1

Galthach hAthevu
Modern (Revived) Gaulish

Tengu in Galth
esi galataca atebia tengua sena indias brogias galatacias auuot inte nouion ris bitou nouiou indos cantaios uoconti cintos Esi Galthach hAthevu tengu sen in bri Galthach, vthu in nhi ri vithu ni in aiuchan gwochon cin. Modern Gaulish is the old language of the Gaulish country, made anew for the new world of the twentyfirst century.

2 Contents Section Page #

Title page Contents Galthach hAthevu Modern Gaulish Ar In Cingen Galthach On Methin I. Introduction The Gaulish Language Reconstruction Historical Survival of Minority Languages Survival of Gaulish Evolution of Gaulish Designation of the Gauls: Keltoi Galatae Keltoi Versus Galatae Designation of the Gaulish Language Language The Language Speaks II. Grammar 1. Orthography 2. Phonology Vowels Consonants 3. Sound Changes Historical changes Changes in Gaulish Diphthongs Consonant clusters Tau Gallicum The group -ctThe group sc Word initial u- before vowel The treatment of s The treatment of -nt The treatment of -nd The treatment of geminate consonants The treatment of semi-vowel i-

1 2 6 6 6 9 9 9 10 10 10 11 11 12 13 13 13 15 15 16 16 17 18 18 19 21 22 22 22 23 23 24 24 24 25 25

3 The treatment of semi-vowel wThe treatment of wordfinal open vowel -o Apocope Graphic representation of sound changes Emphasis Applied example Mutated consonants in word-initial position Schwa Summary of sound changes 4. Personal Pronouns Personal pronoun morphology Personal pronouns as subject or object of a verb Personal pronouns and prepositions Prepositions 5. The verbal paradigm Verb classes Present tense Past tense Future tense Conditional tense Present perfect Past perfect Future perfect Conditional perfect Imperative Verbal noun Direct object of a verbal noun Summary of verbal paradigm The verb to be Summary of the to be paradigm The verb to have Passive or impersonal forms Ablative absolute Progressive form 6. Adjectives Order Natural adjectives Derived adjectives 7. Adverbs Natural adverbs Adjectival adverbs 8. Gender 9. Article 10. Plural formation Standard plural formation Plural after numbers 25 26 26 27 29 30 30 31 31 32 32 33 35 38 39 39 39 41 42 42 42 44 44 44 45 45 46 47 47 49 49 50 51 52 53 53 54 54 57 57 58 58 60 61 61 62

4 Dual plural Collective plural 11. Indication of Possesion Genitive of apposition Possessive pronouns Genitive particle iParticle of quantity u12. Questions Question words Question word compounds Question formulation 13. Negation Declarative negation Interrogative negation Replies to questions 14. Initial Consonant Mutation Mutation of initial consonants in Celtic languages Mutation in ancient Gaulish Mutation changes Context for mutations Contact mutations Grammatical mutation Unmutable words Summary of mutation contexts 15. Word order 16. Demonstratives 17. Locatives 18. Syntax Verbal noun clauses Adjectival clauses Subordinate clauses Conjunction clauses Question word clauses 19. Spatial Paradigm 20. Temporal Paradigm 21. Comparative Paradigm First degree superlative Equative degree Second degree superlative Similarity Sameness Quantity paradigm Diminutive form 22. Word formation Historical compound nouns Noun-noun compounds 63 64 64 64 66 67 68 69 69 70 71 72 72 72 72 72 72 73 75 76 76 78 80 80 80 83 85 86 86 88 90 92 93 93 95 95 95 96 96 97 98 98 100 100 101 102

5 Compound adjectives Prefixes Substantivising suffix for adjectives Prepositions as prefixes Calques Verbal word derivation Abstract noun suffixes: -tlo, -lo, -anto, -eno, -man, -naman, -on Agentive noun suffixes: -iatis, -eto, -ilo Verbal noun formation Agentive form formation Abstract noun formation Verbal adjectives Secondary abstract noun derivation Summary of verbal word formation paradigm Derivation of verbs from nouns and adjectives 23. Numerals 24. Expressions and turns of phrase Liking and loving Wanting and needing Expression of wish Uses of to have The verb to have with a specific subject Expression of obligation Expression of occurrence and presence Reflexivity Basic phrases of courtesy 25. Cultural Background Gaulish calendar Translation of the names of the months Astronomical analysis Pantheon of gods and goddesses 26. Conclusion III. References/ Bibliography Gaulish Welsh Breton Cornish Irish Scottish Gaelic General Appendix I: Galataca Sena Appendix II: The Coligny Calendar by Helen McKay Appendix III: Alternative Gaulish History 104 105 107 108 109 110 110 111 111 113 114 116 116 117 118 118 122 122 123 124 124 125 126 127 127 128 129 129 130 135 137 140 141 141 143 144 145 146 147 147 150 153 170

6 Galthach hAthvu - Modern Gaulish Ar in Cingen Galthach Esi Galatha hol rhanthu en dri ran, plch athrva in Belgh en hon, athrva in Achithan en hal, ach plch athrva en in trthu in tth anwthu Celth gw Galth en s dengu ch, ach Galli en in tengu Rmach. Esi ansamalas tengu, bs ach rith enthers ol. Esi in avon Garun enther in Galth ach in Achithan, ach in avn Mathron ach Schan enther in Galth ach in Belgh. Esi in Belgh drutham in tth-sin ol. Esi s pelam au vli ach au haiedh anchwru in Brvins Rhmach, ach esi s amapis miam tar chweneth en hgha peth anauch en hv blath in tth. Ceth, esi s nesam in Gerwn tr in avon Rin, ach esi cingen enthers aman hol. Esi Galatia olla randatua en tri randa, pelouca atrebant indi Belgi en oina, atrebant indi Aquitani en alla, ac pelouca atrebant en inda trita inda touta anuantua Celti ue Galates en eian tengua ix, ac Galli en inda tengua Romaca. Esent ansamalassi tengua, bessu ac rectoues enteris olli. Esi inda abona Garunna enter indes Galates ac indi Aquitani, ac indas abonas Matrona ac Sequana enter indes Galates ac indi Belgi. Esent indi Belgi drutami indas toutas sinde ollas. Esent sies pelames au beslougi ac au agedoues anueroues inda Provincia Romaca, ac esent sies ambapistues meiames taro ueneti agiou pettias anauacas auuiou blata inda touta. Coettic, esent sies neames indi Germani tre inda abona Rina, ac esi cingeno enteris aman olla. All Gaul is divided into three parts, of which the Belgae inhabit one, the Aquitani another, and the third those who in their own language are called Celts or Galates, and in the Roman language Galli. All these differ from each other in language, customs and laws. The river Garonne separates the Galates from the Aquitani; the Marne and the Seine separate them from the Belgae. Of all these people the bravest are the Belgae. They are the furthest away from the culture and civilized ways of the Roman Province, and are least often visited by the merchants who bring luxuries which tend to make people soft. Also they are nearest to the Germans across the river Rhine and are continually at war with them." On Methin Ponch r hi mi on methin, bri nhi nesWhen I went one morning, new dawn close to-it r hpis mi lerna, -v gnath en I saw a vixen, she had a baby in her r rhdhi mi ch blal, en pen ich I gave her a bullet in her head f f f f f fee fa foo fa fed Ponch r hi mi on methin, en havnal anel bren tree r hpis mi ton crachan, b mr ach sen old r vathi mi ch in nherthach, math ar ben w. I wo m, in a creek, below a I saw a toad then, he was big and I hit him strongly, well on his head

7 f f f f fen fed Ach esi on riem, on rieth you gava ni in canchu, ach rana ni ch it on riem ach on rieth f f f f f-eth foowha one for me and one for you fee fa foo fa we take the treasure and we split fee fa foo fa

and there is one for me, one for

Ponch r hi mi on methin, a ghn in duvr-alv wIwom, for the knowing of the rains r hpis mi ton molth, en dhepri brach arnch I saw then sheep eating malt of wheats r rhithi mi chs di halis, a vrs s gerv I ran them off a cliff, to break their brains f f f f f fee fa foo fa fain Ponch r hi mi on methin, en lhchu gwevrach light r hpis mi cath-chwir, galv, dwiedh, gwrach feral r hli mi m gun grdh, goch en voch crach f f f f fach Ach esi on riem, on rieth you gava ni in canchu, ach rana ni ch it on riem ach on rieth f f f f f-eth foowha one for me and one for you fee fa foo fa we take the treasure and we split wIwom, in amber I saw a tomcat, fat, ugly, I fed my dog his heart, red in his bloody mouth fee fa foo fa fouth and there is one for me, one for

Ponch r hi mi on methin, a gam am har vri mi w.I.w.o.m., t. walk around on flat land r hpis mi ins clus plch r shdhi cenen bri I saw a hole where a foul rabbit sat r ghavi mi grchin, r hvo mi cuchul di I took his skin,I made a good hat fi f f f fi fee fa foo fa fat Ponch r hi mi on methin, a shami ardh ar vr wIwom, to stand high on a hill

8 r hpis mi ins b, en dr wars en lh r v mi s rhimn, can m dhalam im f f f f fch Ach esi on riem, on rieth gava ni in canchu, ach rana ni ch on riem ach on rieth f f f f f-eth foowha I saw there cows, lying in dead land I cut their throats, with my own hands fee fa foo fa fand and there is one for me, one for you we take the treasure and we split it one for me and one for you fee fa foo fa

Ponch r hi mi on methin, in gnas en in hdhan r hpis mi ins cench en hv clus en in lhan r ghavi mi ch an ti, r chwen mi in hisaran f f f f fan feel

wIwom, the earlyness in the smell I s. a plough making holes in a field I took it home, I sold the steel fee fa foo fa

Ponch r hi mi on methin, r lh in dh am hr wIwom, the day it lay around about r hpis mi ins lich, plch r gam mi ar nhather I s flat rock where I stepped on a snake r gni ton m gch, mar adhim en ver he then bit my leg, carrying death to me f f f f fer fee fa fu fa fee Ach esi on riem, on rieth you gava ni in canchu, ach rana ni ch it on riem ach on rieth f f f f f-eth foowha one for me and one for you fee fa foo fa we take the treasure and we split and there is one for me, one for

The two first documents ever produced in Modern Gaulish: the opening paragraph of De Bello Gallico Ar In Cingen Galthach, and On Methin, a song about applied Australian ecology.

I. Introduction The Gaulish language This document presents the revival of the Gaulish language. Gaulish is a Celtic language which was spoken widely throughout Western Europe from at least the seventh century BCE, when it was first attested, until the sixth century CE, when it was last attested. Its range covered the areas west and south of the Rhine, east of the Atlantic, north of the Pyrenees, throughout the Alps from west to east, south of the Alps into the Po valley, and eastwards into the Balkans and the Hungarian Plain. Through immigration it was carried into Asia Minor in the fourth century BCE and survived there at least until the fourth century CE. In the first century CE it was recorded by Tacitus as being virtually identical to and mutually intelligible with the language of the British Isles (referred to as Brittonic), which would later develop into the Welsh, Cornish and Breton languages. In the framework of the Celtic language family, Gaulish is also, but more distantly, related to Irish, and its daughter languages Scottish Gaelic and Manx. The area in which Gaulish was spoken was conquered by the Romans between approximately 150 BCE and 52 BCE, and incorporated into the Roman Empire. The language was put under pressure by Latin, first as used by the administration and military of the Empire, then as used by the state enforced christian religion, and finally by the early medieval christian church. Eventually the language gave way to a variety of Romance and Germanic languages, and ceased to be spoken. Reconstruction

10

The Gaulish language is reasonably well attested, if imperfectly. There is a certain amount of recorded material available, and new material continues to be unearthed at a steady pace (e.g. the Chartres find, discovered early 2012). Drawing on the existent available material, and making use of the surviving Brittonic languages, as well as the Gaelic languages, for support and comparitive studies of such things as vocabulary, semantics and grammatic structure, a modernised version of the Gaulish language is here presented. Departing from the state in which Gaulish was last attested, that is Late Gaulish, the language of circa the fifth century CE, a series of soundchanges, phonetic evolutionary processes and grammatic innovations are postulated. As such, a hypothetical evolution of the language is constructed, the proposed outcome of which is a practically useable modern Celtic language, to be situated in the framework of the modern Celtic languages. While the process of reconstructing or reassembling a language is challenging, it has been done as conscientiously as possible, starting from the original material and attempting to stay as faithful as possible to it, while applying a set of changes which could have been reasonably expected to have happened to the language had it not ceased to be spoken. These changes are based on evolutionary processes which can be observed in the available authentic material, as well as on related processes which have occurred in the related surviving languages. As much as possible, justification for changes and adaptation is provided by drawing from the original material. Creative imagination, or, to put it differently, making up random stuff , has been kept to a minimum. These various changes, adaptations and processes will be discussed in detail in the various sections dealing with them in the body of this document. Historical Survival of Minority Languages The Gaulish language was possibly last attested in the mid sixth century by Gregory of Tours, giving a date of approx. 550 CE. It is easy to imagine that in remote rural areas it survived a lot longer than that; a wild guess might put its demise in those areas into the late 700s. This position is supported by the fact that there is evidence of non-culturally dominant languages in conquest situations that have survived as peasant languages for several hundred years. While the longest historically attested such survival (Elamite) runs up to 1300 years post conquest, there is a certain indication that a survival of 800-900 years is quite common. Several examples of such a survival of minority peasant languages not supported by the culturally and politically dominant classes are known and documented, such as the previously mentioned Elamite (1300 years without written attestation, Central Asia), Ladin (850 years, South Tyrol), Cornish (approx. 850 years), Polabe (775 years, Central Europe), Crimean Gothic (1100-1300 years, arguably; Eastern Europe), and Dalmatian (approx. 1000 years, extinct 1898, descended from Latin, Mediterranean) (Sala & Vintila-Radelescu 1984). Survival of Gaulish

11 Charlemagne gathered a collection of ancient folktales, beliefs and legends from among the people of the countryside. There is speculation that it may well have contained the last records of the language. The collection was burned by one of his successors as being pagan and therefore evil. The time of Charlemagne has been suggested as the period of widespread decline of the language (Piqueron 2006), because of state-enforced Christianity, and there are claims that the last survivals lasted into the 12th century in remote mountainous areas. Suggestions to this effect are found mainly in the work of Hubschmied (1938). Although this has in more recent years been severely criticised by linguists and experts in the field, there is also qualified and informed support for the concept (e.g. Bhrghros, 2010, 2012). Therefore, estimations for the time of extinction of the language range from the most optimistic, 800 years ago, through the cautiously speculative, 1200 years ago, to the conservative, 1500 years ago. Evolution of Gaulish The construction of the modern Gaulish language situates the evolution of modern Gaulish in the cultural continuum of the surviving Celtic languages, and adopts grammatical innovations that are common to all surviving or revived Celtic languages. As such, this work subscribes to the notion forwarded by Isaac (2007) which holds that many of the unusual features of modern Celtic languages are due to linguistic factors demonstrably inherent and innate to the Celtic language family, and are not due to the presence of a presumed and unproven prehistoric Afro-Asiatic substrate language, as suggested and defended most recently by Venneman (2003), building on the earlier work of Morris-Jones, Pokorny and Gensler (in Venneman 2003). Broadly speaking, the hypothetical proposed evolution of Gaulish is characterised by two main processes, both of which are attested in the Gaulish material, and abundantly recorded in the early historical stages of the surviving Celtic languages. These processes, which are best thought of as phonetic erosion, are apocope, or the loss of wordfinal endings and syllables, and lenition of word-internal consonants. The resulting loss of case markings led to profound changes in the grammatical structure of the language. These processes and their outcomes will be discussed in detail in the body of this document. Designation of the Gauls: Keltoi The question of how the Gauls refered to themselves has long been and continues to be the subject of intense debate, heated argument and imaginative speculation. Two parallel terms can be identified as having historical and cultural authenticity. On the one hand, the word Keltoi has been associated with the Celtic people of Western Europe since they were first mentioned by Classical Greek authors (McCone 2006). Caesar indicated that in his day, the Gauls refered to themselves as Celtae at least those Gauls inhabiting the area between the Garonne and Seine rivers (see e.g. McCone 2006). The etymology of the word Keltoi/ Celtae is not widely agreed upon, and is sometimes held to refer to a state of being hidden, or unseen, which in turn is often tentatively linked to Caesars statement that the Celts considered themselves to be descended from their god of the underworld (Caesars Dis Pater). The notion of hailing from the underworld and

12 therefore coming from under the ground, can with some goodwill be conceptually linked with the concept of being hidden, as someone who is under the ground can certainly be said to be hidden from the view of someone above the ground (McCone 2006). Be that as it may, the fact remains that the name Keltoi, whatever its meaning may have been, is historically inextricably linked with the people of Western Europe. Galatae A second appellation that is historically linked with the people of prehistoric Western Europe is that of Galatae (McCone 2006). It has been argued that the term Galatae refered to the young, possibly landless and presumably restless male warriors of Celtic society, and was derived from the root *gal-, meaning vigour, fighting, endowed with the regular Celtic suffix of agency (i)ati. The meaning of Galatae would thus be the warriors, the vigorous ones (McCone 2006, Bernard Mees 2008, pers. com.). This interpretation is reinforced by the documented existence in central Gaul (Auvergne) of a shrine or temple dedicated to or named in honour of Uasso Galatae, a name in which the component Uasso appears to refer to young male(s) (McCone 2006). The term Uasso Galatae appears to have survived to this day in the name of the Central French village Jaude (*galat- > *gald- > *jald- > * jaud-), and the word Galatae has been argued to have led, through a process of linguistic distortion and cultural misinterpretation, to the appellation Galli as used by Caesar (McCone 2006). Keltoi versus Galatae In view of the above information, it would appear that the two terms are mutually compatible and complementary, in as much that it appears that the word Celtae seems to refer to the main body of the nation(s), and that the word Galatae seems to refer to the young males who set forth to carve out a piece of world of their own. The association of the term Galatae with roving warrior bands and groups of mercenaries, such as those who settled in Asia Minor, appears to confirm this notion. A parallel can be drawn with the middle ages, when the crusades were welcomed by certain sections of society as a way of getting rid of young males with nothing to do, no land to inherit and altogether too much time on their hands. As such, it could be said that the word Celtae refers to the core of the people or nations who stayed at home, tended their fields and told stories to their kids, and that the word Galatae refers to the young guns who left home to fight, rape and pillage and generically seek their fortune elsewhere. The above notion also appears to be reinforced by the report of the Sicilian-Greek historian Diodorus Siculus, writing between 60 and 35 BC, who describes the people referred to as Celts as living in the hinterland of Massilia and on the slopes of the Alps and Pyrenees, i.e. the area roughly corresponding to the Roman Province of Narbonensis. According to Siculus, the people North of that region were known as Galatae, all the way North and East to Scythia. In addition, Siculus provides a suitably heroic mythology for the origin of the name, which, tellingly, was based on conquest and expansion by force.

13 Although it may appear that the term Celtae has the older tradition, it seems quite certain that the term Galatae is a native appellation, and one, moreover, which seems to have been associated with rather a larger area, i.e. all of Gaul North of Narbonensis and Galatia in Asia Minor. In addition, it appears to be this term which has led to the use of the word Galli by the Romans, which in turn, handed down through the centuries, has come to be associated with the Celtic people of Western Europe, and which has become absorbed into most if not all Western European languages to designate those people, their culture and their language (Eng. Gallic, Fr. (formerly) Gallique, Ger. Gallisch; Eng. Gaulish and Fr. Gaulois are derived from Germanic *walha-, a term designating a nonGermanic speaking person found living in erstwhile Roman provinces; this term was applied to Romance and Celtic speakers indiscriminately, and is at the origin of the designations Wales, Welsh, Wallonia, Walloon and Wallachia, among others). As such, it seems that the root gal- is linked inextricably with the ethnographic designation of the first recorded people of Western Europe between the Rhine, the Alps, the Pyrenees and the Garonne. In the above context it is also worth noting that the Central Gaulish village now known as Jaude was located in the heart of what was designated by Caesar as Gallia Celtica, indicating either that a) Caesar was a liar; b) Caesar made things up as he wished to suit his own purposes; or c) that the use of the term Celtae was confined to the area South of the Massif Central, i.e. Gallia Narbonensis, or was at the very least used in conjunction with the term Galatae in Auvergne and presumably North from there, if Siculus is to be believed. Then again, while points a) and b) made in reference to Caesar above are certainly valid, it is also worthwhile being aware of the fact that Siculus at various times has been referred to as the greatest liar in history. Designation of the Gaulish language In the 21st century, the Celtic language family is well defined as containing Welsh, Cornish, Breton, Irish, Manx and Scottish Gaelic. None of these languages retain a linguistic link with the word Celtic, nor indeed were they ever linked to the concept of Celtic until 1707 when Edward Lluyd published his linguistic analysis of the languages of Britain. As such, it would seem both misleading and presumptuous to lay claim to the word Celtic to refer to what is known in English as Gaulish. Therefore, for the sake of clarity as well as historical accuracy and continuity, the name of the modern Gaulish language will be derived from the attested term galatia (designating Gaul), an adjectival derivative of which would take the form *galataca. This word, through the application of the regular modern Gaulish sound changes, becomes Galthach in modern Gaulish. In keeping with historical records, the term of galatia itself, rendered as Galatha, will be proposed as the indigenous name of Gaul and will be used as such. The word athevu to designate modern is of contemporary origin and only translates as modern by proxy. Literally translating as again-alive, a better rendition is revived. As such it is a combination of two historically attested words, the prefix ate, indicating a repetition or renewal of something (Delamarre 2003, p. 59), and the word bio-, meaning alive (Delamarre 2003, p. 77). Applying the proposed modern

14 sound changes to the word atebio gives athevu. It is reasoned that while Galthach hAthevu accurately conveys the essential idea of the revived language, the term modern Gaulish is a more straightforward practical working term for discussion of the language in the English language. Therefore, Galthach hAthevu is the modern Gaulish language. Language It may also be appropriate to provide justification for the choice of the word tengu to mean language. While tengu appears to have Indoeuropean cognates in English (tongue although it is contested that this is a cognate W. De Reuse, pers. Com. 2009) as well as in Gaelic (teanga), the word is historically attested in a Cisalpine Celtic inscription from Oderzo under the form pompete(n)guaios, i.e. (son of) he who speaks five languages (Koch 2006, p.969; Coskun & Zeidler 2003, p. 44; Stifter 2008; Bernard Mees 2007 & 2009 pers. comm.). The attested word tengua, leads through a straightforward standard loss of final syllable [a] directly to the form tengu. Conversely, it has been asserted that there is no evidence on the continent of a word Celtic or other - for language that would be derived from the root *yek-, which has produced the Welsh and Breton words for language, ieith and yezh respectively (Mees 2009, pers. comm.). Therefore, Galthach hAthevu is Tengu in Galth, i.e. the language of the Gauls. The language speaks An indication of the phonetical and grammatical evolution of the language may be gleaned from the comparison of a few phrases in both modern and Classical / Late Gaulish. The first is a translation of the lines introducing modern Gaulish to the world. One hypothetical Classical Gaulish version (marked for case) could run like this: esi galataca atebia tengua sena indias brogias galatacias auuot inte nouion ris bitou nouiou indos cantaios uoconti cintos In modern Gaulish this is rendered as follows: esi Galthach hAthevu tengu sen in bri Galthach, avthu in nhi ri vithu ni in aiuchan gwochon cin. Finally, it is possible to produce a representation of the famous last words of Dumnorix, chief of the Aedui, when murdered on Caesars orders in 54 BCE, given here first in the tentatively reconstructed original Classical Gaulish, and followed by its rendition in Modern Gaulish:

emmi uiros rios ex toutia ria

15

esi mi gwir ru e dth ru I am a free man from a free nation

II. Grammar 1. Orthography The Gaulish language has historically been recorded in a number of different scripts, Lepontic, Greek and Latin, with the Lepontic representing the earliest attested inscriptions, the Greek being used in the middle period of attestation, and the Latin being used for the latest records. These latest records are however also the most prolific ones, and the ones providing the most information about the language and the various changes at work in it. Furthermore, all of the important long running texts in the Gaulish language known today are written in the Latin script. Therefore, it seems reasonable to adopt the Latin script used in those texts, with a few adjustments to the orthography to reflect the demands of the modern language. By way of graphic example, a line from the important Chateaubleau text (one of the longest, continuous running, well preserved inscriptions, discovered in 1997) looks like this in the original Gaulish: ne anmanbe gniou ape niteme uee [not by the names which are knowing about ownership may you desire it]

16 The modern Gaulish language is written entirely phonetically, and to represent the sounds of the language it uses a version of the Latin alphabet which contains the following 20 basic characters: a, b, c, d, e, f, g, h, i, l, m, n, o, p, r, s, t, u, v, w. These are augmented by the following consonants: ch, dh, gh, lh, nh, rh, th, gw, chw The vowel inventory is also extended to include the following long vowels: , , , , and the following diphthongs: au, i, i, i, i The various sounds represented by the above are discussed below.

2. Phonology The characters discussed above are used to represent the sounds of the language phonetically, in such a way that a given letter almost always represents a given sound. There are a few exceptions, which will be discussed further below. Vowels The vowels of the modern Gaulish language can be either long or short. These vowels are clear, as is the case in e.g. Welsh and Spanish. The short vowels are: a e o u i a = [a] as in at e = [e] as in hell o = [o] as in fog u = [u] as in book i = [i] as in it The long vowels are: . = [a:] as in father = [e:] as in lay, but without the yod sound

17 = [o:] as in tote = [u:] as in shoe = [i:] as in see The diphthongs are: au, i, i, i, i u = [au] as in how i = [a:j] as in good bye i = [e:j] as in day i = [o:j] as in toy i = [u:j] as in French brouillard All other vowels can occur adjacent to other vowels without combining their sounds into diphthongs, so that each vowel is pronounced independently with its own syllabic peak: gwerach = [gweru:ax] (feral) b = [bo:e:] (cows)

Consonants The basic consonants of the Gaulish language are: p, t, c, b, d, g, f, v, n, r, l, m, s Of these: c is always [k] g is always [g] r is the rolling r of Scottish borrow. s is [s] in word initial and word medial position, and [z] in word final position As mentioned above, in addition to these regular consonants Gaulish also has: ch, dh, gh, lh, nh, rh, sh, th, gw, chw ch = [x] as in Scottish loch dh = [] as in English there gh = [] as in Greek lh = [xl] an /l/ preceded by [x] nh = [xn] as for l rh = [xr] as for l

18 sh = [] as in English shine th = [] as in English thin gw = [gw] chw= [xw] Lastly, Gaulish has the semi-vowels: w, i w = [w] as in English will i = [j] as in English you Of these, w never occurs freestanding: it is always adjacent to other consonants, and never occurs intervocalically. Conversely, the character i indicates the regular vowel [i] when directly following a consonant, or when preceding a consonant by itself: mi = [mi], me imi = [imi], of-me However, i indicates the semi-vowel [j] when following another vowel, when preceding another vowel in word initial position, or when between two vowels: i = [a:j], to go ir = [ja:r], chicken iedh = [a:je], face Many of the various sounds discussed above are the result of a process of sound changes, the beginnings of which can be discerned in the attested Gaulish material. These sound changes are discussed in detail below. 3. Sound Changes Historical changes The process of the changing of the sounds of the Gaulish languages is one of the two main processes which transform it from an ancient language to a modern language, the other one being apocope, or the loss of final syllables, of which more later. The changing of the sounds of a language is best thought of as a process of phonetic erosion: over a certain period of time, by virtue of living in the mouths of people, the sounds of languages change. In a Western Indo European context, sound changes have been historically observed, recorded and extrapolated in the Romance, Germanic and Celtic language families. Very broadly speaking, in the Celtic language families there are two sets of sound changes that have affected the two separate surviving Celtic language families: the Brittonic (Welsh, Cornish, Breton) and the Goidelic (Irish, Scottish Gaelic, Manx). In the

19 Brittonic language family, an important change that occurred was the voicing of intervocalic voiceless stops (* -ata- > * -ada-), and the spirantisation of voiced intervocalic stops (*-aba- > * -ava-). In a nutshell, pre-sound change intervocalic [p, t, c] became post-sound change intervocalic [b, d, g]; at the same time, pre-sound change intervocalic [b, d, g] became [v, , ], the latter of which eventually disappeared; intervocalic [m] also became [v]. A number of changes also affected the Brittonic vowel system, but these are not relevant here. In the Goidelic branch of the Celtic language family, intervocalic voiceless stops were spirantised: [t, c] became [, x] (Goidelic did not have the sound [p]). At the same time, intervocalic voiced stops became spirantised as well: [b, d, g] became [v, , ], as in Brittonic. Goidelic also experienced a shift of emphasis towards the first syllable of a word, which, in words of more than two syllables, typically resulted in the loss of the second syllable, a process referred to as syncope. The two sound changes outlined above are clearly illustrated in the respective treatment by the two language families of the word Caratacos, beloved-like/ loveable, which is attested in Classical Gaulish as well as in Welsh and Irish. Classical Caratacos > Welsh Caradog (voicing of voiceless stops) > Irish Carthach (aspiration of voiceless stops) For a detailed discussion of these sound changes, including this example, see Lambert 2003 (p. 28, 47). Changes in Gaulish The sound changes described above occurred in the respective language families around the 5th-6th centuries CE, again broadly speaking, in a period of considerable political and social upheaval in Western Europe, which appears to have triggered a period of linguistic innovation and change. It can be reasonably expected that the Gaulish language, at that time, was subject to the same pressures that caused sound changes in its neighbouring and related languages, and therefore also experienced a process of sound change. In order to construct Gaulish as a modern language, it is important and essential to discern and to apply this sound change. Therefore, the available Gaulish material has been investigated for clues as to what this change might have been like. Below are presented epigraphic instances of genuine, attested Gaulish material that testify to changes at work in the phonology of the language. Most of these, if not all, can be found in Delamarre 2003, as per individual reference, unless otherwise specified; when available, their primary source is included and follows the page number in Delamarre 2003. aballos > avallo; i.e. [b] > [v] (p. 29; Endlichers Glossary) anman > anuan; i.e. [-nm-] > [-nw-] (p. 50; Larzac, Chateaubleau) iouincos > ioinchus; i.e. [c] > [x] (p. 191, anthroponomy)

20 arganto- > arxanti; i.e. [g] > [] (Lambert 2003, p. 48; Suessons coin) > arganthoneia; i.e. [t] > [] (p. 53; Galatian source) (g)nata > gnatha, i.e. [t] > [] (p. 181; spindle whorl of Saint-Rvrien) litan- > lithan-, i.e. [t] > [] (p. 204; anthroponomy) ate- > atha-, i.e. [t] > [] (p. 57, p. 214; anthroponomy) *sisagsiou > siaxsiou, i.e. [g] > [] (p. 273; Chateaubleau) *agat > axat, i.e. [g] > [] (p. 63; Marcellus of Bordeaux) luge > luxe, i.e. [g] > [] (p. 210; Chamalieres) ambio > ape, i.e. [amb] > [b] (Bernard Mees 2010, Bhrghros 2013, pers. com.; Chateaubleau and Rom) briuo > brio, i.e. [-w-] > [--] (p. 89; Endlichers Glossary) iouinc- > ioinc-, i.e. [-w-] > [--] (p. 191, anthroponomy) magiorix > maiorix, i.e. [-g-] > [--] (Lambert 2003, p. 46; anthroponomy) catugenus > catuenus, i.e. [-g-] > [--] (Lambert 2003, p. 46; anthroponomy) traget- > treide, i.e. [-g-] > [-i-] (p. 300; Endlichers Glossary) [-t-] > [-d-], (p. 300; Endlichers Glossary) *brogilos > breialo, i.e. [-g-] > [-i-] (p. 91; Endlichers Glossary) badio > * baio > baio, i.e. [-d-] > [] > [--] (p. 63, anthroponomy) Stirona > irona > Sirona, i.e. [st-] > [ts-] > [s-] (p. 282, theonomy)

The above instances can be summarised as follows: b > u (v) c > ch (x) d> g > x () g>i g> m > u (w) t > th () t>d amb > b, assimilation of nasal to vowel st > ts > s In the above, the grapheme x is considered to represent the spirantisation of intervocalic g > [] (Delamarre 2003, p. 63, p. 210; Lambert 2003, p. 46, p. 48) An analysis of the above data indicates that the intervocalic consonants of Gaulish were subject to a process of spirantisation or fricativisation, which can be represented as follows: p>p t > th [] c > ch [x]

21 b > u [v] *d>? g > x [], i [j] or m > u [w] > ss [s] [mb > nasal-b] The only exception to this appears to be the rendition of t- as d- in treide, which appears to indicate a voicing of an intervocalic consonant. Lambert (2003, p. 207) considers treide as having been treated as a word of Vulgar Latin and as such as having undergone Romance lenition, where intervocalic voiceless stops become voiced (Western Romance). In support, it is the opinion currently held by scholars in the field of Continental Celtic linguistics that the possibility that treide can be considered as indicative and representative of a genuine Gaulish sound change is negligeable (Bhrghros, Mees, Stifter, Gwinn pers. com. 2009). The remainder of the data appears to uniformly point towards a process of spirantisation of intervocalic consonants. The existence of lenition, the process of weakening of pronunciation of consonants, in this case through spirantisation or fricativisation, in Gaulish has been proposed and defended by Gray (1944), accepted by Fleuriot (in Delamarre 2003, p. 63), rejected and objected against by Lambert (2003) among many others (e.g Watkins 1955), and cautiously and somewhat sceptically regarded as a possibility by Delamarre (2003). Nevertheless, as Eska (2008) emphatically suggests, proposes and defends, the data appear to point in this direction. In addition to the above, there is widespread academic agreement on the disappearance of intervocalic u- and g- (Delamarre 2003; Lambert 2003). In regards to g-, a position is here adopted where g- becomes spirantised when occurring in consonant clusters, and becomes i- in intervocalic position and word finally when preceded by a vowel: -Cg- > -C- (e.g. argant- > arxant-) -gC- > -C-VgV- > -ViV- (e.g. brogil- > breial-) -Vg > -Vi Diphthongs The sound change discussed above gives rise to a number of modern diphthongs which are not derived from inherited Indo European or Proto Celtic diphthongs: Classical Gaulish > Modern Gaulish dago > di (good) aneg- > ani (protect)

22 brogi > bri (country) lugus > li (a god) This therefore produces the diphthongs: i, i, i, i in addition to the vowel inventory discussed previously. The Classical Gaulish diphthong ou-, which itself was the result of an older eu-, is also attested as undergoing an evolution: touta > tota, i.e. [ou] > [o:] (people, tribe, nation) crouco > crocu, i.e. [ou] > [o:] (hillock, hill, heap, pile) This indicates that: ou > o [o:]

Consonant clusters Several examples of the data analysed above provide information on what happens to stops as part of consonant clusters: argant- > arganthoneia: -nt- > -nthargant- > arxant-: -rg- > -rxiouinc- > ioinch-: -nc- > -nchThis indicates clearly that the process of spirantisation was productive across consonant clusters. Tau Gallicum The Tau Gallicum in Gaulish, represented by a wide variety of graphemes including , , , , ds, dd, tth, th, ts, ss and ss with a horizontal bar through it, is a matter of some controversy: some disagreement exists about its actual phonetic value. However, most scholars in the field of Continental Celtic linguistics agree that the Tau Gallicum represented a dental affricate [ts] (Delamarre 2003; Lambert 2003; Mees 2008, 2010; Eska 2008). Regardless of its original phonetic value, there is universal agreement that over the course of its attestation the sound of the Tau Gallicum evolved to become [s], which may become apparent from comparison with modern related languages, as well as with late Gaulish material.

