eyo aoa yeu oto npn 13. . ‘fellow’ 4 TV ‘prayer'IVe'year" sing 1 ‘man’ (m.) IVa ‘fellow’ (m.) “kingdom IVb ‘prayer’ mivoc. ferffir Aylelcé()K | tiath ~—gu(ijde liad oe rim a gen fr etolt tiaithe —gu(i)de —_dlladn(ale at ‘fur cé(ilia faith = gu(di—liadn(at 2 plu. rom fir cel! tiatha gull bliadn(ayt voclacc, iru eilin—tiatha Budi —bliadn(al a Ser céiDle —tiath suilde —bliadn(aje dat, era c&(Dib—tiatayb- © gulldlb-—dlladn(ahb (Classes Ta and both oppose a gen, in-o (later -a) to. form with no ending inthe other cases ofthe singular, the difference being that this is palatal -C” in Hla but non-palatal -C (usually preceded by u) in IM, The plural of Tita (ra) i the same as that of TVb/e but Hb (m.) not ‘only has # distinctive nom, pl. e alongside the ‘Ila’ type ~i and a rarer -a (see da below on reuters) but is also characterised by an acc. pl. -u like that of I. ‘The exclusively neuter lasses Via and b both have a gen sg. ine and a nomJ/ace. sg. with final consonant, palatal -C’ in the former and non-palatal -C in th latte. Both have a det. 53 in-C’ but Via also asa distinctive ‘long’ datin-(a)im(m).In Via-n(n)() appears throughout ‘the plural, while Vib has ~C’e in the nom/ace. and gen. pl. and the ubiquitous dat. pl. ib, The ace., gen. and dat sg. of Wie follow the scheme of V below, the difference being that tke stem-final consonant also appears in the nom. sg. The nom, acc. and dat. pl can also te compared with V but the gen. pl in -e rather resembles that of Ta/b, sing. THTa “eye” (£) TIT ‘shape’ (m) Via ‘name’ VIb ‘plain’ Viet father’ (m.) omiivoe. si cruth ainm mag ath(alir sce. sit coruth ainm = mag ath(a)ir gen. silola.———crothova anm(aje —maige —ath(alr at, sil cruth inn’ malig ath(alir ‘anm(aim(n) plu, nom. sili crothal(aJeH—anman(n) mage —_aithir vocace. sili erwthu ‘anman(n) maige —_athralaithrea en. siile ——croth(aje anman(n) maige _athr(a)elaithre dat, silts erothaib ‘anman(njaib maigib —athr(alib/aithrid ‘The characteristic feature of m/f. consonant stems is that a particular stem consonant appears throughout the paradigm except in the nom. sg. and sometimes in the dat, sg, (see A.4a below onone ofthe few neutersin V unlenden.). Although there are a number of variants ofthe nor. sg. ending (notably -C, -C’, «or -u depending upon type) all of these have in common & lack of the defining stem consonant (cW/d, 1, n and ch/g/c respectively) seen in the rest of the paradigm. This is regularly non-palatal inthe gen. sg. and pl. but palatal in the ace. sg, “long” at, 9g, and nom, pl. The ace. pl. adds ~a and the dat. pl. the usual -a)ib tothe stem consonant. Anumber of sub-classes also display a ‘short dat.sg. lacking the characteristic stem consonant. ‘As can be seen from the above paradigms this form is often identical with the nom. sg. but on occasion differs from it (e.g. folmte above or nom. s. ene ‘ire’, gen. sg. ened, dat. ‘long’ ‘enid sind ‘slon” rein) The ‘short datlves were tending to be replaced quite rapidly by the26 “tong? datives identical with the ace. sing. during the Ok Irish period. Occasionally the reverse process results in a‘short” accusative singular such as ofntu and cathair. veces Ver sing. "Yossi ("hed “ey “epnon’ fudge’) BOE nee Son caale extair ain brite nomoe Ty iuaid caeht cate tala britheman fe. fled ontad coat catch timen, Brean Sith intl cart cata’ ton Brenan Se Ce Cans ayr atti brithem{a)in fom fd caret cathlg foci, flea caivea—cathraca brthenna fie sarot ata briheman Get edad fair eathrac(aib Brithenn(aib .Friner nts on noun delensio. CERIN Ge selon nee oo ae alas doe tos of te © eng eas ne at ot Hower ver rose ae Sa one nom tec lece t Vlonra tm and cnmany, VID a estes ove teen fo te erespndng a cece cues inquesien, Ths exert usted he llowig em teeny ofa p/p mse lanes ante post of avg tyimens oa long are ih asda besiesce - ree ahaa Hla ‘sex! Tb sas* V unlenden oth ne Ce a os ae = Sita ioe = Siw tar Tad de a. sclla) —erde mare rindrenda dia) ee Side ue rie" de Tow, cb mi rend aah sew ve een ced that changes ihn the prdia patilary a sif Som Co ttal palma! cosonmce ad ce vest fen rman yc aint eatvvel fr fr jaro ci ers Se veo sksnaion oti pe sete: betwen ead i 8 Toy Pn nt oad og Toe ante 6b eae tren (ee Tall hog ee pl al se acd cl, sir aan clyde natn betwen Sand dao slg See ee idol) important near hes pestis in mind aoe a eon conaning oe o hese reel ves In Vocrbulay or tons. 2 (©) When the post-toni syllable (ie. the one directly aftr the stressed syllable, which was usually initial; se LB.4) was non-finl, its vowel was normally Iost by syncope (marked "elow), which thus reduced words of three and four syllables to two and three syables respectively. Within a paradigm the addition of a syllable (e.g. dat. pl. -(aib) to a bis ‘wna which is (wont tobe)’, subj. beith ‘may be’ > bes ‘who/which may be’ fut. bieid “ill be'> bias ‘who/which will be’), while in the 3pl. rel, -e ig basically added to abs. palatal (e.g. subst. verb bi ‘are’ > bite ‘wholwhich ae’, subj. beit> bet, fut ieit > bef). An invariable suppletive relative ffeil) ‘wholwhich is/are" corresponds to pres. a:td ‘is and a:taat ‘are’ and the (typically leniting) relative forms ofthe copula display certain peculiarities: eg. pres. 3sg. is ‘is’ > as ‘who/which is", 3p. it ‘are’ > ata “whaiWwhich ae’, 35g. sub. ba, fut. bid> rel. bas or bes, 3pl. sub. [bar?] fut. bit> rel. bete or beta, Apart from mutation afer the copula and the substantive verb’s 3sg rel. boie ‘who'which was" (occasionally reduced to boi even in Old Irish, as in the final stanza of 3e below) corresponding to abs. bof ‘was’ the abs. and rel. forms in the preterite are identical (e.g. da én bétar remib ‘two birds who were in font of them’ in 2a above), ‘The basic word order rules of Old Irish, including the cleft sentences, are very similar to those still applying in Modem Irish. Consequently a student familiar with Modern Irish should have litle difficulty getting a ‘fee! for Old Irish sentence structure. B.3, Reading Practice. (2) The glowing description of Etain begun in A.Sb above continues as follows.so Urthécbal taille ina minmailgib. Ruithen suirge cechar a da rigrose. Tibre diniusa cechtar a da ngriade co n-amilud. di ballaib bihchorcraib co ndeirgifola Iaig » alaill Co solusgili snechtat,.. Ba stré as chéemem , as sildem as chéram...di mndib domuin. (&) A longer description of Conaire's reign than that in A.Se is given later in Togail ‘Bruidne Da Derga by Conaite's exiled foster brothers, whose failure to realise that things have begun to go badly wrong for him is highlighted by ironie use of present I denoting ‘actuality, Part of the letter half runs (with some nocmalisation) as follows (note the typical scribal practice of placing a Koman numberal berween dot): Astaat vii. mai thre i ngiallnai fr fraig inna thig-som fri colmét ind rechta sin, a:té tilaitiv iarna chil, i. Mace Loce. Is inna flaith is bindithir la cach fer guth alaili Detistéta mennchrot ar febus inna edna ; int ide , in chainchomraic fil sethnu inna ‘Brena. Is inna flaith ataat intl baie for Erina barr dias , barr scoth , barr messa. (©) Wo. 5627 hére ts (Dna n-aicet ast ‘Tur, 58a bitddidiu a confesso hisin (that (word) ‘Contfessio') do foistin pecthae, Biid dano do molad. Bid dano do atlugué buide. Do foisitindidiu a:té sunt. ‘Ml. 20613 nif indidit até irascimini be angry (pL.)’)sunt..achtisfoimchomarc asta (@) A mumbers of major medieval Irish collections of annals have come down to us. In essence these record major events such asthe deaths of kings or prominent churchmen, battles or unusual happenings year by year. The style is usually terse and a penchant for [isting the relevant details without undue elaboration can result indistinctly sparing use of finite verbs, as in the passage below which compensates for this to some extent by ‘means of verbal nouns. The Annals of Ulster, which are generally regarded as the most {important collection with particularly eaely roots, survive in effect ina single manuscript, ‘Trinity College Library no, 1282 (or H.1.8). The main entries down tothe year 1489 were ‘writen by Ruaidhri O Luintn, who was succeeded by two other seribes fr three decades (or 50 thereafter and also by various interpolators and glossators. Obviously these were ‘drawing upon older records such as postulated lost ‘fona Chronicle’ down to about 740 ‘AD. and the language of the entries down to the early eighth century is mostly Latin. ‘Thereafter more substantial passages in Irish begin to appear alongside Latin, the account ‘below of exceptionally cold weather in 818 A.D. and some of its consequences being a case in point. The original Old Irish of the entry is obvious enough through a thin layer fof mostly mere orthographic distortion attributable to the later scribe(s), and has been restored here. Aiganaiccentae ,snechtae mar ro:bétar 6 NotlacStélleco hint. Imthech« Boinde cosaib tirmaib , ala n-aile n-abann, Fond oinchumat ind locke. Et, flanlaige iar Loch Echach a (iss alli do thofunn. Solaich daurthige arum é chés iar Lochait i el carat ig chet iar Lochaib Eirnea trib Connacit (c) There follow four stanzas from Félire Oengusso