Page 1 sur 1

Etude comparative du mot temps/durée en Celtique :*amstera

MessagePosté: Jeu 10 Mai, 2018 11:43
de Matrix
Bonjour
en pièce jointe en format pdf et en Anglais, une étude sur l'origine du mot "temps/durée/Intervalle de temps" en celtique, néo-celtique et comparaison avec d'autres langues

Proto-Celtique : am-stera
Gaulois : amman
Brittonique : Moyen Breton Ampser > Breton Amser/Amzer. Cornique Anser. Gallois Amser.
Gaélique : Vieil Irlandais Amm/Aimser > Irlandais Aimsir. Manx Emshyr

j'ai tenté de joindre le fichier mais il est trop gros je fais un copié/collé

http://www.arbre-celtique.com/encyclope ... e-7194.htm


DOI: https://doi.org/10.3318/ERIU.2017.67.4
Ériu LXVII (2017) 1–8 Royal Irish Academy
VARIUM
The Origin of Time*
One of the words for ‘time’ in the Celtic languages is the etymon represented by Old Irish aimser (ā, f), Welsh amser (masc.), Old Cornish anser (n for m probably a scribal error), Old Breton amser, Modern Breton amzer (fem.). In Old Welsh, amser occurs as a temporal and causal conjunction ‘when; because’ (Falileyev 2000, 6). In accordance with a cross-linguistic tendency, these words for ‘time’ can also refer to the ‘weather’,1 thereby evidencing a remarkable semantic shift of reference from a primary abstract2 concept to a phenomenon of daily practical experience.
The Irish and British words manifestly go back to a common ances­tor which is typically mechanically reconstructed as *ammesterā (see, for example, De Bernardo Stempel 1999, 273, 426) or, disregarding the his­torical phonology and morphology of Celtic, as *amstero/ā < earlier *h2emos-tero- (EDPC 33–4) or, even less specific, amm + a suffix *-stero/ā (LEIA A-35, 67). Stokes and Bezzenberger’s (1894, 10) *ad-messerā < *ad-menserā, derived from *mens- ‘to measure’ which is no longer recog­nised as an Indo-European root, is obsolete (cf. O’Rahilly 1950, 339). Old Irish aimser and cognates are evidently related to Old Irish amm ‘time’, apparently originally a neuter, later a masculine o-stem, judging from the attestations quoted in eDIL (dil.ie/3157). Gaulish ạmman, a word attested once on the fragmentary Calendar of Coligny (MacNeill 1928, 34; RIG III) is also believed to be related if—as is conceivable in the context—it refers to a period of time. The first letter of the word is lost except for a tiny portion of the tip (RIG III, 32–33), which is sufficient to identify it as an A. The portion of the inscription before the ạ is missing completely, but most scholars agree that there was no further letter before it as this position coincides with the left-hand margin of the column. Because of this
1 A random collection of languages where the cited words have the double meaning ‘time’ and ‘weather’, or where the two are closely related include: French temps, Italian tempo, Spanish. tiempo; dialectal German Zeit; Serbo-Croat vrijeme, Bulgarian време, Czech čas ‘time’, počasí ‘weather’, nečas ‘bad weather’, Russian год ‘year’, погода ‘weather’; Latvian laiks; Modern Greek καιρός; Albanian kohë; Hungarian idő; Tagalog panahon; Vietnamese thời gian ‘time’, thời tiết ‘weather’. Note finally that Old Irish sín ‘bad/good weather’, Middle Welsh hin ‘weather, bad weather’, and Breton hinon ‘fair weather’ < *sīnā could contain the same root *seh2i- as Middle Welsh hoedl ‘life(time), age’, Middle Breton hoazl ‘age’, and Latin saeculum ‘generation, lifespan’ (Zair 2012, 120), or *seh1i- as in Gaulish siro-, Old Irish sír, and Middle Welsh, Middle Breton hir ‘long’ < *sīro-.