23 e.g.: neamon > nesaf (next - Welsh) mi > mis (month Welsh) Meilus > Medsilus > Messilus (Lambert 2003, p. 46) Caraounios > Carassounios (Lambert 2003, p. 46) adsedo- > aedo- > assedo- (reside, establish, Delamarre 2003, p. 33) stir- > ir- > sir- (star, Delamarre 2003, p. 282) Therefore, the change postulated for the Tau Galicum is: st, ts, > s The group ctThere is no direct evidence of the evolution of ct- in Gaulish. However, the similarity of its evolution in French (due to the Gaulish substrate) and in Brittonic is considered strong enough by Lambert (2003, p. 49) to conclude that its treatment was: -ct- > -xt- > -it-, with i = [j], second element of diphthong. e.g.: lact- > lait (milk, French) > llaeth (milk, Welsh) where -act- > -ait- [a:jt]. This then undergoes spirantisation of the word final stop and becomes aith [a:j]. The group -sc There is in the attested material no indication of an evolution of the group sc in Late Gaulish. Therefore, no soundchange is postulated for this group, which is maintained as it is. e.g. mesco- > mesc (drunk) Word initial u- before vowel There is no direct indication of the evolution of word initial u- before vowel, thought to have been [w] in Classical Gaulish (Delamarre 2003, p. 309; Lambert 2003, p. 46; Mees, Stifter pers. comm. 2009). In the surviving related languages, u- before vowel becomes fin Goidelic, and gw- in Brittonic. A comparison with the treatment of word initial u- before vowel in French, however, gives us some indication of what may be acceptably proposed as a plausible evolution of this phoneme. The French language has known a change of word initial [w] to [gw], often in words of Germanic origin, but not restricted to them; to wit guerre (war), Guillaume (William), g(u)arde (ward), gant (want i.e. mitten), and Gascogne (Wasconia). The French language came to replace the Gaulish language in much of its domain and in the

24 process demonstrably absorbed a certain amount of Gaulish substrate influence (Lambert 2003). Since the evolution of Brythonic [w] to [gw] appears to be partially paralleled in French, it is deemed acceptable to postulate an evolution of word initial [w] to [gw] for Modern Gaulish. While this may appear controversial because of its lack of actual attestation, it is worth bearing in mind that the phenomenon of changing word initial [w] to [gw] through fortition or strengthening is fairly common cross linguistically (B. Mees pers. comm. 2009). While it happened in Brythonic across the three attested languages Welsh, Cornish and Breton, it didnt occur in Breton until the eleventh century (UT n.d.), a date well past the linguistic and cultural unity of Breton with the other Brythonic languages. This appears to indicate that the fortition of word initial [w] to [gw] happened due to internal processes, as opposed to shared innovations across the Brythonic family that could possibly be postulated. Furthermore, it is reasonable to posit that the loss of word final consonants (loss of s, -n, -m attested) would have resulted in a situation where word initial u- ended up in an intervocalic environment, which would have led to its disappearance. It can therefore be postulated that a strengthened gw- developed to prevent the loss of word initial u-. Therefore, in keeping with the arguments outlined above, a sound change of word initial u- before vowel to gw- is proposed for Modern Gaulish. The treatment of s Although there has been controversy around this, it is now widely agreed that s is maintained in Gaulish (D. Stifter, B. Mees, C. Gwinn, Bhrghros pers. com. 2009). There is no indication of it becoming [h] word initially, as in Brittonic. While there have been instances where s is thought to have disappeared intervocalically, it has been argued strongly and persuasively that these were exceptions (Stifter 2009). Therefore, the position is taken here that s is maintained unchanged. However, there is evidence that adjacent to l- s became absorbed into ll- (coslo- > collo-, Delamarre 2003, p. 127). Therefore, in modern Gaulish, s does not occur in clusters with l-. It is proposed here that in its word initial and word medial position s is [s], while word finally it is [z]. Finally, in a mutation context, when s- is followed by a consonant, s- is dropped, and the following consonant remains unmutated. This elision of s- is marked with an apostrophe: e.g. scrvi > crvi (to write). The treatment of ntIn word medial position, it is clear that nt- becomes nth-, as is indicated by arganto- > arganthoneia (Delamarre 2003, p. 53. For its word final treatment, it is postulated here that an intermediate nth evolved to -n through phonetic erosion and for reasons of ease of pronunciation.

25

The treatment of ndThe treatment of nd- is not clearly attested in the available material. However, two instances are known that appear to bear witness to some evolution of nd-, from which tentative clues may be extrapolated. 1. the word andecinga (agricultural measurement, similar to an acre) was absorbed into medieval Latin and hence passed into Old French as ansenge, ensenge. As such, it appears to show an evolution of ande > an, that is an elision of d-. However, this is a highly dubious example, as it testifies to an evolution in Vulgar Latin / Old French, and not in Gaulish. 2. the word andounna (low water, spring, source, Collias inscription, Delamarre 2003, p. 48) has been persuasively analysed by Lambert (1990, followed by Eska, 1992, in Delamarre 2003, p. 48) as ande-udna > ande-unna > andeunna > andounna. In this proposed evolution, the word udna (water, wave, tide) has evolved to unna by assimilation of nasality, of [d] assimilating to [n]. A position is therefore postulated by modern Gaulish that all instances of nd-, word medial and word final, assimilate to n(-). .....The treatment of geminate consonants No pertinent information is available about the evolution of geminate consonants. A position is adopted here that all geminate consonants are treated as single consonants, and as such are subject to spirantisation in word medial and word final position, if they are stops. The treatment of semi-vowel -iIn the attested material, a Latin i longa (long i) is often used to indicate a semi-vowel [j]. It is proposed here that the semi-vowel [j] is represented by the grapheme i, in keeping with its occurrence in e.g. breialo (Delamarre 2003, p. 91). Intervocalic g- will almost always become i-, pronounced [j] e.g.: trageto > trieth [tra:je] (foot) Words starting with g- will, in combination with prefixes, weaken this g- to a i-: e.g.: gar: to call s + gar: good + call = to laugh > *sgar > siar [su:jar] Word initial i- before a vowel will be pronounced as [j]:

26

e.g.: ir: chicken [ja:r] iach: sane, healthy, well [jax] ianth: wish [jano:] However, word medial i- preceded by a consonant and followed by a vowel, is pronounced [i]: e.g.: adhianth: ambition [aiano:] The treatment of semi-vowel wIt is proposed here that in words where u is followed by another vowel and receives emphasis, this u is interpreted as the long vowel [u:]: e.g. arer: to please However, in instances where the emphasis shifts away from u, it becomes the semivowel w: e.g.: arwra mi: I please The treatment of word final open vowel -o In the Late Gaulish material available, it is evident that word final open vowel o [o:] evolves to u [u:] (Delamarre 2003, p. 342; Lambert 2003, p. 62). Although apocope removes most Classical and Late word endings, there are instances where word final open vowel o is retained, most usually for reasons of necessity of distinction between near homonyms. In these cases, this end open vowel changes to u, and is usually short [u]. However, in the case of words ending on nto-, the o is maintained as such, as a long vowel [o:] which follows the cluster nth-. Again, this is done for practical reasons of clarity and disambiguity. e.g.: panto > panth (suffering) Apocope Apocope is the complete loss of word final syllables or word endings, including endings defining case. It is the second major change that has influenced the evolution of both Brittonic and Goidelic and contributed to their becoming uninflected modern languages. In the case of Goidelic, cases were redefined as revolving around the pronunciation of word-final consonants, and as such are retained to this day nominative, genitive and dative, as well as a vocative which does not however involve the pronunciation of the final consonant of a word. However, Goidelic also did lose all case endings, as did Brittonic. In the case of Brittonic, the loss of case endings resulted in the complete loss of

27 all cases, in a parallel with Romance and most of Germanic (excepting German, Icelandic and Faroese). The available Late Gaulish material shows clearly that a process of erosion of word endings was underway by the 4th century CE (see e.g Stifter 2009, 2012). This is mentioned by Delamarre, who indicates the erosion of the case endings (p. 342-43): nominative sing. os > -o (loss of s) accusative sing. on > -o (loss of n) -an > -in/ -im > -i (loss of n) Examples from the attested material include: coro bouido < * coros bouidos (Mees 2010, Chateaubleau) andoedo < * andoedon (Delamarre 2003, La Graufesenque) beni < * benin/ *benim (Mees 2010, Delamarre 2003, Chateaubleau) These are indications that the case system was being eroded. Inevitably, such erosion leads to loss of meaning and the collapse of the case system. It is reasonable to postulate that had Gaulish continued to evolve as a living language, the case system would have disappeared entirely through the phenomenon of apocope, the first stages of which can be observed in the examples cited above. As a result, the Gaulish language would have, quite likely, ceased to be an inflected language with cases, and would have, quite likely, evolved as an analytic language with prepositions. This is exactly what happened to the Brittonic languages. Graphic representation of sound changes The sound changes discussed above need to be able to be represented graphically in the written language. To recap, the graphemes used in the attested material, and their probable pronunciation values, are as follows: p>p t > th [] c > ch [x] b > u [v] *d>? g > x [], i [j] or m > u [w] mb >p = nasal vowel +b > ss [s] In Latin, the orthographic conventions of which were applied to the Gaulish texts in which these graphemes are found, the grapheme u was used to indicate both [u], [v] and [w]. It is commonly held that u for lenited b indicates [v] (Delamarre 2003, p.

28 29, and that u for lenited m indicates [w] (Lambert 2003, p. 46). It is therefore possible to posit: b > v [v] m > w [w] Similarly, the digraphs th and ch used in the Gaulish texts are drawn from the Latin orthographic inventory. They were incorporated into Latin originally to represent sounds of Greek words which were unknown to Latin; as such th and ch represented aspirated t (theta, ) and c (chi, ). While in early antiquity it is thought that these sounds represented aspirated stops, i.e. a [t] and [c] followed by a puff of air (t-huh, chuh), by Late Antiquity the respective pronunciation of these digraphs had evolved to the fricatives [] and [x]. Therefore, it is reasonable to assume that the digraphs th and ch as used in the Gaulish texts are likely intended to represent a fricativised or spirantised sound. It is postulated here that the sounds represented by the digraphs were the fricatives mentioned above. This therefore gives: c > ch [x] t > th [] The changes represented by p > p, and > ss do not represent a problem: they can be taken at face value and be represented by p, and s respectively: p > p [p] > s [s] The change represented by mb > p is interpreted as mb > nas-b, with the grapheme /p/ being used to indicate [b] at a period where /b/ was used to indicate [v] (Bhrghros 2013). It is postulated here that this led to mb > nas-b > nas > /m/, i.e. mb > m, in analogy with the treatment of nd-. This leaves the challenge of how to represent the lenited sounds of g and d. In the case of g, it is agreed (Delamarre 2003, p. 63, p. 210; Lambert 2003, p. 46, p. 48) that the lenited form of g, written as x, is intended to represent the sound []. In analogy with the lenited forms of c and t, i.e. ch and th, used in the attested Gaulish material, it is proposed here that the sound [] of lenited g be written as gh, indicating a spirantised pronunciation of g. This is further supported by the use of the digraph gh to represent an originally similar sound in Irish, Scottish Gaelic and English. The representation of intervocalic g as i, as found in the original material, does not present a problem and can be maintained as such. g > gh [] g > i [j] Finally, the spirantisation of d is not attested anywhere, and only hypothesised by Delamarre (2003, p. 63). For the sake of analogy and conformity, a position is here

29 adopted that d underwent spirantisation to [] as well in intervocalic, word final and consonant cluster position, like all other stops, and that this will be represented by the digraph dh. This latter is used in Cornish to represent the same sound, and as such is deemed acceptable in the framework of Celtic phonological graphic representation. This therefore gives: d > dh [] The graphic representation of post-sound change consonants can now be summarised as follows: p > p [p] t > th [] c > ch [x] b > v [v] d > dh [] g > gh [], i [j] m > w [w] mb > m [m]

Emphasis The emphasis in Gaulish is thought to have been predominantly on the penultimate syllable (see Lambert 2003 for a discussion; also Eska 2008). This is therefore the default emphasis position adopted by modern Gaulish: caran = friend If words receive suffixes, the emphasis is shifted to accomodate for the extended word, so that the emphasis remains on the penultimate syllable: caranach = friendly Exceptions to the penultimate emphasis rule are fusions of prepositions with personal pronouns, where the emphasis falls on the last syllable in order to emphasise the person referred to: canith = with you, emphasis on ith The fact of receiving emphasis can make a vowel change from short to long. This happens in two cases:

30 1. if the vowel in question is followed by a word final open vowel, regardless of syllable position: e.g.: gwel: to want (verbal noun) > gwla mi: I want gar: to call (v.n.) > gra mi: I call 2. if the vowel in question is in the second or subsequent syllable, and is followed by either an open vowel or a spirantised stop followed by a vowel, opened or closed: e.g.: lavar: to speak (v.n.) > lavra mi: I speak 2nd syllable, followed by open vowel car: to love (v.n.) > carthu: loved 2nd syllable, followed by spirantised stop and open vowel > carthach: loveable 2nd syllable, followed by spirantised stop and closed vowel > caran: friend 1st syllable, not followed by spirantised stop or open vowel > caranach: friendly 2nd syllable, not followed by a spirantised stop or open vowel However, there are words where vowel length may be etymologically defined. These will not be subject to these rules: e.g.: gwir: man [gwir] gwr: true [gwi:r] In the above example meaning is solely conveyed by vowel length. Applied example The graphic and emphatic conventions outlined above can now be put into practice and be applied to the example used previously to illustrate the ways in which the two surviving branches of the Celtic languages were affected by their respective sound changes. Thus, Classical Caratacos can be put through the two changes outlined above as affecting Gaulish, i.e. first apocope, loss of the final syllable: caratacos > caratac followed by spirantisation of unvoiced intervocalic stops: caratac > carathach Lastly, the emphasis is placed on the penultimate syllable, which is followed by an aspirated stop and therefore becomes long, indicated by a diacritic: carathach > carthach This can be compared to the permutations of the same word in Brittonic and Goidelic: Gaulish: Carthach (spirantisation of voiceless stops, apocope)

31 Welsh Caradog (voicing of voiceless stops, apocope) Irish Carthach (spirantisation of voiceless stops, apocope and syncope) Mutated consonants in word initial position An important part of the way the modern Gaulish language works is the mutation of word initial consonants, a phenomenon shared among all the Celtic languages. While this will be discussed in great detail in a separate section, it is important, from a phonological point of view and for the sake of clarity, to list here the various sounds that various consonants mutate into, as some of these do not occur elsewhere. Very briefly, the system of changes runs as follows: p t c > b d g > v dh gh m>w f > fh [] (bilabial) In addition, the following are included: l > lh [xl] n > nh [xn] r > rh [xr] s > sh [ ] gw > chw [xw] These are essentially spirantised sounds which, with the exception of chw, only occur word initially, and only in mutation conditions. Schwa It is proposed here that the modern Gaulish features the sound schwa in certain cases, where a word ends on a consonant followed by n, r or l. These are usually the result of apocope. In these cases, a schwa is pronounced between the two final consonants. However, as there is no graphic representation known for schwa in Gaulish, this schwa is never written, and is deduced from the word final consonant configuration: e.g.: attested word sedlo (seat) > modern Gaulish sedhl a schwa is pronounced between dh- and l: sedhl = [sel] petru (four) > pethr = [per] anectlo (protection) > anithl = [ane:jl] louernos (fox) > lern = [lo:ern] In conjunction with the pluralising suffix- this schwa sound disappears: anithl: protections [[ane:jle:] lern: foxes [lo:erne:]

32

As scha only ever occurs as the ultimate vowel, emphasis is always on the vowel before schwa. Summary of sound changes Having explored in detail the sound changes which can be discerned as having been at work in Late Gaulish, and those which may reasonably have been expected to occur had Gaulish continued to this day as a living language, it is now possible to summarise these in table form. This provides the necessary background information for the exposition of modern Gaulish grammar which makes up the remainder of this document. attested Gaulish p t c b d g m n l r s sC (Consonant) u [w] x xt nd modern Gaulish: word initial p t c b d g m n l r s sC gw ch (anywhere) -ith (anywhere) n (anywhere) initial consonant mutation b d g v dh gh w nh lh rh sh C chw word medial, inter vocalically p th ch v dh i [j] gh (rare) m n l r s sC chw word medial, in cons. cluster p th ch v dh gh 1st pos.: m 2nd pos.: w n l r s [s] sC word final

p th ch v dh i [j] m n l r s [z] sC

u [u] if emph u [u] w if n. emph

33 nt Tau Gallicum mb n (anywhere) s (anywhere) m (anywhere)

4. Personal pronouns Personal pronoun morphology The personal pronouns of Gaulish are reasonably well attested in the historical material, and are as follows: English I you he she it we you pl. they Classical Gaulish mi ti es i id ni suis sies Modern Gaulish mi ti ni s s attestation Lambert 2003, p. 69 Lambert 2003, p. 69 < lopit-es, D. Stifter and B. Mees, pers. com. 2009, after Orecilla 1997; avot-e , B. Mees pers. com. 2009 < avot-i, B. Mees pers. com. 2009 < buet-id, affixed 3rd sg neutral pronoun, Delamarre 2003, p. 93 < avot-ni, B. Mees pers. com. 2009 Lambert 2003, p. 160 Lambert 2003, p. 69; 3rd p. pl. f., extended to include masculine and feminine, as is the norm in modern Insular Celtic

For a discussion of attested forms, see Lambert (2003, p. 69, 160) and Delamarre (2003, p. 269, 277). For all intents and purposes, the 3rd pers. sing. fem.and neutre have become one and the same. As such, is used to indicate feminine subjects as well as neutre subjects, i.e. undefined, generalistic subjects such as the weather, or subjects which have not or nor yet been specified. e.g.: cna she sings esi math it [the weather] is fine Personal pronouns as subject or object of a verb The above pronouns are used when they are the subject of a sentence. When used as object in a sentence, the personal pronouns are largely the same as the above, with a few

34 exceptions. Based on the following Classical Gaulish data, it is possible to extrapolate tentative forms for object pronouns: avot-ni: made us avot-ide: made it avot-i: made her avot-e: made him avot-is: made them (in Receuil des Inscriptions Gauloises, from Bernard Mees, pers. com. 2009) This indicates that when used as object (in the accusative case), the Gaulish pronouns take the same form as when used as subject (in the nominative case). An example of this is avot-ni where the form ni, used in the accusative, is the same as the form ni encountered elsewhere in the nominative (e.g. at Thiaucourt, Delamarre 2003, p. 336, B. Mees, pers.com. 2009). This position is further reinforced by the Banassac word tiei, which is translated as at/to you-is, where the accusative form of the second person singular pronoun appears to be the same as the nominative form of the same pronoun, also attested elsewhere (Lambert 2003, p. 69). Nevertheless, the 3rd pers. pl. is attested as -is when used as object: avot-is made them (RIG, B. Mees p.c. 2009) dessu-mi-is I prepare them (Henry 1984, in Delamarre 2003, p. 141; B. Mees pers.com. 2009) This is divergent from the attested third person plural (fem.) sies. While there may be an unattested difference between the third persons plural masculine and feminin, since modern Gaulish no longer distinguishes between masculine and feminin in the third person plural, it is posited here that -is represents the generic third person plural. In view of the above information, a position is proposed for modern Gaulish where the personal pronouns when used as object are the same as the subject pronouns, with the exception of the third person plural, and with the addition of phonetic modifications to the third persons singular and plural when necessary. These are introduced to allow for clear phonetic distinction to be made between two words ending and starting on a vowel where such a distinction is deemed important, and only occur if the word preceding the object pronoun ends in a vowel. In these cases the third persons singular and plural are provided with a semantically empty particle ch- , a practice which is current in modern Breton (see MacAuley 1992). Bearing the above in mind, the modern Gaulish personal pronouns when used as object present as follows: mi me ni - us

35 ti you /ch him /ch - her s you(s) s/chs - them

While in Classical Gaulish both the subject pronoun and the object pronoun were cliticised onto the verb, as shown in the example below, in modern Gaulish these have become detached. Revisiting the previously used example, the evolution sketched above can be illustrated: dessumiis > dsa mi chs I prepare them Further examples are provided below: apsa mi ch I see her apsa mi she sees me batha mi ch I hit him batha mi he hits me The need for the particle ch- as a phonetic bridge is demonstrated below: apsa ch she sees her apsa ch he sees him apsa chs she sees them instead of apsa , apsa and apsa s, which would be phonetically awkward. However, when the preceding word does not end in a vowel, the phonetic bridge particle is not used as it is not required: apsa in gwir the man sees her Unambiguous meaning is further conveyed by the strict VSO word order, which places the subject always before the object in a multiple pronoun phrase. e.g.: apsa mi ti I see you apsa ti mi you see me Personal pronouns and prepositions An unusual feature of modern Celtic languages is the fusion of personal pronouns with prepositions, sometimes referre to as conjugated prepositions. It is a feature not frequently found among modern languages, although it was present in Latin. In the well known Classical Latin expression vade mecum, meaning come with me, the word mecum is a fusion of me- (me) and -cum (with) (W. De Reuse, pers. com. 2011). Forms such as this are no longer found in modern Romance languages; they are, however, common, even ubiquituous, in modern Celtic languages. It seems reasonable to

36 assume that such features are archaic grammatical forms no longer used in many modern languages. In the attested Gaulish material, there is precisely one instance which can be identified as a conjugated/fused preposition. The lead tablet of Chamalieres (Delamarre 2003, p. 337), which is one of the most important continuous texts in Gaulish found to date, features at line 9-10: exsops pissium i tsocanti rissuis onson A reasonable translation of this would be: no-eye willsee-I-it/her with-this for-you (pl.) our ... .....i.e.: blind I will see it with this for you our ... In this line, the word rissuis appears to be quite certainly made up of the preposition ris (for, before) and the 2nd pers. pl. pronoun suis; as such, rissuis is to be taken to mean for-you (pl). This appears to be formed on exactly the same lines as comparative modern Insular Celtic fused prepositions, e.g. Welsh gennyf < *cantimi. From this admittedly modest and humble example of a fused preposition, a system of fused prepositions is devised that is analoguous with those found in the modern Celtic languages. The preposition ris would, under the influence of phonetic erosion of the end of words (apocope), be shortened to ri. To this would be attached the various modernised forms of the prepositions discussed above. For the 2nd pers. pl. of the attested example, this would give: ris + suis > ri + s = ris for-you (pl) The forms for the other pronouns would look like this: ris + mi > ri + mi = rimi + ti > ri + ti = riti + es > ri + = ri + i > ri + = ri + ni > ri + ni = rini + is > ri + is > riis Applying the insertion of the sematically empty phonetic particle ch- as discussed above to the 3rd persons singular and plural would give: ri- > rich ri- > rich

37 ri-s > richs A position is here adopted that holds that for the sake of clarity and unambiguousness, a process of metathesis and vowel differentiation applies to the remainder: rimi > ri-im > riem ri-ti > ri-it > riet > rieth ri-ni > ri-in > rien The fusion of the preposition ri (for) with the personal pronouns would therefore be: riem: for me rieth: for you rich: for him rich: for her / it rien: for us ris: for you (pl.) richs: for them In the above cases the two vowels i and e are pronounced separately, with emphasis on the latter, the word final syllable which receives the emphasis for all the fused prepositions: riem Extending the above principles to the other preposition encountered at Line 10 of Chamalieres, canti (with), would give: canti + mi > cantimi > canthim However, a position is adopted here that holds that the preposition *canti > *canth (with), when used independently from pronouns, was subject to apocope and phonetic erosion, and in analogy with other words on word final nth dropped the final th (see above under treatment of nt-) for reasons of ease of pronunciation. This would reduce the preposition to can. It is postulated that in analogy with the shortening of canth to can, all fused forms of canth also dropped their (in those cases word medial) th-. The result would be the following: canim: with me canith: with you can: with him can: with her canin: with us cans: with you (pl.) cans: with them

38

In analogy with the two examples illustrated above, it is postulated that modern Gaulish applies two models of fusing prepositions with pronouns: 1. prepositions on i follow the first model: -em, -eth etc. e.g. di (off) (Delamarre 2003, p. 143): diem, dieth, dich, dich, dien, dis, dichs 2. prepositions on all other endings follow the second model: e.g. gw (under) (Delamarre 2003, p. 324): gwim, gwith, gw, gw, gwin, gws, gwos. There a small number of prepositions which ended on stops in Classical Gaulish. Due to phonetic erosion, apocope and a desire for ease of pronunciation, these have lost their word final stops in independent usage. However, in their fusion with pronouns these are retained. e.g. ad (to, towards) (Delamarre 2003, p. 31) > a in independent usage > adh- in fusion: adhim, adhith, adh, adh, adhin, adh, adhs e.g. ex (from, out of, Delamarre 2003, p. 169) > e in independent usage > ech- in fusion: echim, echith, ech, ech, echin, ech, echs Prepositions The following prepositions are attested or can be extrapolated from the Classical Gaulish material. While some are clearly used in a prepositional way, others are only attested as preverbal particles or prefixes. However, it is posited here that as Gaulish would have become increasingly analytic in its grammatical construction, such prepositional preverbal particles as are attested would have evolved to fully fledged independent prepositions. Their counter parts are found in both the modern Celtic and Romance language families. The following table provides an overview of such prepositions and their derivations. All are found in Delamarre 2003, unless otherwise indicated. modern Gaulish ri can ar ar ern ur cin s gwer classical Gaulish ris canti are are erno *uritocintus ossiuerEnglish for with on in front of behind against before, first after, last over attestation Delamarre 2003, p.259 p. 103 p.52 p. 52 p. 434 < Proto-Celtic *uritop. 117 p. 244-45 p. 314

39 gw au di e a i u en tre tar am r co chan ithra enther uchel anel uoau diexadiuceni tretaroambi eri co-, com*ex-canti extra enter uxello*anello under away from off out of, from to, toward, at of (poss.) of (quantity) in across through, by around, about around, surrounding than, as without beyond, outside between above below p. 324 p. 60 p. 143 p. 169 p. 31 p. 189-90 p. 308, p. 74-75 p. 163 p. 300 p. 430, 438 p. 41 p. 166 p. 121-22 p. 169 + p. 103 p. 275 (sioxti) p. 163 p. 330 by analogy with uxello-

5. The Verbal Paradigm The verbal paradigm is an area of greyness and uncertainty in the study of Gaulish. While a considerable number of verbal forms are known and understood, no full conjugations are available in any tense, mood or aspect. Therefore, modern Gaulish proposes a verbal paradigm in which verbs are not conjugated, but are instead marked for number and person by way of personal pronouns. This is not unusual in the context of Celtic languages and in the broader framework of western and northern European languages. For instance, the Gaelic languages conjugate only two out of seven persons and number for regular verbs, their copula is uniform across all seven persons and numbers, all Brittonic languages have verbal constructions where al persons use the third person singular form, in the French language the conjugations of all but the first and second plural number are phonetically indistinguishable, western Germanic verbs often only have two or three forms, and continental Scandinavian languages only have one. Verb classes Gaulish shows indications of having had several classes of verbs. Among these, thematic as well as athematic forms can be distinguished. Examples of some of these are: Thematic: uediiu-mi, liiu-mi, iegu-mi

40 Athematic: peta-me, senant, dama, axat Tentative, approximative historical paradigms for the present tense of the above could be drawn up as such for the verbs uedi- (to pray) and peta (to ask): Thematic uediiu mi uedies ti uediet e/i/id uediemu ni uediete suis uedient sies Athematic peta mi petas ti petat e/i/id petama ni petata suis petant sies

It would seem that personal pronouns were used as clitics in combination with conjugation endings. It is posited here that with the evolution of apocope and phonetic erosion, these conjugation endings were eroded and lost, leaving a bare stem or verbal root, followed by personal pronouns to convey information regarding person, number and gender. In modern Gaulish, ten tenses are constructed using this verbal root and a number of pre-verbal particles and suffixes that are attested in the Gaulish material. Present tense It is proposed here that through apocope and phonetic erosion verbal stems are left with a vestigial conjugation vowel, accompanied by a personal pronoun (itself reduced and eroded) to accord information about person and number. For reasons of practicality, unambiguousness and ease of pronunciation, it is suggested here that all verbal stems take the vowel a of the erstwhile athematic ending in combination with the appropriate personal pronoun. Verbal stems ending on a vowel (i or ) therefore drop that vowel and add a: For instance, the verbal stem uedi- goes through the following evolution: uediiu > uedii- > guedi > gwdhi The form gwdhi has thus become the new verbal stem for the verb to pray. This verbal stem is a verbal noun with specific and special properties, which will be discussed in detail below. To mark the new verbal stem gwdhi for person and number in the present tense, the word final vowel i is dropped, and replaced with a: gwdhi > gwdha This present tense form of the verb gwdhi is then endowed with the appropriate personal pronoun to convey information about person and number:

