Old Irish

Old Irish (Goídelc; Irish: Sean-Ghaeilge; Scottish Gaelic: Seann Ghàidhlig; Manx: Shenn Yernish or Shenn Ghaelg; Old Irish: ᚌᚑᚔᚇᚓᚂᚉ), sometimes called Old Gaelic,[1][2] is the oldest form of the Goidelic for which extensive written texts are extant. It was used from c. 600 to c. 900. The primary contemporary texts are dated c. 700–850; by 900 the language had already transitioned into early Middle Irish. Some Old Irish texts date from the 10th century, although these are presumably copies of texts composed at an earlier time period. Old Irish is thus forebear to Modern Irish, Manx, and Scottish Gaelic.[1]

Old Irish is known for having a particularly complex system of morphology and especially of allomorphy (more or less unpredictable variations in stems and suffixes in differing circumstances) as well as a complex sound system involving grammatically significant consonant mutations to the initial consonant of a word. Apparently,[* 1] neither characteristic was present in the preceding Primitive Irish period, though initial mutations likely existed in a non-grammaticalized form in the prehistoric era.[3] Much of the complex allomorphy was subsequently lost, but the sound system has been maintained with little change in the modern languages.

Contemporary Old Irish scholarship is still greatly influenced by the works of a small number of scholars active in the late 19th and early 20th centuries such as Rudolf Thurneysen (1857–1940) and Osborn Bergin (1873–1950).

Notable characteristics of Old Irish compared with other old Indo-European languages, are:

Old Irish also preserves most aspects of the complicated Proto-Indo-European (PIE) system of morphology. Nouns and adjectives are declined in three genders (masculine, feminine, neuter); three numbers (singular, dual, plural); and five cases (nominative, vocative, accusative, dative and genitive). Most PIE noun stem classes are maintained (o-, yo-, ā-, -, i-, u-, r-, n-, s-, and consonant stems). Most of the complexities of PIE verbal conjugation are also maintained, and there are new complexities introduced by various sound changes (see below).

Old Irish was the only member of the Goidelic branch of the Celtic languages, which is, in turn, a subfamily of the wider Indo-European language family that also includes the Slavonic, Italic/Romance, Indo-Aryan and Germanic subfamilies, along with several others. Old Irish is the ancestor of all modern Goidelic languages: Modern Irish, Scottish Gaelic and Manx.

A still older form of Irish is known as Primitive Irish. Fragments of Primitive Irish, mainly personal names, are known from inscriptions on stone written in the Ogham alphabet. The inscriptions date from about the 4th to the 6th centuries. Primitive Irish appears to have been very close to Common Celtic, the ancestor of all Celtic languages, and it had a lot of the characteristics of other archaic Indo-European languages.

Relatively little survives in the way of strictly contemporary sources. They are represented mainly by shorter or longer glosses on the margins or between the lines of religious Latin manuscripts, most of them preserved in monasteries in Germany, Italy, Switzerland, France and Austria, having been taken there by early Irish missionaries. Whereas in Ireland, many of the older manuscripts appear to have been worn out through extended and heavy use, their counterparts on the Continent were much less prone to the same risk because once they ceased to be understood, they were rarely consulted.[5]

The earliest Old Irish passages may be the transcripts found in the Cambrai Homily, which is thought to belong to the early 8th century. The Book of Armagh contains texts from the early 9th century. Important Continental collections of glosses from the 8th and 9th century include the Würzburg Glosses (mainly) on the Pauline Epistles, the Milan Glosses on a commentary to the Psalms and the St Gall Glosses on Priscian's Grammar.

Further examples are found at Karlsruhe (Germany), Paris (France), Milan, Florence and Turin (Italy). A late 9th-century manuscript from the abbey of Reichenau, now in St. Paul in Carinthia (Austria), contains a spell and four Old Irish poems. The Liber Hymnorum and the Stowe Missal date from about 900 to 1050.