relating to Febru it named in fist hal an 1, Mach s,s 170, Pati ef course) pn is feta’ nee ine 295-9) Ser. 2g oveon bomen aoe Shtntened ord suchas th cope renin rot nemaltaon nthe nal sanz sex est oe bol re ty ines fe eplaccnet of np. ta blo by te som fr tt ional dete (ILD Reson Oe ‘nant te easrips wuld ena king as ava moneylae oul ina) rather than a3 a iylable (usual in Ol Ins inorder to retain the required six sylable Feb. Hae dn ind éei, Feb. 11: Mo Gopnat co nglanbail bam Caria are: Im ieire Dé ba hlmain Eiachrae ba fr ferdae maith lis grés da gam is gre da garmaim, bb Frardae aac escopEthcheninmain March: For cola mis Mirat Mach 7: Lasagne ne. nit mérdai fra nguidi apstal Erenn hége, Sentn, Moinenn, Moist Paraic co mit mile Dauid Cille Mun. op ditty a trége. Fpil 289 Cech néeb boi, fl, bias co brit, Brigach fodail, i coimthecht Crist chredail, ro:bet oc mo chobair.52 2 CHAPTER IV PRONOUNS, PREPOSITIONS AND NUMERALS. A. PRONOMINAL ELEMENTS. ‘A. Independent personal pronouns. ‘Since they could neither function as subject (end of II.A.3) or object (see A.6 below) of ‘finite verb nor be governed by prepositions (B.1/2 below), the principal use of the ‘stressed independent pronouns in Old Irish was as predicate of the copula. The basic forms were: Isg. mé, 25g. ri ‘you', 35g, m. (R)é,n. (hJed, ff; Ipl. smi, 2pl. sor sib, 3p We ‘The copula behaved like any other verb in expressing a pronominal subject by means of the appropriate personal ending: e.g. adib cland Abrache ‘you (pl) are the descendants ‘of Abraham’, is adbar rig ar deilb ‘he is the makings of aking on account of appearance’. However, a stressed pronoun was required to substitute for a noun as its predicate: 6. is sssi in tempul sin ‘tat temple is you (p.), ls hé a n-athir far colinn “our father (r.) according ta (the) fleshis he’, i eda dilehinne ‘its reward (n.) is itthat’ is simo irnigde fri Dia ‘ay prayer (f.) to God is itis" ‘A2. Various adjuncts. (@) Od trish hada special set of emphatic particles (variants given are the usual ones afte palatal and non-palatal final respectivelybut-som s widely sed after palatal to): 15g, se) and-sa, 2g. -sluofsw/s0, 35g, /n.-sem/-sium ox som, f.-; 1p. mi, 2p -si, 3p. -sem/-sium of -som. These were used to highlight a wide range of pronominal lemens, notably a verbs subject a expressed by its personal endings (eg biuusa‘Zam (Grontto be)’ até-tom rs), anunstessed objector prepositional pronoun (A.6andB.3 ‘below),apossessive pronoun (A.3 below) and finally a stressed personal pronoun o yield 13g. me-G)se, 2p (su, 39m. (som, £. sist (there is no emphatic form of ed), Ip. (wisn ot s(n)-n)n, 2p. s-()si, 3p. (H)-som). There was aso a simalrly used third person anaphorle particle mesning ‘the aforementioned’: 35g. m..-sde, n/t ade, 3pl-sid ete. An important property ofthese particles isthe need to attach them to a stressed element, Consequently in copula constructions they are attached tothe predicate rather than to the copula itself, eventhough the emphasis applies to the pronominal subject and not to the nominal or adjectival predicate, Indeed, where such a predicate consists of single word, use ofan emphatic particle would seem to have verged onthe obligatory in Old Irish: eg, am cimbid-se‘Tama prisoner’, comba soilse-su ‘so that you (Gg. may bea light, is Diasom “He is God, is ramn-si ‘tis a par’, mad fochrce-som “ifitbe a payment’, adib ereomig-si ‘you (pl) are Believers’, condat anman-som ‘so that they are nouns’ss ee (90% ee et ign citeertonc rs wean e Sacer AB. Possessives. ‘The elements discussed in A.2a/b above ae also used withthe possessive pronouns Ise ‘mo (len. ‘my", 25g. do [len] ‘your’, 35g. msn. a (en) “his its", a {AV-] er’, Ip. ar fas, ‘our’, 2p far or bar [nas.)"your,3pl-a (nas ‘their’. Thus is lobur ar n-irigde-nt ter rye wet (MS np ornate our) name that shall be upon them? (Wb. 42), Coran a ainm-sideCoran his name’ (.2.of the ‘aforementioned druid summoned in Echirae Chornla) tis to be noted that after for ‘upon’ ora preposition ending in a vowel the 1 an¢ 2sg Poms are m and frespectvey, asin iair dum berrad-sa ‘come to tonsureme’ (it. for my tonsuring’,Sety episode from the Book of Armagh, VLA.7. AA. Pronouns and the cleft sentence. ‘Annatural extension of the use of independent pronouns as predicates ofthe copula entails their fronting with the same in a cleft sentence in order to emphasise the pronominal ‘subject or object ofthe following clause, the verb of which is relative in accordance with the rule given in ILB.2b above. The fronted pronoun is often accompanied by an ‘emphatic or anaphoric particle and typical exampes of this construction from the Old Irish glosses include: ¢hé-sidi immurgu beta hicthi “itis they (the aforementioned), ‘moreover, who shall be saved! is hé-side dano as éola ‘its he (the aforementioned) t00 that is knowledgeable’, ished as maith di ‘itis ivthat that is good for you’ tis to be noted hatte cope and te relative va at vaihly in he tid prion in cle constructions ofthis type, the 35g, naturally being used with a fronted singular pronoun ‘and the 3pl with a fronted 3pl, pronoun: eg, ar ba me-se ba sacart oc baitsid ind ‘anmae sin fora athair ‘for it was I who was priest atthe) baptising (of) that name on his 88 father. However, either 3sg. or 3pl (or even a combination of both) can be used with fronted 1 oF 2p pronoun: e.g, snisni ata sonartu (3pl.) itis) we that are stronger’, it sib «ata chomarpi Abracham (3p1.) ‘itis you who are heirs of Abraham’, ni snicet-id:deirgnt sg.) “itis not we who have done it frst’ is sni-sni ara bobes (388. and 3pl.) ‘it is we who are bowes (Lat, ‘cattle’. It is even possible on oceasion to combine @ hanging ‘nominative (see I.B.2a) with cleft pronominal construction, in which case the hanging ‘nominative is resumed by a fronted independent pronoun (in bold italics): .g,,nach gnim tumal bad hé do:ngneith ‘any humble deed, lt it be ivthat that you do’, na maith ro:bé ‘bad hed do:gneid ‘any good thing which there may be, let it be ivthet that you do” AS. Reading practice. ‘Wh. 2626 mad fochricc-som njpa déoladacht acht bid fiach Wh. 4023 is dlmain la edch ar n-orcun-ni. Wo. 4c40 nipat he indi beta thuicsi di iudetb nammé beiteisind inducbail sin acht bieit cit geinti hiressich Wh. Sal9 is ed bunad mo chlainde, Wh. 5017 is mé as apstal geine. Wh. 29d19 naba thoirsech cia béo-sa | carcair ML. 1948 inti as sémit i int! as mindchichthiu Mi, 3014. isindi in, isin did A. 6. Sulfixed and infixed pronouns withthe verb ‘to be’. Being a frequently used form, the verb ‘to be’ displays irregularities in Old Irish as in ‘many other languages such as English. For instance, the compound verb a:té (preverb ad + 1) is replaced by simple *td(i)th “s', which eannot be used on its own, in order to attach the so-called ‘suffixed" pronouns Isg.-()um ‘me, 25g. -(ut ‘you’, 398. mu. (ai ‘him it. -(us “her, i’, 1p. -(um(n)“us’, 2p -(a)ib you’, 3p (Jus “they”. The ‘basic pronoun isin bold italics, the italicised vowel being a glide used to attach all but the 3sg, m/n. (itself a vowel) toa preveding consonant. The bracketed vowels are optional devices, (a) indicating a preceding non-palatal and (i) preceding palatal consonant (see LB.S/6) It is characteristic of Old Irish that object pronouns are invariably ‘unstressed and are attached toa part ofthe finite verb. They usually function as direct objects ofa 33g. verb (c.g beirid ‘caries’ but beir'th- ‘carries it; see V.C.2) but reused as indirect objects with the substantive verb: e.g, tdth-um “there isto me”, boithi ‘there ‘was to him’ (note the preterite’s ‘intrusive’=th- before a suffixed pronoun, doubtless on the model of pres. th- in relation to at), beth-ib ‘there shall be to you’. Since Old Irish thas no separate vecb ‘to have’, this constuction serves that purpose with the result that ‘an expression such as tith-unn lebor ‘there is a book to us" is the equivalent of English “we have a book’ and may be so translated.56 ins can only be used with independent simple verbs. fa verb is compound (Gee VIB) or is dependent by virtue of being preceded by a conjunc particle such a8 Seative ni “no, the object pronoun i then atiached tothe proce element (whether pretonie preverb, on which se V.B.1/2a or conjunc particle) rather than being sffixed fp the ending ofthe verb itself. Since it thus comet to stand between the prociti (see TILA.2-4) and the ret ofthe verb its termed an Infixed pronoun inthis context: eg them “Uhave” (ufixed) but ni-mthé there is nc tome, Hhave not’ (infixed),beth-ib ‘You shall have” (suffixed) but n-b:bia ‘there shall not be 0 you, you shall not have! Gatteay_Attowing fo the foct thatthe’ frequentlyfllow a vowel and so can dispense With the glide w oj the basi forms of the infxed pronouns are the same as those of the affixed pronouns except in the 35g. mun, As nimthd above shows, the Isg, infixed jronoun lenited what followed, So to did the 2s. ¢, while 3 ag. £ and 3p. # were Eptionally followed by nasalistion (e.g. n-s:mbia or nivs:ia ‘there shall not be to ther, sherthey shall not have’). The following mutation is crucial inthe case ofthe tnfixed Jee, ma (oas_]andn.