41 gwdha mi: I pray gwdha ti: you pray gwdha /: he/she prays gwdha ni: we pray gwdha s: you (pl.) pray gwdha s: they pray Exceptions to this model are mono syllabic verbs on a vowel, - or other. Unlike all other verbal stems, these forms do not lose their word final vowel, but instead add the present tense marking word final vowel a on after the vowel: e.g.: b: to cut > ba mi: I cut gn: to know > gna mi: I know Verbal stems ending in consonants simply add the vowel a to the end of the word: e.g.: gwel: to want gwla mi: I want e.g.: pis: to see apsa mi: I see The verb v, the only modern Gaulish verbal stem to end in o, drops its word final and replaces it with the present tense marker a: e.g.: v: to do, to make va mi: I do, I make There are some verbal stems that end in a. These retain their final a and remain unchanged throughout the present tense: e.g.: cara: to love cara mi: I love Past tense The past tense in modern Gaulish is constructed by way of the preverbal particle r placed before the verbal stem. This particle is identified in readdas, (has)placed (Delamarre 2003, p. 255), and is asserted as having been used to construct the past tense by Eska (2008, p. 869). Similar use of this particle is known in the Insular Celtic languages (e.g. Old Irish, Cornish). There is no change to the end of the verbal stem: stems in i or retain these. However, the initial consonant of the verbal stem changes in

42 accordance to the systematic modification of word initial consonants known as Initial Consonant Mutation (ICM). This will be discussed in great detail further below. e.g.: gwel: to want r chwel mi: I wanted e.g.: pis mi: to see r hpis mi: I saw This construction is postulated on the basis of the attestation of preverbal particle re in Gaulish past tenses (e.g. readdas, Lambert 2003, p.66, Delamarre p. 255), which has been identified as giving a perfective value or diverse modalities to verbs (Delamarre 2003, p. 261). Furthermore, the use of re (> *ro) in the formation of past tenses is well attested in insular Celtic, not only in Old Irish (De Bernardo-Stempel n. d., Williams 1908), but also in Late Cornish (Norris 1859, p. 49, Williams 1908) and Old to Middle Welsh (Williams 1908). While re is most often, though not always, associated with the perfective tense in the other languages, it has been identified by Eska (2008, see above) as being used in Gaulish to construct the past tense. Therefore, modern Gaulish employs it as such. An important corrollary of this past tense formation is that it illustrates quite clearly that ancient Gaulish used the process of re-analysis of a preposition as an aspect marker. The pre-verbal, past tense-marking particle re is derived from an inherited Indo European preposition *pre-, meaning before (Indo European *pre- > Proto Celtic *re- > Celtic re-), well attested in other languages (e.g. Latin). As such the phrase r hpis mi, I saw, is semantically derived from a prepositional phrase before I see > I saw. This fact is very important in the discussion about the formation of the progressive form, which will be discussed in detail in the appropriate section below. Future tense The future is constructed by way of the suffix s, from attested *-si(o), which is identified as a marker for the future (Delamarre 2003, p. 251; Lambert 2003, p. 65). The emphasis is put on this marker: apis mi. e.g.: gwels mi: I will want pis mi: I will see Verbal stems ending on vowels add the suffix s to the end vowel: e.g.: rethi: to run rethis mi: I will run v: to do / make vs mi: I will do / make

43 Conditional tense A conditional tense is formed by the combination of the two affixes discussed above: the preverbal particle r and the suffix -s: e.g.: r chwels mi: I would want r hapis mi: I would see r rhethis mi: I would run r hvs mi: I would do / make Present perfect The present perfect formation of ancient Gaulish is reliably attested in a number of instances. The first of these involve the well known and discussed form carnitu, found in the funerary inscriptions of Todi: carnitu artuas Coisis Druticnos = has-built tomb Coisis Druticnos > C. D. has built the tomb > carnitu- = has built (Delamarre 2003, p. 106; Lambert 2003, p. 75-77) The form carnitu is universally accepted as being the present perfect form of a verb carni-, to pile up stones (Delamarre 2003, p. 106). A further attestation of a present perfect form is found at Chateaubleau: iexstumisendi = iexstu-mi-sendi = have-cursed-I-this > I have cursed this (Mees 2010, p. 105) From the above examples it is possible to deduce that the present perfect verbal form in ancient Gaulish was constructed with the suffix -tu, followed by the subject of the phrase: carnitu C. D. = has built C. D. iexstu mi = have cursed I A similar construction is found in the Irish verbal adjective on te, -ta, the, -tha, which is also attested in ancient Gaulish, e.g. nantosuelta. This is translated as vallee ensoleil, sunned valley (Hansen 2012, Bhghros 2012 pers. com.), with the second component suelta literally meaning ensoleill, i.e. .sunned, or, in English, sunny. See also lubitias and caratos (Delamarre 2003, p. 209, 107). Therefore, it is posited here that the present perfect form in modern Gaulish will be constructed with the verbal suffix thu, followed by the subject of the phrase. e.g.: carni: to build

44 carnthu mi: I have built gar: to call: garthu mi: I have called gwelthu mi: I have wanted gn: to know gnthu mi: I have known However, for verbal stems ending on thi, -dhi or s, this suffix is reduced to -. Stems on thi and dhi drop their final vowels i and instead affix . Stems on s affix after the s. In both cases, this perfective marker receives the emphasis. The reason for the use of emphasised rather than regular thu for stems on thi and dhi is that the addition of thu would result in a word that is altogether too fricative for ease of pronunciation: reththu would be awkward to pronounce. Therefore a position is adopted here that holds that the two successive fricatives have assimilated to form just one: *reththu > *reth; *gwedhthu > *gwedh. In the case of verbal stem on s, a perfect form with ending s would be the normal phonetic end-result of the suffixation of s with (pre-sound change) tu, giving *-stu, which, through regular Gaulish sound changes, would assimilate to s (-stu > -tsu > -ssu > -su). Therefore, stems on s take the perfective marker . The use of a perfective marker u is attested in the forms ieuru, iouru, has offered/ has dedicated (Delamarre 2003, p. 188-89) and tioinuoru, has produced (Delamarre 2003, p. 297). e.g.:rethi: to run reth mi: I have run gwdhi: to pray gwedh mi: I have prayed pis: to see apis mi: I have seen brs: to break bris mi: I have broken Past perfect The past perfect is constructed with the preverbal particle r before the present perfect form: e.g.: r garnthu: I had built r chwelthu: I had wanted

45 r rheth mi: I had run re hapis mi: I had seen As with the past tense, the first consonant of the verbal stem undergoes mutation. Future perfect For the future perfect the future marking suffix s is inserted between the verbal stem and the perfective marker thu, which is the only one used, since the insertion of the future marker si between the verbal stem and the perfect marker removes the phonetic issues that give rise to the use of only in the present and past perfect: e.g.: carnisthu mi: I will have built gwelsthu mi: I will have wanted rethisthu mi: I will have run apisthu mi: I will have seen Conditional perfect For the conditional perfect, the preverbal preterising particle r is placed before the future perfect form: e.g.: r garnisthu mi: I would have built r chwelsthu mi: I would have wanted r rhethisthu mi: I would have run r hapisthu mi: I would have seen Imperative The imperative form is well attested in ancient Gaulish (Delamarre 2003, p. 173, 209), and consists of the bare verbal stem, with imperative intonation: e.g.: carni!: build! gar!: call! rthi!: run! pis!: see! Verbal noun In addition to not having any conjugations, Gaulish does not have an infinitive either. Instead it uses the bare stem of the verb, referred to as a verbal noun. The verbal noun in the Celtic languages is a word that can at the same time function as a verb and as a noun, and is an important component of the verbal paradigm. As a verb, it carries an infinitival meaning: gar means to call. However, as a noun gar also has the meaning of the calling.

46 e.g.: gar: to call / the calling pis: to see / the seeing rthi: to run / the running The verbal noun follows modal verbs directly: e.g.: gwla mi pis: I want to see / I want seeing gwla mi rthi: I want to run / I want running Direct object of a verbal noun An important aspect of the fact that the verbal noun is a noun is the treatment of the direct object of a verbal noun: e.g.: gwla mi pis p: I want to see a horse (p < epos horse, Delamarre 2003, p. 16364) This is actually a genitival construction, of which more further below. Because genitive in modern Gaulish is expressed by apposition, the phrase pis p really means the seeing of a horse. Because the genitive in Gaulish is not marked in any way on a noun, this looks and translates exactly the same as to see a horse. However, when personal pronouns are the direct object of a verbal noun, this genitive is expressed explicitly. For this the genitive particle i- is used, derived from the attested forms imon and imi, both meaning mine (masculine and feminine respectively) (Delamarre 2003, p. 18990). This particle is followed by fused forms of the personal pronouns: e.g.: gwla mi pis ith: I want the seeing of you = I want to see you Without the use of the particle i-, the above phrase would be: gwla mi pis ti = I want the seeing you This is not possible. The particle i- is not otherwise used in many other contexts: it only occurs as the genitival marker for the personal pronouns as used as direct object of a verbal noun, and to indicate ownership of objects in conjunction with personal pronouns only. As such, it behaves like a preposition in the way it fuses with the pronouns: im: of-me ith: of-you ich: of-him ich: of her in: of-us is: of-you (pl.) ichs: of them

47

The third person singular and plural forms are formed with the semantically empty particle ch- for phonetic reasons, as in the fusion of prepositions with pronouns. See above.

Summary of verbal paradigm The following table provides an overview of the verbal paradigm in modern Gaulish. Examples are given for verbal stems on a consonant, on thi and on s. All verbal forms are marked for person and number by personal pronouns only; only the first person singular is given. verbal noun gar rthi pis present past future conditional imperative

gara mi rtha mi apsa mi present perfect garthu mi

r ghar mi r rhthi mi r hpis mi past perfect

gars mi rethis mi apis mi future perfect garsthu mi

r ghars mi r rhethis mi r hapis mi conditional perfect r gharsthu mi

gar! rthi! pis!

r gharthu mi

48

reth mi apis mi The verb to be

r rheth mi r hapis mi

rethisthu mi apisthu mi

r rhethisthu mi r hapisthu mi

The verb to beis the only irregular verb in the modern Gaulish language. Its formation is different from that of all other verbs. The verbal noun may be derived from the attested Gaulish word bissiet (Delamarre 2003, p. 76). This can be analysed as containing the future marker si, suffixed to what appears to be the verbal root bis-: bis+si (future marker)+ -et (3rd p.s. ending) (analysis by Lambert, discussed in Delamarre 2003, p. 76). The attested form has a double sswhich does not appear to indicate a Tau Gallicum. A similar double ss- is found in the attested form pissiiumi (Demarre 2003, p. 251). Since in the latter case the word pissiiumi is analysed as consisting of a future marker si- suffixed to a verbal stem pis-, it is reasonable to extend this to bissiet and by analogy to deduce that bis- represents the verbal stem or verbal noun. The present tense is thought to be attested in the form esi (Delamarre 2003, p. 167). While esi is thought to be either a 2nd or a 3rd person singular, a position is proposed here where esi is the verbal stem that is used across the present tense, augmented with personal pronouns to mark for person and number, as is the case with all other verbs in modern Gaulish. The past may be derived from the form buetid, (Delamarre 2003, p. 93-94). Comparisons with Welsh (Modern Welsh bues i = I was; Middle Welsh bu 3rd p. s., was) suggests that it is reasonable to accord a past value to this form. As such, the verbal stem to be used across the past tense will be b. The future form may be found in the attested forms bissiet, bisiete (Delamarre 2003, p. 76), as well as, possibly, in the form biiete (Delamarre 2003, p. 74-75). Again, comparison with Welsh (Modern Welsh bydda i: I will be; Middle Welsh byd 3rd p. s., will be) permits to posit that this form holds a future indicating connotation. As such, the verbal stem to be used across the future tense will be b. A conditional form is not attested, but may be constructed by analogy with the paradigm of all other Gaulish verbs. As such, the preceding of the future form b by the preverbal particle r will denote the conditional form: r v, with mutation of the first letter of the verbal form, as discussed previously. An imperative is attested in Chamalieres as sete, from which a modern form sth, be! can be derived (Mees 2010).

49 Finally, a perfect present may be attested in the form biietutu (Delamarre 2003, p. 75). While this is unsupported academically, a position is proposed here where the component -etu- represents the perfect present form of to be: thu mi = I have been. The above form can further be analysed as a fusion of future stem bi + etu, which would, in analogy with the perfective tense formation of the other Gaulish verbs, therefore amount to the construction of the perfect future: bietu-, becoming bithu mi, I will have been. Again, this is unsupported academically, but is plausible and can reasonably be argued in favour of. An analogy with the above would then permit the construction of a perfect past with the past verbal stem bu-, fused with the perfective etu-. This would give buthu mi, I had been. Once again, there is no academic support for this position, but it is plausible and can reasonably be argued in favour of. Lastly, a perfect conditional would then, by analogy with the construction of other verbs in modern Gaulish, be able to be constructed by means of the preverbal particle r, preceding the perfect future form. This would give r vithu mi, I would have been.

Summary of to be paradigm The paradigm of the verb to be can be summarised in the following table. All verbal stems are marked for person and number by personal pronouns, as with all other verbs. Only the first person is given here. tense verbal noun present past future conditional perfect present perfect past perfect future perfect conditional imperative The verb to have verbal stem bis esi mi b mi b mi r v mi thu mi buthu mi bithu mi r vithu mi sth attested form bissiet esi buetid biiete, bissiete particle re attested bi-etu-tu bietu-tu sete

50

Modern Gaulish, like all other modern Celtic languages, does not have a separate verb to have. Instead, as is the case in its related languages, it uses a construction with the verb to be. This construction is attested in the inscription of Banassac (Delamarre 2003, p. 167; Lambert 2003, p. 142): tiei ulano celicnu, translated as at-you-is the satisfaction of the banquet, where at-you-is translates as you have. The formation therefore of the expression to have in modern Gaulish consists of the appropriate personal pronoun, suffixed by the appropriate verbal stem of the verb to be (bis). The pronoun and the stem will be separated by a hyphen, and internal spirantisation will apply to the first consonant of the stem of bis. e.g.: mi-esi: I have ti-esi: you have -esi: he has -esi: she has ni-esi: we have s-esi: you have (pl.) s-esi: they have The various tenses are constructed by analogy: mi-v: I had ti-v: you will have -rh-v: he would have -thu: she has had ni-vuthu: we had had su-vithu: we will have had si-rh-vithu: they would have had Alone among the verbs of modern Gaulish, this verbal construction has neither a verbal noun, nor an imperative form. Intentional and imperative phrases can be constructed as a subordinate clause, using the relative pronoun o, seemingly attested in Chateaubleau (Mees 2010) and in Marcellus of Bordeaux (Delamarre 2003, p. 268): gwla mi o mi-esi: I want that I have = I want to have gwla mi o ti-esi = I want that you have =I want you to have Passive or impersonal forms Ancient Gaulish had passive (or impersonal) verbal forms on r, analoguous with Brittonic and Goidelic. Attested are uelor (one wants, is wanted; Delamarre 2003, p. 312), nitixsintor (p. 236) and diligentir (p. 144-45). It is widely agreed upon that the suffix or marks the verbal form for the passive construction. Therefore, this suffix is adopted for that purpose. It occurs in concurrence with all other regular verbal particles and suffixes. However, while the perfective suffix thu is added after the passive marker

51 -or, the future marker s- is added after the verbal stem itself, and therefore before the passive marker. This is attested in the form nitixsintor sies (ni-tix-si-ntor + pers. pron. 3rd p.pl.fem., Delamarre 2003, p. 236). The order of suffixes and particles is thus as follows: r [verbal stem]+s+or+thu + personal pronoun An overview, using the verb pis, so see, is provided below: apsor mi: I am seen (one sees me) r hapsor mi: I was seen apisor mi: I will be seen r hapisor mi: I would be seen apisorthu mi: I have been seen r hapisorthu mi: I had been seen apisorthu mi: I will have been seen r hapisorthu mi: I would have been seen apsor!: be seen! An intentional passive phrase can be expressed by means of a subordinate clause: gwla mi och apsor mi: I want that I am seen/ that one sees me > I want to be seen In the above phrase, the particle o is the subordinating pronoun. It receives the semantically empty particle -ch- before a vowel, as is the case with prepositions (see that section). Parallel with the attested passive system on or, it is also possible to achieve a passive construction by using the appropriate form of bis (to be) combined with the perfect present participle, exactly as in other Western European languages: esi mi apis: I am seen b mi apsi: I was seen b mi apis: I will be seen r v mi apis: I would be seen thu mi apis: I have been seen buthu mi apis: I had been seen bithu mi apis: I will have been seen r vithu mi apis: I would have been seen sth apis!: be seen! Both forms can be used at will. There are no semantic differences between them. Ablative absolute

52 In the important long running text of Late Gaulish known as the tablet of Chateaubleau the following verbal form is attested: anmanbe gniiou: the names which know sini siaxsiou: these which seek sue cluiou: you (pl.) who hear (Mees 2010) It appears that the verbal form with suffix iou, used in the above context, refers back to a predicate subject in a way which has been interpreted by Mees (2010) as resembling a relativising or subordinate construction. The form appears to fulfill a function that is very similar to that fulfilled by a Latin ablative absolute (Bellovesos 2012, pers. com.). Furthermore, it bears a close resemblance to the suffixed relative pronoun io well attested in Gaulish: the suffixed particle iou appears to act in the same way as the attested particle io. A position is adopted by modern Gaulish where, due to regular evolution of the diphthong ou [o:w] to o [o:], the ablative absolute suffix iou and the relativising pronoun io have amalgamated in the modern suffixed particle . As such, - functions as a verbal suffix to refer back to a predicate subject immediately preceding it, and can be used to construct subordinate clauses: apsa mi p reth: I see a horse that/which runs This will be further explored in the section on subordinate clauses, further below. Progressive form A progressive verbal form expresses an action that is ongoing, happening right at the moment of discussion. Within the attested Gaulish material, there are two indications of how such a form might be constructed. While the interpretation of these attestations is not uniformly agreed upon, there is nevertheless a degree of consensus regarding their plausibility and historical acceptability (Bhrghros 2012, pers. com.) and their practicality of use in terms of clarity of expression (Bellovesos 2012, pers. com.) for the modern language. One potential attestation of a progressive form in the attested ancient Gaulish material is the word atenoux, found on the Coligny calendar (see Lambert 2003, p. 111-18). While there is contention over the interpretation of this word, as there is over everything remotely connected with the Coligny calendar, Lejeune (1995, In Delamarre 2003, p. 58) proposes the following etymology: atenoux = ate-en-oux, again-in-rising

53 This was supported by Schmidt (1999, in Delamarre 2003, p. 58) and is considered plausible by Delamarre (2003, p. 58) and Bhrghros (2012, pers. com.). A second potential attestation of a progressive form may be found in the etymology of the Gaulish names Enistalus (Delamarre 2003, p. 163) and Enissa (Gwinn 2012, pers. com.). These words have been linked to the widely accepted etymology for the insular Celtic word inis, meaning island (Finsen 2012, pers. com.; Wiktionary, etymology of inis): inis = *enisti (standing in [the water]/ in-standing), from eni- (in)+ *steh2 The above can be interpreted as eni-sti = in-standing (Strachan, in MacBain 1982, innis). This would provide a construction identical to at-en-oux. The preposition eni-, in, is attested in ancient Gaulish (Delamarre 2003, p. 163). Based on these admittedly scant and contentious indications, a position is proposed here where modern Gaulish uses the preposition en to construct a progressive verbal form in conjunction with the verb bis, to be. Direct parallels are found in all Celtic languages, and in western Germanic. Because en is a preposition, it causes mutation on the first letter of the verbal noun following it. This will be discussed in greater detail in the section dealing with initial consonant mutation. e.g.: esi mi en garni: I am building esi mi en ghar: I am calling esi mi en rhthi: I am running While the periphrastic construction thus achieved is not attested in the ancient Gaulish material, its inclusion in modern Gaulish may be deemed acceptable for two reasons. First, such forms are by no means unusual or exceptional in modern Western European languages (see Isaac 2007) and may as such be considered a development that may reasonably be expected to have occurred. Second, and more importantly, the use of a preposition to form a verbal form requires a process of re-analysis of a preposition as a verbal aspect marker (Borsley, Tallerman & Willis, 2007). This process is attested precisely as such in ancient Gaulish, with the reanalysis of the ancient IE preposition *pre-, before, via Proto-Celtic *re- to Gaulish re-, used as a pre-verbal particle to mark the past or preterite aspect of a verbal form. This shows that the process of re-analysis of a preposition as a verbal aspect marker was internal to the Gaulish language, as it manifestly is to its related languages, all of which have used the same process. As such it is perfectly reasonable to posit this process happening to construct a progressive verbal form in modern Gaulish. 6. Adjectives Order

54 Adjectives are clearly and abundantly attested in ancient Gaulish. While the oldest material shows adjectives preceding their head nouns, the later material indicates a shift towards head nouns preceding adjectives. Attestations of both include: Ueru-lamium Lama-uerus: wide-hand > generous (Delamarre 2003, p. 317) Attestations of adjectives following head nouns are: agedo-uiros: face-man: face like a man Amarco-litanos: view/sight-wide: view/sight that is wide/broad/vast (Delamarre 2003, p. 317) Perhaps the most incontestable attestation of an adjective following its head noun is found on the spindle whorl of Autun (Lambert 2003, p. 125; Delamarre 2003, p. 331), which reads: nata uimpi curmi da = girl beautiful beer give Continuing the trend observable in late ancient Gaulish, modern Gaulish proposes a situation where adjectives follow their head nouns: e.g.: p: horse carthach: friendly > p carthach: a friendly horse ti: house (< tegia, Dlamarre 2003, p. 294) gwin: white > ti gwin: white house Adjectives qualifying plural nouns are not marked for plural. However, adjectives that follow feminine nouns, undergo a mutation to their first letter: e.g.: aman: time sr: long > aman shr: (a) long time blth: flower coch: red > blath goch: (a) red flower These mutations will be discussed in more detail later. There are two broad categories of adjectives: natural ones and derived ones. These are discussed below. Natural adjectives

55

Natural adjectives are words that are inherently adjectival. A representative selection is given below: e.g.: mr: big mi: small sr: long bir: short di: good druch: bad math: fine, favourable anwath: unfavourable In addition to these, there are also adjectives derived from other sources. Derived adjectives 1) adjectives constructed with suffixes In the attested Gaulish material it is possible to identify five different suffixes that were used to produce adjectives. These are used in modern Gaulish and are listed below, with examples: a) the suffix ach, derived from attested aco (Lambert 2003, p. 34, p. 39). Attested examples are: Anualonacu = Anualos sanctuary (Lambert 2003, p. 39) Parisiaci = [people] of the Parisi (a tribe) (id.) Caratacos = beloved-like (< *carato- loved + -aco) (Delmarre 2003, p. 107) This is the most productive and most readily used adjective marker in modern Gaulish. It is used to derive adjectives from any type of noun: e.g.: nerth: strength nerthach: strong panth: pain panthach: painful caran: friend caranach: friendly achaun: stone achaunach: stony duvr: water duvrach: watery b) the suffix ich, derived from attested ico (Lambert 2003, p. 34). It is attested as follows:

56

uertamicos = excellent, superior quality (Delamarre 2003, p. 317) The above is a compound *uer-tam-icos, derived from uer = over, tam = quality, ico- = adjectival suffix. Modern terms using this suffix are: gwerthamich: superior quality, excellent donich: human, adj. (attested, Delamarre 2003, p. This suffix, which is similar to the ach discussed above, is used as a general adjectivising suffix only for words ending in a, for which it is impractical to use ach, and for words ending on a diphthong in -i: gr: sand (Delamarre 2003, p. 183-84) grich: sandy ti: house (Delamarre 2003, p. 294) tich: domestic, pertaining to the house or house related matters c) the suffix in The suffix in is attested in one instance only, in the word bledinos (Delamarre 2003, p. 78-79). It appears to lend to the word bled-, wolf, the meaning wolfish. As such, it may be proposed that this suffix be used to describe the physical and metaphysical characteristics and attributes of animals only, in a direct parallel with the English ine (derived from the Latin suffix inus/a, only used with Latin derived words): e.g.: bldh: wolf bldhin: wolfish, wolf-like (lupine) cun: dog (Delamarre 2003, p. 132) cunin: dog-ish, dog-like (canine) p: horse pin: horsey, horse-ish, horse-like (equine) ernu: eagle (Delamarre 2003, p. 166) ernin: eagle-ish, eagle-like (aquiline) d) the suffix -dhu The suffix dhu is derived from the attested suffix (i)do, attested in Chateaubleau, Larzac and La Graufesenque (Mees 2010). In the context, it appears to pertain to cows: coro bouido = cow contract In this instance bouido appears to translate as bovine, or as of the cows. It is proposed here that this suffix, of which is little known, be used in modern Gaulish as a

57 general adjectivising suffix dhu that is to be used with words ending on ch, for which ach would be phonetically impractical: e.g.: bruch: heather (Delamarre 2003, p. 92) bruchdhu: heathery, heather-like coch: leg (Delamarre 2003, p. 128) cochdhu: leggy, pertaining to legs 2) adjectives constructed with prefixes Adjectives can be made from nouns and verbal nouns by means of the prefixes su-, good, and du-, bad. e.g.: car-: verbal root pertaining to love (Delamarre 2003, p. 107) suchar: popular, i.e. well-loved duchar: unpopular, disliked If the prefixes su- and du- precede a vowel and do not receive emphasis, they become swand dw-: e.g.: iedh: face, appearance (Delamarre 2003, p. 34) swiedh: beautiful, good looking, handsome = good-face/appearance dwiedh: ugly = bad-face/appearance 3) verbal adjectives Verbal adjectives consist of the verbal form of the present perfect, applied as an adjective: e.g.: rani: to divide (Delamarre 2003, p. 164-65) bri: land (Demamarre 2003, p. 91) bri ranthu: a divided land cara: to love don: a person (Delamarre 2003, p. 176) don carthu: a beloved person (cf. caratos, Delamarre 2003, p. 107) 4) verbal nouns as adjectives Verbal nouns can be used as adjectives. They follow their head noun and undergo mutation of their first consonant if the head noun is feminine. e.g.: cun: dog bi: to fight/ fighting (Delamarre 2003, p. 63-64) cun bi: fighting dog gs: spear (Delamarre 2003, p,. 174)

58 aghri: to hunt/ hunting (Delamarre 2003, p. 35) gs aghri: hunting spear ben: woman cra: to love ben gra: loving woman If verbal nouns end in a vowel this is dropped when used in conjunction with prefixes: e.g.: gwidhi: to understand > verbal root gwidh> suchwidh: intelligent (good understanding) > duchwidh: stupid (bad understanding) 7. Adverbs Natural adverbs Natural adverbs are words that are by definition adverbial (although some also can be used in adjectival function), i.e. they complement adjectives, nouns, verbs or other adverbs. A considerable number are attested in the ancient Gaulish material. Examples are: ach: and ath: again lu: a lot, many eth: more ceth: also duch: therefore Adjectival adverbs The way to construct adjectival adverbs is clearly attested in ancient Gaulish, in the running text of Lezoux: inte nouiio = newly (Delamarre 2003, p. 191) The construction of an adverb from an adjective by means of the particle inte has perfect parallels in all surviving Celtic languages. In modern Gaulish, inte becomes in and causes mutation of the first consonant of the following word: e.g.: nerthach: strong in nherthach: strongly mr: big, great in wr: greatly tech: beautiful

59 in dech: beautifully 8. Gender Three genders are attested in ancient Gaulish: masculine, feminine and neutre (Delamarre 2003, 342-46). A position is adopted by modern Gaulish where the neutre has amalgamated with the masculine, leaving only two genders. This is analoguous with the situation in the surviving Celtic languages, and with neighbouring languages (e.g. Romance). As such, modern Gaulish has masculine and feminine gender. Words with a final vowel e/, -o/, -u/ and au are usually (but not always) masculine; words with a final vowel a/ and -i/ are usually (but not always) feminine. The following table provides examples: masculine bech: beak coch: leg cun: dog achaun: stone cernu: horn feminine aval: apple ci: hedge lam: hand br: hill cich: flesh, meat

However, in cases where gender is semantically explicit the above rules dont apply, and gender is as implied semantically: e.g.: map: son gwir: man, rich: king geneth: girl swor: sister ben: woman In addition to these two genders, modern Gaulish has retained a vestige of a neuter gender, which is only manifested in the neuter pronoun (from attested id > , Delamarre 2003, p. 93) and which is used solely to express indefinite concepts such as the time, the weather, or an unknown and unspecified subject. e.g. esi di: it (i.e. the weather) is good Masculine words that have a feminine counterpart can construct this by the addition of word final vowel a. The occurrence of word final vowel a to indicate feminine gender in words is well attested in ancient Gaulish: e.g.: bena = woman geneta = girl

60 While modern Gaulish postulates a near-total loss of word endings, it proposes that the word ending a is retained in some cases to construct the feminine counter part of a masculine word, especially for animated subjects (e.g. people, animals): cun: dog cuna: bitch p: horse pa: mare lern: fox lerna: vixen dridh: scholar, teacher dridha: female scholar, teacher caran: friend carana: female friend In addition, words that can be either gender can be qualified by the words gwir (man) or ben (woman) as the second component of a compound, where the second component receives the emphasis. The second component effectively acts as an attributive adjective, and constructs a meaning of male/female. In these cases, the gender is as indicated by the suffix. The first consonants of these suffixes undergo the regular word internal sound changes: E.g.: p: horse (masculine, general term) p+ gwir > epchwir: stallion, male horse (m) p+ben > epven > ven: mare, female horse (f), with assimilation of pv- > -vcun: dog (m) cunchwir: male dog (m) cunven: female dog, bitch (f) caran friend (m) caranchwir: male friend, boyfriend (m) caranven: female friend, girlfriend (f) Support for these constructions is found in the attested word banolucci (Delamarre 2003, p. 72), which translates as woman-wolf, i.e. she-wolf or female wolf. While an alternative translation could be woman for wolves, the former interpretation is adopted here for practicality, in analogy with such usage in the Goidelic languages. 9. Article Insular Celtic languages have a variety of articles (W y, yr/r, I/S/M an/na, C an, Br an/ar/al un, ur, ul) where they are thought to have been derived from the demonstrative *sindos (Delamarre 2003, p. 274). *Sindos is attested in ancient Gaulish (sinde, Delamarre 2003, p. 274). While there is no real attestation of the evolution of any word into an article in the available ancient Gaulish material, there is nevertheless evidence of

61 the gradual transformation and mutation of the word sinde in a grammatical position which appears to have been favourable for the development of sinde into an article. In Larzac, there is evidence that sinde evolved into indas, as in indas mnas (Delamarre 2003, p. 274), the translation of which is generally accepted as being these women (Delamarre 2003, p. 274, David Stifter, 2009, pers. comm.). indas would be a form derived from sinde with loss of initial s- in unstressed proclitic, pretonic position, a process which is documented in Old Irish (David Stifter 2009, pers. comm.). A semantic shift from these/this to the is not unusual in Indo-European languages, is well documented, and is for instance the process that has given the English language the demonstrative this, which evolved through doubling of the original demonstrative after it had semantically weakened to beome the definite article the (Bernard Mees 2009, pers. comm.). The process of doubling of demonstratives appears to be attested in ancient Gaulish, namely in Larzac, which has insinde se bnanom, as well as elsewhere (e.g. sosin, sosio, Delamarre 2003, p. 279). In light of the above, modern Gaulish postulates a hypothetical evolution of the word inda(s) into a definite article. As such, the following evolution is proposed: inda(s) > *inda > *ind > in The above is a straightforward phonetic evolution, through loss of final syllable, and through the standard modern Gaulish soundchange of nd > -n. Therefore, modern Gaulish uses the word in as the article; it is definite and unmutatable, and not affected by gender or number. In analogy with most modern Celtic languages (except Breton), modern Gaulish does not use an indefinite article. e.g.: pen: a head in pen: the head cun: a dog in cun: the dog When feminine nouns, singular or plural, are preceded by the article in they undergo a mutation of their initial consonant: e.g.: geneth: a girl in gheneth: the girl br: a mountain in vr: the mountain 10. Plural formation Standard plural formation

62

In ancient Gaulish, most word stems ended on vowels, and word function and meaning, including the plural, was conveyed through a variety of case endings. These case endings varied according to the final vowel of stems and according to the case the word was in. There is clear indication in the attested material that word endings were eroding (see discussion under apocope section), which would inevitably result in the loss of meaning of case and therefore, eventually, in the loss of case. As a direct result of the loss of word endings, a majority of words would have come to end in consonants instead of vowels. Under the old case system, word stems ending in consonants in the nominative case took the ending es: e.g.: eurises = dedicators, donators (Nautes Parisiaques Pillar, Delamarre 2003, p. 169)) sies = they 3rd p. pl. fem. (Larzac, 3x, Delamarre 2003, p. 338-39) tidres = three pl. fem. (La Graufesenque, Delamarre 2003, p. 302) (for discussion see Delamarre 2003, p. 345; Lambert 2003, p. 63). It is posited here that with the loss of case endings, the nomimative consonantal plural ending es was retained and spread out across the word paradigm to form a single allpurpose pluralising ending. This plural ending -es would then be further phonetically eroded to e, which, it is posited here, was then lengthened to -. As such, modern Gaulish adopts the single, all-purpose pluralising suffix -: p: horse p: horses cun: dog cun: dogs This suffix is also applied to nouns ending on vowels: b: cow b: cows tru: bull tar: bulls cuna: bitch cun: bitches The only exception to the standard plural formation is the plural of ben, woman (< ancient Gaulish bena), which is attested under the forms bnanom, mnanom (2x), mnas (2x) at Larzac (Delamarre 2003, p. 338). With regular loss of case endings nom and s that gives a form mna-, which with regular lengthening of emphasised vowel in modern Gaulish becomes mn. This form will be the regular standard plural of the word ben, woman.