2 What can be more abstract, that is, less visualisable as an object, than time? For the difficulties that even theoretical physicists have with defining what time is, see, as one example among many, Carroll (2010).
* This article was written as part of the research project Chronologicon Hibernicum. Chronologicon Hibernicum has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agree­ment No. 647351). My acknowledgements go to Elliott Lash, Corinna Salomon and Michael Weiss for help and suggestions, as well as to two reviewers of Ériu. The etymologies discussed here were first outlined in SnaS 18 (‘time’; 31.5.2016; available at: https://www.facebook.com/chronhib/posts ... 659945371; http://dhprojects.maynoothuniversity.ie ... e-snas-18/ (accessed 5 April 2018). All errors are mine.
2 DavidStifter
lack of further space, Olmsted’s (2001, 45) idiosyncratic reconstruction [S]AMMEN, which he connects with Celtic *sam- ‘summer’, is unconvincing. The ending -an of amman has all the appearances of the expected Gaulish ending of the nominative singular of a neuter n-stem. Despite their similarity, amm and amman do not, therefore, form an exact equation with each other, but synchronically they continue different stem forma­tions based on a common root.
The further etymology of amm and aimser is commonly regarded as unclear (cf. LEIA A-35, 67; EDPC 33–4). One etymology that has been pro­posed connects it with Hittite ḫamešḫant- ‘spring’3 (RIG III, 422; EDPC 33–4; Stüber 1998, 79), deriving both from a root *√h2em-, whose mean­ing, however, cannot be ascertained independently since it is not attested outside of those two branches of Indo-European.4 Also, the unlenited m of Celtic remains unexplained by this proposal. Zavaroni (2007, 32) takes recourse to a fanciful, idiosyncratic root ‘*HṃB “cum, par, simul” > “join, even”’. To avoid the circularity of an explanation obscurum per obscurius that besets all previous suggestions, a new solution is proposed in this arti­cle. It will be argued that a more satisfactory etymological and morpholog­ical analysis is possible, and that this group of words can be connected with semantically more proximate words outside Celtic.
The Proto-Indo-European root *√h2et- ‘to walk, wander’ (LIV 273), which occurs as a verb only in the rare Vedic at- ‘to go, walk’, underlies Italic and Germanic words for ‘year’, that is Latin annus; Oscan loc.sg. akeneí, acenei, gen.sg. aceneis, gen.pl. acunum; Umbrian acc. sg./pl. acnu; and Gothic aþn(s)* and ataþni*, all ultimately from a preform *h2et-no- (De Vaan 2008, 43–4; Kroonen 2013, 40). The semantic motivation behind the occurrence of this root in a word for ‘year’ is probably the idea of the ‘cycle/course/perambulation’ that the sun performs across the sky from one solstice until it reaches the same position again. Although superficially only the initial a- seems to relate the words in these two language families with the above-mentioned Celtic lexemes, on a deeper level a preform can be set up from which all forms can be derived in equal measure.
Starting from this premise, two slightly diverging reconstructions for the immediate precursors of Old Irish amm and Gaulish amman are possible, both meaning approximately ‘the going/course (of time)’.
3 Kloekhorst (2008, 280–1) derives hamesha(nt)- from the root *√h2meh1- ‘to mow’ (LIV 279), since spring, which could also be written with the sumerogramme Ú.BAR8 ‘har­vest’, was the time of harvest in Anatolia and in the Near East. The three months of April, May and June are the time of bloom and of rain in the Anatolian highlands (Tischler 1983, 143–4). This semantic link with mowing or harvesting is ruled out for Celtic by agricultural facts. Kloekhorst accordingly does not mention the alleged connection with Celtic at all.
4 RIG III, 422 refers to the dictionary entry for ḫamešḫ(a)- in Tischler (1983, 143–4) who, however, makes no mention of the Celtic words. In the discussion of the etymology of the Hittite word, Tischler critically reports a suggestion by Čop according to which it derives from an Indo-European root *Hem-/Hom- ‘hot’. Other Indo-European scholars have not accepted the existence of such a root.
varium3
1.