-a [len], which elide a preceding o: e.g, ro:boi “has been’ (Gee ILA $a), ro-:(m)Bot ‘has been to herthem,sbethey have had’, r-a:mboi “has boon fe him, be has had’, r-a:boi ‘has been to iit has bad’. The 3sg, mn. infix is fully sbsorbed by a preceding negative i, the upshot being that only the mutation (not always ‘ible in spelling, of course) remains to show its presence: e.g. ni:bia fi bia/ ‘will nt ve but nisbia fal viel*will not be toi, twill mot have’ and nf:mbja [rd mia! “will not be whim: he will nt have" (in normal Old Irish writing nt formally distinguished fom tnem:bia ‘will not be to me, shall not have"). Note that, although ro has its basic “perfet” sense in r0-:(m)Bo, r-a:mbo!sbove, itis sometimes used witha form ofthe sabstantive verb beginning with b-solely inorder to attach an infixed pronoun, inthe case of fut, ro-sbia and ro-t-bia in A.7 below. Saffixed pronou “The special relative fei(e)‘wholwhich is’ of present td has already been mentioned in IIB 2b, Apart for this anda rare archaic or archaising use of independent fl (-8, “fv in te last example in A.7 below), the main Suction of (iin Old Irish sas the ‘ppletve dependent form of a:td ‘isin non-possessive constructions (as apposed 10 sMiuncthd “there is not to me, I do not have" above withthe dependent td required by the ‘possessive construction). Bkymologically fei iste imperative ofan old verb meaning Msee” (ef Of “poet’ originally ‘seer’ and Middle Welsh gue-et to see”) and can be ougily compared with French expressions such 8 vo-c ‘here s(t. ‘see here’), vo 13 “theres” (it see there’). A peculiarity rooted in these origins isthe fat that Ole. (ffeil does not show normal personal endings but is invariable and has its subject in the Seeusative (¢€ French me voici "here Iam, le voll ‘there itis" et.) 8 a: ben and ‘atnisfilmnal and there is (not) a woman there’ a:t6 for Iongats but ni-m.fl for longats “Tam (not) in exile’, a:taatoca cluichi but nfl ca cluichiu “they ae (not) at thet play’ 7 ‘plo examples of an enpti pl igh 2 pl ging a ot pronoun int Sect erstaleome sal aa ithe ee =e mt Fave nour te ity Gcope cn ar pl lass C’ infixed pronoun, on which see VI.B.7), = AT. Reading practice. ‘The followings selection of short sentences select tain ori ive foots combine wise ed arid pono and te wo. Tels exes toma short ovethcety ory pling sigh exe oo lenge han he min nests of Ish loses, The nt vo examples seat aoa oleic’ ofthe pronoun ‘namipaionoffllowingnou obj aon o yt Eg hin, policeman’ ote ike, Roba athe begining fhe eighth entre tas the "ont O10 ishyraceofemployings3sp ve withafellowing singular noun subject ven wen i i llowed ya iter objet of ithe ter, whether nl coordinated by ocus ‘an, the later beingin effect regarded as an afterthou ee Gonchubar oes Ci Cain oc bi fidehileConsbarand Ci Chuan were sg was playing deh Bch for pes. bid plus sued pronoun seats ‘we supplementary development, arly he contraction of ential hats vowel due waclicrazaiaimearseaeseas eesemeae ee ith cole mat Tihs mdr math Tika art othe Bema mo gue Biheller la orahar cm Ro-t:bia grédugud. . a os bind oa bad cw ecl Nektar tama debbioise Pisa nas esa al ot Ibo hrs rambal exch math (Wb. 203) inte pr 03) eam mgrecptarm tah se Nicon:fil nack n-intliucht n-aile and (ML. 18¢ i. — Mijn re dig ba coe Nef ath ia fora in58 Fll-us daneu tre chenéle martre ata légmara le Dea (Camibrai Homily - see VIB.6). B. PREPOSITIONS AND THEIR PRONOMINAL FORMS. B.L. The basie system. id Irish prepositions governed the accusative or the dative case, afew being used with either (see end of ILA1) The pronouns governed by them were unstessed in Old Irish fand so formed a single unit with a preeeding prezostion as is still the case in Modern Irish, The resultant combinations are usually known as prepositional prowouus o: ‘conjugated prepositions. The frst ané second person pronominal suffixes are forthe ‘most part formally identical with those used with verbs, namely 1s Jum ‘me’, se, put you, Ip. -(Dum(n) “us? 2p. (ib ‘you’, and do not distinguish between dative and accusative forms. However, two prepositions taking the dative hevea special palatal form ofthe 25g, namely do “to” with duit ‘to you' and tia ‘from’ with lait ‘from you’, land the second of these also has a palatal form of the 1sg. Giatmfm) ‘ftom me") and ‘optionally the Ipl iain) beside ian(n). The thiré person forms are not only different from theit verbal counterparts but also distinguisk between accusative (3, f.-[8]¢, Pl. “hen ative Gop. 3p. (ib) forms, The] ofthe 3sg.f.and3 plac. isnever preserved as such but had certain indirect effects,namely: (1) devoicing of a preceding Yoiced stop in the case of 3p. dat. indi in’ div’ ‘in them’ but ace. infu ‘into them Ta'tw (in(o)), 356. mn. cuo(e(a) (up) to hint” ug (Moa. Ir. chuige) but & ‘cui)ce up) to hevit” nk’ ef (Mod. Ir. chuice; co “(up) to’) and 38g. man. imbi “about birt im’ but pe ‘about her’ fim’ pe! (imm around” dV-)is due to the influence of pronominal to/ar-, while conversely la and a ( a) presumably influenced 2sg. lar and 3sg, as for expecred “lt, "es. Similarly, Ol. cen for expected *can (see V.B.2b on e toa in procitics) must be due to pressure from cen- in the stessed pronominal forms, The 3sg.n. con(a)e ‘without it’ had also developed an. adverbial sense “besides, moreover’. Ths form was still represented as unlenited in We. ‘and MA but, doubtiess under the influence of the compound olchenae ‘besides’, normally appears as lenited chen(aje in Sg. (40211, 140b3, 15:1, 212a1 1, 23801; see too the last line of verse 6 in V.C.4a), where chen ‘without’ then also appears as the base form (Se. 7Sal, 7862, 14763). From here the lenited initial apparently began to spread to the ‘pronominal forms of co, which also had initial c-, to yield isolated 3sg. m. chucai (MI 6c) and £ chucae (Sg. 19182). The preposition fran its pronominal forms were also atleast sporadically affected i the Glosses, possibly witha view to simplifying the initial consonant cluster, on the evidence ofthe following unambiguous spellings (see LB.8): re (MLA4bA), rissa (with neut. art; Ml. 3002), 2sg. fri (Se. 15143) and 2pl rian (MI. ‘54a3), Interaction between the pronominal forms of different prepositions no doubt also played a role on occasion: eg, 1p. lime was probably influenced by Ip frinn and 3sg. ‘v/a. and ‘in it’ may have contributed tothe cretion of as ‘out of it’, Factors such as these tended to increase complications as did some confusion between ace. -e and dat. lin the 3g, £ (eg. (Whiade beside tadi esse beside e)s(0i and air for expected airre) ‘as well as slight encroachment of dat. pl. -aib upon acc pl -()u (asin samlatb and cenaib ‘bove) and an isolated instance ofthe reverse in occu (Wb. 13c1) for acai. tis probably ‘easier to lear to recognise the basic prepositional steams and suffixes plus the 33g. m/n. ‘than to lear whole paradigms with all their variations by heart B3. Further adjuncts. Emphatic particles (A.2a above) are frequently added to the pronominal forms of ‘prepositions: eg. nita chumme-sefriu-som ‘Tam notsimilar to them’, ar is gloria dib-si {n for that is glory to you’, ni samlid s6n din-ni “that is not likewise for us’. However, rather than attaching the anaphoric particle -sid/-ade to such forms Old Irish usually 6 employs the appropriate form ofthe fully stressed anaphoric pronoun suide (IV but with invegular nom Jace. sg. neut.sodain): eg. hi suldiu ‘in that/the aforementioned, therein’, {a sodain ‘with thatthe aforementioned, thereupon’. Note to fully stressed ace. sé, dat siu ‘this’ after prepositions (e.g. co ssé ‘up to ths, hitherto’, re siu ‘before (his)") and sin “that” inthe common expression far sin after that, thereafter, then’, although the latter ‘was otherwise usually attached without stress to the 3sg. n. pronominal form of a preposition: e.g, is samlaid sin ‘it is in that way’, is air! insin “its on account ofthat, ‘ant sin “in that, thee’. Otherwise so and sin are stressed when preceded by the article alone (eg. tn 0 ‘this’, im sin “that but, aa in the previous sentence, sometimes used for plain sin) but unstressed when modifying article plus noun (e.. in fer so ‘this man’, ix ben sin that woman’) Bl, Reading practice. Wh. 103 is mér a thorbe dé din-ni Wh. lat is ola-side, nif nach dichlith art Wh. 2216 ni forr neck alailiv an. Wh. 2b1 is Fd leu teistiu fuile. Wb. 2627 is hé a amomus lem. Wh. 3630 is deidbir ib, cid mebul lib, ataldmet. ‘Wh. 3¢34 até dethidenfuiricene. Wh. 4a6 cia beid Crist ndibst te foisitn hirise in baptism et (‘in baptism and) is béo ind anim tri sodin, is marb in corp immurgu trisna senpecthu Wh. 4dl2 lour leu gnima rechto dia firidnugud. Wh, 420 ba wisse hirnaigde erru ba liach a n-epelty, Wh. 5c7 it carit dom-sa immurgu. ‘Wh. 11210 is giélim-sa ro-m:bia biiaid, ‘Wh. 1669 nf Indrdigne lib cin-n:fil i, a tdi maithi cene. (C. NUMERALS. CA. Cardinals, (2) Old rish numerals (cardinal and ordinal) are placed after the definite article, ifpresent, and before their noun: e.g, da mac ‘two sons” (nom. du. m.; see ILA.4d and ILC.1), for-sna cbic riga ‘one the five kings’ (ace. pl. m.). Gen-‘one” isa