63 ben: woman mn: women Plural after numbers There are a number of instances attested in the ancient Gaulish material where a noun is preceded by a number: trimarcisia, trinoxtion, petruroton, petrumantalon, pempedula, decamnoctiacis. These can be analysed as follows: trimarcisia = tri+marcos+ia three+horse+substantivising ending, instr. singular trinoxtion = tri+noxt+ion three+night+subst. end., nom. neut. singular petruroton = petru+roto+n four+wheel+subst. end., nom. neut. singular petrumantalon = petru+mantalo+n four+road+subst. end., nom. neut. singular pempedula = pempe+dula five+leaf, nom. singular decamnoctiaca = decam+noct++io+iac+a ten+night+rel. pron.+adj.mark.+subst. end., dat-abl. neut. plural LATIN (from Demamarre 2003, p. 302, 302-3, 250, 250/216, 248, 137) In five of the six examples listed above, the compound consists of a number, followed by a noun in the singular, followed by a substantivising ending. Translations are, respectively: trimarcisia: three horse (three horse riders) trinoxtion: three night (three nights of celebration petruroton: four wheel (wagon with four wheels) petrumantalon: four road (crossroads of four roads) pempedula: five leaf (flower with five leaves) decamnoctiaca: ten night-like-s (ten nights of celebration) For five out of these six cases, the cardinal number is followed by a noun in the singular. The ending of the sixth is thought to be adapted to the Latin case system (Delamarre 2003, p. 137). While it is possible that the singular was employed because the compound word in question was considered a word in the singular, the fact remains nevertheless that these numbers are followed by nouns in the singular. Also, there are no attestations of numbers being followed by nouns in the plural in the ancient Gaulish material. Therefore, it is posited here that modern Gaulish does not use the plural of nouns after cardinal numbers. This is furthermore in parallel with the situation in the Brittonic languages, where numbers are followed by nouns in the singular.

64 e.g.: tri march: three (riding) horses pethr roth: four wheels pethr manthal: four roads pimp dul: five leaves dech nith: ten nights Dual plural It is unclear at this stage whether ancient Gaulish had a dual number, due to the incomplete nature of the attestation of the language (Lambert 2003, p. 51). Nevertheless, it is known that Old Irish has a dual number, pertaining to things that appear in the world in natural pairs (Lambert 2003, p. 51). Furthermore, the very close relationship between ancient Gaulish and Old Irish is becoming increasingly evident (Mees & Stifter 2012, pers. com.). Therefore, an analogy is made with the situation in Old Irish, and a dualplural is proposed for modern Gaulish here. As such, things that occur naturally in the world in pairs form their plural with the prefix d-, meaning two. e.g.: lam: hand dlam: hands coch: leg dchoch: legs p: eye dp: eyes However, these plural formations only pertain to situations where these subjects naturally occur in pairs. Therefore, dchoch applies to the legs of a human, because they come in a pair of two, but not to the legs of a horse, because there are four of them: dchoch: legs (of a human or other bipedal animal, e.g. a bird) coch: legs of a horse, or other animal with more than two legs dlam: hands of a human (a pair of two) lam: hands of a clock (three hands: hours, minutes and seconds) Collective plural In the ancient Gaulish material a word slougo- is attested. This word, meaning group, troop, gathering, crowd, assembly (Delamarre 2003, p. 276) is used in the Brittonic languages to form a collective noun. Although it is not attested as such in ancient Gaulish, it is proposed here that this same word fulfills the same function in modern Gaulish. Regular modern Gaulish sound changes render slougo- as sli. Furthermore, in suffix position the initial s- of slougo- is absorbed into the l-, giving a hypothetical ancient Gaulish ll-, reduced to l- in modern Gaulish. slougo > -li This can now be applied as a suffix to such entities as display collectivity:

65

e.g.: sir: star (< stir-, Delamarre 2003, p. 282) sirli: constellation (group of stars) br: mountain (< briga, Delamarre 2003, p. 87) brli: mountain range 11. Indication of possession Genetive of apposition In ancient Gaulish, genetive was indicated by word endings. Exceptionally clear and unambiguous examples of this are found in the long running text of Larzac (Delamarre 2003, p. 338-39; p. 344): Adiega matir Aiias = Adiega mother of Aia Aia duxtir Adiegias = Aia daughter of Adiega In the above example, the genetive relationship is expressed by the word ending ias. If we apply the postulated loss of case endings, which is a well attested, well agreed upon and important factor in the formation of the modern Gaulish language, to these two phrases, we are left with the following: Adiega matir Aia Aia duxtir Adiega These translate as follows: Adiega mother Aia Aia daughter Adiega It is posited here that the main genetive formation of modern Gaulish is derived from these constructions: the left-overs after case endings disappeared. With the loss of the case endings, the only thing that is left to define the genetival relationship formerly specified by word endings is the word order. Therefore A mother B means A mother of B. This is known as genetive of apposition: two nouns are put next to each other, and the first belongs to the second. This principle is applied throughout the modern Gaulish language to indicate possession of one thing by another: p Belo: horse of Belo > Belos horse ti Gwina: house of Gwina > Gwinas house cun Garghan: dog of Garghan > Garghans dog In the example above, a noun is followed by a name, which therefore owns that noun. The same principle can be applied to two nouns:

66 coch p: leg of a horse pen gwir: head of a man brun ben: breast of a woman The principle of genetive of apposition means that when a noun is followed by another noun, the first noun is owned by or belongs to the second noun. This principle can be extended to include the article in. Instead of using a non-specific entity (a horse), we can use a specified entity: the horse. coch: leg in p: the horse coch in p: leg of the horse The above phrase means the leg of the horse. To express the notion a leg of the horse, the indefinite can be described as being one leg: on coch in p: a/one leg of the horse If the possessor is indefinite, this becomes: on coch p: a/one leg of a horse or simply: coch p: a leg of a horse For strings of more than one item that is possessed, only the article of the possessor is used, all other ones are omitted: e.g.: lu: colour in lu: the colour bar: top in var: the top coch: leg in coch: the leg p: horse in p: the horse To translate the phrase the colour of the top of the leg of the horse all the articles except the last one are dropped: lu bar coch in p: the colour of the top of the leg of the horse the horses leg top colour Possessive pronouns

67

Possession can be expressed directly by use of possessive pronouns. Of these, a number can be identified in the attested material: 1st pers. sing.: in mon derco = in my eye (Marcellus of Bordeaux, in Delamarre 2003, p. 139) moni gnatha, gabi buddutton imon = my girl, take my cock (spindle whorl Saint-Rvrien, Delamarre 2003, p. 228-29) mon gnat ixs = my son my-own (Chateaubleau L-90, Delamarre 2003, p. 181 nd 2 pers. sing.: mentobeto to diuo = remember your god (Life of Saint Symforien, in Delamarre 2003, p. 181) cara uimpi, to caranto = dear beautiful, your friend (spindle whorl Amiens, in Stifter 2009) st 1 pers. pl.: onson = our (Chamalieres, Lambert 2003, p. 160) This can be summarised as follows: mon, moni = my, mine to = your onson = our The other forms are not attested. The form onson is debated and doubtful, and its translation as the 1st pers. pl. possessive is unsure. This therefore leaves mon/moni and to. It is posited here that these forms have been taken as the example for the entire paradigm of possessive pronouns in modern Gaulish. By analogy with to and conforming to the general loss of word final letters all forms end in o. All of the possessive pronouns cause mutation on the first letter of the following word, except the 3rd pers. sing. fem., where the lack of following mutation is the feature that distinguishes the form from the 3rd. pers. sing. masc., which is otherwise identical, and the 2nd pers. pl. which is distinguished from the 3rd pers. pl. in the same way: m: my t: your : his, with mutation : her, without mutation n: our s: your (pl.), without mutation s: their, with mutation e.g.: cun: a dog m gun: my dog gun: his dog cun: her dog n gun: our dog

68 s cun: your (pl) dog s gun: their dog Genetive particle iA genetive particle i- can be observed in the Saint-Rvrien spindle whorl inscription mentioned above: moni gnatha, gabi buddutton imon This can be translated, litterally, as my girl, take cock mine/of-me This same particle is also in evidence in the spindle whorl inscription of Sens (Lambert 2003, p. 126; Delamarre 2003, p. 189-90): geneta imi daga uimpi = girl mine/of-me good beautiful and possibly in the inscription of Rom (Delamarre 2003, p. 189, 341) derti imon = my skin, i.e. my girl It seems that imon refers to a male object (buddutton, penis), and imi to a female object (geneta, girl). It is therefore possible to tentatively identify an element iwhich appears to have been employed in ancient Gaulish to construct a post-posed genetive pronoun. It is posited here that possessive pronouns are no longer post-posed, but instead that the possessive particle i- is used in fusion with pronouns. These are used to indicate the object of a verbal noun, as well as to indicate ownership of objects in conjunction with personal pronouns only: im: of-me ith: of-you ich: of-him - etc. See section on Direct object of verbal noun, p. 43, for full list. esi mi en hpis ith: I am seeing you (I am in the seeing of-you) esi mi en ghar is: I am calling you (pl) (I am in the calling of-you (pl.)) esi sin cun: this is a dog esi im: he/it is mine (i.e. the dog) Particle of quantity uA particle indicating the dealing with a quantity of something is attested in the inscription of Lim (Delamarre 2003, p. 187):

69 ibetis/ uciu/ andecari/ biiete = drink (2nd. pl.)/ of-this/ very-loveable/ you-will-be In this phrase it is possible to identify the particle (or preposition) u-, of, in conjunction with what may be the adverb this in dative or instrumental case (-ciu, cf. ci- in ciallos, Delamarre 2003, p. 116). u+ci+u= of+this+dative/instrum. ending The modern version of the above form would be uch, of-this. It is proposed here that, considering the context of this inscription, the particle u- is used only in association with a quantity of something, not in association with possession or ownership. It causes mutation of the first letter of the following word: pan: glass gwn: wine pan u chwn: a glass of wine lithr: litre curu: beer lithr u guru: a litre of beer

12. Questions Question words Question words are not well attested in Gaulish. However, from what is attested it is possible to deduce clues which can be used to construct a question word paradigm. The least controversial is: ponc = when (Delamarre 2003, p. 252) This is straightforward, and regularly becomes ponch in modern Gaulish. Further indications may be derived from the following : peti VX riuri = ? (Coligny, Delamarre 2003, p. 249) peti sagitiontias = save those who seek? how much they seek? (Larzac, id.) petidsiont sies = they will save ? (Larzac, id.) While interpretations vary and there is no agreement on the exact meaning of these phrases, Delamarre (2003, p. 249) nevertheless identifies the component peti-as meaning how much. This is an important assertion, especially in combination with the following analysis:

70 eti = the adverb more (Delamarre 2003, p. 167-68) Therefore: peti = how much eti = more In light of the analysis of eti as more, it is possible to analyse peti as being a combination of the adverb eti with a hypothetical question word starting with p-, which is the regular Gaulish reflex of inherited Indo European kw-, ubiquitous in the formation of qustion words throughout the spectrum of the Indo European languages (Latin quot, quom etc, Spanish/French que etc., Irish c, Welsh pa): peti > *p-+eti If word internal assimilation of vowels is assumed, it is possible to postulate a form: peti < *pe+eti = what-more, i.e. how much The above hypothetical analysis would indicate that the Gaulish word for what may have been pe. Comparison with the other Celtic languages offers support for this theory: Welsh: pa Irish: c Using this question word pe, what, rendered in modern Gaulish as p, it is possible to construct the following forms: lch: place (< leucutio-, Delamarre 2003., p. 200) p+lch > plch: where (what place) ri: for p+ri > pri: why (what for) The word for who may be arrived at by analogy, by comparison with neighbouring related and IE languages: Welsh: pwy, Cornish pyu/piu, Breton piv, French qui, Latin quis, proto Celtic kwei-jo From the above forms it may be concluded that a form containing the vocalism [i] may be deemed appropriate; as such, it is posited here that: pi: who

71 Lastly, a suitable word for how may be derived from the Proto Celtic root *pod- (< IE kwod-, Bhrgros 2012 pers. com.): *pod- > podh: how The above attestations, constructions and derivations provide a full question word paradigm, which may be summarised as follows: p: what pri: why pthi: how much plch: where ponch: when podh: how pi: who Question word compounds The above question words can be used in conjunction with other words to ask specific questions. If this is the case, the words directly following question words undergo a mutation of their first letter. e.g.: cun: dog p: what/which p gun a hesi : what dog is it? r: hour p: what/which p hr a hesi : what time is it The question word pthi, how many, inquires about a number of something. Therefore, as with numbers, nouns following pthi are in the singular form: e.g.: bledhn: year bledhn: years pethi vledhn: how many years Question formulation All the surviving Celtic languages currently use or have at a previous stage of their development and evolution used interrogative particles to introduce and mark questions. While there is no attestation of any such particles in the ancient Gaulish material, it is nevertheless posited here that they be included in modern Gaulish, on the basis that it is useful and practical to have unambiguous indication of the interrogative mode. Such indication by particles of phrases is not uncommon cross-linguistically, to wit the French phrase est-ce que ....

72

A certain variety of particles and phrases is attested in the surviving languages (W. a, ai, oni, onid, Corn. a, Br. ha(g), daoust ha(g), Ir. an, Sc. Gael. am). It is proposed here that a single particle a be used in the modern Gaulish language. This particle causes a mutation of the first letter of the word following it. The particle introduces questions without question words, and immediately follows question words when these are used: e.g.: gwla mi i: I want to go a chwla ti i: do you want to go? gwla mi vo peth nep: I want to do something p a chwla ti v: what do you want to do? This particle is also used in indirect, embedded questions: e.g.: gn: to know a ghna ti p a chwla ti v: do you know what you want to do?

13. Negation Declarative negation Negation is abundantly attested in the ancient Gaulish material; examples can be found in Lezoux (ne regu; ne dama; ne curri; ne papu; ne tetu), Larzac (ni tixsintor; ne lissatim; ne liciatim; ne rodatim) and Thiaucourt (ni exuertinin) (in Delamarre 2003, p. 233). It appears to be clear that negation in ancient Gaulish was formed by placing the negating particle ne immediately before a verb (ne dama = dont suffer/yield, Delamarre 2003, p. 135). Following the attested pattern discussed above, modern Gaulish constructs its negation by placing a negating particle n immediately before a verb, causing mutation of its first letter: e.g.: gwla mi i: I want to go n chwla mi i: I dont want to go apsa mi: I see n hapsa mi: I dont see Interrogative negation

73 In interrogative negation the negating particle n takes first position in the sentence, preceding the interrogative particle, which does not get mutated itself but which does cause mutation on the word it is followed by: e.g.: n chwla mi i: I dont want to go n a chwla ti i: dont you want to go? n hapsa mi: I dont see n a hapsa ti: dont you see? Replies to questions In analogy with common practice in modern Celtic languages, questions are answered by repeating the main statement either affirmatively or negatively, as the case may be. There is no indication of words for yes or no being used in ancient Gaulish, so therefore modern Gaulish follows modern Celtic practice as described above, and does not have words for yes or no: Q.: a chwla ti i: do you want to go? A.: gwla mi: I want Q.: n a hapsa ti p: dont you see a horse? A.: n hapsa mi: I dont see 14. Initial Consonant Mutation Mutation of initial consonants in Celtic languages Initial consonant mutations have been referred to several times in the preceding sections. It is a phenomenon that is particular to the modern Celtic languages: the first letter of a word systematically changes its sound in certain contexts. These sound changes sometimes, though not always, carry grammatical and semantic meaning. In the context of the modern Celtic languages, it is thought that the systematic mutation of the first consonant of a word (when followed by a vowel) happened when it was preceded by a word ending in a vowel, thus creating an intervocalic environment, when considered across word boundaries. This phenomenon, which is not unusual crosslinguistically, is known as sandhi (sometimes referred to as trivial), and is perhaps best thought of as advanced slurring of one word into another. It is closely akin to the liaisons of the French language, and is widespread in western Germanic languages. In this context, it is worth bearing in mind that it is precisely those languages that replaced the Gaulish language through most of its erstwhile range, and that therefore may have absorbed a certain amount of substratal influence from the Gaulish language, of which the tendency to engage in advanced slurring may have been one aspect. The process is aptly illustrated by an example from the Welsh language:

74

Brittonic *sindos tatos = the father Brittonic *sinda mama = the mother In the first example, the initial t- of *tatos does not find itself in an intervocalic environment, since the article *sindos ends in s. Therefore, no sandhi effect took place on the word *tatos. In the second example however, the initial m- of *mama finds itself in an intervocalic environment, because the article *sinda ends in a vowel. As a result, the initial m- of *mama was weakened, and became [v]. This gives, in modern Welsh: tad: father y tad: the father mam: mother y fam: the mother [vam] The occurence of these mutations, once established, was later endowed with grammatical meaning and became systematical. Mutation in ancient Gaulish It is commonly accepted wisdom that the ancient Gaulish material shows no evidence of the systematic mutation of the first letter of a word depending on context. This is often considered to be an indication that the phenomenon only occurred in the Insular Celtic language group (see e.g. Isaac 2007 for a discussion). However, this is overlooking two important factors: 1. It is thought that the process of development of the system of initial consonant mutations did only fully develop in the 5th-6th centuries CE. The latest Gaulish attestation dates to the late 4th or early 5th century; therefore, a development that might still have been in process might not have shown up in the written ancient Gaulish record. 2. More importantly, mutations were not graphically recorded or represented in the insular languages until many hundreds of years after they first developed as a phonetic and grammatical phenomenon: Old Irish (500-900 CE) and Old Welsh (800-1100 CE) show no mutations, even though the modern languages manifestly feature them, and even though their incorporation into the languages evidently goes back to a period predating the first written records. Mutations only first started to be included in the written record, and imperfectly at that, in the Middle period of both languages. The Breton language did not start to record its mutations until the 17th century. In view of the above, it is clear that if Gaulish was developing a system of mutations, which are a phenomenon of the spoken language, there is a high likelihood that this would not have been reflected in the written record of the time. Nevertheless, it is worth analysing the attested Gaulish material for possible indications of mutation-like processes. It is also important to bear in mind that some leading scholars in the field

75 believe that the attested Gaulish material does indeed show indications of a process of mutation (e.g. Schrijver on Chateaubleau, from David Stifter 2012, pers. com.). There are a number of instances which appear to indicate a variation in the consonant of a word within one and the same document. These are: cutio > gutio (Coligny, Delamarre 2003, p. 133) cutio > qutio (id.) cantlos > gantlos (id.) flatucia > ulatucia (Larzac, Delamarre 2003, p. 338-39) bnanom > mnanom (id.) The context of the first document, in which the first three examples are found, can not reveal anything about the phonetic environment in which these words were found, as the document is a calendar and does not feature continuous writing. The second document however, the inscription of Larzac, is the longest extant running text in the ancient Gaulish language. An examination of the contexts in which the examples quoted above are found yields the following: banona flatucias potiti ulatucia In the above cases, both words are preceded by words ending in a vowel; however, the case apparently differs. insinde se bnanom biontutu se mnanom biontutu indas mnas bietutu se mnas biontutu se mnanom In the above case, it would be possible to isolate the following: se bnanom > preceding word ending on vowel > intervocalic environment > initial b- indas mnas preceding word ending on consonant > non-intervocalic environment > initial m However, the word mnas also occurs three times preceded by se, the same word which precedes bnanom, which has not apparently brought on a mutation of the initial consonant. Furthermore, the consonant of the radical bena, woman, is b-, not m-, so it would be expected that the situation would be the other way around. It is therefore

76 not possible to state that these examples indicate a systematic mutation of initial consonants analoguous with the model operative in Brittonic. Nevertheless, in view of the examples analysed above it is possible to suggest that there were situations in ancient Gaulish where the exact phonetic value of initial consonants was considered either ambiguous or unsure, and was subject to change under some circumstances. This situation appears comparable with the situation prevailing in the Brittonic and Goidelic languages in the 4th-5th centuries CE, which brought forth the systmematic mutation of initial consonants. Therefore, it is posited here that Gaulish would have evolved a system of initial consonant mutation similar to those found in Brittonic and Goidelic. Mutation changes Based on the apparent voicing of c- to g- in Coligny and of f- to, possibly, v- at Larzac, it is proposed here that the following changes take place in mutation contexts: radical p t c b d g m n r l s < vowel (sV-) s < consonant (sC-) gw vowel (V-) semi-vowel imutation b d g v dh gh w nh rh lh sh (shV-) (C-) chw h- (hV-) ch-(chi- = [xj]) phonetic process involved voicing voicing voicing spirantisation spirantisation spirantisation spirantisation ch-prothesis, fortition + spirantisation ch-prothesis, fortition + spirantisation ch-prothesis, fortition + spirantisation palatalisation elision spirantisation and devoicing h-prothesis, aspiration ch-prothesis, fortition + spirantisation

The phonetic values of the above graphemes has been discussed under the section sound changes. As regards to s- before a consonant, it is dropped, and marked in writing with an apostrophe. In the spoken language, the appearance of a radical in a mutational context indicates the elision of s-, therefore indicating the mutation of s-. The various mutations of the modern Celtic languages are categorised according to the phonetic processes which occur, e.g. softening, hardening, nasalisation etc. All of the modern Celtic languages have different kinds and combinations of mutations, although there are deep fundamental resemblances among all of them. The mutations proposed for

77 the modern Gaulish language consist of a combination of phonetic processes, mostly voicing and spriantisation (see above). Therefore, it is easiest to describe this mutation as a mixed mutation. However, its description is irrelevant, as there is only one change for each sound. Contexts for mutation The contexts for mutation have become apparent throughout this document. It is posited that mutation has become established as a system for assisting with the conveying of grammatical meaning in the wake of the loss of case endings and therefore meaning, as discussed previously. As such, they represent a simple switch from marking the end of words for meaning to marking the start of words for meaning. This position is supported by Isaac (2007) who argues that VSO languages have a greater innate tendency to modify words word initially than word finally. There are two kinds of contexts for mutation: contact mutation and grammatical mutation. The contact mutation category is the largest one. In contact mutation, a mutation is triggered by a lexeme immediately preceding the word affected. In grammatical mutation, a mutation is bestowed because of a grammatical condition or requirement, which may or may not be caused by a lexeme immediately preceding the word affected. Contact mutations do not necessarily convey grammatical meaning in themselves; instead they usually assist in the marking of such meaning. However, such meaning would not be apparent in the absence of the lexemes triggering the mutation. As such, it can be said that a large part of the mutations featured in modern Gaulish are phonetic mutations rather than grammatical mutations. This situation is similar to the one prevailing in modern Breton, Welsh and Cornish, where some mutations occur for no apparent reason and carry no apparent grammatical meaning. Contact mutations 1) after preverbal particles The following preverbal particles cause mutation: r, en, a, n, a r: past tense marker: cana mi: I sing r gan mi: I sang en: ongoing tense marker: delgha mi: I hold esi mi en dhelgh: I am holding a: intentional form marker: depri: to eat gwla mi i a dhepri: I want to go to eat n: negation marker: gara mi: I call

78 n ghara mi: I dont call a: interrogative marker: gara ti: you call a ghara ti: do you call? 2) after adverbial particles The following adverbial particles cause mutation: in, r, r, ma in: adverbial adjective marker: tech: beautiful in dech: beautifully r: intensive marker: mr: big r wr: very big r: excessive marker: r wr: too big ma: conditional marker: gwla mi: I want ma chwla mi: if I want 3) after question words Normally question words are followed by the interrogative particle a, which causes mutation on the following word, and is not in itself susceptible to mutation. However, in some cases a question word may be directly followed by a different word, which will then undergo mutation: p: what, which cun: dog p gun: which dog? caman: road, way p gaman: which road, which way? 4) after prepositions All prepositions cause mutation: ar: at, on mr: sea ar wr: at sea gw: under pren: tree gw bren: under a tree

79 can: with caran: friend can garan: with a friend Grammatical mutations 1) feminine nouns after the article Feminine nouns both singular and plural mutate when preceded by the article in. Thus the gender of nouns is effectively marked by the initial consonant mutation. This is not a contact mutation, becaue the article does not trigger mutation in masculine nouns. e.g.: ben: woman in: the in ven: the woman mn: women in wn: the women br: hill in vr: the hill in vr: the hills gr: sand in ghr: the sand in ghr: the sands 2. adjectives qualifying feminine nouns Adjectives qualifying feminine nouns, whether it be one or several, are marked by mutations: ben: a woman tech: beautiful ben dech: a beautiful woman br: a hill ardhu: high br hardhu: a high hill gr: sand gwin: white gr chwin: white sand gr chwin dech: beautiful white sand 3. after possessive pronouns

80 Words following possessive pronouns regularly undergo mutation. As this mutation is triggered by the presence of the possessive pronouns, it could be said that this is straightforward contact mutation. However, these mutations differ from contact mutations in two ways: a) the mutations serve to mark the distinction between the 3rd pers. s. masculine and feminine, and between the 2nd and the 3rd person plural: e.g.: cun: dog m gun: my dog t gun: your dog gun: his dog ICM cun: her dog no CM n gun: our dog s cun: your (pl.) dog no ICM s gun: their dog ICM

b) the mutations are transferred to any words preceding the noun that is possessed, but also across them to the possessed noun, even though it is no longer in contact with the mutation trigger: e.g.: cun: dog pethr: four m bethr gun: my four dogs Unmutatable words Some words can not be mutated, and also block mutation from taking place. These are: 1. the article in The article blocks a contact mutation, and does not mutate itself: e.g.: pen: head ar: on ar ben: on a head ar in pen: on the head 2. the possessive pronouns While the possessive pronouns cause mutations (see above), they are not mutatable themselves: e.g.: pen: head m: my ar: on ar m ben: on my head

81

Additional words that do not mutate are the adverbs n (not) and ma (if). However, contexts that could potentially lead to their mutation are rare. Summary of mutation contexts In summary, mutation is caused by: 1. pre-verbal particles 2. adverbial particles 3. question words 4. prepositions 5. possessive pronouns 6. the article before feminine nouns 7. adjectives qualifying feminine nouns 15. Word Order A variety of word orders is attested in ancient Gaulish, due, it is commonly thought, to the flexibility afforded to sentence structure by the grammatical case endings lending meaning to words regardless of sentence position (Lambert 2003, p. 70-72): e.g.: 1). Licnos Contextos ieuru Anualonnacu caneco sedlon L. C. has dedicated to A. the golden seat Subject Verb Object

2). billicotas rebellias tioinuoru Siluanos billicotas very beautiful has produced S. Object Verb Subject ccc 300

3). sioxti Albanos panna extra tuss added A. vessels beyond batch Verb Subject Object

4). ratin Briuatiom the fort of the Briuates Object

Frontu Tarbetisconios ieuru F. T. has dedicated Subject Verb

5). Bratronos Nantonticnos Epaatextorici leucutio suiroebe logitoi B. N. for E. the place with the sisters has established Subject Object Verb

82

(examples from Delamarre 2003, p. 331-34) The above examples illustrate the wide variety of word orders that is attested in the ancient Gaulish epigraphy: SVO, OVS, VSO, OSV, SOV. However, most of these examples draw on short inscriptions, mostly of a dedicatory nature. Turning to the long continuous texts, it is possible to discern the following: andedion, uediu-mi Nether-ones, pray-I [address] V. S. diiuion of-the-gods ... (Chamalieres, Delamarre 2003, p. 337)) O

buetid ollon reguccambion may-be everything bone-deformation (id.) (? uncertain) V S O exops pissiu-mi soccanti blind will-see-I this-with-it (id.) V S O bisset luge it/you-(pl)-will-be by-the-oath (id.) V O dessu-mi-is prepare-I-them V S O luge by-the-oath (id.)