In addition to contemporary witnesses, the vast majority of Old Irish texts are attested in manuscripts of a variety of later dates. Manuscripts of the later Middle Irish period, such as the Lebor na hUidre and the Book of Leinster, contain texts, which are thought to derive from written exemplars in Old Irish now lost and retain enough of their original form to merit classification as Old Irish.

The preservation of certain linguistic forms current in the Old Irish period may provide reason to assume that an Old Irish original directly or indirectly underlies the transmitted text or texts.

The consonant inventory of Old Irish is shown in the chart below. The complexity of Old Irish phonology is from a four-way split of phonemes inherited from Primitive Irish, with both a fortis–lenis and a "broad–slender" (velarised vs. palatalised) distinction arising from historical changes. The sounds /f v θ ð x ɣ h ṽ n l r/ are the broad lenis equivalents of broad fortis /p b t d k ɡ s m N L R/; likewise for the slender (palatalised) equivalents. (However, most /f fʲ/ sounds actually derive historically from /w/.)

Some details of Old Irish phonetics are not known. /sʲ/ may have been pronounced [ɕ] or [ʃ], as in Modern Irish. /hʲ/ may have been the same sound as /h/ or /xʲ/. The precise articulation of the fortis sonorants /N/, /Nʲ/, /L/, /Lʲ/, /R/, /Rʲ/ is unknown, but they were probably longer, tenser and generally more strongly articulated than their lenis counterparts /n/, /nʲ/, /l/, /lʲ/, /r/, /rʲ/, as in the Modern Irish and Scottish dialects that still possess a four-way distinction in the coronal nasals and laterals. /Nʲ/ and /Lʲ/ may have been pronounced [ɲ] and [ʎ] respectively. The difference between /R(ʲ)/ and /r(ʲ)/ may have been that the former were trills while the latter were flaps. /m(ʲ)/ and /ṽ(ʲ)/ were derived from an original fortis–lenis pair.

Old Irish had distinctive vowel length in both monophthongs and diphthongs. Short diphthongs were monomoraic, taking up the same amount of time as short vowels, while long diphthongs were bimoraic, the same as long vowels. (This is much like the situation in Old English but different from Ancient Greek whose shorter and longer diphthongs were bimoraic and trimoraic, respectively: /ai/ vs. /aːi/.) The inventory of Old Irish long vowels changed significantly over the Old Irish period, but the short vowels changed much less.

1The short diphthong ŏu may have existed very early in the Old Irish period/but not later on.

/æ ~ œ/ arose from the u-infection of stressed /a/ by a /u/ that preceded a palatalized consonant. This vowel faced much inconsistency in spelling, often detectable by a word containing it being variably spelled with au, ai, e, i, or u across attestations. Tulach "hill, mound" is the most commonly cited example of this vowel, with the spelling of its inflections including tulach itself, telaig, telocho, tilchaib, taulich and tailaig.[6][7]

Archaic Old Irish (before about 750) had the following inventory of long vowels:

1Both /e₁ː/ and /e₂ː/ were normally written é but must have been pronounced differently because they have different origins and distinct outcomes in later Old Irish. /e₁ː/ stems from Proto-Celtic *ē (< PIE *ei), or from ē in words borrowed from Latin. e₂ː generally stems from compensatory lengthening of short *e because of loss of the following consonant (in certain clusters) or a directly following vowel in hiatus. It is generally thought that /e₁ː/ was higher than /e₂ː/.[8] Perhaps /e₁ː/ was [eː] while /e₂ː/ was [ɛː]. They are clearly distinguished in later Old Irish, in which /e₁ː/ becomes ía (but é before a palatal consonant). /e₂ː/ becomes é in all circumstances. Furthermore, /e₂ː/ is subject to u-affection, becoming éu or íu, while /e₁ː/ is not.