leniting prefix forming ‘acompound witha following singular noun (e.g. 6 ofnll from one seed’), “Two? (da/d), “three* (eritenir) and ‘four’ (ceth(a)ir/cethéoir inflect for gender and case: meee mn. f mn Nom. da {len} da [nas] di [len] tri teuirfteoir ceth(ajir cethéoir ‘Ace. da [len] da {nas} di (len) i téora_—cethri_—_cethéora Gen. da flen} da [nas] da {len} ri [nas] séora {nas] cethi [nas] cethéora {nas} Dat. aid [nas] aib nas} dib [nas] trib tdoraib —cethrib —_cethéoraib[Note thatthe ace, fem, forms t4ora, cerhéora are frequently used as nom. oo in place of teuir, cethéoir. "Two" governs special “dual” forms of nouns (e.g. with nasalising dat. “fab; see ILA 4d), while numerals from ‘three’ to ‘ten’ are followed by a plural and are invariable fom five upwards: edfe (len) ‘five’, sé (HV-)'six’, seche [nas] ‘seven’, acht [nas] ‘eight’, of [nas} ‘nine’, deich [nas] ‘ten’. Apirt ftom expected naselisation in the gen pl, the forms of three’ and "four" caused no mutation (except presumably /h-/ before ‘vowel in the ease of ace. cethri and nom/ace. tri, ora, cethéora et least) although the Early Olt. nomJ/ace. neut re ‘three’ did cause the expected lenition ofa following noun fon the evidence of wre chenéla) ‘three kinds in the Cambrai Homily (see VI.B.6a) and this property of the neuter seems to have been coninued by its Ol. replacement vi, to judge ftom riche ‘three hundred() witha ‘shor’ pl. (see IL.A.4a) quite prone to be used ‘with numerals . The mumerals 1-10 may be used as virtual adjectives with a following ‘noun (e.g. nom. du. n. da n-orpe ‘two inheritances) or on their own, in which case den iso longer compounded and the rest are preceded by an appropriate form of the article if definite and by @ (RV-) ifindefinite two" and ‘three’ also having the special stressed forms dw/d6 (non-dat.) and tri (mn. non-dat.)€.g,ofn d airchinchib Assi ‘one ofthe leaders of Asia’, a déu ‘two", « (hoch ‘eight (Mod.Ir. a hoch), inna ockt ‘the eight (things), hbnaid di fromm the two", The numbers ‘eleven’ to ‘nineteen’ are produced by placing disyllabie deae (older deec; Mod. I. déag) after the relevant number from ‘one’ {o ‘nine’ plus its noun, if present: e.g, d mill déec ‘twelve thousand’, in da apstal deac “the twelve apostles’, coieblladni deac ‘fifteen yes’, a ocht deac ‘eighteen’ (b) Various other types of numeral couldbe derived from the foregoing forms. Apart from irregular dias ‘two people, couple’ (I; note dias ar fchit “twenty-two people’, lit. “two ‘people in front of twenty”, but da fer deac “twelve men’), composition with fer ‘man ‘produced of denar ‘one person’, triar, cethrar, cbicer, seisser, mér-feisser, ochtar, rndnbor, deichenbor (all normally I, m.) ‘three/fourfve'sin/seven (it. “big six'Veight! ninerten people’. It is to be noted that these forns are frequently used in the dative ‘without a preposition and with or without a possessive: e.g, denur ‘on his own, alone’ in C.34 below and m denur ‘on my own, alone (it ‘with his/my one’, ediciur ‘one of five? (it. “in five’) in V.E.2b, The suffix -de was used for groups of things: dé(i)de ‘pit’, tré(lde, cethard(a)e, ediede, sé(ide, sechiae, cchtae, noide, deichde (all IVa, ».) “hree/fourlfvelsin/ sevenveighvnineiten things’. (€) The devads from ‘twenty" to ‘ninety are fiche ‘twenty’ tricho/a ‘thirty’, ethorcho/a four’, eoleo/a “ffty’, sesco/a ‘sixty’, sechtmogo/a ‘seventy’, achtmogo/a ‘eighty’, nécho/a ‘ninety'(al V unlenden, m.). These, cét hundred” (I, n.) and mile ‘thousand? (IVb) are nouns (like ‘score’ in English) and consequently @ noun qualified by them stands in the genitive plural: e.g, el fichier ‘thee score men’ (fer I,m.), fiche miled “twenty soldiers” (mil, V lenden., m; cf. Eng. ‘a score of soldiers’), dic o¢t fer ‘five hundred men’. In combination with the numbers 29 plus anoun the decads appear inthe ‘gen, sp, a further hundred or more being introduced inthe dative by the preposition ar “in front of se. da lebur fichet ‘22 books’ (it. ‘two books of twenty’), di ir fichet ‘22 letters", ee capt asa sescat ar trib cétaib ‘365 noble chapters’. In combination witk ‘one’ the noun plus ar sufficed, as in bd ar fichit ‘21 cows’ (it. ‘a cow in front of twenty’), and the same construction applied to numerals used absolutely, whether ‘preceded by a(hV-) orby the neuter definite article: e.g,,an-6en ar fichit'21", nna deich ar dib cétaib'210", a ocke deae ar chét 18" C2. Ordinals ‘The ordinals are cét- (compounding prefix, ening) or eétmae (IV) or sometimes roisech “first? (I/I), tdnaise or aile (both IV; also proclitic ala or compounding prefix all-) “second! tris ortres third’, cethramad fourth’, cdiced ‘fifth’, se()ssed ‘sixth’, sechtmad ‘seventh’, ochtmad ‘eighth’, ndmad ninth’, dechmad ‘tenth’, fichetmad ‘twentieth’ (all VM) and so on up to céemad ‘bundredth’. Apart from toisech and aile (as opposed tothe procltic variant ala) these also come after the article, if present, and before their noun: in cétipersin} (standard abbreviation for Lat. uel ‘or’ = Olt. nd) in ténaist ‘the first, ‘person or the second’ (acc. sg.) isind pistilofsech ‘in the first letter’ in tris diltud ‘the third denial’, im eres chuit deac the thrteenth portion’, seckimad rann cethorchat 's forty-seventh part’ isind fichetmad blladain ar ché ‘inthe hundred and twentieth yea’ C3, Reading practice. (@) The opening ofthe saga Seélae Muiece Mac Da Ths “The Tale of Mac Da Th6’s Pig introduces the brivgu or ‘hospitaller' named inthe title and his hound before going on tc lis the other four hospitallers important enough to possess a bruiden or ‘hostel’. Boi righriugu amrae la Laigniu, Mac Da Thé a ainm. Bot ci occa... llbe ainm in chon ‘ocus ba lan Eriu di airdircus in chon... Is si sin in chéced bruden boie i nErinn isin ‘aimsir sin ocus bruden Da Derga i crick Cialann ocus bruden Forgaill Manaig ocus ‘bruden Maic Da Réo i mBréjni ocus bruden Da Choea n-farthur Midi. Secht ndorus six ‘bruidin ocus echt sligi tree ocus secht tellach indi ocus secht cori. Dam ocus tinme cach (p) In three separate sections the legal text Crith Gablack (see ILA.Sb) recognises the following main categories of the aristocracy, of kings and of a general call-up of able-bodied free males termed a slégad or ‘hosting’ Cérusflatha, i. flath 5 déis co rig. Cs lr fullechta for suidtb? A secht. Cateat? Aire désa, aire écha, aire ard, aire tiiseo, aire forgll, tinaise rig , ri. Cis lr fodlai for rigaib? Teoirfodlai. Cateat? Ribenn, ribuiden ri bunaid cack cinn... Cis lr slégad aa chérai do rig do giull fora thiatha? 4 tr. Cateat? Slégad hi crich i medén fr indnaide6 slégaidtairse;slégad co hor eriche fx foiresin ir ,dligid...;slégad tar eric. (©) The early Irish law tract on status known as Uraicecht Bec or the ‘Little Primer” offers the following list ofthe different grades of poet and then goes onto discuss the briugu ar “hospitaller’ ‘echt ner ied: Comdir ollam frig n-Sentiaithe trthugud mis dé» hochair a lin. Gens, divefochaccan Gentaaathurthugud , Biathad dese db. Ti séot do mac uirmidy Vath tr trdgud telve. Cite seit do dis» blathad cothratr 48 ‘urthugudcdtedi Secht sett do chanait,Bathad ses thagud seckrne.Deich Séoit do chi Biathadochtar urthugud dechmaid Fiche st do dnrad ,Bathad defor ieee »tarthugud cbc laa ndea... Lethohit etch dia mnai no dia gormac no dia ‘rechtairiu no dia Seonapaid. Comgréidbriugusrifath dia‘mbé diablad cena lai cach {7did de thir tread.» Nibi briugu radi chetach.. ls comdive frig taithe Briugu Tettech,diablad tockusa la side coir ainsi as ruta las. (@ Another legal text concemed specifically with poets, namely Uraicecht na Riar or the “Primer ofthe Grades’, enumerates the grades from dnruth etymologised ingeniously, if fanciflly, as sruth dn) downwards and certain oftheir key attributes as follows (see MILA.3 above on prepositions like la ‘with’ plus relative -s)a [nas] ‘whom, which’) Anruth tré, i dnruth a athair , a Senathair , énruth fodeisin. An a airchetal, én a “frithgnum. Tri colcatt dréchiae; lethdréchtae las. Fiche sé a dire. Da’ fer deac dé for ‘hatha, mérfelser oc acru, cethrar for coi lari. Cli, deich séoit a dire. Ochtar alin for ‘athe, cethrar oc acr, triar for cot la rig. Secht nréchta ochimogat lais do dréchitb Cano, sechtséoit a dire. Seiser alin for tiatha trar oc acru, dias for coi lari Sesca do dréchaib lis. Dos, ete Séoit a dive. Cethrar an for tiatha, tiar oc acru, dias for ‘co la rig. Coica dréchtae lis. Mac fuirmid, cethirséoita dire. Triar a lin for tiatha, dias 0c acru, a demur dé for col lari... Cethorcha dréchiae las, Fochloc, et; lethiét a dire. Oin'fer las for tiatha ; a denur oc acre , for cot la rig... Tricka dréchiae las. Taman, fiche dréchiae lais. Lethscrepul a dire ai héola i fedaib. Drisiuc.. Deich indréchtalais,serepul.a dire. Oblaire, dic dréchia lais. Letkscrepul a dire... Ocus fer Tasa:mbiairchetal nin, filedacht én ,ni:bi lia athair nack a senathair dé, cia ainm in ‘grdid sin? NI hansae, topar. Ocus a mac-side, cic ainm? Ni hansae,tiprae. Ocus a ‘mac-side 1a-sa:mbi airchetal n-dn , filedacht dn, cia ainm? Ni hansae, sruth dn Gniruh, Foglaim n-in ,airchetal nin la cach mai (©) In one episode of her Old