Nemna lu-mi beni [address] denounce-I a-woman (Chateaubleau, Mees 2010) V S O iegu-mi sini curse-I her (it/this) (id.) V S O iexstu-mi sendi (id.) have-cursed-I this V S O nitixsintor sies duscelinatia (Larzac) not-will-be-bewitched they by-bad-omen V S O biontutu se mnanom (Larzac, Delamarre 2003, p. 338-39) may-hit these women

83 V S

biontutu indas mnas (id.) may-hit these women V S petidsiont sies will save they V S se rinoti this sells O V peti sagitiontias (id.) how much the-seeking-ones O? (S?) XXXV (Rze, Stifter 2012) 35

Secundo dinarilu S. for-dinars S

se tigi prino Ascanius are boletu XV (id.) this contract buys A. for ? 15 O V S These extracts from the most important continuous texts from the ancient Gaulish corpus show clearly that the Gaulish language frequently used constructions where the verb was followed by the subject, often in turn followed by the object (VSO word order). While there is variation on the theme, and while various translations continue to be fought over bitterly without consensus, there is no doubt that a word order placing the subject after the verb was in common usage. Furthermore, the Chateaubleau text, which is the most evolved Late Gaulish text available to date, is virtually exclusively VSO (see Mees 2010 for the parsing and translation). The above observation is supported by Isaac (2007), who asserted that Gaulish had developed a tendency towards VSO word order by the 1st cent. BCE (Isaac 2007). The examples clearly show that verbal forms are followed either by a cliticised subject pronoun, apparently in accordance with Wackernagels Law and Vendryess Restriction, or, when pronouns are not used, by what is clearly recognisable as a subject (e.g. Rze, Larzac). In light of the observations made above, it is deemed reasonable to posit VSO word order as the predominant and default word order for the modern Gaulish language. This is furthermore in analogy with the situation in the modern Celtic languages, where VSO is the standard word order (notwithstanding apparent variations in e.g. Breton, Cornish and Middle Welsh; whilst presenting as SVO, these are historically derived from VSO forms; see e.g. Press 1986). e.g.: apsa mi ti: see I house > I see a house V S O prina mi p: buy I horse >

84 I buy a horse V S O gra mi m gun: call I my dog > I call my dog V S O 16. Demonstratives Demonstratives are well attested in ancient Gaulish under a variety of forms: se, sinde, inda, sini, sendi, sondios, so, sosin, sosio se: se mnanom, se bnanom (4 x at Larzac, Delamarre 2003, p. 338-39) se rinoti (Reze, Stifter 2012) se tigi prino (Reze, Stifter 2012) se tingi (Chateaubleau, Mees 2010) se dagisamo cele (Chateaubleau, Mees 2010) so: so adsagsona (Larzac, Delamarre 2003, p. 338-39) so adgarie (2 x Chartres, Stifter 2012) so cantigarie (Chartres, Stifter 2012) sinde: insinde (Larzac, Delamarre 2003, p. 338) sindiu (today, Coligny, Delamarre 2003, p. 274) > indas mnas (Larzac, Delamarre 2003, p. 338) iegumi sini (2 x Chateaubleau, Mees 2010) sendi: iexstumi sendi (2 x Chateaubleau, Mees 2010) sondios: sondios adgario (Chartres, Stifter 2012) sosin: sosin celicnon (Vaison, Delamarre 2003, p. 279) sosin celicnon (Alise-Sainte-Reine, Delamarre 2003, p. 279) sosio: sosio legasit (Seraucourt, Delamarre 2003, p. 279) Of these, it is likely that sinde, sendi, sondi are variations on the same theme sinde. Likewise, sosin seems to be a combination of so with a variant of sinde, and sosio a variation of so. The form so is also associated with the 3rd pers. pl. (see Delamarre 2003, p. 279), and in modern Gaulish is reanalysed as the possessive pronoun for the 2nd and 3rd pers. plural. This makes its use as a demonstrative impractical. That leaves the forms se and sinde to be used as demonstratives. It is posited here that the form sinde regularly becomes the form sin in modern Gaulish, after regular loss of final de. That leaves the forms se and sin to be used as demonstratives. The attested form sindiu from Coligny (Lambert 2003, p. 112; Delamarre 2003, 274) is widely agreed upon as meaning this-day < sinde-diu (Delamarre 2003, p. 274). This permits the deduction that the form sinde means this. Therefore it is proposed here

85 that the remaining form se be interpreted as that, permitting the construction of a demonstrative paradigm for modern Gaulish. In the attested material, the demonstrative is placed before the noun: e.g.: se mnanom = those women-GEN se tigi prino Ascanius = A. bought that contract se dagisamo cele = that best companion All modern Celtic languages have evolved to move the demonstrative after the noun, in combination with the definite article before the noun. Bearing in mind that word order of those languages is default VSO (Verb-Subject-Object) and that this word order is instrumental in conveying grammatical meaning to phrase components, it is easy to see why this happened. It is not only an expression of the tendency of the Celtic languages to have heads precede qualifiers, and to be strongly right branching, but it is also necessary to be able to convey precise semantic meaning. It has been established that ancient Gaulish was evolving a VSO word order, and that modern Gaulish adopts VSO as its default word order (see section on word order). This means that, if the demonstratives were pre-posed, as in ancient Gaulish, the following phrases would be ambiguous: esi sin ti gwin: is this a house white > this is a white house esi sin ti gwin: is this house white > this house is white If the demonstrative precedes the noun it refers to, it is not possible to discern whether the exact meaning of this phrase is this house is white or this is a white house. However, if the demonstrative is post-posed and combined with a pre-posed article, the meaning becomes unambiguous and clear: esi in ti-sin gwin: is the-house-this white > this house is white esi sin ti gwin: is this a-house white > this is a white house Therefore, it is posited here that in modern Gaulish the demonstratives sin and s, meaning this and that, follow the noun they refer to, in combination with the article preceding the noun; they are further cliticised onto the noun by means of a hyphen. This construction is refered to as bifurcated demonstratives, and it is a feature that is characteristic of the Celtic language family. These demonstratives are not marked for gender or number. in ti-sin: this house (m) in p-s: that horse (m) in lham-sin: this hand (f) in ghr-s: that sand (f) Any adjectives refering to the noun being specified will follow the noun and will precede the demonstrative; in this case, the hyphen is dropped:

86 gwin: white tech: beautiful in p gwin tech s: that beautiful white horse 17. Locatives It is posited here that the demonstratives discussed above also serve to construct the locatives, in conjunction with the article. All the surviving Celtic languages construct locatives in this fashion: Ir.: an: the sinn: that seo: this ansinn = an + sinn: there anseo = an + seo: here W.: y: the ma: this na: that yma = y + ma: here yna = y + na: there Br.: an, ar: the ma: this se: that ama = *an + ma: here aze = *an + se: there While a locative is not clearly attested in the ancient Gaulish material, the inscription of Larzac features the form insinde, the very opening word. The form insinde apears to be constructed in exactly the way the surviving Celtic languages construct the locatives: insinde = *inda-sinde The word insinde is most usually translated as in-this (Delamarre 2003, p. 274). However, it is posited here that the modern Gaulish language has reinterpreted this word as meaning the-this and hence here, in an exact parallel with the surviving Celtic languages. As such, the attested insinde is reanalysed as here, and its unattested counterpart inse will be there. With regular loss of word endings and application of modern Gaulish phonology, that gives: insin: here ins: there The concept of over there (i.e. at a point further away than just there) can be expressed with the attested Gaulish word pel (< pelos, Delamarre 2003, p. 247-48): ins pel: over there (there far)

87

18. Syntax The interpretation of the syntax of the available ancient Gaulish texts is to a large extent a matter of speculation and debate, as they are often poorly understood and hotly debated. Nevertheless there are a number of features that can be observed and which can be used in the modern Gaulish language to construct sentences. These will be detailed below. Verbal noun clauses Verbal nouns form one of the corner stones of modern Gaulish syntax. Unfortunately they are not well attested in ancient Gaulish, or, if attested, they are not well understood. According to Mees (2010) the text of Chateaubleau, which is the latest and most evolved example of continuous Gaulish writing available today, features several examples of verbal forms which can be thought of as verbal nouns. They are described by Mees (2010) as deverbalised forms or deverbal constructions and appear to be used throughout the Chateaubleau document in a uniform manner, regardless of their various morphologies and markings for number, aspect or tense (Mees 2010, p. 103). As such, they display a clear tendency to appear after clitics either to negators or verbs (Mees 2010, p. 103). This situation is refered to as a situation of collapsing-cum-generalsing of verbal nouns of various origins (Mees 2010, p. 103). Examples of such verbal noun forms and their usage are as follows: lu mi beni ueionna = I denounce a woman wishing anmanbe gniou = names knowing sue resetesi iegiinna = may you fix her a cursing siaxsiou beiassu n biti = seeking punishment let her not be cluiou se dagisamo cele = hearing this best companion (from Mees 2010) Of the above, the verbal nouns are: ueionna, gniou, iegiinna, siaxsiou, cluiou These display a wide variety of suffixes and morphological features, including a future marker (-si- in siaxsiou) and possibly an adverbial marker inna, among others (the interpretation of these suffixes is debatable). A position is proposed here where the modern Gaulish language strips these various deverbalised verbal noun constructions of their various suffixes, leaving just the root or stem of the verb: uei = to wish gni = to know ieg- = to curse

88 siag- = to seek clui = to hear These forms will be considered the verbal noun form of modern Gaulish verbs: the bare, stripped back root, denuded of any suffixes. These are the forms used in the section on verbs (see further above). As such, these verbal nouns are the equivalent both of an English infinitive (cli = to hear) and of an English gerund (cli = the hearing). In keeping with Bernard Mees observation that the verbal nouns of Chateaubleau display a clear tendency to appear after clitics either to negators or verbs (Mees 2010, p. 103), these verbal nouns are placed immediately after the personal pronoun which is in clitic position to an absolute verb: e.g.: gwl: to want mi: I i: to go > gwla mi i: I want to go pis: to see in p: the horses > gwla mi pis in p: I want to see the horses Therefore, the first verbal noun used in a complex sentence always follows immediately after the absolute verb, without undergoing any changes to its structure. If the verbal noun used requires a preposition, as is the case with the verbal noun go (> go to), then this preposition will cause a mutation of the following word. If the following word is another verbal noun, then this will mutate like any other noun would: e.g.: a: to i: to go pis: to see > gwla mi i a hpis in p: I want to go to see the horses. Since a, to is a preposition, the following word, the verbal noun, is marked by ICM. It is important to bear in mind that the verbal noun is a deverbalised nominal construction, i.e. a noun, not a verbal form. pis means the seeing. Therefore gwla mi pis: I want the seeing Because a genetive relationship is expressed by apposition of two nouns, the phrase below is genetival: pis in p: the seeing of the horses > gwla mi pis in p: I want the seeing of the horses > gwla mi i a hpis in p: I want the going to the seeing of the horses

89

cli: to hear / hearing can: to sing / singing ethn: birds > gwla mi cli can in ethn: I want the hearing of the singing of the birds > I want to hear the birds sing Because the relationship between the verbal noun and the following object is genetival, if the object is a personal pronoun it must be used in conjunction with the genetival particle i-, used with personal pronouns only (see section on pronouns): ti: you i-: of > gwla mi pis ith: I want the seeing of you > I want to see you > gwla mi i a hpis ith: I want the going to the seeing of you > I want to go to see you Adjectival clauses Verbal noun phrases such as the above can be further specified by the addition of an adjectival clause, which complements a preceding object. This requires a specific object to be stated: gwla mi i a hpis in p gwla mi cli can in ethn In the above phrases, p and ethn are objects. Adjectival clauses can be used to provide further information about the preceding objects of a main clause. To do this, a verbal noun is used. Verbal nouns in adjectival relationships to preceding objects are attested in Chateaubleau at several reprises: ne-i anmanbe gniou = not by the names knowing it iegumi sini, siaxsiou beiassu ne biti = I curse her, seeking punishment not let it be beiassu sete sue, cluiou se dagisamo cele = punishing may be you, hearing this best companion (in Mees 2010) In these examples, the adjectival verbal nouns are gniou = knowing siaxsiou = seeking cluiou = hearing

90

From these attested forms it is possible to deduce that in ancient Gaulish a verbal form expressing an adjectival relationship with a preceding object was constructed by means of the suffix iou. Following regular Gaulish sound changes, this suffix would become i (-ou > -o:). A comparison with the attested Gaulish relativising particle io is immediate. This particle is attested in several instances: gobedbi dugiiontiio ucuetin = the smiths who honour (or fashion/shape) Ucuetis (Alise-Sainte-Reine, Delamarre 2003, p. 153-54) secoui toncnaman tonsionto = the victors who swear an oath/who destin a destiny (Chamalieres, Delamarre 2003, p. 298) scrisumio uelor = that I spit is wanted (Marcellus of Bordeaux, Delamarre 2003, p. 268) The resemblance between the relativising particle io and the adjectival verbal suffix iou may be coincidental. Nevertheless, they appear to perform very closely related functions. As such, it is posited here that in modern Gaulish the two have collapsed into one particle - through assimilation, and are used only in adjectival phrases with verbal nouns refering back to an immediately preceding object. e.g.: rethi: to run ar: on in: the trith: beach (< traxta) > gwla mi i a hpis in p reth ar in drith: I want to go to see the horses that run on the beach in ven: the woman can: to sing in dech: beautifully lavar: to speak > lavra mi can in ven can in dech: I speak with the woman who sings beautifully In both the above examples, the suffix is used with a verbal noun that immediately follows the noun that the adjectival clause provides information about. Essentially, the suffix - provides a dummy subject for the verbal form, referring back to the preceding object: lavra mi can in ven can in dech translates as: speak I with the woman sing-that in beautiful

91 As such, the relative particle - is in subject position following the verbal noun. The first example above can also be constructed using the progressive aspect, using the preposition en: gwla mi i a hpis in p en rhethi ar in drith > I want to go to see the horses running on the beach Subordinate clauses Subordinate clauses that have a separate subject embedded within them do not use the suffix discussed above, as the subject in a modern Gaulish phrase always follows the verb. An adjectival verbal noun can not refer to a subject that follows after it. The following phrase I speak with the woman whose daughter sings beautifully can NOT be constructed as follows: dithir: her daughter > lavr mi can in ven can dithir in dech This can NOT convey the meaning I speak with the woman whose daughter sings beautifully. As the particle effectively provides a dummy subject refering back to the previously stated noun, the phrase above would have two subjects, which can not be. To construct the above phrase a separate particle introducing a subordinate clause with embedded subject is required. In the ancient Gaulish text of Chateaubleau it is possible to identify a particle that appears to fulfill this function: iexsetesi sue regeniatu o quprinno = may you curse the family that is purchasing may-curse you family that purchases / is purchasing (in Mees 2010, p. 101) It appears that the particle o fulfills a subordinating function in this phrase. Therefore, it is posited here that the modern Gaulish language will use this particle o as a single, all-purpose, non-referential relativising subordinating particle. The use of non- referential relative and subordinating particles is uniform across the modern Celtic languages (Isaac 2007). e.g.: lavra mi can in ven o cna dithir in dech I speak with the woman that sings her daughter beautifully i.e. I speak with the woman whose daughter sings beautifully

92 Since a subject is embedded in the subordinate clause, the main verb is in absolute form, not in verbal noun form: can: to sing cna dithir: her daughter sings The relativising particle o is augmented by the semantically empty phonetic clitic-ch if the following word starts with a vowel, for ease of pronunciation, as is the case in Modern Breton. gal: to be able to pis: to see di: good > gala mi pis och esi di: I can see that it is good Using the above particles complex compound sentences can be constructed: pethi: to ask adhim: to-me ma: if gwel: to want r chwels mi: would want I laman: dance (lam hand + man move = moving holding hands > to dance) techi: to leave r dechi: left : her caranchwir: her boyfriend (caran friend + gwir man/male) geneth: girl al: other > lavra mi can in ven o r bethi adhim ma rh chwels mi laman can dithir o r dechi caranchwir can gheneth hal I speak with the woman who asked to-me if I would want to dance with her daughter whose boyfriend left with another girl The example used in the section on adverbial clauses can also be expressed using the relativising particle o: lavra mi can in ven o cana in dech: I speak with the woman who sings beautifully Conjunction clauses The above example shows that subordinate clauses can also be headed by conjunctions, such as: ma, if:

93

petha in ven adhim ma chwla mi laman can dithir: the woman asks to-me if I want to dance with her daughter gw, or: a chwla ti laman gw n a chwla ti?: do you want to dance or dont you? ach, and: gni: to know gwla laman canim ach n ghna mi ma chwla mi: she wants to dance with me and I dont know if I want to ithr, but: gwla laman canim ithr gwla mi vi curu: she wants to dance with me but I want to drink beer Any other conjunctions can be used in a similar fashion.

Question word clauses Question words and their attendant interrogative particles can be used to head subordinate clauses in a way similar to that described above: pi: who esi: is > n ghna mi p a hesi dithir: I dont know who her daughter is p: what r: hour : it > a ghna ti p hr a hesi ?: do you know what time it is? (do you know what hour it is) plch: where > n ghna mi plch a hesi dithir: I dont know where her daughter is podh: how > n ghna mi podh a hesi : I dont know how she is ponch: when

94 techi: to leave > n ghna mi ponch a rh dechi : I dont know when she left pr: why > n ghna mi pr a chwla laman canim: I dont know why she wants to dance with me pethi: how much / many pan: glasses u: of curu: beer > n ghna mi pethi pan u guru a hivthu : I dont know how many glasses of beer she has drunken 19. Spatial paradigm Terms defining the spatial paradigm are well attested in the ancient Gaulish corpus. It is possible to identify a number of pairs of opposing terms which between them define the elevation aspects of the spatial paradigm. All terms are featured in Delamarre (2003) or Lambert (2003). gwer: over uch: up uchel: above ardhu: high / gw: under / aner: down / anel: below (by analogy with uxellos, Delamarre 2003, p. 330) / th: low

These concepts are attested as being used to describe geophysical features: uchn: waterfall, i.e. water from above (by analogy with andounna) ann: source, spring, i.e. water from below (< andounna, Delamarre 2003, p. 48) They are also attested as being applied to define metaphysical concepts: uchedh: superior, better (Delamarre 2003, p. 330) anedh: inferior, worse (Delamarre 2003, p. 46-47) Similarly, they are attested as having been used in combination with the term tam, level, to construct a scale of appreciation of worth or value (Lezoux, in Delamarre 2003): anamich: worst, bad, poor (quality) (< andamica, Delamarre 2003, p. 46) mthamich: mediocre, ordinary, average (quality) (< mediotamica, p. 222) gwerthamich: best, good, excellent (quality) (Delamarre 2003, p. 317) This allows for the construction of two parallel comparitive value systems:

95 a) The grammatical system, using the attested prefix gwer (Delamarre 2003, p. 317) and the attested suffix am (Mees 2010 p. 104; Delamarre 2003, p. 233) di gwer dhi disam (< dago-uer dago-dagisamo) good better best mes gwer wes mesam (< messamobi, Delamarre 2003, p. 226) bad worse worst b) The metaphysical system, as defined above: math uchedh - gwerthamich fine superior excellent druch anedh - anamich bad inferior poor Both systems can be used freely as preferred. The existence of more than one option to express a concept only adds to the richness of language, and is essential to avoid staleness and restriction of expression.

20. Temporal paradigm The temporal paradigm, used to measure the passing of the time is not comprehensively attested in the ancient Gaulish corpus. However, based on the attested material constructions can be made to complete the paradigm, in conjunction with some loan words from other languages the existence of which in surviving Celtic languages appears to justify their inclusion here, and combined with re-constructed speculative Proto-Celtic words. All terms listed below are derived from Delamarre (2003) and Lambert (2003) unless otherwise specified. aman: time sthl: generation iu: age r: hour (modern loan) minuth: minute (id.) sechon: second (id.) pimdhech minuth: fifteen minutes, quarter of an hour sim r: half hour (< sim half, attested) d: day sin: today

96 nith: night sinith (by analogy with sin < sindiu > sindenocta > sinith) lthi: period of 24 hours, daytime, day and night bri: dawn (< proto-Celtic *ba:re:gom, loss of om & g-, -e > i cf. Lambert 2003, p. 43) methin: morning (< Latin matina, cf. Br. mintin, C. metten, Ir. maidin) mdhi: midday (< medh middle + d) swdhi: afternoon (< s after + medh + d) nesnith: evening (< nes near, close to + nith medhnith: midnight (< medh + noith) avri: tomorrow (< a to, at + bri) s havri: after tomorrow des: yesterday (< proto-Celtic *gdijes) cin dhes: before yesterday? sithnith: week (seven-night, by analogy with e.g. trinoctia) pensithnith: weekend (< pen head + sthnith) ms: month bldhn: year penvldhn: anniversary, birthday sonching: period (year?) sam: summer meth: autumn (harvest < *met- to harvest) gam: winter gwis: spring trinith: three-night feast dechnith: ten-night feast 21. Comparative Paradigm First degree superlative Enough of a variety of comparative forms can be discerned in the attested Gaulish material to permit the construction of a comparative paradigm. The inscription of Cajarc shows: redresta in uertamon nantou = climb to the summit of the valley (Delamarre 2003, p. 317, 332, 256) in which can be seen that the preposition uer, over, beyond, on, is used in combination with the word tam, level, to indicate a notion of superior quality. The same construction is found in the word

97 uertragos = over foot, super foot > fast feet > a hunting dog Delamarre (2003, p. 145, diuertomu) indicates that in both these cases as elsewhere (diuertomu) a superlative notion is constructed with the preposition uer. Therefore it is posited here that modern Gaulish constructs the first degree of its comparative paradigm with the preposition gwer, which, being a preposition, causes a mutation on the following word: sr: long gwer shr: longer mr: big gwer wr: bigger Equative degree Attestation of an equative degree of comparison can be discerned in the form Comarus (Delamarre 2003, p. 122), which is identified by De Bernardo-Stempel and Slocum (n.d.) as meaning as big [as] or equally big (De Bernardo-Stempel & Slocum, n.d., Old Irish Online). This equative form is constructd with the preposition com. Therefore, modern Gaulish adopts the use of the preposition co to construct the equative degree of comparison. It is reiterated after the adjective to tie up the equation. As a preposition it causes mutation both times: c shr c shin: as long as this c wr c hp: as big as a horse Second degree superlative A second degree superlative is well attested in ancient Gaulish: dagisamo = best (Chateaubleau, Mees 2010) belisama = most powerful (theonym, numerous attestations, see Delamarre 2003, p. 71-72) messamobi = the worst (Lezoux, Delamarre 2003, p. 226) nessam = the nearest (Banassac, Delamarre 2003, p. 233) tragisama = the quickest, most fleetfoot (several river names, Delamarre 2003, p. 300) brigiomu = the briefest (Coligny, Delamarre 2003, p. 88) diuertomu = without-highest (Coligny, Delamarre 2003, p. 145) Delamarre (2003, p. 233) identifies the second degree superlative suffix as being samo, rendered as omu in Coligny. The same suffix is found in Brittonic and Gouidelic. Therefore, it is posited here that the modern Gaulish language uses a second degree superlative suffix sam, to be reduced to am after consonants: di: good

98 disam: best sr: long sram: longest (cf. Welsh hiraf) pel: far pelam: furthest (cf. Welsh bellaf) nes: near nesam: nearest (cf. Welsh nessaf, Old Ir. nessam) This paradigm can be summarised as follows: mr: big c wr (c): as big (as) gwer wr (c): bigger (than) mram: biggest di: good c dhi: as good, i.e. as well gwer dhi: better disam: best (see also p. 92 for a discussion of the metaphysical comparative system). Both the equative and first superlative degrees make use of the preposition c to link the things that are being compared to each other: e.g.: esi in p-sin c wr co n p s: this horse is as big as that horse esi in ven-sin gwer dech cn ven-s: this woman is more beautiful than that woman This same preposition is also used adverbially to provide the equivalent of the English emphatic so: e.g.: ghru: cold aman: weather (also time) esi in haman c hghru: the weather is so cold Similarity A word describing similarity is not attested in the ancient Gaulish corpus. However, the word *samalis is attested across all the modern Celtic languages, is deemed to be of sufficient antiquity to warrant the hypothesis that it would have been included in the ancient Gaulish language, and is widely accepted as such (Stifter 2012, pers. com.). As such, it is posited here that the modern Gaulish language uses the word samal to indicate similarity, in the following fashion: dithir: daughter mthir: mother

99 esi in dhithir samal mthir: the daughter is like her mother The tablet of Lezoux provides a term which may be used to accord an added degree of similarity (Delamarre 2003, p. 115): messamobi molatus certiognu sueticon = the worst praises born of certainty wellsufficient ... In this phrase, the word certiognu (born of certainty?) contains the stem cert-. This stem is translated as right, correct, exact, true, just (Delamarre 2003, p. 115). It is posited here that this stem is used by the modern Gaulish language to convey an additional degree of similarity: esi in dhithir samal mthir in gerth: the daughter is exactly like her mother In the above example the term exactly is used adverbially. Sameness The ancient Gaulish corpus does not contain anything that could be translated as the same. However, for this purpose a simple and ingenious expression can be borrowed from modern Welsh which does not necessitate borrowing words as such: W. yr un = the one This can easily be rendered in modern Gaulish as: in on: the one, i.e the same > esi in dhithir ach in wthir in on: the daughter and the mother are the one, ie. they are the same Quantity paradigm A word for more can be deduced from attested Gaulish material. The words: coettic (Larzac, Delamarre 2003, p.167) = with-and-more eti (La Graufesenque, Delamarre 2003, p.167) = more etic (Alise-Sainte-Reine, Chamalieres, Larzac, Delamarre 2003, p.167) = and-more peti- (Larzac, Delamarre 2003, p. 249) = what-more > save (how much) petidsiont (Larzac, Delamarre 2003, p. 249) = they will save all contain the stem eti, which is identified by Delamarre (2003, p. 167) as the term more. Therefore, it is posited that modern Gaulish uses the word eth to indicate an increase of a quantity: curu: beer

100 > gwla mi curu eth: I want more beer The opposite less can be constructed from this word by means of the adjectival/adverbial negating prefix an-, attested at Coligny and Larzac (Delamarre 2003, p. 43): matu = favourable (Coligny) anmatu = unfavourable (id.) andogna = indigenous (Larzac) anandogna = non-indigenous (id.) This gives modern Gaulish a word for less: eth: more aneth: less > n chwla mi curu aneth: I dont want less beer The attested word coettic (see above), constructed from the preposition co- + eti + -c, translates as with-more-and. It will be used in modern Gaulish under the form ceth: also, furthermore A word for a lot, much, can be identified in attested onomastic material: eluo- = numerous, many, a lot (Delamarre 2003, p. 162) This can be used in the modern Gaulish language as lu: much, a lot, many vi: to drink > gwla mi vi curu lu: I want to drink a lot of beer rdhi: to ride > gwla mi rdhi p lu: I want to ride a lot of horses A word for enough can be found in the Gaulish corpus at Lezoux, where Fleuriot identifies the term extincon with sufficiency (Delamarre 2003, p. 340). Therefore a modern term ithinch can be proposed for enough: mi-thu curu ithinch: I have had enough beer The quantity paradigm for the modern Gaulish language can therefore be summarised as follows: lu: a lot, much, many th: more aneth: less

101 ceth: also ithinch: enough 21. Diminutive form The word bardala, attested in glossary (Delamarre 2003, p. 67), is understood to indicate a songbird with a crest (alouette huppe), and is widely thought to be a diminutive formed on the well known Gaulish word bardos (bard, singer, poet, musician): bardala = bard- + al + a The suffix al evident in the above example may be interpreted as a late Gaulish development of an earlier suffix ilos, attested elsewhere: brogilos > breialo = small area of land or forest Both the above terms are attested (Delamarre 2003, p. 91-92). They show the evolution from the earlier term on -ilo to the late Gaulish form on -alo. It is posited here that the modern Gaulish language uses the suffix -al as a diminutive marker: p: horse pal: foal (cf. Welsh ebol, Cornish ebal, Breton ebeul, all foal) cun: dog cunal: puppy avon: river avnal: creek naus: boat nausal: small boat 22. Word Formation The formation of words is a crucial aspect of the functioning of any language, and particularly so for the revival and reconstruction of a language that is imperfectly and erratically attested. The ancient Gaulish corpus bears witness to a variety of ways in which words were constructed. The following sections deal with these various ways, and indicate how they might be applied to the modern Gaulish language. Historical compound nouns Compound words combining two or more nouns, verbal nouns or adjectives are well attested in the ancient Gaulish material. Two components are joined by a composition vowel, and endowed of a case ending, treating the whole as one unit. Random examples are: Noun + noun:

102

argantodanos > argant-o-dan-os = silver + magistrate > magistrate in charge of minting silver coins (Delamarre 2003, p. 54) blatomago > blat-o-mag-o = flower + field > flowerfield, field of flowers (Delamarre 2003, p. 78) teutoboduus > teut-o-bodu-us = people + raven > people raven > raven of the people? (Delamarre 2003, p. 81) Noun + verbal nouns: uidubion > uidu + bion = wood + cutting > axe, saw (Delamarre 2003, p. 75) namantobogi . namanto + bogi = enemy + breaking, hitting > enemy hitter (Delamarre 2003, p. 81) Noun + adjective: glisomarga > gliso + marga = white, shining + clay > shiny clay (Delamarre 2003, p. 181) uerolamium > uero + lamium = large + hand > generous (Delamarre 2003, p. 195) Many of the compounds attested in the corpus of the Gaulish language, including all of the above, conform to the standard Indo-European compound formula Modifier + Head. However, it is possible to discern an evolution in Late Gaulish forms: Uerolamium > uero + lam- = large + hand > generous (uero = large; lama = hand) Lamauerus > lama + ueru- = hand + large > generous (Delamarre 2003, p. 195) Attested synomyonous forms such as this are thought to indicate (Delamarre 2003, p. 99) that Late Gaulish was undergoing a shift from the standard IE construction Modifier + Head towards the construction Head + Modifier. Another example is found in calliomarcos > calli-o-marc-os = shoe-horse > horse shoe Another word for horseshoe is attested as epocalium > epo-calium = horse + shoe It is thought that the latter, following the standard IE compound noun construction rules, is more ancient than the former, which is thought to be Late Gaulish (Delamarre 2003, p.

103 99). This indicates that Late Gaulish appeared to be undergoing a shift away from Modifier + Head constructions and towards constructions of Head + Modifier. Noun-noun compounds As the shift described above mirrors the shift in adjectival constructions (see section on adjectives), it is posited here that the concept of compound nouns is reinterpreted in the modern Gaulish language as an attributive adjectival noun-noun compound. As such, nouns and verbal nouns are interpreted as attributive adjectives modifying a preceding head noun. To indicate the fact that the two belong together they are pronounced together with emphasis on the second component. If the second component has only one syllable, this will receive the emphasis. If the second component has two or more syllables, the emphasis will be on the penultimate, as is habitual. They are further phonetically linked and fused together by the application of the regular modern Gaulish word-medial sound changes to the initial consonant of the adjective, as per the rules outlined in section 3 (sound changes). Therefore, these changes only affect t, c, b, d, g, m and gw. P, n, r, l, s and vowels are unaffected. To recap, these changes are: t > th c > ch b>v d > dh g > gh after C, i after V m > w after C, m after V gw > chw These changes happen irrespective of the gender of the head noun. Plural is marked at the end of the compound word. e.g.: caliomarcos > loss of composition vowel and word ending > calmarc > regular modern Gaulish sound changes > calwarch: horse shoe > plural calwarch: horse shoes (emphasis in bold) Examples of such noun-noun compounds include: gwir: man (m) pen: head > gwirpen: head man cernu: horn (m) tru: bull > cernuthru: bull horn

104 ti: house (m) curu: beer tichuru: pub p: horse rithi: to race > prithi: racehorse cal: shoe (f) march: horse > calwarch: horseshoe march: horse (f) cathu: battle > marchathu: battlehorse (warhorse) (< march-chathu, fusion of identical phonemes) ben: woman (f) lauthi: to wash > benlauthi: washer woman In the above examples, second components on p, r, l do not undergo changes. Plurals are marked regularly: gwirpen > gwirpen: head men cernuthru > cernuthar: bull horns tichuru > tichur: pubs prithi > prith: racehorses calwarch > calwarch: horse shoes marchathu > marchath: war horses benlauthi > benlauth: washer women The noun-noun compound construction discussed here closely resembles the appositional genitival construction, but is distinguished from it by the internal fusing of the initial consonant of the second component: pen: head b: cow > penv: cow head > pen b: head of a cow As is the case above, for some words the distinction will be clear due to the internal sound change: cal: shoe march: horse calwarch: horseshoe

105 cal march: a shoe of a horse However, in cases where the second component of the compound starts with n, r, l, p or s it wil not be possible to distinguish phonetically between a compound word and a genitival noun phrase: ben: woman lauthi: washing > benlauthi: washer woman ben lauthi: woman of washing The compound and the noun phrase above sound exactly the same. However, this is not an issue as the meaning of the two phrases is identical. Compound adjectives If the second component of a compound is a true adjective, it is possible for the whole compound to be interpreted as a true adjective: lam: hand gwru: broad > lamchwru: wide hand > generous > lamchwras: generosity For the above example, an opposite is attested in corolamus = closing-hand > avaricious (Delamarre 2003, p. 195) In this case not a true adjective is used, but a verbal noun acting as an adjective. Nevertheless the entire compound becomes adjectival: lamchor: avaricious > lamchras: avarice This is a good example of the ability of a verbal noun to express adjectival qualities, and of the flexibility of the Celtic languages, ancient and modern. Prefixes The attested material shows a number of prefixes were used in ancient Gaulish to construct words. Examples are: s: good < suauelo- = su + auelo wind = good wind > welcome (Delamarre 2003, p. 284) d: bad < ducarius = du + car- love = unpopular, unpleasant (Delamarre 2003, p. 157)

106 di: un-, off, from < diacus = di + acu- speed = slow, lazy (Delamarre 2003, p. 143) ath: re-, again < ate- < atespatus = ate + spa- saying = reply, answer (Delamarre 2003, p. 57) an: un-, im-, non- < anandogna = an + andogna indigenous = non-indigenous (Delamarre 2003, p. 43) an: very < ande < anderoudus = ande + roud- red = very red As is clear from the above examples, these prefixes were used in combination with nouns (auelo, acu), verbal stems/nouns (car-, spa-) and adjectives (andogna, roud-). It is proposed here that modern Gaulish does exactly the same thing, and uses these prefixes with nouns, verbal nouns and adjectives to construct words as required. Because these prefixes become fused onto the words they combine with and form one contiguous word with the first consonant of the word they fuse onto becomes subject to the rules of word internal sound changes detailed in section 3 (sound changes). These are as below: original initial consonant t c b d g gw m after su-, du-, di-, ath-, an th ch v dh i chw m after anth ch v gh chw w

All other consonants remain unchanged. These prefixes carry the following connotations: 1. s- expresses favourable notions of quality. It is used with nouns (swiedh) and verbal nouns, denuded of any word final vowels (suchwidh), and creates adjectives. swiedh: good looking < su-agedos < agedos face = good-face schwidh: clever < su-uidu < uidu to understand = good-understanding 2. du- expresses unfavourable notions of quality. It is used with nouns (dwiedh) and verbal nouns (duchwidh) and creates adjectives. dwiedh: ugly < du-agedos = bad face dchwidh: stupid < du-uidu = bad understanding 3. ath- expresses a repetition of something, comparable to English/French re-, again. It is only used with verbal forms and creates new verbs.