The first one starts from pre-Celtic *h2et-mo- > *atmo- > Proto-Celtic *ammo- for Old Irish amm, and mutatis mutandis from *h2et-mn̥ for Gaulish amman. This etymology implies an ad hoc rule that Proto-Indo-Europeam *‑tm- resulted in *‑mm- in Celtic. There is, to my knowledge, no other example in support of this change, nor is there a counter-example. Origi­nal *-dm- behaved differently. In Old Irish ammus ‘attempt’ < *ad-med-tu- and in verbal forms such as con·ammalt ‘has ground/consumed’, where ad functions as perfective augment, it seems to have been assimilated to -mm-, but this behaviour may be morpho-phonological. In analogy to other con­texts where the dental of ad assimilated regularly to a following consonant, yielding a geminate sound that resisted lenition, for example Proto-Celtic *ad-teku̯o- > Old Irish attach ‘refuge’, or *ad-bati̯on- > *abbatii̯on- > apthu ‘death’, this behaviour of preverbal ad could also have been extended to the position before m. Otherwise the retention of the cluster *dm can be observed. In formations such as Old Irish maidm ‘breaking’ < *mad-man- and feidm ‘effort, work’ < *u̯ed-man-, the d may have been retained under influence from paradigmatically related verbal forms such as the preterite memaid or the present tense feidid. No such analogy is possible, however, in the case of the hapax gen.sg. deidmea ‘law, usage’, nom. dedm*, cognate with Middle Welsh deddyf/deddf ‘id.’, Old Breton dedm* in annedmolion · anomala, Modern Breton (artificial?) dezv ‘decree’ < *dedmā < Proto-Indo-European *dhe-dhh1-mo‑, cf. Greek θεσμóς, Doric τεθμóς ‘law, custom’ (Zair 2012, 184). I conclude from this evidence that the regular treatment of *dm is its retention.5 The different behaviour of *-tm- vs. *-dm-, and the former’s greater propensity for assimilation to the following resonant, could be due to the lesser sonority of the voiceless occlusive.6
2.
Phonologically neater, but adding to the morphological complexity, is the second possible reconstruction, namely pre-Celtic *h2et-s-mo-, which effort­lessly leads to the desired result, via *atsmo- > *asmo- > *ammo- (for the treatment of Proto-Celtic clusters involving s, see Stifter forthcoming a). In this solution, the suffix *-smo- could either be the ‘sigmatic’ variant of *-mo-, replacing the latter by a common process of substitution,7 or *-mo- has been added to an underlying s-stem *h2etes- ‘the going’, for which, however, there
5 This point has also been made by O’Rahilly (1950, 339), who ascribes the assimilation in ammus to the influence of the cognate commus ‘power’.
6 Other examples of the divergent treatment of voiced and voiceless stops in Irish are rare, but not completely lacking. The sequences *-akR- and *-agR- gave different results: *dakru- > dér ‘tear’ and *maklii̯ā > mélae ‘shame’ vs. *agrā > ár ‘slaughter’ and *maglo- > mál ‘prince, noble’. *-tr- yielded -thr- and -thar, for example *aratrom > arathar ‘plow’, whereas *-dr- gave either -r- with compensatory lengthening when there was a clear morpheme boundary between the two sounds, for instance *ad-rīmā > áram ‘counting’, or it resulted in -ddr- elsewhere, such as *kredri- > cretair ‘sacred object’.
7 See Stüber (1998, 52–3) and De Bernardo Stempel (1999, 241–3, 265–7) for doublets of the suffixes *-men/*-smen and *-mo-/-smo- across the Indo-European languages and even within a single branch. This morphological variation seems to entail no functional difference.4 David Stifter
is no independent evidence. The disadvantage of reconstructing the Celtic forms in this way is that it removes the direct comparison with the Italic and Gothic forms, which are clearly non-‘sigmatic’, cf. Umbrian acnu or Gothic aþn(s)*.