Irish Life (see HL.A.Se) the young Brigit went to visit her sick mother and work inher stead churning milk. Despite her generous provision forthe ‘poor, plenty of her produce remained over forher mother’s master and mistress, who were ‘soimpressed by the miracle that they freed her mother and converted to Christianity. Note 6 thatthe dative of comparison discussed in I.B.3 and illustrated in I.B.4 was not the only possibilty there being an alternative way of expressing ‘than (is/was et.)" by means of proclitic [nas] in regard to) what’ or af [nas.] ‘beyond what’ plus anasalised relative ‘form of the substantive verb (e.g. ol-daas ‘than (is)) with its subject in the nominative as expected. ‘She (Brigit) would divide a torad indi ratnn deac cona gruth ,no:bith in tres chultdeac ‘I mmedén , ba mé-side i-mboie cach cut, ‘Ced torbae latin sin?" ol int arae. ‘Niansae’, ol Brigit ®. ro:batar da apstal deac lassin Cotmdld , héssom fessin in tres deac. Ro-m:bia-sa la Dia (that) thirteen paupers will come one day in the same number as (Christ and hs apostles. (O)Inmedieval Ireland genealogy was more than the mere tracing of pedigrees. Itnot only. provided crucial justification for the status of groups and individuals but could also express actual or desired political relations between them Its centrality is indicated by the survival of large genealogical collections, usually in a mixture of Irish and Latin, tracing ‘countless groups of varying importance back via many intervening (and often fictitious) stages to distant alleged ancestors whose origins could then be traced back to the Flood by linking them with biblical figure. As in the Bible, genealogies could be presented in either ascending or descending order, both of which are seen in the following typical passage from the collection inthe Laud 610 manuscript preserved in St. Joln’s College, Oxford, After giving Tiathal Techtmar's pedigree in ascending order, this traces a line ‘of supposed kings of Tara and Irland from him via Conn of the Hundred Battles to Niall of the Nine Hostages, while noting various significant offshoots. iiathal Techomar mac Flachach maic Feradalg maic Craumthainn maic Lugdach Riab Derg. Da mac airegdai leis i. Flachu Mér , Feidlimid Rechtaid. Feidlimid Rechtaid Jmmrgu, secht mai lel. Conn Cétchathach, a quo (Lat. ‘from whom’) Leth Cun, » Cll , Crinnae , Crosse; Eohu Find Fitathnairt, is iad Fotharta, de quibus (Lat ‘of ‘whom’, pl.) Brigit; Flachrae Soguide a quo na Déisi; Flachrae Réeda, a quo Corco Réeda. Conn, tri male laiss i, Connlae Céem, Crinnae, Art Oenfer. Genmac Air Cormac. Cathir maic Cormaic i. Carpre, Muredack, Cellach, Dare. Tri maic Carpri Fiachu Sraptine, Echaid, Echu Doimlén, aquo Airgialla. Da mac Fiachach i. Muredach Tirech, Domnall, Sen Ua Maine. Oenmac Muredaig Echu Mugmedén. Céic maic Echack 1. Mall, a quo Hui Néill, Brin, a quo Hul Brin, Sil Muredaig, Fiachrae Foltsndthech, ‘quo Hui Fiachrach, Ailill, Fergus Céechén, « quo Hl Chdechén.66 (@) Below isthe verse from Félire Gengusso on January 24th Do-mm anmain, do-mm chorpde, op mir ar cech merblén Babill, ruth dir frlén, ‘con-a thriur ded. (h) The following gloss (Ml. 242) is concerned with numerical aspects of the psalms _ sloss (MI. 242) is concemed with al aspects of the psa Amal it da leburficet, it di lr flchet dano indir fichet hsin, ind rin, ind etercert ili sui, is ()t bed immaircidefrisa nnuiadnise air a n-ben ar fichit, iti secht sn 4. secht n-ernadman sin file isind saltatr. CHAPTER V THE PRESENT STEM, SIMPLE AND COMPOUND VERBS, OBJECT PRONOUNS. A. THE PRESENT INDICATIVE OF WEAK, STRONG AND HIATUS VERBS. ‘Aa. Some bunts. (@) The Old Irish present indicative basically corresponds to the English habitual or ‘general present of the type ‘(s)he takes’ (Olt. gaibid). Sine, however, the periphrastic progressive type a:td oc gabail ‘(s)he staking’ described atthe end of IN.A.2¢ was stil ‘no more than optional in the Old Irish period, an Olr. form like gaibid may on occasion ‘be more appropriately rendered ‘(s)he i taking’. (b) The tense/mood system of the Old Irish verb and the basic classification of the stems ‘associated with it has already been presented sKeletally in IILA.1 as a prelude to the ‘treatment of the substantive verb and the copula. Old Irish verbs are classified int three ‘main groups termed WEAK, STRONG end MlATUS respectively with regard to thei basic conjugation, (©) Ashas already been pointed outin ILA.3, there isan important distinction in Old Irish verbal inflection between the so-called INDEPENDENT forms of a verb and the corresponding DEPENDENT forms used when itis preceded by a so-called CONJUNCT PARTICLE such as @ negative like nf ‘not’ or ma-n! ‘if not, the interrogative particle in [nas.]*?, various conjunctions such asco (nas. ‘so that’ as well as its negative co-(wna [iV-] ‘30 that. not, les’, and prepositional relatives such as ar-a (nas.] ‘on account of Which, dia [nas.) to/from which’, fr-s(s)a [nas] ‘to(wards)/against which’, la-s()a [nas] “with which’ or iregular [nas] ‘in whieh’ (which sometimes also develop into ‘conjunctions, whence ara (nas) “(in order) that’, dia [nas] “if, when’, {[nas.] when’) as well as'their negatives formed by substituting -na [AV-] for -a [nas]: e.g, arena ‘on account of which., mot or ‘(in order) that. not, lest’, as-na ‘with which.. not In the case Of the simple verbs dealt with in this section, independent and dependent forms use different verbal endings in most persons, namely ABSOLUTE ENDINGS with independent ‘and CONJUNCT ENDINGS with dependent forms, asin 3sg. gaibid ‘(s)he takes’ and 3pl ‘gaibit ‘they take’ (independent with absolute endings) versus 38g. nfzgaib ‘(s)he doesnot take" and 3pl n:gaiber ‘they donottake” (dependent with conjunct endings). Inthe tables ete, below, the absolute endings are unmarked (e.g. Ipl. marbm(a)) but the conjunct endings are preceded by a hyphen (indicating that a closely connected conjunet particle ‘must always precede them (¢.. Ipl.-marbam) and a bracketed hyphen is placed before :6 forms used in both absolute and conjunet (e.g. 198. ()é(ieiu or ()é{i)cim). It is absolutely vital thatthe five fandamental terms placed in bold small capitals inthis paragraph and their implications be fully understood as they will be encountered ‘again and again in this and subsequent chapters. A2, Weak, strong and hiatus presents plus paradigms. (@) Whereas the present stem of all strong (e.g. be(i(-), g0i8(:), ben() and alsaost all ‘weak verbs (eg. marb/-), Iéic-)) ends in a consonant, that of all hiatus verbs (€.8. Substantive verb fa, bi- in THLA.3 and so-) ends ina vowel. As wil be seen in subsequent chapters, this three-way division has important implieations forthe way in whicha given. verb forms its subjunctive, preterte and future stems but here only the present stem will ‘be considered. The crucial difference inthe presen indicative relates to the 3sg,conjunct form: weak verbs add an extra vowel -@ or ~ to the stem (e.g. ni:marba “does not Kill’, ni:léci does not leave"), whereas strong verbs add nothing (e.g l:beir ‘doesnot camry’, ‘nlegaib ‘does not take’ and ni:ben ‘doesnot strike”) and hiatus verbs typically occupy an Jntermediate position whereby the stem vowel is either lengthened (see LB4) or ddiphthongised (eg, a:td*is', nisi ‘is not wont to be’, nisot "does not turn’) Each of the three main categories is then subdivided on the basis of further differences in inflection. Weak verbs are divided into W1 with finl -a and W2 with final fin the 3se, pres. con, the stem final of W1 being fundamentally non-palatal throughout and thet of |W? being regularly palatal where # following vowel has been lost by syncope (e.. 1p. abs, 16m "we leave’, ri)d'mi "we say’ youtotherwise palatal e.g. Ipl.con}.-lé()cem ete.) or non-palatal (e.g, Ipl. con. -rddam etc.) throughout depending upon the basic shape of the verb involved. W2 verbs such as dl()id ‘distributes’ with a non-pelatal final consonant in unsyncopated forms show a limite tendency topalatalisethisasin 3p. con diet beside normal -délat in conformity with the straightforward pattern ofthe (Jed type. Strong verbs aze divided into S1, which displays an alternation between ‘non-palatal and palatal stem-final consonant in the different persons of the present indicative (most commonly non-palatal inthe Isg. with w, the 1pl. and the 3pl. versus palatal in the 2s, 3sg, and 2pl, as in the case of be()r(-) below), $2 with basic palatal Stem final throughout the present and $3 with baste non-palatal n throughout. Hiatus ‘verbs are subdivided on the basis ofthe stem vowel into HH (-a-), H2 (+) and residual HS (o-,-e- or -u-) They are so called because a hiatus disyllable results when an ending consisting of vowel plus eonsonant such as 3s. abs.~idis added tothe vowel ofthe stem (ee 3pl a:taat and 35g. bild in TILA.3). The pres. ind. of H2 is fully illustrated by the paradigm of biid, -bi ‘is wont to be’ in TILA, walle that of HI may be edequately illustrated by supplementing the conjunct forms of a:id given in TL.A.3 with abs.conj ‘3g, baid-bd ‘dies? and 3pl. baivbaat ‘die’, For 13 39g, sofa)id/-sol ‘tums’ and 3pl. ‘so(a)it/-soat ‘tra’ will suffice (but note the exceptional 3sg.conj.