107

Before a vowel the final is dropped: ath + v to do > athv: to repeat, do again ath + pis to see > athpis: to see again Before a consonant the is retained: ath + men to think > athmen: to rethink ath + gar to call > athiar: to recall, call again ath + brs to break > athvrs: to re-break, break again ath + rethi to run > athrethi: to re-run The word ath is also used by itself and means again. This provides two alternative ways of expressing a repetition: gwla mi v ich ath: I want to do it again gwla mi athv ich: I want to do it again 4. an- expresses an intensification of something. It is only used with verbal nouns and adjectives, and creates new verbs and adjectives: an + car to love > anchar to love very much > to adore an + bi to hit > anvi to hit very hard > to smash an + mr big > anmr very big > huge an + mi small > anmi very little > tiny 5. di- is the equivalent of English un-, im-, non-, and expresses the absence or opposite of something. It is used with verbal nouns, nouns and adjectives and creates new verbs, nouns and adjectives. di + i to go > dii: to come di + anthi to end > dianthi: to begin di + sami to stand > disami: to fall di + techi to leave > dithechi: to arrive di + menvethi to remember > dimenvethi: to forget di + anth end > dianth: beginning di + panth suffering > dipanth: enjoyment di + lnu full > dilnu: empty di + achu fast > dichu: slow di + len happy > dilen: unhappy

108 4. an- is synonymous with di- discussed above, and expresses the opposite or absence of something. It translates as un-, im-, non-, but unlike di- it is used with adjectives only. an + math fine > anwath: bad, poor an + gwr true > anchwr: untrue an + anghn indigenous > ananghn: non-indigenous an + carthach loveable > ancharthach: unloveable Substantivising suffix for adjectives In the section dealing with adjectival formation it was discussed how nouns and verbal forms can be turned into adjectives by means of a variety of suffixes (see section 6 adjectives). It is also possible to turn an adjective into a noun. The Late Gaulish text of Chateaubleau includes the following: suante ueiommi petamassi Papissone = for wanting I desire perdition for Papissona siaxsiou beiassu ne biti = let her not be seeking [my] punishment (in Mees 2010) In the above examples, petamassi (with Tau Gallicum) is analysed as representing an abstract (accusative) form (Mees 2010, p. 101). Comparison is made with the Irish suffix as and the Gothic suffix assus (Mees 2010, p. 101). Of the word beiassu it is suggested that its suffix assu (also with Tau Gallicum) also seems to represent an abstracting form(Mees 2010, p. 102). In the first case the root pet- is thought to be derived from pant-, suffering, hence the translation of petamassi as perdition. In the first case the root bei- is thought to be related to bei-, strike, and beiassu is translated as punishment. While it is true that in neither of the two cases above illustrated an adjective is involved in the formation of these abstract forms, a suffix similar to the one attested above is nevertheless found in the modern Celtic languages, where it creates an abstract form from an adjective. Examples are: Breton levenez: happiness Cornish lowena: happiness (<*lowenas (?)) Irish sonas: happiness cairdeas (friendship). While it it not entirely sure that the suffixes apparent in the above modern languages are descended from the same formation as -assu, it is nevertheless posited here that modern Gaulish will use the suffix -assu to form abstract nouns from adjectives, on the basis that it is necessary to be able to do so, and that the attested suffix -assu seems to fit the

109 bill and will make do. It is therefore posited here that the modern Gaulish language will use the suffix -as (< -assu) to form abstract nouns from adjectives: e.g.: len: happy > lanas: happiness sr: long > sras: length lithan: wide > lithanas: width ardhu: high > ardhas: height mr: big > mras: greatness, size gwr: true > gwras: truth ru: free > ras: freedom (intervocalic u- is not lost because it is an emphasised vowel, not a semi-vowel w-) caran: friend > caranas: friendship Prepositions as prefixes The use of a variety of prepositions in word construction is well attested in the ancient Gaulish corpus. They are found in word initial position and combine with verbs, nouns and adjectives to create new words. e.g.: adret- = attack < ad towards + ret- run (Delamarre 2003, p. 31) conrunos = confidant < con together + run- secret (Delamarre 2003, p. 123) cantipisontias = witnesses < canti with + pisontias those who see (Chartres, in Stifter 2012) diacus = slow < di off, away from + acus speed (Delamarre 2003, p. 145) (di is preposition as well as prefix; see previous section). exobnos = fearless < ex without, out of + obnos fear (Delamarre 2003, p. 170) It is clear from the above examples that any preposition can be prefixed to a noun, verbal noun or adjective. It is therefore proposed here that the modern Gaulish language does exactly that, and thus creates new words. e.g.: can: with echan: without (< ex + canti) i: to go di: to come (see section on prefixes further above) ur: against gar: call urghar: to argue con: with, like iedh: face coniedh: similar > conaidhas: similarity, resemblance

110 > conaidhi: to resemble Calques For the purpose of constructing words necessary for a modern language, it is proposed here that calques be made of words in other languages, if appropriate and justifiable. A calque is a translation of the various components of a complex word, to arrive at a complex word in another language with the same meaning. e.g.: pwysfawr (Welsh) = important > pwys + mawr = weight + big pwys < pois (French) weight This idiomatic expression for the abstract concept important is semantically related to the English adjective weighty, used e.g. for a weighty decision to make, which clearly infers a notion of importance. While the ancient Gaulish corpus does not appear to contain a word for heavy, it does feature the word luxtos, which is translated as loaded (Delamarre 2003; p. 212). It is possible to posit the following: luxtos > lith, according to the regular soundchanges of modern Gaulish lith: load > lithach: heavy > lithachas: heaviness, weight The use of the word component maros to phrase a concept is attested in ancient Gaulish: attested iantomaros, with big desires, is identical to the Old Irish etmar, which translates as jealous, zealous. This indicates that the component maros can be used to lend an extended adjectival quality to the whole constructed word, not just to a preceding component. Therefore the following is posited: lith + mr > lithwr: important > lithwras: importance It is clear that the process of forming calques on words from related languages can be of considerable assistance in the construction of a sufficiently varied and diverse vocabulary for the modern Gaulish language. 23. Verbal Word Derivation The area of derivation of words from verbal stem is particularly rich and important for the construction of vocabulary. A number of forms exhibiting a variety of verbal suffixes which construct a number of different concepts are attested in the Gaulish corpus: anextlo = protection cantlo = song, music?

111 sedlo = seat caranto = friend arganto = silver namanto = enemy arueriatis = which gives satisfaction orget = murderer cinget = warrior orgen = murder popillos = cook menman = thought garman = call toncnaman = oath/destiny glon = obstruction, thing that sticks (all from Delamarre 2003) From the above it is possible to deduce a number of verbal form substantivising suffixes and their meaning. Abstract noun suffixes: -tlo, -lo, -anto, -eno, -man, -naman, -on -tlo: aneg- > an-eg = to go-along > to accompany > to protect > + tlo = protection -lo: sed- = to sit > + -lo = seat -anto: car- = to love > caranto = friend (someone who loves / is loved) arg- = to shine > arganto = silver (something that shines / is shiny) nama- = to dislike > namanto = enemy (someone who dislikes / is disliked) (-anto is an ancient present particple suffix that has lost its verbal connotations and has become fixed as a substantivising suffix; see Delamarre 2003, p. 107) -eno: org- = to murder > orgeno = murder -man: men- = to think > menman = thought gar- = to call > garman = call

112 -naman: tonc- = to swear, pledge > toncnaman = oath -ion: gli- = to stick, get stuck, obstruct > glion = something that sticks, obstruction Agentive noun suffixes: -iatis, -eto, -ilo -iatis: aruer- = to please > arueriatis = that which pleases -et: org- = to murder > orget = murderer cing- = to go to war (to advance) > to fight > cinget = warrior -ilo: pop- = to cook > popilo: cook Using the above information it is possible to compose a verbal word derivation paradigm based on verbal morphology. From each verbal root will be derived: a verbal noun an agentive form an abstract noun adjectives if possible nouns derived from these adjectives The above suffixes will take the following modern Gaulish forms: -tlo > -thl: loss of final vowel, spirantisation of stop -lo > -l: loss of final vowel -anto > an: loss of final vowel and wordfinal t (> th) after -n -eno > en: loss of final syllable -man > u: lenition of man to uan, loss of final syllable (*-man > *-uan > *-ua > *-u), cf. attestation of garuo, Delamarre 2003, p. 176 -naman > -na: loss of final syllable -on > -on: no change -iatis > -ath: loss of final syllable, spirantisation of stop -eto > -eth: loss of final syllable, spirantisation of stop -ilo > -il: loss of final syllable Verbal noun formation

113 It is proposed here that the use of the above suffixes will be determined by the morphology of the verbal stems they complement. A number of verbal noun classes can be outlined: 1) verbal stems on th, -dh, -ch, -p, -v form their verbal nouns on i: ret- > rethi: to run sed- > sdhi: to sit derc- > derchi: to watch pop- > popi: to bake gab- > gavi: to take Justification for this is found in attested forms on i, e.g. gabi, lubi, exugri, carni-. While several of these examples are imperative forms, it is posited here that this imperative ending has spread to the verbal noun by analogy. While verbal nouns are sufficiently attested in the ancient Gaulish corpus (e.g. at Chateaubleau and Chartres), their formation is heterogenous (see Mees 2010) and poorly understood. Nevertheless, the following forms are commonly proposed and accepted (Bello 2012; Bhrghros 2012): gabi, imperative, *gabion, verbal noun > *gabion > *gavion > gavi: to take (loss of final syllable) lubi, imperative, *lubion, verbal noun > *lubion > *luvion > luvi: to adore (loss of final syllable) 2) verbal stems on Consonant+g form their verbal nouns on : org- > orgh: to murder delg- > delgh: to hold cing- > cing: to go to war, fight, march, advance The above is based on the attestation of orge as the imperative of a verbal stem org-, to murder, which is broadened to include all verbal stems on Consonant+g. 3) verbal stems on Vowel+g form their verbal nouns on Vowel+i: ag- > i: to go log- > li: to establish aneg- > ani: to protect 4) verbal stems on n, -r, -l, -m, -s form their verbal nouns without changing the root: men- > men: to think gar- > gar: to call uel- > gwel: to want dam- > dam: to endure, tolerate, accept

114 apis- > pis: to see 5) verbal stems on a form their stems without changing their root: cara- > cra: to love ama- > ma: to like nama- > nma: to dislike peta- > petha: to ask 6) mono syllable verbal stems on i dont change: gli- > gl: to stick, get stuck, obstruct gni- > gn: to know lig- > l: to lie (down) Agentive form formation The three agentive suffixes ath, -eth and il are distributed according to the morphology of the verbal nouns: 1) verbal nouns on n, -r, -l, -m, -s, fricative stops +i, Vowel+i, and mono syllabic i take the suffix ath: men > menath: thinker gar > garath: caller gwel > gwelath: wanter dam > damath: endurer pis > apisath: see-er rethi > rethath: runner sedhi > sedhath: sitter derchi > derchath: watcher gavi > gavath: taker i > iath: go-er ani > aniath: protector li > liath: establisher gn > gnath: knower 2) verbal nouns on - take the suffix eth: orgh > orgheth: murderer delgh > delgheth: holder cing > cingeth: warrior 3) verbal nouns on pi take the suffix il:

115

popi > popil: cook 4) verbal nouns on a and - drop the final vowel and take the suffix ath : cra > carath: lover petha > pethath: asker v > vath: do-er, maker Abstract noun formation The suffixes thl, -l, -an, -en, -u, -na, -on are also distributed according to the morphology of the verbal nouns: 1) verbal nouns on n, -r, -l, -m take the suffix u: men > menu: thought gar > garu: call gwel > gwelu: will(power) dam > damu: endurance 2) verbal nouns on thi, -vi, -pi, -s and drop their final vowels and take the suffix an: rethi > rethan: run gavi > gavan: taking, take popi: to cook > popi > popan: cookery pis > apsan: sight v > van: deed, action, act 3) verbal nouns on Vowel+i take thl: i > ithl: go, trip ani > anithl: protection li > lithl: establishment 4) verbal nouns on dhi take l: sdhi > sdhl: seat gwdhi: to pray > gwdhl: prayer 5) verbal nouns on chi take -na tonchi: to swear > tonchna: oath

116 rinchi: to need > rinchna: need, necessity 6) verbal nouns on a take the suffix n or s Some verbal forms on a have attested nouns formed on ancient anto that do not quite correspond with the paradigm proposed here. These forms are fixed and are retained as such. For these verbs secondary abstract nouns can be formed using the suffix (a)s: cra: to love caran: friend > cras: love nma: to dislike naman: enemy > nmas: dislike argha: to shine arghan: silver > arghas: shine This suffix is the same as the one used to make nouns out of adjectives, see that section for discussion. Verbal forms on a for which no attested ancient abstract nouns on anto are known form an abstract noun regularly on an: petha: to ask > pethan: question 7) mono syllabic verbal nouns on i take on gn: to know > gnon: knowledge gl: to obstruct, stick > glon: obstruction fr: to spread (Bhrghros 2012) > fron: spread Verbal adjectives The past participle can be used as a verbal adjective: cra: to love > carthu: loved (Delamarre 2003, p. 107)

117 Further adjectives can be derived with the regular adjectival suffix ach: carthu: loved > carthach: lovely, loveable (Delamarre 2003, p. 107) caran: friend > caranach: friendly nma: to dislike > namthu: disliked > namthach: unlikeable > naman: enemy > namanach: hostile Secondary abstract noun derivation Secondary abstract nouns can be constructed using the suffix as, also used for substantivising adjectives (see that section): caran: friend caranas: friendship naman: enemy namanas: hostility

Summary of verbal word formation paradigm verbal noun verbs on -n, -r, -l, -m men: to think gar: to call gwel: to want dam: to endure verbs on s, -thi, -vi, - pis: to see rethi: to run gavi: to take v: to do verbs on Vowel+i agentive form -ath menath: thinker garath: caller gwelath: wanter damath: endurer -ath apisath: see-er rethath: runner gavath: taker vath: doer -ath abstract noun -u mnu: thought gru: call gwlu: will(power) damu: endurance -an apsan: sight rethan: run gavan: take, taking van: deed -thl

118

ani: to protect verbs on dhi sdhi: to sit verbs on chi tonchi: to swear verbs on gh orgh: to murder verbs on pi popi: to cook verbs on a, noun attested cara: to love verbs on a, noun not att. petha: to ask verbs on mono syllabic -i gn: to know

aniath: protector -ath sedhath: sitter -ath tonchath: swearer -eth orgheth: murderer -il popil: cook -ath carath: lover -ath pethath: asker -ath gnath: knower

anithl: protection -l sedhl: seat -na tonchna: oath, pledge -en orghen: murder -an popan: cooking -s cras: love -n pethan: question -on gnon: knowledge

All verbal nouns ending in Consonant+i lose the final i before the abstract noun ending. Verbal nouns on a lose the final a before the agentive ending. Derivation of verbs from nouns and adjectives It is posited here that the modern Gaulish language can form verbs from nouns by the simple expedient of adding the verbal ending i to a noun or adjective: mr: big mri: to grow (get bigger) sr: long sri: lengthen maru: death mari: to die boru: hot spring bori: to bubble > to boil

119 brau: mill braui: to mill If a word ends in a vowel other than i, the verbal ending i is added to this vowel: bulgha: bag bulghi: to bag (put in a bag) > bulghia mi: I bag 23. Numerals Numerals are reasonably well attested in the ancient Gaulish material. A complete set of ordinals is attested at La Graufesenque (see Delamarre 2003; Lambert 2003), and cardinals can be deduced from indirect attestations (glosses, personal onomastic material, toponomy). While the attestation is imperfect, enough is known to be able to reconstruct a full set of practically useable cardinals and ordinals. The following are attested: cardinals 1: oino 2: 3: tri (treis, tidres) 4: petru (petuar) 5: pimpe (pempe) 6: suex 7: sextan 8: oxtu 9: 10: decan 14: 20: uoconti 30: triconti 100: canton (in Delamarre 2003). From the above attested forms it is possible to reconstruct the ordinal number nine: nametos = ninth na + met + os > nine + ordinal marker + ending > na = nine The ordinal number two can not be reconstructed from the above data. It can however be deduced by comparison with the surviving historical Celtic languages. As such, it is posited here that: ordinals 1st: cintus 2nd: ciallos (allos) 3rd: tritos 4th: petuarios 5th: pimpetos 6th: suexos 7th: sextametos 8th: oxtumetos 9th: nametos 10th: decametos 14th: petrudecametos

120

two: d Furthermore, the number fourteen is attested: petrudecametos = fourteenth > petrudecan = fourteen This permits the conclusion that the numbers from 11 to 19 were formed as follows: petru-decan = four-teen > number+ten It is therefore possible to deduce the formation of all the missing numbers from 11 to 19: oinodecan dadecan tridecan petrudecan pimpdecan suexdecan sextandecan oxtudecan nadecan uoconti It is also possible to observe that starting from the number seventh all ordinals are formed with the suffix metos. Therefore, the ordinal marker for numbers upwards of seven for the modern Gaulish language will be: -metos > -met > -weth after consonants -meth after vowels e.g.: ith: eight > ithweth: eighth n: nine > nmeth: nineth Applying the regular modern Gaulish sound changes to the above data gives the following: cardinals 1: on 2: d 3: tri 4: pethr 5: pimp ordinals 1st:cin 2nd: cal 3rd: trthu 4th: pethar 5th: pimpeth

121 6: swech 7: sith 8: ith 9: n 10:dech 11:onech (ondec- > onec-) 12: ddhech 13: tridhech 14: pethrdhech 15: pimdhech (< pimp+dech: loss of p-) 16: swechdhech 17: sidhech (< sith+dech: loss of th-) 18: idhech (ith+dech: loss of th-) 19: ndhech 20: gwochon (loss of final th after n-) 100: can (as above) 6th: swechu 7th: sithweth 8th: ithweth 9th: nmeth 10th: dechwecth 11th: onechweth 12th: ddhechweth 13th: trdhechweth 14th: pethrdhechweth 15th:pimdhechweth 16th: swechdhechweth 17th: sidhechweth 18th: idhechweth 19th: ndhechweth 20th: gwochonweth 100th: canweth

To construct numbers above twenty, it is proposed here that the modern Gaulish language employ the vigesimal way of counting. This is based on the fact that such a method of counting is widespread among and apparently indigenous to the modern Celtic languages, and that is is also found in the French language, heir to the Gaulish substrate. As such, multiples of twenty are constructed. The numbers d, tri and pethr fuse onto the following word, undergoing word internal spirantisation following the regular rules of compound word formation: gwochon: twenty dachwochon: forty (two-twenty) trichwochon: sixty (three-twenty) petherchwochon: eighty (four-twenty) After twenty or a multiple thereof the numbers 1 through till 19 are added onto the multiple of twenty, until the next multiple is reached, after which it starts again from the start: 20: gwochon 21: gwochon on 22: gwochon d 23: gwochon tri 24: gwochon pethr 25: gwochon pimp 26: gwochon swech 27: gwochon sith 28: gwochon ith 41: dchwochon on 50: dchwochon dech 51: dchwochon onech 59: dchwochon ndhech 60: trichwochon 61: trichwochon on 69: trichwochon n 70: trichwochon dech 71: trichwochon onech

122 29: gwochon n 30: gwochon dech 31: gwochon onech 32: gwochon dadhech 33: gwochon tridhech 34: gwochon pethrdhech 35: gwochon pimdhech 36: gwochon swechdhech 37: gwochon sidhech 38: gwochon idhech 39: gwochon ndhech 40: dchwochon 79: trichwochon ndhech 80: pethrchwochon 81: pethrchwochon on 89: pethrchwochon n 90: pethrchwochon dech 99: pethrchwochon ndhech 100: can 101: can on 110: can dech 139: can gwochon ndhech 200: da can 500: pimp can 1000: mil 2000: d mil

To construct the ordinals of numbers, the suffix weth/-meth is added to the last number of the series: 79: trichwochon ndhech 79th: trichwochon ndhechweth 100th: canweth 2000th: d milweth Cardinal numbers precede the noun, as is attested in the Gaulish corpus: trimarcisia = three horseriders (Delamarre 2003, p. 302) decamnoctiaca = three night feats (Delamarre 2003, p. 137) > tri march: three horses > dech nith: ten nights It is also attested that ordinal numbers follow the noun the refer to: tuos nametos = ninth batch > tus nmeth (Lambert 2003, p. 132) 24. Expressions and turns of phrase Identifying turns of phrase and expressions in the Gaulish corpus is an enterprise that is fraught with danger and incertainty, as very few phrases can be translated reliably enough to identify such a thing as expressions or turns of phrase. Therefore, the following section will consist of educated conjecture, suggesting and proposing phrases that may plausibly be used in the modern Gaulish language.

123 Liking and loving A number of modal verbs can be identified that appear to convey these sentiments: cara: to love (p. 107) ama: to like (derived from namanto- = ne+ama+nto, p. 231) nama: to dislike (p. 231) luvi: to adore (< lubi p. 209) arer: to please/give satisfaction (< arueriiatis p. 56-57) (All page numbers refer to Delamarre 2003). These verbal forms can be used as straightforward transitive verbs: cara mi ti: I love you ama mi ti: I like you nama mi ti: I dislike you luva mi ti: I adore you arwra i mi: it pleases me To express being pleased by a complex subject, the preposition a is used in its conjugated form. It precedes the subject, thus effectively forming a VOS clause: arwra adhim depri esc: pleases to-me eating fish > I like to eat fish arwra adh vi curu: pleases to-her drinking beer > she likes to drink beer The verb arer is used in the expression please: ma harwra i ti: if it pleases you > please (see also further below) The same verb can also be used to express preference: arwra mi eth: it pleases me more > I prefer it arwera mi aneth: it pleases me less > I avoid it The abstract noun arwru can be used to construct the conceptual nouns for the above phrases. They are adjoined with the nouns pen, head and los, tail in noun-noun compounds: arwrupen: head/main/first-pleasure > preference arwrulos: tail/last-pleasure > avoidance

124 Wanting and needing There are a number of verbal forms attested which express the notion of wanting, desiring, needing etc. These are listed in order of strength of expression, as much as such can be derived from their attestation: uel-: scrisu-mi-o uelor = spit-I-that one-wants/is-wanted (Delamarre 2003, p. 312) > that I spit is wanted / one wants that I spit / I want to spit > gwel: to want uei-: beni ueionna in coro bouido = woman wishing into cattle contract suante ueiom-mi = for fancying I wish uiro iono ueo-bie = true [and] just [your] wish may-be (Mees 2010, p. 96-104) > gwi: to wish iant-: onomastic component only, no context. iant- = desire (Delamarre 2003, p. 186) > ianthi: to desire suant-: suante ueiom-mi = for fancying I wish [punishment] suant- is an exact match of Welsh chwant: covet, lust, desire and of Breton choant: fancy, feel like, avoir envie de > suant- appears to be ambiguous and can have negative overtones in Welsh, and conveys wistful longing, which can be but need not be envious (Mees 2010, p. 101) > swanthi: to fancy rinc-: rinci-tuso = need, get the advantage of (Delamarre 2003, p. 258) > rinchi: to need, require e.g.: gwla mi i an drith: I want to go to the beach gwia mi arghan eth: I wish [for] more money iantha mi ben: I desire a woman swantha mi cerdhl in tiern: I fancy the job of the boss rincha mi depri: I need to eat Expression of wish The expression of a wish, intention, desire, blessing or salutation is, in many languages, typically expressed with a subjunctive construction. However, the attested Gaulish material shows indication that the subjunctive mood was being aligned or confused with the future construction, both using the suffix/infix si(e)/o- (see Lambert 2003, p. 65; Delamarre 2003, p. 298).

125 Therefore, it is proposed here that the modern Gaulish language, to give expression to a subjunctive notion, uses the future form of a verb in combination with the relative particle o, which translates as that. Parallels for such a construction can be found in other Celtic languages (Irish and Scottish Gaelic) as well as in Romance (e.g. French) and Germanic languages (English). e.g.: o b ti len: that you will be happy > may you be happy och uris su tanch: that you-pl. will find peace > may you-pl. find peace To construct sentences which in English would be formed with lets, the imperative is used: i ni a nam: lets go swimming! cluthi ni in chwras: lets hear the truth! Uses of to have The expression for to have, attested at Banassac (tiei ulano celicnu, Delamarre 2003, p. 323) can be used to express a wide variety of notions: 1) a physical attribute, permanent or non-permanent: mi-esi derch bi: I have blue eyes mi-esi gwolth sr: I have long hair 2) a physical state, permanent or non-permanent mi-esi trieth bris: I have a broken foot mi-esi d coch: I have two legs mi-esi panthpen: I have a headache mi-esi oghr: I have cold > I am cold mi-esi tes: I have warm(th) > I am warm mi-esi nanu: I have hunger > I am hungry (< Lezoux L-66, Delamarre 2003, p. 340) mi-esi onu: I have thirst > I am thirsty (< Banassac, Delamarre 2003, p. 242) mi-esi achas: I have speed > I am in a hurry 3) a state of possession, physical or intellectual mi-esi cuchul: I have a hat/ hood mi-esi ti: I have a house mi-esi mnu: I have an idea (a thought) mi-esi ulanu: I have satisfaction > I am satisfied (Banassac, Delamarre 2003, p. 323) mi-esi in bes: I have the habit > I am used to

126 mi-esi gwras: I have truth > I am right 4) a state of desire, intention or wistful longing mi-esi san a hi an ti: I feel like / fancy going home mi-esi san a gan: I feel like singing The verb to have with a specific subject The question of how to use the form for to have with a subject other than a personal pronoun is a tricky one. In the phrase tiei ulano celicnu of Banassac (Delamarre 2003, p. 323), ulano [celicnu] is the subject of the verb ei and ti is the indirect object of the phrase, the recipient. The phrase transliterates as: to/ at-you is the satisfaction ..., i.e. the satisfaction is to-you. Therefore the indirect object, which is in the accusative case, precedes the verb, which is in turn followed by the subject. This gives a sentence structure of O-V-S: tiei ulano celicnu OV S This structure is not uncommon in the Gaulish corpus: eso ieuri rigani rosmertiac = this dedicated the Queen and Rosmerta O V S (Lezoux, Delamarre 2003, p. 334) se tigi prino Ascanius = this contract buys Ascanius O V S (Rze, Stifter 2012) scrisu-mi-o uelor = that I spit one wants/ is wanted O V(S) It is clear that an object can be preposed to a verb if required. Therefore, it is posited here that when a specific, complex or multiple entity is the recipient of the form to have, this entity, which is is the indirect object of the phrase, precedes the form to have. The personal pronoun featured in the form to have will agree with the number and gender expressed in the object phrase: -esi: he has (to-him is) boch mr: a big mouth -esi boch mr: he has a big mouth in cun: the dog in cun -esi boch mr: the dog to-him is a big mouth > the dog he has a big mouth > the dog has a big mouth

127

in wn-s: those women in wn-s s-esi coch sr: those women they have long legs > those women have long legs Gwina: a female name Gwina -esi gwolth duv: Gwina has black hair Subordinate clauses can thus be the recipient of the form to have: in don o n chwla can: the people who dont want to sing panthpen: headache in don o n chwla can s-esi panthpen: the people who dont want to sing have a headache in don o s-esi panthpen: the people who have a headache ethn o gwla can: a bird that wants to sing in don o s-esi panthpen s-esi ethn o gwla can: the people who have a headache have a bird that wants to sing Expression of obligation Something that ought to be done can expressed as would be right/just/correct for someone to do, making use of the attested form cert-, translated as just, right, correct: certh: just, right, correct an ti: to/ at the house > home > r v certh riem i an ti: it would be just/right for me to go home > I should go home Expression of occurence and presence The notion of there is can be translated by the verbal form esi, without pronoun or subject: naus: boats esi naus ar in mr: there are boats on the sea (boats are on the sea) aman: time, weather druch: bad dii: to come esi aman dhruch en dhi: there is bad weather coming (bad weather is coming) n hesi neveth a hv: there is nothing to do (not is nothing to do)

128 esi aval ar in pren: there are apples on the tree (apples are on the tree) Reflexivity A reflexive pronoun sue may tentatively be identified in the Gaulish corpus: suebreto = by ones own judgement > sue = self (Meid in Delamarre 2003, p. 285) It needs to be pointed out that the term sue also occurs at three reprises at Chateaubleau, where it is interpreted as the 2nd pers. pl. personal pronoun you (Mees 2010, p. 99). Nevertheless, this pronoun is attested at Chamalieres as suis (Delamarre 2003, p. 337). It is posited here that for the purposes of the modern Gaulish language the form suis is used as the personal pronoun you (pl.) and the form sue is used for reflexive constructions. The latter has an obvious cognate in Romance sui- (sui-cide = self-killing, Fr. soi-meme = ones self). Therefore in the modern Gaulish language reflexive forms will be constructed with the pronoun sue, in conjunction with the relevant personal pronoun and suffixed to it. It does not change for gender or number. apsa mi mi-se: I see myself moltha ch-se: he praises himself lautha ch-se: she washes herself esi in gwir en chii -se: the man is cursing himself The same form is used to construct emphatic statements: avthu mi ch mi-se: I did it myself apis s ch s-se: you (pl.) have seen it yourselves Basic phrases of courtesy A number of basic phrases of courtesy and simple conversation can tentatively be constructed, using either attested or semi-attested forms (e.g. brathu) or using the attested material in a manner consistent with authentic syntactic characteristics (e.g. adjective following head) or with general pan-Celtic and/ or Western European standards and/ or conventions (e.g. d wath). All words are found in Delmarre (2003) unless otherwise indicated. d wath: goodday methin dhi: good morning swedh dhi: good afternoon nesnoith di: good evening noith di: good night

129

brathu: thanks (< bratou decantem, Delamarre 2003, p. 85) m harwra ti: please (if it pleases you) iach: well, healthy, sane, fit iachas: wellbeing/ health sln: healthy slnas: health > iachas dhi: good health > cheers, drinking toast > slnas: health > goodbye swel: good wind > welcome podh a hesi ti: how are you esi mi in dhi: I am well (with adverbial particle in) mi ceth: me too, me as well nep: any, some, neither > n hva dias nep: it doesnt do any good > esi duvr nep en in ban: there is some water in the cup > n hesi on nep al: it is neither one or the other rithu: right, law, entitlement > mi-esi in rithu: I have the right certh: right, correct > esi mi certh: I am right (also mi-esi gwras) ancherth: wrong > esi mi ancherth: I am wrong sru: fault > mi-esi sru: I have fault > I apologise > esi m shru: id. li: to leave, let be, let off (< [prinni] laget, Delmarre 2003, p. 253) > li mi nach: leave me alone 25. Cultural background This section aims to provide a small amount of cultural background for the modern Gaulish language. While this is not strictly dealing with grammatical aspects of language, language is nonetheless the vehicle through which a peoples cosmology and ontology are expressed. As such, it is considered justifiable to provide such information as can be gleaned from historic sources, and to shape it in such a way that it provides an appropriate setting in which the language can be situated. Gaulish calendar The Gaulish calendar, refered to as the Coligny Calendar after the place where a nearcomplete version was unearthed in 1897, is the source of much heated debate, bitter

130 argument and endless controversy. The calendar is a lunar one, following the lunar year, but it aims to achieve agreement and balance between the lunar year and the solar year. The model presented here is the only one to date that achieves stunning astronomic accuracy of an unparalleled degree and is distinguished by its high degree of relevance and plausibility. It is the work and intellectual property of Helen McKay (2013) and is reproduced here with explicit permission. The translation of the names of the months suggested here is one based on a combination of the etymologies of the names of the months, of the climatic conditions prevalent in Western Europe in the proposed corresponding modern months, and of historically attested cultural information. A modest attempt is made to match it up against 21st century astronomical considerations in an effort to find approximate matches with present day reality.The calendar is approached from a linguistic, cultural, historical and geophysical point of view, and what is proposed here is what is considered to be a plausible interpretation. Substantiation for the tentative translation of the names of the months can be found in Delamarre (2003) and draws on contributions from Gwinn, McKay and Cassell (2011-12). Many of the names of the months have several alternative translations. The ones given here were selected based on feasibility, plausibility, probability and overall best fit. There is no official academic consensus on when the Gaulish year started and finished, and opinions and theories vary wildly and are hotly contested. The solution proposed here was arrived at as a result of extensive collaboration and consultation between Helen McKay, Michael Cassell, Bhghros, Bellouesos Isarnos and myself (AtedugyonYektisKeltika 2012-13), with explicit reference to Delamarre (2003) and Lambert (2003). As such, The interpretation of the Coligny Calendar presented here starts the year around May-June, a month before the summer solstice of 21 June. There is no historical support for this, as there is for no other possible starting date. However, taking account of the meaning of the names of the calendar and of their potential relevance to seasonal, climatic and cultural conditions, it is considered reasonable to do so. Support for this theory is provided by Gwinn (2010) and McKay (2013), although significant opposition to this notion can be found with e.g. Cassell (2010-13). It is suggested in the historical record that time reckoning started on the 6th day of the moon. However, it is unclear exactly which phase of the moon that refers to. While it is historically recorded that time was reckoned by nights rather than days and that these started at nightfall, no indication is recorded pertaining the start of seasons and years. In view of the above, and in view of the fact that this calendar goes to great lengths to achieve agreement and harmony between the lunar year and the solar year, the notion is espoused here that months were reckoned in lunations, as is obvious from the calendar itself, and that the four solar events of the year (the solstices and equinoxes) were significant and were taken into account. Though no specific information regarding the use of solar dates has been recorded or has been handed down to this day, the fact that the calendar marries the solar year with the lunar year is here taken to provide sufficient justification for the accordance of relevance to the solar events.