2A similar distinction may have existed between /o₁ː/ and /o₂ː/, both written ó, and stemming respectively from former diphthongs (*eu, *au, *ou) and from compensatory lengthening. However, in later Old Irish both sounds appear usually as úa, sometimes as ó, and it is unclear whether /o₂ː/ existed as a separate sound any time in the Old Irish period.

3/ou/ existed only in early archaic Old Irish (c.700 or earlier); afterwards it merged into /au/. Neither sound occurred before another consonant, and both sounds became ó in later Old Irish (often ú or u before another vowel). The late ó does not develop into úa, suggesting that áu > ó postdated ó > úa.

1Early Old Irish /ai/ and /oi/ merged in later Old Irish. It is unclear what the resulting sound was, as scribes continued to use both and to indicate the merged sound. The choice of /oi/ in the table above is somewhat arbitrary.

The distribution of short vowels in unstressed syllables is a little complicated. All short vowels may appear in absolutely final position (at the very end of a word) after both broad and slender consonants. The front vowels /e/ and /i/ are often spelled ae and ai after broad consonants, which might indicate a retracted pronunciation here, perhaps something like [ɘ] and [ɨ]. All ten possibilities are shown in the following examples:

The distribution of short vowels in unstressed syllables, other than when absolutely final, was quite restricted. It is usually thought that there were only two allowed phonemes: /ə/ (written a, ai, e or i depending on the quality of surrounding consonants) and /u/ (written u or o). The phoneme /u/ tended to occur when the following syllable contained an *ū in Proto-Celtic (for example, dligud /ˈdʲlʲiɣuð/ "law" (dat.) < PC *dligedū), or after a broad labial (for example, lebor /ˈLʲevur/ "book"; domun /ˈdoṽun/ "world"). The phoneme /ə/ occurred in other circumstances. The occurrence of the two phonemes was generally unrelated to the nature of the corresponding Proto-Celtic vowel, which could be any monophthong: long or short.

Long vowels also occur in unstressed syllables. However, they rarely reflect Proto-Celtic long vowels, which were shortened prior to the deletion (syncope) of inner syllables. Rather, they originate in one of the following ways:

Stress is generally on the first syllable of a word. However, in verbs it occurs on the second syllable when the first syllable is a clitic (the verbal prefix as- in as·beir /asˈberʲ/ "he says"). In such cases, the unstressed prefix is indicated in grammatical works with a following centre dot (·).

As with most medieval languages, the orthography of Old Irish is not fixed, so the following statements are to be taken as generalisations only. Individual manuscripts may vary greatly from these guidelines.

The Old Irish alphabet consists of the following eighteen letters of the Latin alphabet:

The following table indicates the broad pronunciation of various consonant letters in various environments:

Angle brackets ⟨⟩ here indicate graphemic differences to the unmutated consonant.

When the consonants b, d, g are eclipsed by the preceding word (always from a word-initial position), their spelling and pronunciation change to: ⟨mb⟩ /m/, ⟨nd⟩ /N/, ⟨ng⟩ /ŋ/[11]

Generally, geminating a consonant ensures its unmutated sound. While the letter ⟨c⟩ may be voiced /g/ at the end of some words, but when it's written double ⟨cc⟩ it's always voiceless /k/ in regularised texts; however, even final /k/ was often written "cc", as in bec / becc "small, little" (Modern Irish and Scottish beag, Manx beg).

In later Irish manuscripts, lenited f and s are denoted with the letter h ⟨fh⟩, ⟨sh⟩, instead of using a superdot ⟨ḟ⟩, ⟨ṡ⟩.[11]

When initial s stemmed from Primitive Irish *sw-, its lenited version is ⟨f⟩ [ɸ].