At this stage of the comparison, irrespective of which of the two solu­tions is adopted, the Celtic and the Italo-Germanic words only make a root-equation, that is to say they are built from the same root *√h2et-, but their morphology is separate, one showing the suffix *-no-, the other the suffixes *‑(s)mo- and *-(s)men-. As Michael Weiss has pointed out to me (pers. comm., 31 May 2016), it is in fact possible to unite all forms in a com­mon morphological framework. Starting from the neuter men-stem, which only survives in Gaulish, allows combining all attested words in all three language groups, Celtic, Italic and Germanic, in a single derivational com­plex. From the neuter verbal abstract *h2etmn̥ ‘the wandering’ (pre-Celtic *atman > Proto-Celtic *amman > Gaulish amman; see Stüber 1998; and 2015, 114–15 for the type of formation), a thematic derivative *h2etmno- ‘having wandering, that which wanders = year/time’ can be derived. The complex cluster *-tmn- subsequently underwent simplification in the indi­vidual branches in accordance with a general tendency in Indo-European to avoid such sequences (Mayrhofer 1986, 159).8 In Italic and Germanic, the medial consonant was suppressed to yield *atno-, whereas in pre-Celtic the thematic derivative, perhaps under analogical pressure from the verbal noun *atman, the preform of Gaulish amman, lost the third consonant,9 viz. *atmo- > Proto-Celtic *ammo- (notwithstanding the question of whether the suffixes *-man and *-mo- had been replaced by *-sman and *-smo-, a question that is of no consequence for the etymology as such). With this solution, a common Western Indo-European word for ‘year, (period of) time’, *h2etmno-, can be reconstructed.
How does Old Irish aimser, Welsh amser, etc. fit in this picture? The word is most effortlessly derived from the Celtic verbal abstract *amman via fur­ther suffixation by means of the complex suffix *-stero/ā, that is *amman­stero/ā. The difference in gender (feminine in Irish and Breton, masculine in Welsh) is unclear. Maybe it hints at an earlier adjectival status of the word, or the feminine served originally as a collective to the masculine/neuter.
8 Another possible example for this type of simplification may be found in Old Irish gein ‘birth’. Proto-Indo-European possessed a verbal abstract *g̑enh1men- ‘birth’, of the root *√g̑enh1- ‘to be born’, as evidenced by Vedic jániman- and Latin germen. The Irish cognate can be explained through loss of the laryngeal in a first step, that is, *g̑enmen-, and then through loss of the middle nasal in those parts of the paradigm where the suffix stood in the zero grade before a vowel, for example the inherited instrumental, that is, *g̑enmneh1 > Proto-Celtic *genn°. From there the new en-stem *genen could have been created. Doric Greek γέννᾱ ‘descent’ < *g̑enh1-mn-eh2 shows a similar behaviour (NIL 140).
9 This must have occurred at a very early, perhaps pre-Celtic, date. In the case of Gaul­ish acaunos*, attested in the placename Acaunus (Saint-Maurice-en-Valais, Switzerland; the ancient name is preserved in the name of the monastery Abbaye de Saint-Maurice d’Agaune) and in glossaries under the form agaunus ‘rock’, which arguably continues *h2ek̑m̥no-, a the­matic derivative of Proto-Indo-European *h2ek̑mon- ‘stone’ (NIL 292), this simplification did not occur. varium5
Welsh amser and its other British siblings continue the preform *amman­stero/ā regularly, with syncope of the middle syllable (Jackson 1953, 651–6). In Irish, tautosyllabic *an became *en, and the syncope of the front vowel then yielded regularly the attested aimser with a palatalised cluster.