131

Translation of the names of the months The names of the months found on the Coligny Calendar are listed here, with their ranking and their rendition into modern Gaulish: number month 1 2 3 4 5 6 7 8 9 10 11 12 ancient Gaulish name Samonios Dumanios Riurios Anagantio(s) Ogronios Cutios Giamonios Simiuisonna Equos Elembiuos Edrini Cantlos modern Gaulish name Samon Duman Rur Anian Oghron Cuth Giamon Simison chu Elem dhrin Canthl

The translation and interpretation of these names is uncertain, fraught with controversy and the subject of much bitter debate. There is no widespread consensus among experts in the field. The following interpretations are offered here after significant debate and consultation. Primary linguistic analysis is according to Delamarre (2003) and Lambert (2003), with additional interpretations provided by Gwinn (2010), Cassell (2010-13), Bhrghros (2010-2013), and Bellouesus Isarnos (2010-2013). Some of the persuasive arguments in favour of these interpretations and their figurations in time are provided as follows: 1) Samonios: summer All the surviving Celtic languages make reference to summer months in their calendars. It is proposed here that this month heralds the start of summer, and is situated around MayJune. 2) Dumanios: smoke, fumigations It has been suggested that the concept of smoke might be refering to the tradition of lighting bonfires at midsummer night, a custom which is widespread across western and northern Europe and which is thought to be of great antiquity. This supports the notion of Duman containing the summer solstice and fits with a placing of Dumanios around JuneJuly. 3) Riurios: harvest

132 It is suggested that this month is associated with the god Lugus because of the notations deuor lug riuri found accompanying it on the calendar. The word riur has also been interpreted as containing the name rivo-, which some claim is a synonym of Lugus. The traditional Irish calendar features a festival in honour of Lugus, Lunasa, and a month named after it (August). Alternatively, the word riur has been analysed as refering to times of plenty, the fat of the land. All of these interpretations support an analysis of Riur as a harvest month, which can therefore plausibly be situated around July-August. 4) Anagantios: ablutions, washing This is one of several possible translations suggested. A concept of cleaning up, washing, tidying up, can be linked with preparations for winter time, putting affairs in order while it is still possible to do such things, before the winter weather makes it hard or impossible. An August-September timeframe might suit this notion. 5) Ogronios: cold Oghron is one of the very few month names that expert opinion is unanimously in support of a single translation of: it is widely accepted that Oghron refers to cold. Suggested as being situated around September-October, this month would include the autumn equinox. 6) Cutios: difficulty? Based on an etymology linking Cutio with modern Breton kudenn, it can tentatively be proposed here that Cutios may refer to difficulty or problems. If considered a month of difficulty it could fit in with an October-November slot, refering to a time of inclement weather and difficult living conditions. However, support for this notion has not as yet been found. 7) Giamonios: winter It is universally accepted by experts in the field that giamonos refers to winter. At six months remove from Samonios, this month can be considered to introduce a winter half of the year, just as Samonios introduces a summer half of the year, thus dividing the year into two 6-month semesters. Support for this suggested bipartite division is found in the structure of the months themselves, which manifestly consist of two halves of fifteen or fourteen days. A proposed November-December timeslot appears suitable for this month. 8) Simiuisonna: half-sun It is proposed here that this word is analysed as a compound of simi-, half and sonna, sun, translating roughly as half sun, weak sun. This admittedly does not account for the syllable ui-. However, a description of a half-sun or weak sun suits

133 this months proposed situation around December-January perfectly, and could as such be interpreted as refering to the winter solstice: the point of the year when the sun is at its lowest and weakest point. 9) Equos: animal foot Equos is a bone of contention due to its apparent featuring of a preserved inherited IE labio-velar, which in Gaulish regularly became [p] historically at least 700 years before the estimated date of the Coligny calendar. The etymology of Equos/ Echu adopted here is proposed by Bhrghros (2013): equos < *pekupos, animal foot, with regular Celtic loss of ()p- and maintenance of syllabic u- after c-, i.e. not a labiovelar. It is tentatively suggested here that a reference to animal tracks may be intended, possibly in the context of snow covering the land, rendering the land dead, sterile, unyielding. This appears reasonable for a month following the winter solstice, being situated around January-February. 10) Elembiuos: deer It is widely accepted that this word refers to deer. A context of deer hunting, or, conversely, of the end of deer hunting may be imagined. In modern France, the hunting seasons of all kinds of native deer (red deer, roe deer, chamois etc.) close in February. While this is obviously a modern arrangement, it is entirely possible and likely that this happens in accordance with deer behaviour, breeding patterns and ecological factors which may not have changed significantly since the time of the calendar. This provides a close match for the proposed timeslot of February-March for this month. There is an equivalent similarly named Greek month which introduces the start of the deer hunting season.

11) Edrini: flame, fire The suggestion that this word contains a reference to fire, flames or warmth is reasonably widely accepted. Since it is suggested that this month is situated around March-April, it can be imagined that it announces the start of the warmer weather. This month would also include the spring equinox, another event traditionally associated with bonfires. 12) Cantlos A reference to song or music is commonly agreed upon for the name of this month. It has been suggested that it refers to the presence of birdsong in the natural world, which provides a match with the timeslot proposed for this month, April-May.

134 The following table summarises these translations and compares them with realistic conditions and/ or practices at corresponding times of year that are proposed and adopted here. CGM stands for Corresponding Gregorian Months. # 1 2 3 4 5 6 7 8 9 10 name of meaning of month name Samon summer Duman Riur Anaian Oghron Cuth smoke plenty, fat, reference to god Lugus ablutions, washing cold difficulty CGM May-JuneJune-July July-August AugustSeptember SeptemberOctober OctoberNovember NovemberDecember DecemberJanuary JanuaryFebruary FebruaryMarch MarchApril April-May reality check start of summer yes solstice celebrations, bonfires at midsummer night - possible harvest time, time of plenty of food, festival of Lunasa (of god Lugh) - yes time to wash and clean before winter, preparation for winter, activitywinding down - possible start of cold season, autumn equinox yes time when life starts to get difficult, inclement weather yes start of winter yes sun at lowest and weakest point of year, winter solstice yes animal tracks visible in snow, land frozen, covered in snow, time of scarcity possible time when the deer move around the forest to mate, start of new activity, end of deer hunting season, start of new life possible start of the warm season, spring equinox, bonfires yes time of birdsong in the natural world, celebration of life yes

Giamon winter Simison half-sun, weak sun Echu animal tracks Elem deer

11 12

dhrin Canthl

warmth, fire, singing, song, music

The above table indicates that the proposed schedule is feasible, possible and reasonable. In the spirit of the Gaulish habit of naming things in opposite pairs also evident in the onomastic material (see Delamarre 2003, p. 351-352), it is possible to tentatively propose a grouping of the names of the months in opposing pairs, and a divsion of the year in a summer half and a winter half. In this model, each of the months finds its perfect opposite in the exact same spot in the opposite half of the year, as is illustrated in the following table.

135 Summer Half month name meaning Samon start of summer Duman smoke and fire Rur harvest, times of plenty, yield Anian finishing activity, winding down Oghron start of cold Cuth time of difficulty Winter Half month name meaning Giamon start of winter Simison weak, watery sun chu snow cover, times of shortage, sterility Elem renewed activity, starting up dhrin start of warmth Canthl time of celebration

# 1 2 3 4 5 6

# 1 2 3 4 5 6

While the above interpretation is certainly possible and feasible, it is important to bear in mind that it is only one interpretation and makes no claim to being anything other than a reasonably educated guess. In summary, the year and its months can be described as follows: Summer half of year: Samon: May-June, start of summer Duman: June-July, smoke (bonfires), solstice 21 June Riur: July-August, harvest, month of plenty, month of Lugus/Riuos Anian: August-September, ablutions, cleaning up Oghron: September-October, start of cold, equinox 23 September Cuth: October-November, start of difficulty, inclement weather Winter half of year: Giamon: November-December, start of winter Simison: December-January, half-sun, weak-sun, solstice 22 december Echu: January-February, animal tracks [in the snow] Elem: February-March, month of deer Edhrin: March-April, flame, start of warmth, bonfires, equinox 21 March Canthl: April-May, singing, music, birdsong Astronomical analysis The Gaulish calendar ran over a period of five years, during which time intercalary months were inserted to match the solar year to the lunar year. The year varied in number of days over the five years. The beginnings and ends of the various months oscillated to and fro to the extent that can be seen below. The length of months varied between 29 and 30 days, varied from year to year, and were designated at either favourable (mat, math in modern Gaulish) or unfavourable (anmat, anwath in modern Gaulish). The brilliance of Helen McKay (2013) lies in the fact that by adjusting the length of the months of Echu, which are not clearly recorded, to a length that is in keeping with the

136 rest of the months, she has produced a calendar model that achieves perfect balance and agreement between the lunar year and the solar year. As such, it is proposed that the Coligny calendar is a Metonic calendar, lining up the lunar year with the solar year every 19 years. This is achieved by adjusting the length of Echu as appropriate (see model) and by repeating three cycles of five years complemented with a fourth cycle from which the first year is deleted. In this way the lunar and solar year are perfectly aligned every 19 years. While there is a certain amount of speculation regarding the deletion of cycles and the adjustment of the length of Echu as required, the bronze plaque that constituted the calendar was fitted out with a complex series of peg holes. It is thought that these pegholes served the exact purpose of indicating these adjustments. Furthermore, the bronze calendar plaque under discussion would have been able to function as a perpetual calendar: thanks to the peghole-adjustment system, it would have been possible to keep it in synch with the real world perpetually. As such, it is without a doubt the most sophisticated and advanced calendar of the Classical world, and is a monumental testimony to the knowledge, achievement, understanding and organisation of the Gaulish people. For a detailed discussion of the workings of the calendar, see appendix The Coligny Calendar. Below is the layout of the Coligny Calendar (reproduced with permission). The names of the months are given in the original ancient Gaulish, as per Helen McKays work.

Metonic cycle of the Coligny Calendar

137 The following table shows the configuration with EQUOS values as 30/29/29/29/30, and shows the full Metonic cycle as laid out in the Coligny calendar by using 4 cycles end to end, but omitting the very first year:

In year I and V the month of Echu has 30 days In year II, III and IIII Echu has 29 days Months that are math (favourable): Samon, Rur, Oghron, Cuth, Simison, Edhrin, Echu intercalary 1 Months that are anwath (unfavourable): Duman, Anian, Giamon, Elepi, Canthl, Echu intercalary 2 The extent to which this model was able to keep in synch with real time is shown in the following table, showing the fraction of days by which the timekeeping deviated from true. Note that at the second intercalary the difference is exactly one day. This model allowed the Gaulish people to keep track of time accurately and, over a timespan of 19 years, never be out more than one day, which ultimately got caught up after 19 years.

138

While it is posited here that the calendars time reckoning starts around May-June, this is in actual fact irrelevant: the time reckoning could start anytime, and the model would still hold true. No truer, more accurate calendar or system of time keeping has been recorded anywhere within antiquity or pre-modern times, and it was not until the scientific advances of the 20th century that the Gregorian calendar was able to come up to the standard set by the genius of the Gaulish people. Pantheon of Gods and Goddesses It is generally agreed upon that there was no such thing as an organised Gaulish Pantheon, or, that if there was, that information about it has not been accurately recorded. What follows here is a synthesis, interpretation and manipulation of authentic recorded information (see Delamarre 2003; Griffin 2007), brought together in a conceptual framework that while it aspires to be plausible is nonetheless completely arbitrary. Gods and Goddesses are organised in 4 groups of three triads. The first group reunites the phenomenological deities, the second one the deities governing the natural world, and the third one those deities concerned with the activities of everyday human life. The final group deals with the relationship between humanity and the universe through supplication. It is generally felt that the gods and goddesses in the first group are in a position of hegemony over those in the lower levels, with the exception of those in the last triad, who instead are overlords over sections of the Pantheon and, by inference, over the other gods and goddesses. This lends a circular aspect to the web of relationships. I. Gods and Goddesses of the phenomenological world 1) Sky: Gran, Lord of the Air Thunder: Taran, Lord of Sound Lightning: Belen, Lord of Fire

139 2) Sun: Li, Lord of Plant Growth Moon: Belisama, Lady of the Tides Stars: Sirna, Lady of Time 3) Life: Cernun, Lord of Animal Growth Death: Rosmertha, Lady of Abundance Fertility: Damna, Lady of Prosperity II. Gods and Goddesses of the Natural World 1) Mountains: Garghan Oceans: Anchauna Land: Ltha 2) Rivers: Nansueltha Lakes: Borv Forests: Ardhwna 3) Fish: Divna Mammals: Epna Birds: Taru Tri Garan III. Gods and Goddesses of human activity 1) Horticulture: Avel Farming: Suchel Hunting: Avnva 2) Metalwork: Govan Crafts: Berechinia Trade: su 3) War: Camul Warriors: Anartha Victory: Seghw IV. The Gods and Goddesses of humanity, supplication and the universe 1) Men: Tthath Women: Mathrna Children: Mapon 2) Healing: Avtha Learning: Nemethna Language: Oghw

140 3) World of the Gods: Alvrich World of the Living: Bithrich World of the Dead: Duvrich The latter triad refers to the vertical structure of the universe. It is conceived as a Tree of Life (Pren Bith) that has three levels. The upper level is the Land of the Gods (Bri Alvi), the bright and shiny world of the sky, represented by the crown of the tree. The middle level is the Land of the Living (Bri Bith), the organic world of the surface of the Earth, above ground, represented by the trunk of the tree. The lower level is the Land of the Dead (Bri Duv), the underworld, the deep and dark world under the ground, represented by the roots of the tree. Alvrich presides over Bri Alvi, Bithrich presides over Bri Bith, and Duvrich presides over Bri Duv. Logically, these are therefore the overlords of all the other Gods and Goddesses mentioned above, each in their own domain. To reach the underworld, the dead have to complete a threefold journey across all three levels of the universe. On each leg of the journey, they are accompanied by a different God or Goddess whose job it is to see them safely through to the next stage. 1. They need to be taken across the sky in a lunar chariot by Oghw, the talker, full of wind 2. They need to be accompanied overland by Epna, the horse, the beast of burden of overland transport 3. They need to be ferried across an underground expanse of water by Suchel, the farmer, who digs down into the earth for a living In addition to this three-tiered cosmology however, there is Bri in Iinch, the Land of the Young. The Land of the Young is a paradise-like place that exists on the middle level, alongside of the land of the Living, but invisible and inaccessible to it. In the Land of the Young, time is suspended: there is no decay, only enjoyment. Entry can only be gained through magic, and often only through specific geographic locations, where temporary passageways between the worlds can be opened: hilltops, cavemouths, stonecircles; or at certain times of the year (e.g. the turn of the new year).

26. Conclusion As this document represents the reinvention of the Gaulish language, driven into extinction by acts of military aggression, imperialism and genocide, it seems fitting to conclude with a translation both into hypothetical Late Ancient Gaulish and the new modern Gaulish of Article One of the Universal Declaration of Human Rights: Esent doni olli gentui rii ac cotamici en balcassu ac rextoues. Siesesent condo ac conuissu, ac re bisiet certo rissies adauuo canti doni alli en menman braterassa. Esi don ol genthu ru ach cothamich en valchas ach rith. S-esi conu ach

141 conchwisu, ach r v certh ries adhv can dhon al en wnu brathras. All human beings are born free and equal in dignity and rights. They are endowed with reason and conscience and should act towards one another in a spirit of brotherhood. The invasion of an independent sovereign nation by an aggressive, expansionist, imperialistic state bent on conquest for reasons of political and personal profit, led to the murder of hundreds of thousands of men, women and children, and the exile into slavery of hundreds of thousands more. It resulted in the obliteration of an erstwhile autonomous peoples culture, identity and language. Language can be revived. People cant. Lest we forget. Steve Gwiru Mrghnath Hansen, Australia April 2013

142 III. References/ Bibliography Gaulish Albin, Jacques (2011) La langue Gauloise la tuile de Chteaubleau Academia.edu http://www.academia.edu/1747376/La_langue_gauloise_-_la_tuile_de_Chateaubleau Anderson, Carl (2006-2013), personal communications. Arbre Celtique, http://www.arbre-celtique.com/encyclopedie/langue-gauloise-2.htm AtedugyonYektisKeltika Yahoo Group online (2011-2013), discussion forum. Ball, Martin J. & Fife, James (1993) The Celtic Languages. http://books.google.com/book?Ball+Fife+Celtic+Language Bauer, Brigitte L. M. (1996) Language loss in Gaul: socio-historical and linguistic factors in language conflict, Southwest Journal of Linguistics, v. 15, no. 1-2, pp. 2344. Bellouesus Isarnos (2006-2013), personal communications, http://bellovesos.multiply.com/ Bhrghros (2006-2013), personal communications. Cassell, Michael (2006-2013), personal communications Celticaconlang Yahoo Group online (2006-2011), discussion forum Continentalceltic Yahoo Group Online: Bernard Mees, David Stifter, Chris Gwinn, and many others (2006-2013) Coskun, Altay & Zeidler, Jurgen (2003) Covernames and Nomenclature in Late Roman Gaul: The Evidence of the Bordelaise Poet Ausonius, http://www.books.google.com.au Delamarre, Xavier (2003) Dictionnaire de la Langue Gauloise, Editions Errance, Paris. Eska, Joseph (2008) Continental Celtic, in The Cambridge Encyclopedia of the Worlds Ancient Languages, pp. 857-78. Finsen, Lars (2006-2013), personal communications. French-English Dictionnary Online, http://www.french-linguistics.co.uk/dictionary/

143

French-English Dictionnary Online, http://www.wordreference.com/fren/ Fournet, Arnaud (2011) About the ethnolinguistics of Gaulish people: the case for a Kartic substrate, The Macro-Comparative Journal, vol. 2, No. 1. Glozelic, http://www.glozel.net/shortHistory.html Gray, Louis H. (1944) Mutation in Gaulish, Language, v. 20, no. 4, pp. 223-30. Hubert, Henry; Dobie, Maryat R.; Mauss, Marcel; Lantier, Raymond & Marx, Jean (1934-2002) The Rise of the Celts. http://www.books.google.com/books? id=henry+hubert+the+rise+of+the+celts&source=web-etc. Hubschmied, J. H. (1938) Sprachliche Zeugen fur das Spate Aussterben des Gallischen, Vox Romanica, III, pp. 48-155. Isaac, Graham R. (2007) Celtic and Afro-Asiatic, in The Celtic Languages in Contact, http://opus.kobv.de/ubp/volltexte/2007/1568/ Koch, John (2006) Celtic Culture: A Historical Encyclopedia, http://www.books.google.com.au Lambert, Pierre-Yves (2003) La Langue Gauloise, Editions Errance, Paris. McKay, Helen (2006-2013), personal communications. Mees, Bernard (2008) The women of Larzac, KF 3, pp. 169-88. Mees, Bernard (2010) Words from the well at Gallo-Roman Chteaubleau, pp. 89-106. Mees, Bernard (2011) Wackernagels, pro-drop and verb-second in syntax in Continental Celtic, Academia.edu. http://www.academia.edu/1701275/Wackernagels_law_pro-drop_and_verbsecond_syntax_in_Continental_Celtic Mullen, Alex (2007) Evidence for written Celtic from Roman Britain: a linguistic analysis of Tabellae Sulis 14 and 18, Studia Celtica, XLI, pp. 31-45. Piqueron, Olivier (2006) Yextis Keltika, Yahoo Celtica Conlang Files, http://tech.groups.yahoo.com/group/celticaconlang/ (note: file access for members only) Stifter, David (2000) The Chateaubleau Tiles, https://listserv.heanet.ie/cgi-bin/wa? A2=ind0006&L=oldirish-I&O=D&P=12796

144

Stifter, David (2008) Old Celtic languages, http://www.univie.ac/at/indogermanistik/ Stifter, David (2009) Notes on Chteaubleau (L-93), KF, pp. 229-244. Stifter, David (2012) New Gaulish inscriptions, http://www.rootsofeurope.ku.dk/.../New_Gaulish_Inscriptions_Copenhagen_p... Stifter, David (2012) Gallo-Latin inscriptions, Old Celtic Languages, Spring 2012. http://www.rootsofeurope.ku.dk/kalender/arkiv.../Gallish_2_WS_2010.pdf/ The Gaulish Language http://en.wikipedia.org/wiki/Gaulish_language The Gaulish verb to be, http://www.angelfire.com/me/ik/tobes.html Watkins, Calvert (1955) The phonemics of Gaulish The Dialect of Narbonensis, Language, v.31, no. 1, pp. 9-19. Woodard, Roger (2008) The ancient languages of Europe. http://books.google.com.au Welsh A Welsh Course, http://www.cs.cf.ac.uk/fun/welsh/ Borsley, R. D., Tallermn, M. O. & Wills, D. W. E. (n.d.) Historical syntax, in The Syntax of Welsh, http://people.ds.cam.ac.uk/dwew2/diachrony.pdf Borsley, R. D., Tallermn, M. O. & Wills, D. W. E. (2007) The Syntax of Welsh, http://books.google.com.au/books Celtic Literature Collective (n.d.) Welsh texts (Old and Middle QWelsh), http://www.maryjones.us/ctexts/index_welsh.html Falileyev, Alexander (2000) Etymological glossary of Old Welsh, http://npu.edu.ua/!ebook/book/djvu/A/iif_kgpm_Etymological_Glossary_of_Old_Welsh.pdf Gareth King (2003) A Comprehensive Grammar of Modern Welsh, http://books.google.com.au Glosbe English-Welsh dictionary, http://glosbe.com/en/cy/ Middle Welsh Grammar, http://www.mit.edu/~dfm/canol/contents.html Schafer, Roland (2012) A grammatical sketch of Middle Welsh, http://hpsg.fuberlin.de/~rsling/downloads/pubs/Schaefer_AGrammaticalSketchOfMiddleWelsh.pdf

145

SiaradCymraeg.com, http://www.siaradcymraeg.com/intro.html Some Basic Rules of Welsh Grammar, http://www.bbc.co.uk/wales/learnwelsh/pdf/welshgrammar_allrules.pdf Welsh/English/Welsh dictionnary, http://www.geiriadur.net/ Welsh grammar, http://www.bbc.co.uk/wales/learnwelsh/pdfwelshgrammar_all_rules.pdf Wikipedia for Welsh, Breton, Irish and Cornish grammar, http://wikipedia.org/wiki/Welsh_language/&/Breton_language/&/Irish_language/&/Corni sh_language Willis, David (n.d.) Old and Middle Welsh, http://people.ds.cam.ac.uk/dwew2/diachrony.pdf Kevin J. Rottet and Rex A. Sprouse; Tag Questions in Welsh, 2008; John Benjamins Publishing Company; http://benjamins.com/series/dia/25-1/art/03rot.pdf Breton Breton http://www.agencebretagnepresse.com/dico/dico.cgi/ Breton Language, http://fr.wikipedia.org/wiki/Projet:Langues/Breton Breton Verbal Inflection, http://johannes.heinecke.free.fr/dict/brezhoneg/verbs.html Brezhoneg Daou, http://en.wikiversity.org/wiki/Brezhoneg_Dao%C3%B9/Lesson_3 Dico Geriadur: Breton Dictionary, http://www.agencebretagnepresse.com/cgi-bin/dico.cgi Freelang Breton-English dictionary, http://www.freelang.net/online/breton.php?lg=gb Geriadur.com: Dictionnaire Evolu Francais-Breton, http://www.geriadur.com/ Glosbe English-Breton dictionary, http://glosbe.com/en/br/if Hewitt, Steve (n.d.) Breton, a tale of two conditionals, http://www.academia.edu/785418/Breton_A_tale_of_two_conditionals Kervaker Breton Lessons Online, http://www.kervaker.org/en/lessons_01_toc.html Laurent, Loeic [n.d.] De lapparition du language au Breton, forme moderne du

146 Gaulois, http://www.agencebretagnepresse.com/fetch.php?id=8079 Preder Embannaduriou: Geriadur ar Brezhoneg, http://www.preder.net/klask.php (Breton Dictionnary) Press, Ian (1985) A Grammar of Modern Breton, http://books.google.com.au/ The Breton verb besan, to be, http://www.smo.uhi.ac.uk/saoghal/miouchanain/brezhoneg/ The direct object personal pronouns in Lorient area Breton, http://www.academia.edu/964814/The_Direct_Object_Personal_Pronouns_in_Lorient_A rea_Breton Timm, Lenora A (1990) Some observations on the syntax of the Breton verbal noun, in Celtic Linguistics / Ieithyddiaeth Geltaidd: Readings in the Brythonic Languages, eds. Ball, Fife, Poppe and Rowland. http://books.google.com.au/books University of Toronto, History of the Breton Language. http://www.utm.utoronto.ca/~w3cevans/mst2039/HistoryofBretonlge.pdf Cornish Agan Tavas, Our Language, http://www.agantavas.org.uk/ Cornish Grammar, http://www.cornishgrammar.com/?n=Verbs.Home Cornish Lessons, http://www.bbc.co.uk/cornwall/connected/stories/new_cornwall_language.html Cussel an Tavas Kernuak, http://home.btconnect.com/htm_cornwall/ Cussel an Tavas Kernuak, Cornish grammar, http://www.moderncornish.net/beginnersnotes/section-3.html English-Cornish dictionary, http://home.btconnect.com/htm_cornwall/articles%20translations/mamgerlever.pdf Jenner, Henry (1904) A Handbook of the Cornish Language, http://www.aolib.com/reader_26192_78.htm Kernewek Dre Lyther, http://www.kd/cornish.freeserve.co.uk/firstgradecourse/kdl16.pdf Norris, Edwin (1859) A Scetch of Cornish Grammar, http://books.google.com.au

147

Omniglot, http://www.omniglot/writing/cornish.htm Warlinnen, the Cornish Language Fellowship Online, http://www.cornishlanguage.org/English/Dictionnary.asp Williams, G. P. (1908) The pre-verbal particle re in Cornish, PhD thesis, http://www.archive.org/stream/preverbalparticl100willrich/preverbalparticl100willrich_dj vu.txt Irish An Chrannog: Focloir Bearla-Gaeilge, http://www.crannog.ie/ttt.htm (Irish dictionnary) Basic Outline of the Grammar of Irish for Unreconstructed Linguists http://fiosfeasa.com/bearla/language/grammar1.htm Comparison Irish - Scottish Grammar, http://www.geocities.com/ecosse_mon_coeur De Bernardo-Stempel, Patricia [n.d.] Old Irish, http://www.utexas.edu/cola/centers/Irc/eieol/iriol=TC-X.html eDIL, electronic dictionary of the Irish language, http://www.dil.ie/ Gaelic Dictionnaries, http://www.ceantar.org/Dicts/search.html (Irish, Scottish & Manx) Gaelic Lessons, http://www.maths.tcd.ie/gaeilge/lessons.html Gordon, Randall Clark (2012) Derivational morphology of the early Irish verbal noun, Proquest, http://gradworks.umi.com/34/97/3497395.html Gramadach na Gaeilge Irish Grammar, http://nualeargais.ie/gnag/gram.htm Irish Dictionnary, http://www.englishirishdictionnary.com/dictionnary Irish Dictionary Online: http://www.irishdictionary.ie/home Irish Grammar www.daltai.com/grammar Irish Initial Consonant Mutations, http://www.answers.com/topic/irish-initial-mutations Irish, www.smo.uni.ai.uk/gaidhlig/faclair/sbg/lorg.php Learn Irish, Eurotalk Interactive, CD-ROM

148 Learn Irish Gaelic, http://www.erinsweb.com/gaelic19.html Word Initial Mutations in Celtic Languages, http://everything2.com/index Scottish Gaelic Robertson, Boyd & Taylor, Iain (1993) Teach Yourself Gaelic, A Complete Course for Beginners, Hoddor, London (book & cassette) Stewart, Thomas (2004) Mutation as morphology: bases, stems and shapes in Scottish Gaelic (PhD thesis), www.ohiolink.edu/etd/send-pdf.cgi?osu1086046888 General Alphabetizer, http://alphabetizer.flap.tv/index.php Ball, James & King, Gareth (1993) The Celtic Languages, in The Handbook of Morphology, eds Andrew Spencer and Arnold M. Zwicky, http://books.google.com.au Chao Li (2004) On verbal nouns in Celtic languages, Proceeding of the Harvard Celtic Colloquium, eds. Jones, Alberro, Bempechat. http://books.google.com.au/books Cimbri http://www.davidkfaux.org/Cimbri-Chronology.pdf Clitics, www.wordforms.net/Clitic/encyclopedia.htm Conjugation of verbs, http://www.allverbs.com/verbtable.php?langid=1949&verb=lenn (Breton/Welsh/Cornish/Irish/General) Continuous and progressive aspects - Wikipedia English-Proto Celtic Word List, University of Wales, http://www.wales.ac.uk/Resources/Documents/Research/CelticLanguages/EnglishProtoC elticWordList.pdf Formation of Future Tenses, http://en.wikipedia.org/wiki/Future_tense Green, Anthony Djubach (n.d.) The Independence of Phonology and Morphology: the Celtic Mutations, http://roa.rutgers.edu/files/652-0404/652-GREEN-O-O.PDF Hewitt, Steve (2007) Remarks on the Insular Celtic Hamito-Semitic question, Academia.edu, http://www.academia.edu/283231/Remarks_on_the_Insular_Celtic_HamitoSemitic_question

149

Hopper, P. J. & Closi-Traugett, E. (1993) Grammaticalisation, Textbook on Linguistics, Cambridge University Press, http://www.21stcenturyfogey.com/language/linguisticcomplexity.htm Indo-European Copula, Wikipedia (2007), http://en.wikipedia.org/wiki/IndoEuropean_copula Indo-European s-mobile, http://en.wikipedia.org/wiki/Indo-European_s-mobile Macauley, Donald (1992) The Celtic Languages. http://books.google.com.au Matasovic, Ranko, Etymological Lexicon of Proto Celtic, http://www.scribd.com/doc/61167870/An-Etymological-Lexicon-of-Proto-Celtic Matasovic, Ranko (2011) Problems in the reconstruction of Proto Celtic, http://exadmin.matita.net/uploads/pagine/644703631_handouts_pavia_matasovic.pdf Pokorny Root Index, http://homepage.ntlworld.com/richard.wordingham/pok/pok_index.htm ProtoIndoeuropean Pronouns and Particles, http://en.wikipedia.org/wiki/Proto-IndoEuropean_pronouns_and_particles Proto Indo European Verbs Wikipedia. Pyatt, Elisabeth J. (n.d.) Relativised Domains in the Celtic Languages, http://www.personal.psu.edu/ejpb/Pyattpll26.pdf Ronan, Patricia (2006) Aspects of verbal noun constructions in medieval Irish and Welsh, with reference to similar constructions in Basque, NUI Maynooth, http://eprints.nuim.ie/3519/1/VNThesis.pdf Russell, Paul (1990) Celtic wordformation: the velar suffixes. http://books.google.com.au Sala, M. & Vintila-Radulescu, L. (1984) Les Langues Du Monde, Petit Encyclopdie, in Dark Ages Observations. http://www.darkages.espadana-walker.com/spip.php?article7 University of Wales, http://www.wales.ac.uk/resources/documents/moe-pcl.pdf (ProtoCeltic Database) Venneman, Theo (2003) Languages in Prehistoric Europe North of the Alps, http://www.scribd.com/doc/8670/Languages-in-prehistoric-Europe-north-of-the-Alps Watkins, Calvert (1962) Indo-European origins of the Celtic verb: the sigmatic aorist.

150 http://books.google.com.au Willem De Reuse (2009), research fellow in linguistics, University of Texas, personal communications.