-a of H3 epd. do-goa 6 ‘ehooses’). Now for the present indicative paredigms of the basic weak and st subypes. — a WI abs, conj, «WE abs, con Sing. 1 (marb(a}im (see 16C)cim 2: (Jmarb(aji lei 3. marb(a)id -marba léfijeid = Le(iJct Plus. 1. marbm(a)i -marbam jemi “léfileem 2. marbth(a)e -marb(a)ld Wé(ijethe — -16(i)cid 3. marb(ajie -marbat Ljete—“Teioe SI abs conj.$2 abs. conj. $3. abs. con Sing. 1. birwhe(irim-biur— ()gaibiu/-jgaibim (benta)im 2 birt bir Ceaibi Cbentay 3 be(rid “Beir —gaibid gai —benfayid ben Pur. 1. berm(aji -beram —gaibmi -gaibem — benm(a)i_ -benam 2.” befirhe -betirid gaibthe “gaibid —bent(aje_-ben(a)id 3. Ber(ajit —-berat_ —gaibit. ~—-gaibet_ —ben(a)it —-benat (©) The sg, abs con ending -im(m) was originally found only in WI aiid $3 but was so slechtth in popul , niet guth ison. ‘Mace sacle én tosfitel dele - nip hon, nip anim, xip at, nip galar, nip eri cruack, nip loch lack, nip aupaith Lit gréne,frs:ben at, bexith gala. (@ Wo.167 imerddat iri cld maith as dént!nertitadénum in maid sn. Con:nessat Jimmurgu in ngnim n-olee ;ar-a:ngairet ‘Wh. 506 n-m:charar-sa ind fir hore pridchim soscéle do gentib ‘Wh. 7021... no-b:guidet i. no-b:bendachat, Wb, 1107 4, hére no-n:bendacha-ni. Wo.27429 ni issiansib spritdldib molt. Wh. 31b14 no-s:moider i scélaib et senchazsaib rec(hjto et geintlec(hte, Wh 32416 ma nuom-gatbi ar chart ~ (€) Batly Irish syllabic poetry is based, as has already been seen and as the name itself suggests, upon a line containing a fixed number of syllables, most commonly (but by no ‘means always) seven, anda fixed cadence (e.g. a final monosyllable ora final syllable throughout), These lines tend to be organised in quatrains with arhyme between the last swords of lines two and four (so-called rannaigecht metres etc.) or one rhyme between lines one and two as well as a further one between lines three and four (So-called deibide ‘metres - see A.3b above). Full rhyme demands agreement in the quality and quantity of the stressed vowel (usually that ofthe first syllable), in the class (basically four: stops, Ienited consonants, unlenited nesals or liquide, and ina clas of its own) and quality (basically palatal or non-palatal) of any subsequent consonants, and inthe quality and ‘quantity of any final unstressed vowels present. The two quatrains below are from the ‘margin of a folio ofthe St. Gall codex. In this charming occasional poem the scribe has taken time off from glossing Priscian’s Latin grammar and finds or imagines himself out ‘of doors. Even so he feels himself comfortably enclosed, being surrounded by wood and ‘song while writing and thinking of God as if till within the monastery. Each quatrain has four lines of seven syllables each with a monosyllabic cadence in all but the third, which ‘end on a disyllable (schematically 7! 7'7°7', The finals ofthe second and fourth lines rayme (chél with én, both with Iong e followed by a non-palatal lenited consonant, and dass with ross, both with short o followed by non-palatal s) in accordance with «basic rannaigeckt scheme. This, however, usually requires further omament. In the present case there isa so-called “aici!” rhyme between the final word of line three and an internal ‘word in lin four (limech and trireck; coima and roida, with diphthong oi, lenited m /7/ and d /8/ respectively, and final -a), while the final word of line one (fal, mass) makes so-called ‘consonance’ with those of two and four (chél and én, doss and ross). This entails stressed vowels ofthe same quantity bu different quality (here long d versus long and short a versus short 0) and consonants of the same class and, if (but only if) final, {quality too, a condition clearly met by the final non-palatal -l of fail and the final rnon-palatal -s(6) of mass. A further omament is provided by alliteration (indicated by ‘underlining below) between the initials of stressed words, unstressed words not counting ‘and initial mutations being disregarded, Poetry sometimes allows pattems of word order that would be inadmissible in ordinary prose, a feature manifested here by the inversion cof normal fl fidbaide, ross roida (head noun with following genitive) and by the ‘tmesis" for sundering ofthe proclitic from the stressed part of standard no-m:choima in the second {quatrain, Note that nad:chél ‘which I shall not conceal’ i a sg. é-future (IX.3b) of SL Ce(ilid ina so-called leniting relative clause (VILB.3; here neg. nad + lenition instead of ‘non-leniting main-clause ni). () Do-msfarcai dbaide fl Fo-m:chain old lun, liad nad:chel ‘as mo lebrn ind inech Fosm:chain trirech inna nén. (i Fo-mmzchain cof menn, medair mass, hi mbrot glass de dindgaib dos. Débrad, no-m: Choimmdia -cotma, CCain:seribaimm fo roida ross.80 D. THE IMPERATIVE. ‘Dal. Basie inflection, ‘The imperative is used for giving commands and is typically formed by adding a special set of endings, one per person with no dfferentistion of absolute and conjunet, to the present stem. For obvious reasons the 1sg. is rare (although there area few attestations Of tiag ‘Tet me go’), The 2s. isthe bare stem (and hence identical withthe 3s. conj. in the case of strong and hiatus verbs butacking te final ior ofthe 38g. con}. in the case fof weak verbs), the 33g. ending is -e ur ~ud (aftera palatal consonant and enon palatal ‘consonant or a vowel respectively) and the plural endings are identical with the ‘corresponding conjunct endings of the present indicative. The imperative endings are incompatible with a suffixed pronoun and othervise independent simple verbs in the imperative must use no (see C.3a above) plus the appropriate class A infixed pronoun: eg, novmcléic ‘release (5g,) me!’ n-a:n-oirg ‘kil (s.) him”, no-s:ngaibid “take (pl) them" “The conjugation ofthe substantive verb given in IE-A.3 includes an imperative paradigm that will serve toillustrate tis formation in H2 andthe other hiatus classes. The following are the imperative paradigms corresponding to those of the present indicative ofthe main ‘weak and strong subtypes given in A.2a above. wi wr st s2 3 Sing. 2. (Jmarb — (JIéic—Jbeir_— gab — (ben 3. (Imarbad (lé(ioed ()be(ired ()gaibed (-Jbenad Pur. 1. (Jmarbam (-1é(Jcem (Jberam (jgalbem ()benam 2. Cmarb(alid ()teGJeid (Jbe(rid C-)gaibid (Jben(a)id 3. Cmarbat (Jle(cet ()berar— (-)gaibet (Jbenat D2, Compound forms and negation. ‘Unless the presence of an infixed pronoun makes deuterotonie form unavoidable, the imperative of independent compound verbs takes the prototonic form. Thus sg, ip. tabair ‘give!’ vs. 38g. pres. ind. da:beir ‘gives’ 3s ipv.foimled ‘et him/her eat’ vs. 358 ‘pt (see E.1 below) do:me(i)led ‘(s)he used to eat” and 2p. ipy. eprid ‘say! vs. pres. ind ‘as:be(iJrid ‘you (p.) say" but arbiguous d-a:ber (ether 2s. ipv. ‘give it” or 3sg pres. ind, ‘gives i), a-t:be(rid (ether 2p. ipv. ‘say it!” ot 2pl. pres. ind. ‘you say it’) nd ddo-s:mejled either 35g pv. "let hinver eat them’ or 3. fpf ‘(she used to eat them’). [Needless to say, such formal ambiguities can usually be resolved from the context in Which the form in question occurs, and a special negative form na (sometimes nd under the influence of n clearly distinguishes imperatives such as 2sg.na:tabair ‘donot give!” 2pl. na:benaid “do not strike!'and 3pl, nactcimlet ‘let them not eat!” from the corresponding present indicativesni:abair ‘does not give’, nicbenaid ‘you (p.) do not strike’ and nitoimlet ‘they do not eat. The infixed pronouns used with negative na (as also with -na, on which see A. above, and rel, nad) basically resemble class B (see C.3¢ above) except for the peculiarity of having initial ch rather than a dental: 1sg, rna-cham(m), Ipl. na-chan(n), 2sg, na-chat, 2pl. na-chab, 3sg. m./o. na-ch ( nas.en, respectively), 36g, £/3pl ma-cha. D3, Reading practice. @) Inthe first brief passage below ftom the Old Irish Life of St. Brigit (see ILA Se) her father gives hera welcome order. In the somewinat modified second a sickrelative orders his servants o send a cow to Brigit and she then bids her followers give it to bishop Mel ‘Gaib-siu rd calle, ammo ingen, ar ised t'accobar. Fodail dano in trebad so do Dia ; ini.” Dia Mairt arabérach boi fer maith ifochratb, coibdelach do Brigt. Bliadain lén dé i seurc. ‘Berid dam’, ol suide ‘boin bes dech bes i mo indis indiu do Brige , guided Dia ‘ncerum daus imba shin’, Berait in mboin ,as:belr Brigit friu:‘Berid fo chétiar do Mel. (©) Wb. 5416 taibred cdc airmitin di alaiiu. Wh. 6220 tatbrith a techie ndoib Wb. Gott na:taibred dimiccim foi, Ml. 72411 ducm:em-se , deich tarm chenn. (©) Below is a passage from Tecosea Cormaic in which Cormac utters a series of injunetions appropriate to the king who would be good, Torramad lubru, Lessaiged trun Techtad fir. Cariged got. Carad irinni. Dinged om Bidad bibdada. Bred frvetha,Bithad ech mat, Cagad cech sd. Crenad mat Lessaiged anmain. Aisndéided cach réilbreith.Inbed fina seéo med. Canad cack ir. Ar is tre fr aithemon do:beir Dia in sn ule (@) In Jmmram Brain or ‘Bran’s voyage’, which is one ofthe oldest surviving Irish tales, ‘woman in strange clothing appears and utters along poem tothe hero and his friends. At the end ofthis she addresses Bran in particular and urges him in the final deibide stanza (with fourth line modified) below to set off over the sea to find the ‘Land of Women’. ‘Na:tut fri lige leseae Na-chat:tréethad do mescae. Tindscan imram tar muir nglan Dochum tire inna mban.