The suffix *-stero/ā is very rare in Celtic, if not isolated. In the section on suffixes (VKG ii 22), Pedersen mentions only aimser as an example for the suffix -stero-, -sterā-. However, in the phonological section (VKG i 80), Gaulish Epostero-uidus, Epotsoro-uidus is cited as a parallel formation, albeit the only one and without an etymological analysis. In fact, the variant with -e- is a ghost form, the inscription (CIL 13, 1036; Santons) has Epotsorouidus, which Delamarre (2003, 282) and Raybould and Sims-Williams (2009, 153) regard as a compound involving *storo- ‘firm, solid, forceful’ < Proto-Indo-European *ster- ‘rigid, stiff’. A superficially similar-looking ending occurs in the pair Old Irish sinser ‘the elder, the eldest; a senior; pl. elders, ancestors, forefathers’ and ósar ‘one who is younger, a junior’. This pair is unlikely to contain the same suffix as aimser. Sinser has been persuasively explained as a formation *sen-is-tero-, where the ‘Oppositivsuffix’ (oppositional suffix) *‑tero- has been added to the comparative stem *sen-is- (with zero-grade of the comparative suffix) of sen ‘old’ (LEIA S-114–15; De Bernardo Stempel 1999, 425–6). Ósar ‘junior’ is similar, albeit less straightforward. The oldest form appears to be ósar with a non-palatalised s;10 later palatalisation spreads there as well, perhaps in parallel to sinser, and forms with initial s- arise, that is óiser, sósar and sóiser. Ósar cannot go back to a preform *i̯ou̯-is-tero- since this should have resulted in **oíser in Old Irish. Instead it presupposes something like *i̯ou̯-s-tero- or *i̯ou̯-ā(i̯)s-tero-, which had arisen under influ­ence from the comparative oä ‘younger’ < *i̯ou̯-ās < *i̯ou̯-āi̯ūs of oäc ‘young’ (McCone 1994, 124; LEIA O-3). In any case, the origin of the suffix -sar seems to be the same as that of -ser in sinser.
There is thus no formation in the Celtic languages that exactly parallels the second part of aimser. De Bernardo Stempel (1999, 273 fn. 104) floats the idea of an obscured compound with the verbal root *√sterh3- ‘to strew, spread out’ (LIV 599–600), but does not expand on it. Nominal formations from this root are typically o-grade verbal nouns with preverbs as first members in Old Irish, for instance cosar ‘strewing; bed, couch; slaughter’, essair ‘strew­ing; litter’, fosair ‘strewing; litter’, osar ‘litter, bed’ (all cited following eDIL) < Proto-Celtic *kom/eχs/u̯o/uds-storā, although semantically ‘(temporal) expanse/ spread’ would make sense as an etymon for ‘time’.11 Breton gouzer ‘litter’, the cognate of Old Irish fosair, seems to show the e-grade, but since it stands beside the variant gouzel, it has no evidential value. As an alterna­tive, one may speculate if the second element of aimser, etc. is Proto-Celtic *ster- ‘star’, cf. Welsh ser ‘stars’, Old Irish ser ‘star’ < Proto-Indo-European
10 Once-attested úaser (Colmán’s Hymn 5b, MSS anuas … /anóser; Thes. ii, 300), rhyming with úasal, may contain an archaic spelling of the second syllable.
11 Cf. Old Church Slavonic prostorъ ‘extent, space’ for a semantic parallel, but again with a preverb as first member and o-grade of the root.6 David Stifter
*h2ster- ‘star’ (see Stifter forthcoming b), but semantically this idea is not appealing. The provenance of -ser remains obscure for the time being.