151 Appendix I: Galataca Sena Galataca Sena is an unintentional and accidental spin-off product of the development of the modern Gaulish language. It was devised in response to calls from interested parties to write modern Gaulish texts in a way that would be intelligible to those with general background knowledge of the ancient language, but not of the modern version. Therefore, Galataca Sena was conceived as a hypothetical point somewhere between ancient Gaulish and modern Gaulish, but much closer to the former. As such, it represents a stage of the language after the erosion and collapse of the case system and after the onset of initial consonant mutations, but before the modernising sound changes. The hypothetical timeframe for Galataca Sena would be approximately the 6th - 7th century CE, a century after ancient Gaulish was last attested in writing. It is convenient here to sketch a broad outline of the evolution of the language and to situate the various stages in it: I. Ancient Gaulish: 1). Early ancient Gaulish: appr. 7th century BCE-2nd century BCE. Centres around the Lepontic and Greco-Gaulish inscriptions. 2). Middle ancient Gaulish: 2nd century BCE 2nd century CE: the age of Vercingetorix, Ambiorix and the invasion 3). Late ancient Gaulish: 2nd century CE 6th century CE: the language of Chateaubleau, gradual erosion of case system and start of lossof final syllables II. Old Gaulish: Galataca Sena, which translates literally as Old Gaulish. 6th 7th century CE. The state of the language after the loss of the case system, but before the sound changes. It is a hypothetical state, not attested, but the onset of which can be discerned in the language of Chateaubleau. The stage prior to the extinction of the language. Estimates for the language vary from the 8th century, time of the enforcement of christianity as state religion by Charlemagne and the erosion and loss of the foothold of the language among the rural population of Gaul a a result (a notion considered plausible by e.g. Mees 2010), to the 12th century. The notion that Gaulish survived in the high Alps between Helvetia and Gaul (Haute-Savoie) is supported by some, e.g. Bhrghros (2012). III. Modern Gaulish The revived language of the 21st century. Subject of this document.

Brief Outline of Galataca Sena

152

Without going into too much detail, GS is basically a preliminary version of Galthach hAthevu which features the same syntax, but which has retained 1) open vowel word endings which do no longer determine case 2). fully conjugated verbal paradigms 3). pre-sound change phonology 4). mutations which are only marked by an apostrophe Verbal conjugation follows a thematic pattern, as opposed to the athematic pattern which became the default position in GhA: to carry, present tense: beru mi beres ti beret e/i/id beremu ni berete suis berent sies passive: beror All tenses are derived exactly the same way as in GhA, but retain the present tense endings throughout: Past Tense re beru mi re beres ti re beret e/i/id re beremu ni re berete suis re berent sies past participle: bertu Future Tense bersiu mi bersies ti bersiet e/i/id bersiemu ni bersiete suis bersient sies

to be:

153

present: emi es ti esi e/i/id esemu ni esete suis esent sies

past bue mi bues ti buet e/i/id buemu ni buete suis buent sies

future bie mi bies ti biet e/i biemu ni biete suis bient sies

past participle: etu And so on. In terms of morphology, masculine words end in o and feminine words end in a. There are some attested as ending in i in ancient Gaulish, these retain the i ending. Adjectives follow the nouns and are declined for gender both with an apostrophe at the start of the word to indicate mutation which is not written, and at the word final syllable which takes o or a as required. All phonology is as per pre-sound change. Pural is marked as per nominative plural, but without the final s if there was one. Verbal nouns are conceived as neuter nouns and end on on, -ion for stems on stops. Example text: Galataca Sena: Esi id uiro o ne esent doni elui lavarentiou inda Galataca inte dago, extero esemu ni en cerdion ad auuion sinde uer dago. Galthach hAthevu: Esi gwr o n hesi don lu en lhavar in Galthach in dhi, ithr esi ni en gerdhi a hv sin gwer dhi. English: It is true that there are not a lot of people who speak Gaulish well, but we are working to make it better. Galataca Sena is only ever used as a companion language to GhA to provide an easily understandable bridge between the old and the new.

Appendix II: the Coligny Calendar by Helen McKay (2013)

154

Reproduced with permission of the author.

THE COLIGNY CALENDAR AS A LUNAR CALENDAR The pattern of calendar months and their lengths, as proposed on this page.

IS THE COLIGNY CALENDAR A LUNAR CALENDAR?

All the evidence points to the expectation that the Coligny calendar is a complex and fine-tuned lunar calendar. While we can still argue about how the solar years were aligned through periods longer than the 5 years of the bronze plaque, the one thing that is without question is that the calendar concentrates on the lunar cycle, and that its internal notations are aligned both between months of the lunar year, and between days of the months across the years. The daily notations of each month must have significance in relationship to the lunar phase in which they occur. This is supported by masses of evidence, both in history and folklore, that the Celtic (and European) peoples held the moon in great awe, and used its various phases to regulate their own earthly activities. It makes no sense that a lunar calendar with complex and detailed notations focused on certain points of the lunar month should be allowed to get out of sync with the actual moons phases. Certainly other lunar calendars of the time were not very precise it is true, and needed constant adjusting, but the Coligny is different in its complex web of daily notations which make best sense as tied to a point in the lunar phase. Given the long history of the western European peoples watching and measuring the sun and moon since Neolithic times, we should expect the calendar to keep in sync with the lunar cycle with great precision.

HOW BEST TO SYNC THE CALENDAR AND LUNAR MONTHS

155
Here we shall examine the 5 year cycle on the calendar plaque that we have, to see how its calendar months fit within the same 62 synodic months of the moon. The calendar is organized into 5 years of 12 lunar months each, with 2 intercalary months, the first at the lead into year 1, and the second half way through year 3. This gives a total of 62 lunar months in 5 calendar years.

One synodic lunar cycle Lunar months on the calendar Total number of days in 62 synodic months

29.53 days 62 1830.94 days

In other words, if the 5 year period of the calendar is to stay in sync with the moons phases overall it must contain 1831 days, the closest possible day count to 1830.94.

Strangely, this count of 1831 days has given most scholars a headache, because there has been difficulty in fitting this count to the calendar.
If we are to presume though that the calendar does hold the correct 1831 days, then because the calendar is constrained to count in whole days, in each 5 year cycle it will fall behind the moon by 0.06 days, (1831 1830.94), and, if the calendar is simply repeated every 5 years, this would require a day to be removed from the calendar every 83 years, which as we will see below can easily be done by adjusting the length of the variable month EQUOS in one year when required. However, if the Coligny plaque is part of a Metonic cycle calendar, then it will stay in sync almost indefinitely - one day in 219 years - as the difference between sun and moon after19 years is a matter of a couple of hours, which is less than the possible variation in a lunation more on this later.

THE STRANGE CASE OF EQUOS We know with reasonable certainty the length of most calendar months. But EQUOS presents a problem as it shows an exceptional pattern to the normal MAT months of 30 days and ANM months of 29 days. Firstly, EQUOS is marked as ANM, the notation which usually dictates 29 days, and in its patterns of daily notations EQUOS does indeed act as an ANM month. However, in years 1 and 5 we can see that EQUOS is 30 days long. Year 3 can also be confidently said to contain 30 days in EQUOS, as the script attached to the second intercalary month which occurs in the middle of the third year, states that this year contains 13 months and 385 (CCCLXXXV) days. Most scholars have accepted that this means that EQUOS in years 2 and 4 is shorter, probably 28 days long in year 2, and 29 days long in year 4. Year 4 seems to require 29 days to keep its average months reasonably in step with the lunar phase length. There is also an inconsistency in the length of the IVOS run in year 2, which some have proposed is because EQUOS is shorter in that year. This however doesnt seem to be the case. [LINK] However, accepting these proposed values of EQUOS as 30/28/30/29/30, gives us a total count of 1832 days, not 1831 days. We seem to have a day too many. This run over is for example taken as the jumping off point by Olmsted for his theory that the calendar was allowed to go out of sync with both moon and sun, and that this then required extra computation to track.

Zavaroni, on the other hand, looks at the individual lunations, and maintains EQUOS with 30 days across all months in order for it to always contain a full lunation. For full details, please see his publication.
I recall here the reason of my thesis according to which Equos has always 30 days. Even if one does not consider the yearly lunation pattern in Table 4 from which one may

156 infer that Equos lunation starts 35.2 hours before the first day of the month the Equos lunation would commonly start more than 24 hours before the end of Simivisonna because it follows a semester with 178 days and a month with 30 days which cause a large advance (2-08 29.53x7 = 1.29 days). Hence the Equos lunation is not wholly contained within the month because it always starts at least 24 hours before the beginning of the first solar day of the month. This systematic being out of the beginning of Equos would be the reason why it was considered anmat. But, there is another month which is in question, INTERCALARY ONE. There is very little present in the fragments of this intercalary month, including the bottom section. If we were to say that INTERCALARY ONE was 29 days long (rather than the 30 days which has been generally accepted up till now), then we have a perfect lunar calendar in sync with the actual lunar cycles. And this 1831 day calendar is what is presented in the table above. INTERCALARY ONE with 29 DAYS So why has Intercalary One been thought to have 30 days, rather than the necessary 29 days? There are a couple of reasons. The first is that INTERCALARY TWO does indeed have 30 days, and it is easy to assume the same for both intercalary months. The second is that line 3 of the top fragment of INTERCALARY ONE starts with MATV (then the fragment breaks), which has been taken to be an extension of the M(AT) of other months. But, because all the days of an intercalary month are either copied or dragged from other months, it has no internal MD/D notations of its own to provide supporting evidence as to whether or not the month is being used as a MAT month. Even should this be true, as we see from EQUOS, the naming of a month as MAT or ANM doesnt necessarily dictate its length under some circumstances. On the final Intercalary One fragment, there is a break line running after the last X of the list of day numbers in Roman numerals, and only one other X visible on the fragment above it. Some have claimed that there is the downstroke of a V just visible after the X, giving XV (day 15a, i.e. the 30th day). This may be so, but it is very difficult to assert with any degree of certainty. Even if this turns out to be correct, the day XV may have just been marked as the usual DIVERTOMV which occurs on the virtual 30th day of AMN months of 29 real days length. Usually an XV is not marked on the 30th day with DIVERTOMV, but sometimes it is, for example ANAGANTIOS year 4 reads XV DIVERTOMV. So, even should this turn out to be XV for day 30, this does not guarantee that Intercalary One has 30 real days. So, is there any reason to consider that INTERCALARY ONE instead has only 29 days apart from the necessity of keeping the calendar in sync with the moon? The answer is simple: yes. Each intercalary day is copied from a particular month of another year, and should INTERCALARY ONE have a 30th day, then this day would need to be copied from DAY 30 of CANTLOS. But, CANTLOS is an ANM month with only 29 days there is no day there to be copied. Therefore it isnt possible for INTERCALARY ONE to have a day 30.

157 Another reason why Intercalary One should only hold 29 days, is that if it held 30 days, then by the end of the first SAMONIOS half of year 1, there would be 5 months of 30 days and only two months of 29 days. At which point the calendar month would be already hopelessly out of sync with the lunations by the amount of nearly one and a half days. And this misalignment would continue to pass down through the rest of the 5 year cycle. In other words, if Intercalary One had 30 days, it would throw the calendars months significantly out of sync from the very start of the 5 year cycle. On the other hand, 29 days keeps the calendar in perfect sync for the 5 years.

COMPARISON OF THE EFFECT of 29 or 30 days in INTERCALARY ONE in the FIRST YEAR

INTERCALARY ONE AND THE METONIC CYCLE If the Coligny calendar is part of a larger cycle of years, then the Metonic cycle is a likely candidate. The Metonic cycle is 19 solar years long and 235 synodic months, a total of 6940 days. If we take 3 cycles of the Coligny calendar, 15 years, this contains 5493 days, leaving 1447 days for the remaining 4 years, and a year 13 lunar months to drop. This is exactly the first year of the calendar that we now have an intercalary of 29 days plus 12 lunar months of 355 days. The amazing thing about this is that the Coligny plaque can be used for all four cycles of a metonic cycle, without ever replacing it!

158 Below is one way this can be arranged, by dropping the entire first year, including its 29-day intercalary, and then continuing to cycle through the next 19 years. THE COLIGNY PLAQUE AS A METONIC CYCLE Each box below represents the 5-year Coligny plaque. If the first year of 12 lunar months plus a 29-day intercalary is dropped (highlighted in yellow), then the 4 cycles of the 5-year calendar create a perfect example of a Metonic calendar.

To summarise the case for INTERCALARY ONE being 29 days, rather than 30: 1) This brings the 5 year cycle into sync with the 1831 days required by 62 lunations, and the calendar keeps in perfect time with the moon over a long period. 2) If the extra day was removed instead from EQUOS year 4, that would cause year 4 to have lunations significantly too short 3) We have no evidence that INTERCALARY ONE holds 30 days 4) There is no day 30 in CANTLOS which the intercalary month can borrow to form its days notation at day 30. 5) A month of 29 days is needed to balance the run of months in the first half of year 1 which would otherwise hold 5 of the first 7 months as 30 day MAT months. A length of 30 days for the intercalary would push the calendar months out of sync immediately with respect to lunations. 6) An intercalary of 29 days is needed if the calendar is part of a Metonic cycle calendar. If Intercalary One has 29 days, then by ignoring the first year in the first cycle to give 4 years, followed by three 5-year cycles, we have a perfect Metonic cycle calendar, and a bronze plaque which displays all 19 years and never needs replacing by a different version. In other words, the

159 Coligny calendar provides us with the complete calendar of the Gauls, not just a 5-year section as has been previously thought.

SYNCING INDIVIDUAL CALENDAR MONTHS WITH LUNATIONS

So now we have a 5 year lunar calendar that has the correct number of days to keep it in alignment with 62 synodic moons. But, this is not good enough just by itself. If we are to use the internal notations on the calendar as indicative of a certain phase or day of the moon, then we must also keep each individual month as closely aligned to the synodic month as possible. That is, the lengths of months in the 5 year cycle must show a pattern that is quite rigidly controlled, so that each calendar month starts as closely as possible to the start of a lunation. We do have a very good indication of what the tolerance of this syncing might be. If we accept that day 1 of a month is the quarter moon, then the triplets 7-8-9 are the days around the full moon, while the triplets 7a-8a-9a are the three days of the lost moon. This gives us a wiggle room of approximately 1.5 days either side of the actual point of the full moon (midway through day 8) in which it is acceptable for the full moon to appear. Or expressed another way, 1 day either side of day 8. If the full moon appeared outside this range then the dragging notations of days 7-8-9 would no longer refer to the full moon phase and presumably the activities normally associated with this triplet would no longer be appropriate. The reason for the calendar months to be aligned to the moons phases is purely to keep daily notations at a certain point in the moons phase, in particular to keep the triplet 7-8-9 at the full moon. It has no impact at all on the pattern of each notation, as these are determined by patterns that take no concern with the number of days a month has. For instance, D AMB occurs on days 5 and 11, regardless of how many days a month holds. The IVOS and TII counters also show us that all months are considered to have 30 days, even if the DIVERTOMV day of 29-day months is a virtual day used only for pattern purposes. CHOSING THE BEST FITTING EQUOS VALUES FOR YEARS 2 and 4 The following table shows the way the calendar months track with the synodic months. The first table shows the situation where EQUOS in years 2 and 4 are 28 and 29, while the second table shows when years 2 and 4 are 29 and 28. An interesting point to note about the first table is that it gives an exact value (difference 0) half way through the 5 year cycle at Intercalary Two.

160

EQUOS 30/28/30/29/30

min -1.41 max 0.91

EQUOS 30/29/30/28/30 min -0.69 max 1.63

Although it may be difficult to see from the above tables, the first table where EQUOS has the values 30/28/30/29/30 tracks considerably closer to the lunar cycle. The table below is a graph of the first table above, it simply gives a more graphic expression to the variation of the differences.

EQUOS 30/28/30/29/30

min -1.41 max 0.91

161

EQUOS 30/28/30/29/30

min -1.41 max 0.91 (sorted)

The following table is the same as above, with values sorted. Here we can see how the difference values waver around the zero (central) value, but are generally within the expected variation allowed of 1.5 days either side. In fact most values are within 0.5 days either side. Note also that the values are reasonably equally divided either side of the average value (0.00).

In contrast to the table above, the next table shows how the EQUOS 30/29/30/28/30 is within acceptable limits on the negative side, but pushing past the acceptable value of 1.5 days on the positive side. But perhaps more significantly the values are skewed into the positive side, illustrating that the EQUOS values here are not as close a fit to the month as the 30/28/30/29/30 above.

EQUOS 30/29/30/28/30

min -0.69 max 1.63 (sorted)

162

In summary, we can see that with INTERCALARY ONE of 29 days length, and with EQUOS of 30/28/30/29/30, that the Coligny calendar presents a pattern of calendar months which closely follows the 62 synodic lunar months of the 5 year cycle, maintaining itself within a day either side of the exact lunar phase. The Coligny calendar is not just a lunar calendar, but a lunar calendar which is designed to track the moon so closely that it is generally well within a days tolerance of the actual lunation at the beginning of each calendar month. Which makes it one of the most extraordinary calendars of its time, without a doubt. THE UNEVEN PATTERN OF MAT And ANM MONTHS The SAMONIOS half of the year holds 4 MAT (30 day) months and 2 ANM (29 day) months. The GIAMONIOS half holds the opposite, 4 ANM months and 2 MAT months, the exception being to the length of EQUOS, as weve seen above. On the surface, this is a strange arrangement. On a lunar calendar, it seems more straightforward to arrange months in 30/29 pairs, so that they average out to a length of 29.5 days, close to the lunar length of 29.53. However, such a simple arrangement would see the calendar drift out of sync by nearly 2 days over the 5 year period. The other difficulty is that 1831 days cannot be equally divided into two halves. So that one half of the calendar will contain one day less than the other. In this case, the first half of the calendar holds 915 days and the second half 916 days. And this difficulty will also push the calendar further out of sync. The fascinating thing here is that the Gauls have managed to slightly rearrange the months and their lengths, so that the two halves of the 5 year cycle each stay in sync with their 31 lunar months (2.5 years), to the point that GIAMONIOS in year 3 is exactly in line with Intercalary One. They have done this in part by pushing an extra MAT (30) month CUTIOS up into the SAMONIOS half of the year.

163 This extra MAT month in the SAMONIOS half has other advantages, one being that it further complicates the seeming randomness of the calendars notations to the casual observer. And it provides more MAT (good, auspicious) days and notations in the summer half of the year when most social activities are taking place. If we were to need proof that the Gauls were superb astronomers, mathematicians, and philosophers, and all three together, then this is possibly it. CHECKING THE FULL MOON TOLERANCE All the above tables are calculated using the end of a full calendar month. But there is one more thing to check before we can be sure that the above solution is correct, we must check whether the triplet 7-8-9 always contains the full moon. Because we cant tell at what precise point in time the calendar is aligned exactly to the moon to begin its first year with, I have allocated the very first day 8 of Intercalary One as zero. By doing this, we can again watch how the values vary around this point. The table below shows that the picture presented by the full moon of day 8 is not too different from the previous table showing the ends of each month, and that the full moon will always occur within the triplet of days 7-8-9.

FULL MOON (day 8) TRACKING using EQUOS 30/28/30/29/30

ZAVARONI and LUNATIONS There are two factors that we should also look at more closely before accepting the above pattern of calendar months. The first has already been mentioned: that the 5 year calendar, if recycled constantly, slips out of sync by 0.06 days, requiring a day to be deleted roughly every 83 years. If on the other hand, if it is part of a Metonic cycle, it slips out of sync by one day in every 219 years. The other factor is one that has been looked at in some detail by Zavaroni, the fact that a lunation itself is not a steady value but oscillates around the average lunation value used above, that of 29.530589 days, or 29 days 12 hours 44 minutes and 3 seconds. Actual lunations vary from this average figure in a complex pattern which is dictated by the

164 eccentric orbits of the moon and earth. Although most lunations fall within a few hours of the average value, an extreme lunation can be up to nearly 15 hours on either side of the average. That is, 0.6 of a day. This chart is taken from http://individual.utoronto.ca/kalendis/lunar/index.htm#trend Each dot on the curve is the value of an individual lunation as it oscillates around the average value of 29.53.

There is no easily definable pattern of lunations, each lunation in history is today given a certain number by different cultures, and has its own unique value. What it means for this discussion, is that every so often in cycles of about 111 months the the length of a lunation might be pushed out by up to nearly 15 hours either side of the average value. If we return to the full moon bar chart (last above), we can see that the full moon would still occur easily within the positive side of the lunation, but if an extreme lunation of -15 hours occurred with one of the 4 most negative values, then the full moon would actually occur just over 2 days previously, which is nearly half a day outside our tolerance limits. This would be a very rare occurrence, but it would still happen. However, if this occurred, would the people be able to tell the difference between a full moon occurring within half a day? I dont think this is possible using the naked eye. It is often impossible to tell the difference between two days around the full moon. But, practiced astronomers would presumably be able to discern the difference. Or, would the Gaulish astronomers be able to predict these rare negative extremes and adjust the calendar accordingly? These negative extremes occur when EQUOS is given 28 days in year 2, so it would be relatively easy to make a small adjustment to the EQUOS pattern to avoid these moments, by giving EQUOS 29 days in year 2 and removing a day from an adjoining year, either year 1 or year 3. But all this is assuming that the Gauls could calculate and predict, and Im not sure that this was possible, or if possible, appropriate for a lunar calendar to be concerned with to that minute level. The main reason they might have been concerned to this level, is that it is crucial to know when the moon is at its full or dark point in order to help predict potential eclipses. So my suspicion is that the Gauls would certainly have known how to calculate lunations. But is this the role of the Coligny calendar? I suspect not. I think that the calendar must track the lunar phases closely so that the activities and omens indicated by the daily notations made sense, but it is not designed to provide astronomers with complex calculations. These calculations have been done elsewhere, and the calendar created

165 from them, not the reverse (as Olmsted and Zavaroni seem to infer). I need also point out, that while Zavaroni understands the calendar to be focusing on the lunar cycle, he also believes that the daily notations are concerned with lunations, rather than omens, which is the main point of difference between our views. Still, it would not be a good thing if an eclipse happened on an extreme lunation, and it fell outside the eclipse triplets of full moon 7-8-9 and dark moon 7a-8a-9a. VARIABLE EQUOS ANOTHER IDEA And, just when the story all seems to be coming together, something else entirely pops up. Lets say that the city of Melbourne has commissioned me to create a massive, expensive, very complex, modern calendar on a huge bronze plaque, for public display at the town centre. Of course, there is one small problem, if I were to make it with February having 28 days, then I would have to pull it down every fourth year and make a second hugely expensive calendar with the extra day of February 29 to put in its place. Well, that would be unacceptable. So what would I do? I would create the plaque with 29 days in February, and because it is a PEG calendar, if the year actually only has 28 days, then Ill place a special peg in the hole of the 29th day to indicate that it isnt being used this year. So, what if the Gaulish artisans did exactly the same thing? Would it be so surprising if the bronze plaque shows EQUOS with 30 days carved across all years? This would then allow the astronomers to mark certain instances of EQUOS with the 30th day to ignore, simply by putting a special peg in its hole. And this pattern of when EQUOS was either 29 or 30 could then be adjusted according to how the actual lunatations of that year are moving. For example, if a lunation is predicted to be in the extreme negative end, then allowing that years EQUOS to be 29 days long would make for a very simple and flexible system of adjustment. So, even though we can see that years 1, 3, and 5 on the calendar are carved with 30 days, it may not necessarily be the case that these months were normally considered to contain 30 days, but could have been reduced to 29 days. And it also takes care of something else which has never felt right that if EQUOS really did have 30 days in years 1, 3 and 5, then this requires EQUOS year 2 to only have 28 days. Which in a calendar that otherwise moves to adjust itself through a variation of 29 and 30 days, seems really very odd and uncomfortable indeed. But now we are free to allocate two 30 day months and three 29 day months to EQUOS in whatever pattern is required across the five years. As weve seen above, the 5 year calendar as we have on the bronze plaque of Coligny is in general a finely-tuned lunar calendar. But, rarely, but every so often, it will require EQUOS to gain or lose a day, in order to keep it closely in sync with the variable lunations or with the unavoidable slippage of one day every 83 (or 219) years. If the Gauls were creating a new calendar every 5 years this potential adjustment might be taken care of each cycle when deemed necessary. But there is a problem with that. The Coligny calendar shows distinct signs of being copied from a much earlier calendar. Mistakes such as TRINO for PRINO show that the carvers no longer fully understood these notations. There are also parallex mistakes where the correct notation has been miscopied onto the wrong line, and so on, again suggesting that the Coligny calendar is a copy of a far older one.

166 But now it seems possible for the Gauls to traditionally have had just the one calendar (as we do), but with the flexibility built into it to readjust the values of EQUOS as the lunations fall, by the simple device of allowing EQUOS to be carved with 30 days on all years, and a special peg to put on EQUOS day 30 when it wasnt needed Should this be the case, then a lot of what is said above will need to be rethought. For instance, despite EQUOS showing 30 days in year 1, it might have effectively only been allowed 29 in most years. We could be looking at EQUOS values of 29/29/30/29/30 for example. This would eliminate the most negative differences in the graphs above when EQUOS is given the unique and strange value of only 28 days. Note that while the pattern of whether EQUOS has 29 or 30 days is flexible, it stills means that Intercalary One is best with 29 days, and that EQUOS needs to have the same total of days over the five years. I have now looked at how best the values of EQUOS would sit around an average lunation, and compared it to the values settled on above, 30/28/30/29/30 (series one). It turns out that there are only two likely candidate patterns that are better and that track the moon closely enough to work with the average lunation value. Those values are 30/29/29/29/30 (series two) and 29/30/29/29/30 (series three). EQUOS PATTERN FROM TO TOTAL RANGE Series 1 30/28/30/29/30 -1.41 0.91 2.32 Series 2 30/29/29/29/30 -0.69 1.00 1.69 Series 3 29/30/29/29/30 -1.03 1.00 2.03 The two graphs below compare the differences of series 2 and 3 respectively to our original series 1. COMPARING LUNAR DIFFERENCES Series1 EQUOS 30/28/30/29/30 Series2 EQUOS 30/29/29/29/30

167

This pattern of EQUOS oscillates the closest around the average lunation value. It has much the same spread in the positive, but shaves approximately 0.7 days off the negative. This has the advantage of allowing for the extreme lunation values in either positive or negative values without exceeding the tolerance of a total of three days.

COMPARING LUNAR DIFFERENCES Series1 EQUOS 30/28/30/29/30 Series3 EQUOS 29/30/29/29/30

168

This pattern of 30/29/29/29/30 also is a tighter fit to the average lunations. Again it shaves off those troublesome large negative values which are created with EQUOS of 28 days in year 2. The negative side of Series 3 is not as close a fit as the pattern of Series 2 above, but it is still comfortably within the range which will allow the extreme lunation values to be added without exceeding the tolerance value of 3 days. There is little to chose between these two series of EQUOS values, although Series 2 EQUOS 30/29/29/29/30 is strictly speaking the best fit. Which means, in the end, I still cant tell exactly what the usual EQUOS pattern would be. But in assuming that the plaque is carved with 30 days in EQUOS in all years, a wonderful flexibility is built into the calendar by which it can be finely adjusted according to current lunation patterns. And it means that the bronze plaque could well be the only physical calendar that the Gauls ever needed, as it can be adjusted on the fly, simply by inserting a peg. There is one significant reason, though, to choose Series 2 as the norm. It is that by having the first year of the calendar start with an intercalary of 29 days, and EQUOS of 30 days, this means that we find ourselves with a METONIC CYCLE inbuilt into the calendar. The 19-year Metonic cycle would run starting in year 2 for the rest of the 4 years of the cycle, then through three more full cycles of 5 years. And if the metonic cycle was used by the Gauls as a larger cycle of time, then they could also use the same bronze plaque for all years, simply by marking the first year to be ignored during the 4 year cycle of the 19 year cycle. [LINK here to SOLAR] So I end with another possible version of the shape of the calendar, incorporating Series 2. After all this, there is only one difference with the shape that I started with - EQUOS years 2 and 3 are now both 29 days long, rather than 28 and 30 respectively.

169

And I will now repeat here, with EQUOS as 30/29/29/29/30, the full Metonic cycle as laid out in the Coligny calendar by using 4 cycles end to end, but ignoring the very first year.

170 METONIC CYCLE of the COLIGNY CALENDAR

171 Appendix III: Alternative Gaulish History The following is a pastiche of what an alternative history of Western Europe might have looked like, creating an environment in which the Gaulish language might have survived and flourished. 52 BCE: Uercingetorix (Gwerchingethrich) wins the battle of Alesia and the war. The Romans are decimated at the battle of Alesia, and Caesar is captured. The tribes of Gaul unite and set up a federation of nations, which also includes the people of Britain and Ireland, that brings into being and maintains a common armed force based on voluntary contributions alone and operates as a professional unit. The natural borders of Gaul are reinforced and provided with armed forces to keep safe; Gaul is defined as the area between the Rhine, the Alps, the Mediterranean, the Pyrenees and the Atlantic. The Federation of Gaulish, British and Irish Tribes (commonly referred to as FGBIT; in English that is) hold a tribunal which is attended by all people who wish to do so, and which is administrated by druids. Caesar is tried fairly under the native laws of the land, and is found to be a murdering sadist genocidal bastard. A Roman delegation is invited to witness the tribunal and the verdict. Caesar is executed in some satisfyingly horrific fashion. The Romans are told to piss off and don't come back. They bugger off and retreat behind the Alps, where they spend the next 400 years fighting, murdering, backstabbing and poisoning each other merrily with not a care in the world. The Germanic people fidgeting behind the Rhine, being pushed in the back by the movements of miscellaneous horseriding invaders from the East, find they can not get across the Rhine and past the armed forces of FGBIT. Instead they turn south and follow the Rhine's east bank in through Eastern Switserland and down into Italy. They spend a certain amount of time happily murdering, poisoning, massacring and, at a stretch, back stabbing each other as well as various Romans. They catch a boat from Sicily and make it to Northern Africa; some poke up back across from Gibraltar and manage to stay in Andalusia long enough to give it its name. They rule Spain and Italy, but not Gaul or Britain. The Frisians, Angles and Jutes are shoved back into the North Sea and are told to stay away if they know what's good for them, and Vortigern was drowned at birth after a seeress looked into the innards of a freshly sacrificed chicken and saw, not surprisingly, a bloody mess. His parents weren't too fussed, because he used to wake up ten times a night and scream their ears off, so that was all right. And they had another 12 kids anyway, so it didn't really matter. Gaul, Britain and Ireland were left alone to develop their own distinctive culture. Their languages continued to flourish and develop into the instruments of great learning and understanding they were in ancient times, continuing a tradition tentatively recorded in Gaul and featuring such things as the development of the investigative inclinations of the druidic mob of people, having yearly congregations in the area of Chartres to discuss

172 science, philosophy and the conspicuous consumption of beer (back in the real world, prior to the 16th century the vernacular language in which the single biggest body of medical literature was written, outside of Latin and Greek, was Irish). As an aside I think it would probably be useful to include that the library of Alexandria was never burned down, continues to flourish to this day as a centre of great learning, connected to the university of Alexandria, and that christianity as a religion never got any further than an insignificant mystical nonsense-sect shared by no more than a handful of stoned paranoid fanatics crawling around the Judean deserts, eating mushrooms and cockroaches, and spending altogether far too much time in the midday sun without a hat and sunscreen. Their incoherent ramblings and ravings were commonly tolerated and indulged by their country people, who, out of the kindness of their hearts, would give them food and water sometimes and bring their kids around to watch them scratch around in the dirt trying to find their own arse, just for a bit of mild entertainment. Islam, on the other hand, never got off the ground at all. On the day Mohammed was supposed to flee to Medina, he slept in because he had a late night the night before, indulging himself in sex with underage camels and partaking a bit too heavily of datewine. Therefore he forgot to set his alarm, and instead of fleeing to Medina he got arrested for his long term refusal to pay the outstanding parking fines for his camels. As a result he spent the rest of his days tied to a yoke and walking around in circles pulling buckets of water out of desert wells, with as his only company a donkey, a mule and a camel who, to Mohammed's deep and everlasting regret, steadfastly refused his sexual advances. As his was a mind naturally given to esoteric and mystical religious contemplation, spending every waking hour walking around in circles and dragging his feet through the dirt without ever getting anywhere suited him perfectly, and he lived out his days in ecstacy, dying peacefully at the ripe old age of 93 1/2, in his bed surrounded by all his favourite camels, and one goat, because his sexual appetites had become somewhat deviant in his old age. As a result, untold millions of people missed out on being butchered, slaughtered, tortured, murdered, subjugated, indoctrinated, exploited, enslaved, castrated, burned and dehumanised. In spite of this obvious handicap, the whole of the middle east, northern Africa and Eastern Europe passed though a 2000 year period of prosperity, cooperation and tolerance, which, I am happy to say, continues unabatedly to this day. All thanks to Vercingetorix. Good on him, I say.