The slender (palatalised) variants of the 13 consonants are denoted with /ʲ/ marking the letter. They occur in the following environments:

Although Old Irish has both a sound /h/ and a letter h, there is no consistent relationship between the two. Vowel-initial words are sometimes written with an unpronounced h, especially if they are very short (the Old Irish preposition i "in" was sometimes written hi) or if they need to be emphasised (the name of Ireland, Ériu, was sometimes written Hériu). On the other hand, words that begin with the sound /h/ are usually written without it: a ór /a hoːr/ "her gold". If the sound and the spelling co-occur, it is by coincidence, as ní hed /Nʲiː heð/ "it is not".

The voiceless stops of Old Irish are c, p, t. They contrast with the voiced stops g, b, d. Additionally, the letter m can behave similarly to a stop following vowels. These seven consonants often mutate when not in the word-initial position.

In non-initial positions, the single-letter voiceless stops c, p, and t become the voiced stops /g/, /b/, and /d/ respectively unless they are written double. Ambiguity in these letters' pronunciations arises when a single consonant follows an l, n, or r.[11] The lenited stops ch, ph, and th become /x/, /f/, and /θ/ respectively.

The voiced stops b, d, and g become fricative /v/, /ð/, and /ɣ/, respectively—identical sounds to their word-initial lenitions.

In non-initial positions, the letter m usually becomes the nasal fricative //, but in some cases it becomes a nasal stop, denoted as /m/. In cases in which it becomes a stop, m is often written double to avoid ambiguity.

Ambiguity arises in the pronunciation of the stop consonants (c, g, t, d, p, b) when they follow l, n, or r:

After m, the letter b is naturally a stop /b/. After d, l, r, the letter b is fricative /v/:

After n, l, or r, the letter g is usually a stop /g/, but it becomes a fricative /ɣ/ in a few words:

The letters l, n, r are generally written double when they indicate tense sonorants and single when they indicate lax sonorants. Originally, it reflected an actual difference between single and geminate consonants, as tense sonorants in many positions (such as between vowels or word-finally) developed from geminates. As the gemination was lost, the use of written double consonants was repurposed to indicate tense sonorants. Doubly written consonants of this sort do not occur in positions where tense sonorants developed from non-geminated Proto-Celtic sonorants (such as word-initially or before a consonant).

Geminate consonants appear to have existed since the beginning of the Old Irish period, but they were simplified by the end, as is generally reflected by the spelling. Although, ll, mm, nn, rr were eventually repurposed to indicate nonlenited variants of those sounds in certain positions.

Written vowels a, ai, e, i in poststressed syllables (except when absolutely word-final) all seem to represent phonemic /ə/. The particular vowel that appears is determined by the quality (broad vs. slender) of the surrounding consonants and has no relation to the etymological vowel quality:

It seems likely that spelling variations reflected allophonic variations in the pronunciation of /ə/.

Old Irish was affected by a series of phonological changes that radically altered its appearance compared with Proto-Celtic and older Celtic languages (such as Gaulish, which still had the appearance of typical early Indo-European languages such as Latin or Ancient Greek). The changes were such that Irish was not recognized as Indo-European at all for much of the 19th century. The changes must have happened quite rapidly, perhaps in only one or two hundred years around 500–600, because almost none of the changes are visible in Primitive Irish (4th to 6th centuries), and all of them are already complete in archaic Old Irish (8th century). A capsule summary of the most important changes is (in approximate order):[12][13]

The following are some examples of changes between Primitive Irish and Old Irish.

These various changes, especially syncope, produced quite complex allomorphy, because the addition of prefixes or various pre-verbal particles (proclitics) in Proto-Celtic changed the syllable containing the stress: According to the Celtic variant of Wackernagel's law, the stress fell on the second syllable of the verbal complex, including any prefixes and clitics. By the Old Irish period, most of this allomorphy still remained, although it was rapidly eliminated beginning in the Middle Irish period.