The Celtic words do not add decisive evidence to answer the question whether the Proto-Indo-European root *√h2et- ‘to go’ had a root-final laryngeal, that is *√h2etH-, a possibility implied by Vedic átithi-, Avestan asti- ‘guest’ < *h2etHti- (LIV 273 fn. 1). According to the complex rules of laryngeal developments in Celtic, as described in Zair (2012), *h2etH-men- should have resulted in Proto-Celtic **atamen- (> Old Irish **athaim), cf. Middle Welsh adaf ‘hand, talon’ < *φatamā < *pth2-meh2- (Zair 2012, 193–201). In the thematic derivative *h2etHmno-, on the other hand, the laryngeal would have been regularly lost in the position CHCC (Zair 2012, 160–8), resulting in *h2etmno- which would then have developed as described above. From this derivative, the allomorph without a laryngeal could have then been analogically introduced into **atamen- → *atmen- as well, thus obscuring the trace of the putative laryngeal. It is less certain what *h2etH-smen- would have yielded, but it is likely that in accordance with the general trend just mentioned the laryngeal would have been lost.
If the etymology for amm, etc. suggested here is correct, it implies a cycli­cal, not a linear concept of time12 for Proto-Celtic, a concept that has also been suggested for the Italic languages and thus makes a minor contribu­tion to the idea of Italo-Celtic (see Zair forthcoming for the concept of Italo-Celtic). The Slavic languages provide a semantic parallel for cyclical time. Proto-Slavic *vermę, Old Church Slavonic vrěmę ‘time’ continues the Indo-European verbal abstract *u̯ert-men- ‘turning’, cf. Vedic vartman- ‘track, course’. A similar notion may underlie the etymon Proto-Indo-Euro­pean *Hi̯eh1ro/ā, which designates the ‘year’ or related concepts in a range of languages (Germanic *jēran, Avestan yārə ‘year’, Proto-Slavic *ěrъ/a ‘spring’), but which can also be used more generally for ‘time’ in Luvian āra/i- or Greek ὥρα, if this noun is connected with the verbal root *√h1ei̯- ‘to go’ (LIV 232–3). Tocharian A pukäl, Tocharian B pikul ‘year’ could continue the root *√ku̯elh1- ‘to turn’, but an alternative and more widely accepted etymology derives it from *√peku̯- ‘to cook, mature’ (NIL 549–50).
DAVID STIFTER
Maynooth University
12 The cyclical perception of time is a psychological illusion suggested by the observable reiteration of natural processes such as the courses of the sun and the moon, or the annual return of the seasons, not an intuitive understanding of the fundamental structure of the world. Time is directional, that is, it constantly progresses. The arrow of time is ultimately linked with the Second Law of Thermodynamics and the increase of entropy, and thus with the ensuing irreversibility of physical processes in the macroscopic world (Carroll 2010, 135 ff.). It has even been suggested that entropy itself is the cause of which time is only an effect. varium7
Abbreviations
eDIL Greg Toner (dir.), electronic Dictionary of the Irish Language. Available at: http://edil.qub.ac.uk/ (2007– ).
EDPC Ranko Matasović, Etymological dictionary of Proto-Celtic. Leiden Indo-European Etymological Dictionary Series 7 (Leiden and Boston, 2008).
LEIA Joseph Vendryes et alii, Lexique étymologique de lʼirlandais ancien (Dublin and Paris, 1959– ).
LIV Lexikon der indogermanischen Verben. Die Wurzeln und ihre Primärstammbildungen. Zweite, erweiterte und verbesserte Auflage, bearbeitet von Martin Kümmel und Helmut Rix (Wiesbaden, 2001).
NIL Dagmar S. Wodtko, Britta Irslinger, Carolin Schneider, Nomina im Indogermanischen Lexikon (Heidelberg, Winter 2008).
RIG III Paul-Marie Duval et George Pinault, Recueil des inscriptions gauloises. Volume iii, Les calendriers (Coligny, Villards dʼHéria). Gallia Suppl. 45 (Paris, 1986).
SnaS David Stifter, Sanas na Sengoídilce. Available at: https://www.facebook.com/chronhib/ (2015– ).
References
Carroll, Sean, 2010: From eternity to here. The quest for the ultimate theory of time. London.