(¢)Some quatrains from Félire Oengusso provide asvitable conclusion to this section. The first two commemorate Aed alias Méedéc of Fems and Brigit of Kildare. Note that inthe two verses from the epilogue the medial consonant of the third line’s final word is nnon-palatal, whereas the medial consonants ofthe finals of lines 2 and 4 with which it cconsonates are palatal (see C.4e above. Jan. 31: Skuind ded fortrén Fernae, Feb. 1: Mébrait calaind Febrat ‘Méel-anfaid ainm remdin: fross martar mar nglédenn, DBenalt co mBrig roméir Brigit bin bale ncslann barr find for sig End. conn eid caillech mErenn Feb. 17: Dlomiths cc o hollmér {fail Chormaice co nglanbal (ail Fintan fli Chiiona Eanigadbat Epil. 193-6; Istdiuscud lac, Epil, 309-12: Mierere ih, (Lat. “have mercy upon me") ci dirdaea cri; a rig lath fore; do:formaiga ance, @ fou noscharu: Deirtht i laith nime a Dénéir, no-tguldiu E. THE IMPERFECT INDICATIVE. E.l. The forms. ‘The Ole Irish imperfect indicative basically corresponds to the English past habitual ofthe type ‘(he used to take’ (Ole. no:gaibed) but sometimes seems (as in E.2a below) to dnote a repeated past action best translated into English as ‘kept taking” te, Moreover, since the periphrastic progressive type bof oc gabail (she was taking” described atthe end of IMLA.2c above was stil no more than optional in tie Old Irish period, an Olr. form lke rno:galbed may on occasion be more appropriately rendered ‘(s)he was taking” (e.g. some ‘mperfects in 2a below). As pointed outwith reference tothe substantive verb in L.A.3, the imperfect endings ae always conjunct. There is thus only a single set and it is impossible fora simple verb with imperfect ending ostand alone: where no other conjunct ‘particle is present, the meaningless dummy particle ro is used in order to keep such forms in ‘conjunct’ position, In C3a above it was seen how the empty element no could be used ‘with an otherwise unaccompanied simple verb in order to support an infixed pronoun. In the case of the invariably conjunct imperfect endings no has a similar ‘quasi-compound! function but this time an infix is not an essential tigger of its use: eg. 358. ipf no:be(ired “used to carry" or n-a:mbe(ired ‘used tc carry him’ and ni:be()red “used not to carry’ or ni-mbe(ired ‘used not to carry him’. No, of course, can never be used with & compound verb in Old Irish and these display the usal alternation between deuterotonic 5 and prototonic forms in the imperfect .g, 3 sg pf. do:be(i)red or d-a:mbe(I}red "used to bring (him) and n:taibred ‘used to bring (him)’. These aspects are illustrated in the first three columns below and itis to be noted that, since the imperfect (as well as the similarly formed past subjunctive and conditional; ef. I.A.1 above) was always preceded by aprocitic (no if all else failed), it could never take a sufixed pronoun. Wh st ‘82 83 1 sg. no:marb(a)inn do:be(irinn “gain -benfajinn 2sg. no:marbtha do-be(irthea — -gaibthea -benta Beg. mo-murbud —nlsived —do.be(red ——-gaibed —~benad pl. no:marbm(ajis —ni:léicmis do:be(ijrmis 2pl. no:marbih(aje —niléicthe do:be(irthe ——-gaibthe 3pl. nocmarbifajis —niléictis dovbe(iris, = —gaibtis -gaibmis -benm(a)is “bent(aje “dent(ais “The imperfect paradigm of the substantive verb in IILA.3 will serve to ilustrate hiatus formation. The occurrence of palatal and non-palatl consonance conforms to that ofthe corresponding present in all ofthe classes above except, where basic palatal stem-fnal ‘consonance is found throughout the imperfect paradigm of verbs with eorvocalism such as do:beir with palatal -r-in all persons. This contrasts with th alteration between palatal in some persons and non-palatal in others of the corresponding present indicative and imperative (see A.2a and D.1 above). As noted in A.2b above, S1 verbs with o and a vocalism such as org/aid slays" or can(a)id ‘sings’ resisted palatalisation wherever the ‘vowel ofthe following syllable was retained, This trait would be expected to produce an alteration in their imperfect between anon-palatal consonant in the unsyncopated 1/3 sg ‘anda palatal consonant where a following vowel had been syncopated as usually happened in the other persons, a prediction duly supported by the ipf. 35g. fris:orcad vs. 3p _fris:oire tis “used wo offend (ris: org) with non-palatal an palatal /rg respectively in the ‘Milan Glosses. See A.2b above on the deleition of dental -th- to -t-in the 2sg./pl. ending after the final nasal dental of ben-. Note that 3p. /-dis’/ is sometimes written -dfa)is rather than -f(ais after a consonant (whence -marbdfa)is, -Iécdis, -be(rdis ete.) in conformity with the spelling rule given in LB.1. E.2. Reading practice. (2) The two short passages below are taken from the famous story entitled Aided Com inna Cordae “The Death ofthe Smith's Hound’ from the series of flashback narratives in Tain Bd Ciiailage known as Macgnimrada Con Culainn ‘Ci Chulainn's Boyhood Deeds’. This tells how Sétantae acquired the name Cii Chulainn ‘Hound of Culann’ when only six years old by slaying Culann the smith’s ferocious watchdog. In the first excerpt Séantac keeps defeating the est of king Conchobor’s hundred and fifty foster-sons in an early form of hurling. Conchobor is so impressed that he invites the boy to a feast at Culann’soy homestead. After finishing his games Sétantze sets out after the other guests but ‘meanwhile his imminent arrival ha been forgotten, Culann's gates closed and the mighty ‘hound set loose to defend the property. The second excerpt depicts Sétantae’s remarkable lack of eoncem as he carries on playing while th: savage dog is bearing down upon him. Intan ba hdin phil do:gnids no:linad-som in poll di-aliathréitib,nf-cumeatis in maice 4 irchlige, In tan batar hé-som uili do:bidetis in poll arca‘cliched-som a denur ‘conmartéiged cid denlfathrbit ind Ticin gillae fo todain. F-a:n-opair in ci. No:fethed-som a cluiche colic. Fo:cerded a athrdit ,fo:cerded a loirg in-adiad co:mbenad in liathrdit, Nibo mé in band ol-daas a chéle, Ocus fo:cerded a bunsaig in-a ndiad di-a ngabail re tothaim. (©) The first passage below isa slightly modified version of the account in Togail Bruidne ‘Da Derga (plus four druids ftom the similar one in Serglige Con Culainn) ofa tarb-feis “bull-feast’ to determine the nextking of Tara, Tre second from Tecosca Cormalc depicts the king's youthful behaviour, which tums out have been rather more impulsive than that urged by him in later years. See IV.C.1b on the use of personal numerals in the dat 88 No:marbtis tarb snocithed denfer aiéithdé ,novibed aenbruithe ,no:caintl celthirdruld 65 firinne fir inca ligiu. Ad:cld fer in-a chotld , is é bad ri. A-t:baildis a béol in tan as:beired got, No:gonainn muic, noslenainn lore mba m’enur. Noccinginn ar chuire cbicir -mba céiciur, Basa oirgnech L-mbsa dechenborack. Basa indredach i-mbsa fichteck. Basa cathach tmbsa cétach, a5 CHAPTER VI THE PRETERITE ACTIVE STEM. AND THE RELATIVE MARKERS. A. THE PRETERITE ACTIVE. AA. The spreterite, ‘The past (indicative) tense or preteriteis typically use in Old Tish tn cemate an action ae having happened in the past without more ado anda such corresponds quite closely tothe English past tense seen in expressions of the type he went home (yesterday) or our team won (the match last Sunday), The formation used by all weak verbs is known as an s-preterte forthe simple reason that it basically adds a suffix -s)s- to the stem vowel appearing inthe 3sg. conjunct ofthe present indicative WI -marba and W2 -lé(el etc. (see V.A.2a). Hence the underlying preterit stems of these verbs may be represented as ‘marb-as()(-) (and so on forall other WI verbs) and léi)e-is(s)- (and so on far most W2 verbs) Except forthe 3,, the endings added to these stems are exactly the same as those of the Si present illustrated by be()rid in V.A.2a, bearing in mind that the innovatary Ise. mm) was strictly confined tothe present indicative. In the 39g. abs. the stem-final-(s) was merely palatalised (no ending -id being added) and in the 3sg. conj. the stem-final syllable was usualy lost completely, although in W2a the final vowel -i was occasionally retained (cg -rai ‘says’ or ‘said’). Needless to say, when an extra syllable was added as a personal ending, the vowel of a second non-fnal syllable was syncopated before the (9) in accordance with the rule in ILA.de. These features can be clearly seen in the following typical paradigms, where the 2pl. abs, slot has been left blank because of the lack of attestations and some doubt as to what the Old Irish form would have been (probably marbast(a), iicest(ae, to judge from s-fut. gigeste in IX.6d), WI abs. conj. — W2. abs. con Ise. marbsu -marbus eosin -lé(eius 2se. marbs(a}i smarb(ajis Leet é(eis 35g. marb(a)is -marb leas leet) tpl ‘marbs(alim(m)i_ -marbsam lédesim(mjt—-16(esem 2p. -marbs(ajid éQesid Spl marbs(ajit ——-marbsat le(pesit é(ieset Although all W2 verbs form their present in the same basic way, it becomes necessary to ‘make a further division into two sub-groups, here termed W2a and W2b, a regards the \way in which the other tense or mood stems are formed. W2a verbs such as Iéicid retain the same basic form of the root throughout, e.g lép’/ in both present and preterite (and subjunctive too; see VIIL.2a) with no fundamental change either to the vowel orto the I i6 quality ofthe final consonant. In W2b, however, the root vowel is typically u followed by palatal consonance (although this may be non-palatal in certain circumstances, asin the ‘case of W2a; see V.A.2a) inthe present but o followed by non-palstal consonance inthe preterte (and subjunctive), except for palatalisation where the following vowel was lost bby syncope or apocope (cf. V.A.2a): e.g, pres. 358, scurid, scuir ‘unyokes, releases, Finishes’, pret. 38g. scorais, ~scoir, 3pl. scoirst, -scoirset ‘unyoked’ and 3sg. pres. do:lug(a}t (or do:lu()gi) ‘forgives’, pret. do:loig forgave’ In addition to being the characteristic formation ofall weak verbs, the s-preterite is also ‘used by most H3 verbs and the two strong verbs with a root ending ina labial, namely ‘gaibid ‘takes’ (and its various compounds; stem gab-as(s)-) and Ibid ‘drinks’ (stem {b-is(9)): eG Pret. 398.204, -so,3pl soist, -soiset“tumed’ (HB sod, soi “tuns’), 38. situ, -élai, Spl. as:lulset, -élaiset ‘escaped’ (H3 as:lu,-éla ‘escapes, absconds"); Is. sgabsu, -gabus, 38g. 26(a)is,-gab, Ipl -gabsam, 3p. -gabsat “took” (S2 gaibid,-gaib); 5sg. ibis, i, Il. -tbsem, 3p ibset ‘drank’ (SI ibid, ib). ‘The s-pretrite, then, is easy to recognise outse the 3sg. conjunct on account of @ characteristic -1(-) following the otherwise virtully unchanged verbal root, sometimes ‘with an intervening vowel. The 35g. conjunct is nore difficult since it lacks -s(-) and so isnot quite so straightforwardly distinguished from the corresponding suffixlesspreterite in A.Afbelow, Since the (oot-)final consonant ofthe suffixless preterite is always palatal, the s-pretertes of W1 and gaibid with non-palatal (root-}final consonant in the 3sg. ‘conjunct are distinctive, a more general diagnostic being thatthe root ofan s-preterite is typically has the same form as the present (including in H3 and allowing for o instead of ‘vin W2b), whereas that ofthe suffixless preterteis liable to significant modifications such ‘as reduplication or change of vowel. Moreover, if verb is weak, it is bound to have an s-preterite, the suffixless preterite being confined to certain types of strong and hiatus verb (HB being a slightly prey area with mostly s-preertes but a couple of suffxless formations 100). A. Ther-preterite, ‘This isthe characteristic formation of strong verbs with root-final ror suchas beirid “carries (ber-) and ini “grinds (mel), S2 gaiid ‘alls’ (gar). A suffix 1) is added direct tothe root to produce pet. bert), mell(),gart() and soon, the singular endings having basic affinities with those of the SI present except forthe 39) andthe s-pretrit, “There are sporadic examples of the expected corresponding plural endings, eg. 3pl ‘con:geltat ($1 con:gellgrazes jointly") and fo:gelta (SI forgel‘grazes (on), but the usual plural endings are the same as those ofthe sflnlss preterit in AAf below and as such do not distinguish formally between absolute and conjunc. Ii tobe noted that invariably in 3 and also in a few SI verbs -n(-)is a present marker only and does not “appear inthe preterit, subjunctive and fture tems. Thus 83 sern(jid-sem ‘sews’ a nd SI marn(a)id-mairn “betrays” have basic roots ser- and mer- with a final + caling forat-prterie, the 35g. conj.of whichis-sert‘strewed? and-mert betrayed respectively A similar relationship applies to SI a-:bail(, -ep(a/il() ‘dies’ (present (J) L! from "by assnilation, end a asi root Bet: el compounded with prever et it ‘throws it ouvaway') with its preterite 36g. a-:belt, pelt ‘died’. The following are typical paradigms. - . 81 bid cae’ ple) 2 dosairn gai promises (ep; deat. fms sn birte bit ee tee eee Be hint in doa gin See bt er aineget Ipl— (Jbertam(m)ar 2pl.— (-)bert(alid 3pl— ()bertatar (ove. syncopated (-)bertar) do:air-n-gertam(mjar docairn-gert(ajid do:air-n-gertatar (also syncopated do:air-n-gertar) Some ston ves wth nln and alse a prem rh thn sues pete nis eatgory (ce A in tatese tena occ ine tue (ten cf Ole el, Mod and nergy Cent ‘nies g's esa Gove och vee hte So soot “puted out spe (1 ders, at sem our ach Trove (Shane ocean ve’; -anch rece” agi oem. rey -oF ew Se ones “eh Shon rotorg)satarecht she pe erect ey woe ‘sathey raised henelfemssvesy SI eel ep tow fey ence do,formacht ‘increased! (Si do-for-maig, cpa. to + for + root mag- ‘grow"). : ‘The eee is party easy to deny as itis lay clearly characte by sui nea ny ocr fa eb shang wih ool on Conéitions must be met. aS cae A3. Reading practice, Below is slighty moied version ofthe opening Ol rh ha of the episode Brgi’s Lien IDA Se, See V-A.2¢ on fei fom Fechtn-and birt Dubthach ini Brit di-areiceririg Laigen, fri Dim n reicefririg Laigen, fri Dinlang, do chumail fognama dé hiaire co-ta:sdid-si a les-méthair bn no:gatad-si cach nisin tg do chet Dé. Fo-s:n-dcab-si Dubthach inna charput oce-a chomés i falthgi in dine , ficcab a chailg na comair. Do-s:bert-sido chlam... As:bert Dubthack frisinn rig: ‘Cren dim-sa mvingin dochum fognama duit, ar a-t:roiliset a bésa’.ss ‘Add, The suffixless preterite. (a) Almost all strong verbs not inthe foregoing catzgories (and a couple of H2, notably ‘compounds of -ci ‘see’, as well asthe H3 pair do:goa ‘chooses’ and fo/a)id ‘passes the night, sleeps (with)’) display this formation, whichis o called because its characterised by a special set of endings as well as modifications to the verbal root itself and not by @ specific suffix such as those seen in te s- and - preterites. Although realised somewhat differently inthe singular of vowel-final and consonant-fnal roots, the inflectional pattern ‘sfundamentally dhe same throughout, namely 1/2sg.-C (or stressed ~d, unstressed-a), 33. (C’ (ora diphthong'vowel other than -d/-a such a8), 1p -(am)mar, 2p. -(aid, 3p “(aliar without differentiation of absolute and conjunct (see the paradigms inf below). However, several different subtypes are tobe distinguished on the basi ofthe manner in ‘which the root is modified. () The reduplicated preterite is typically formed by repeating the initial consonant of the root (underlined) and placing a vowel afterit e.g, 388. S2 maid “breaks? > memaid ‘broke’, SI ligidlicks'> lelaig ‘licked’, rigd binds > reraig bound”, sligid ‘cuts down’ > selaig, dingid “presses, crushes’ > dedaig ‘pressed, crushed’, gonaid ‘slays’ > geguin ‘slew’, canaid ‘sings’ > eachain ‘sang’, hongaid ‘breaks’ > bobaig ‘broke’, S3 lenaid “sticks* > lf ‘stuck’, eenaid ‘sells’ > rie sold”. Since the reduplicate syllable ends in 2 ‘vowel, the initial consonant of the root is lenited, as in ca-chain or the more extreme serlaig (< *se-laig). Ifthe reduplicator contains ¢ front vowel e ori, its consonant is ‘predictably palatal (c.g. memaid {rv’e¥86’D) but in SI the inital consonant of the root itself is typically non-palatal, even if it is pelatal m the present (e.g. ligid [L'iy’@8"], ddingid(a'in'g 90") but lelaig (elOy’', dedaig [deb }). Sometimes the reduplicator ‘was lost by regular syncope: e.g, the -cf compounds 3sg. prs. do:éc(a)t, - -d, the 3sg active of mostpresens,subjuncives and fires), palatal -s (the 358. active ofthe s preterit, subjunctive and s-futre) ot -r (non-palatl inthe | and Spl sufixess and t-preterite; plata in the deponent and passive, on which see VID, the relative ending was-e: eg, 398 pres. (ig) it‘ goes” withrel. ete ‘who goes’, pret. birt ‘bore’ with re. erte “who bore, which he bore’, suffxess pret. bo! “as id “went, gid “prayed, geguin slew’, ir “sold? with re. bole “who was’ vide “who went gdde “who prayed, which he prayed/whom he besought’, gence ‘who slew, whom/which he slew’, rire “who sold, which he sold’; 1 pl. pres. guldmi ‘we pray/beseech’, tagmat ‘we go' with rel. guidme ‘which we pray/whom we beseech’, Fagmae “which we go (to), spre. celebirsimml ‘we aid farewell” ( /O/ realised as [a] between two non-palatal consonants or, as evidenced by the ‘classical’ equivalent do:esmet of rucesmor below, e] between palatal and anon-palatal; see IB.6). The voicing of final -th to -d after an unstressed vowel (see IILA.3) is not seen in the Cambrai Homily (no examples inthe passage below but diltth for ‘classical’ diltud ‘denial’, proclitc cith for ld ‘though it be, céssath for céssad ‘suffering’ and so on elsewhere inthe Homily; see VIIL.66), while ocuis below does not display the depalatalisation of consonants that is typical of Old Irish proclities such as ocus (cf. proclitic dochum in the vocabulary with nnon-palatal -ch- and n/m) versus is stressed counterpart fo())chim(m) with palatal -ch- and -m(m)). Sometimes Cambrai has more moder spellings (e.g. ocus below ot Ipl i