Among the most striking changes are in prefixed verbs with or without pre-verbal particles. With a single prefix and without a proclitic, stress falls on the verbal root, which assumes the deuterotonic ("second-stressed") form. With a prefix and also with a proclitic, stress falls on the prefix, and the verb assumes the prototonic ("first-stressed") form. Rather extreme allomorphic differences can result:

The most extreme allomorphy of all came from the third person singular of the s-subjunctive because an athematic person marker -t was used, added directly onto the verbal stem (formed by adding -s directly onto the root). That led to a complex word-final cluster, which was deleted entirely. In the prototonic form (after two proclitics), the root was unstressed and thus the root vowel was also deleted, leaving only the first consonant:

In more detail, syncope of final and intervocalic syllables involved the following steps (in approximate order):

All five Proto-Celtic short vowels (*a, *e, *i, *o, *u) survived into Primitive Irish more or less unchanged in stressed syllables.

However, during the runup to Old Irish, several mutations (umlauts) take place. Former vowels are modified in various ways depending on the following vowels (or sometimes surrounding consonants). The mutations are known in Celtic literature as affections or infections such as these, the most important ones:[24]

Nominal examples (reconstructed forms are Primitive Irish unless otherwise indicated):

The result of i-affection and a-affection is that it is often impossible to distinguish whether the root vowel was originally *e or *i (sen < *senos and fer < *wiros have identical declensions). However, note the cases of nert vs. fiurt above for which i-affection, but not a-affection, was blocked by an intervening rt.

Proto-Celtic long vowels and diphthongs develop in stressed syllables as follows:

The Old Irish diphthongs úi, éu, íu stem from earlier sequences of short vowels separated by *w, e.g. drúid- "druid" < *dru-wid- "tree-knower".

Most instances of é and ó in nonarchaic Old Irish are due to compensatory lengthening of short vowels before lost consonants or to the merging of two short vowels in hiatus: cét /kʲeːd/ ‘hundred’ < Proto-Celtic kantom (cf. Welsh cant) < PIE *kṃtóm.

See Proto-Celtic for various changes that occurred in all the Celtic languages, but these are the most important:

Old Irish preserves, intact, most initial clusters unlike many other Indo-European languages.

Many intervocalic clusters are reduced, becoming either a geminate consonant or a simple consonant with compensatory lengthening of the previous vowel. During the Old Irish period, geminates are reduced to simple consonants, occurring earliest when adjacent to a consonant. By the end of the Old Irish period, written ll mm nn rr are repurposed to indicate the non-lenited sounds /L m N R/ when occurring after a vowel and not before a consonant.

Lenited stops *x *ɣ *θ *ð generally disappear before sonorants *r *l *n *m, with compensatory lengthening of the preceding vowel. Many examples occur in reduplicated preterites or words with consonant-final prefixes (such as ad-):[9]

However, *θr, *βr, *βl survive: críathraid "he perforates" < PCelt *krētrāti-s; gabur "goat" < PCelt *gabros (cf. Welsh gafr); mebul "shame" (cf. Welsh mefl).

Nouns decline for 5 cases: nominative, accusative, genitive, prepositional, vocative; 3 genders: masculine, feminine, neuter; 3 numbers: singular, dual, plural. Adjectives agree with nouns in case, gender, and number. The prepositional case is called the dative by convention.

Verbs conjugate for 3 tenses: past, present, future; 3 aspects: simple, perfective, imperfective; 4 moods: indicative, subjunctive, conditional, imperative; 2 voices: active, and passive; independent, and dependent forms; and simple, and complex forms. Verbs display tense, aspect, mood, voice, and sometimes portmanteau forms through suffixes, or stem vowel changes for the former four. Proclitics form a verbal complex with the core verb, and the verbal complex is often preceded by preverbal particles such as (negative marker), in (interrogative marker), ro (perfective marker). Direct object personal pronouns are infixed between the preverb and the verbal stem. Verbs agree with their subject in person and number. A single verb can stand as an entire sentence. Emphatic particles such as -sa and -se are affixed to the end of the verb.

Prepositions inflect for person and number, and different prepositions govern different cases, sometimes depending on the semantics intended.