De Bernardo Stempel, Patrizia, 1999: Nominale Wortbildung des älteren Irischen. Stammbildung und Derivation. Buchreihe der Zeitschrift für celtische Philologie 15. Tübingen.
Delamarre, Xavier, 2003: Dictionnaire de la langue gauloise. Une approche linguistique du vieux-celtique continental. Paris.
De Vaan, Michiel, 2008: Etymological dictionary of Latin and the other Italic languages. Leiden Indo-European Etymological Dictionary Series 7. Leiden and Boston.
Falileyev, Alexander, 2000: Etymological glossary of Old Welsh. Buchreihe der Zeitschrift für celtische Philologie 18. Tübingen.
Jackson, Kenneth, 1953: Language and history in Early Britain. A chrono­logical survey of the Brittonic Languages 1st to 12th c. a.d. Edinburgh University Press.
Kloekhorst, Alwin, 2008: Etymological dictionary of the Hittite inher­ited lexicon. Leiden Indo-European Etymological Dictionary Series 5. Leiden and Boston.8 David Stifter
Kroonen, Guus, 2013: Etymological dictionary of Proto-Germanic. Leiden Indo-European Etymological Dictionary Series 11. Leiden and Boston.
Mayrhofer, Manfred, 1986: Indogermanische Grammatik. Band I, 2. Halbband: Lautlehre. Heidelberg.
MacNeill, Eoin, 1928: ‘On the notation and chronography of the Calendar of Coligny’, Ériu 10, 1–67.
McCone, Kim, 1994: ‘An tSean-Ghaeilge agus a réamhstair’, in Kim McCone, Damian McManus, Cathal Ó Háinle, Nicholas Williams, Liam Breatnach (eag.), Stair na Gaeilge in omós do P[h]ádraig Ó Fiannachta. Maigh Nuad.
Olmsted, Garrett S., 2001: A definitive reconstructed text of the Coligny Calendar. Journal of Indo-European Studies Monograph 39. Washington D.C.
O’Rahilly, Thomas F., 1950: ‘Varia II. 6. The prefix ad-’, Celtica 1, 337–41.
Raybould, Marilynne E. and Patrick Sims-Williams, 2009: Introduction and supplement to the corpus of Latin inscriptions of the Roman Empire con­taining Celtic personal names. Aberystwyth.
Stifter, David forthcoming a: ‘58. Celtic Phonology’, in Jared S. Klein and Brian Joseph (eds), Comparative Indo-European linguistics. An inter­national handbook of language comparison and the reconstruction of Indo-European. Handbücher zur Sprach- und Kommunikationswissen­schaft. Berlin and New York.
Stifter, David forthcoming b: ‘The stars look very different today’, TBC.
Stokes, Whitley and Adalbert Bezzenberger, 1894: Wortschatz der keltischen Spracheinheit. 4. Auflage. Göttingen.
Stüber, Karin, 1998: The historical morphology of Celtic n-stems. Maynooth Studies in Celtic Linguistics 3. Maynooth.
Stüber, Karin, 2015: Die Verbalabstrakta des Altirischen (2 vols). Münchner Forschungen zur historischen Sprachwissenschaft 15. Bremen.
Tischler, Johannes, 1983: Hethitisches Etymologisches Glossar 1. A–K. Innsbrucker Beiträge zur Sprachwissenschaft 20. Innsbruck.
Zair, Nicholas, 2012: The reflexes of the Proto-Indo-European laryngeals in Celtic. Brill’s Studies in Indo-European Languages and Linguistics 7. Leiden and Boston.
Zair, Nicholas, forthcoming: ‘Italo-Celtic’, in Jared S. Klein and Brian Joseph (eds), Comparative Indo-European linguistics. An international hand­book of language comparison and the reconstruction of Indo-European. Handbücher zur Sprach- und Kommunikationswissenschaft. Berlin and New York.
Zavaroni, Adolfo, 2007: On the structure and terminology of the Gaulish calendar. BAR International Series 1609. Oxford.