This is an excerpt from my new etymological dictionary of Basque. You can get a full copy from one of the following sites: https://www.createspace.com/4812609 http://amzn.com/1499595468 http://www.amazon.it/dp/1499595468 http://www.amazon.de/dp/1499595468 http://www.amazon.co.uk/dp/1499595468 A First Etymological Dictionary of Basque as an Indo-European Language – Basque Native, Basic Lexicon by Gianfranco Forni TABLE OF CONTENTS 1 Introduction Origin of this dictionary This dictionary is an evolution of a discussion paper published in The Journal of Indo-European Studies (JIES) in 20131. That paper was designed as a step-by-step demonstration that Basque is an IndoEuropean language. For this reason, many of its sections dealt with detailed methodological issues and statistical analyses of findings, while actual etymologies were scattered across various sections, according to historical phonology and other criteria. Furthermore, the discussion paper was followed by my reply to the critics2, where I amended some etymologies. As a result, it is a fairly laborious task to extract Indo-European etymologies of Basque native, basic lexicon from that paper, so I thought it might be useful to re-organize those etymologies, in order to make them easily accessible to a wider audience. That’s why I decided to write the present dictionary. Besides re-arranging entries alphabetically, I have made several further changes with respect to my original article:       the introductory section on historical phonology (i.e. sound laws) has been simplified, so it’s now easier to understand; a new theory on the loss of plosives is proposed3; every single etymology has been revised; some have been changed; a few have been discarded; a few new etymologies have been added; each entry has been significantly expanded by adding Basque dialect variants, additional Indo-European cognates and additional references to sources; several indices (including a semantic one) have been added. 1 Gianfranco Forni – Evidence for Basque as an Indo-European Language. In: J. P. Mallory (Ed.) - The Journal of Indo-European Studies, Volume 41, Numbers 1 & 2, Spring/Summer 2013. 2 Gianfranco Forni – Evidence for Basque as an Indo-European Language: A Reply to the Critics. In: J. P. Mallory (Ed.) - The Journal of Indo-European Studies, Volume 41, Numbers 1 & 2, Spring/Summer 2013. 3 see “Historical Phonology” section for details. 2 Introduction Acronyms Basque dialects and variants: A Aezkoan Basque B Bizkaian Basque c common Basque G Gipuzkoan Basque HN High Navarrese Basque L Lapurdian Basque LN Low Navarrese Basque R Roncalese Basque S Salazarese Basque Sout Southern dialect of Alava Z Zuberoan Basque Other acronyms: adj. adjective C any consonant fem. feminine IE Indo-European masc. masculine n. noun PIE Proto-Indo-European V any vowel For bibliographical acronyms, see next section. 3 Introduction Bibliography AD Douglas Q. Adams – A Dictionary of Tocharian B – Rodopi 1999 – ISBN 978-90-420-0435-1 B Robert Beekes, Lucien van Beek – Etymological Dictionary of Greek – Brill 2010 – ISBN 978-90-04-17418-4 DERK Rick Derksen – Etymological Dictionary of the Slavic Inherited Lexicon – Brill 2008 – ISBN 978-90-04-15504-6 JIES Gianfranco Forni – Evidence for Basque as an IndoEuropean Language. In: J. P. Mallory (Ed.) - The Journal of Indo-European Studies, Volume 41, Numbers 1 & 2, Spring/Summer 2013 K Alwin Kloekhorst – Etymological Dictionary of the Hittite Inherited Lexicon – Brill 2008 – ISBN 978-90-04-16092-7 KR Guus Kroonen – Etymological Dictionary of Proto-Germanic – Brill 2013 – ISBN 978-90-04-18340-7 M Ranko Matasović – Etymological Dictionary of Proto-Celtic – Brill 2009 – ISBN 978-90-04-17336-1 M-A James P. Mallory, Douglas Q. Adams – The Oxford Introduction to Proto-Indo-European and the Proto-IndoEuropean World – Oxford University Press 2006 – ISBN 978-0-19-92668-2 4 Introduction MART Hrach K. Martirosyan – Etymological Dictionary of the Armenian Inherited Lexicon – Brill 2010 – ISBN 978-90-0417337-8 MICH Luis Michelena – Fonética Histórica Vasca - 3rd edition, 1985 – San Sebastián: Publicaciones del Seminario de Filología Vasca ‘Julio de Urquijo’ de la Excma. Diputación de Guipúzcoa O Vladimir E. Orel – Albanian Etymological Dictionary – Brill 1998 – ISBN 90-04-11024-0 T Robert L. Trask – The History of Basque – Routledge 1997 – ISBN 0-415-13116-2 T-E Robert L. Trask – Etymological Dictionary of Basque – 2008 (published online as a PDF file) V Michiel de Vaan – Etymological Dictionary of Latin and the other Italic Languages – Brill 2008 – ISBN 978-90-04-167971 Of these, the most important sources are T, T-E, M-A and M – plus, of course, JIES. 5 Introduction Scope Only Basque native, basic lexicon4 of Indo-European origin is included in this dictionary. Readers interested in the etymology of other Basque terms (including e.g. loans from other languages, Basque-internal etymologies, etc.) should refer to Michelena’s and Trask’s work – especially T and T-E. The present dictionary can be seen as complementary to T-E, as it proposes Indo-European etymologies for many entries labelled as “OUO” (“of unknown origin”) in T-E. Methodology The present dictionary strictly follows the standard comparative method, i.e. it derives Basque lexemes from Proto-Indo-European via regular sound laws, as described in the next section. For an extensive discussion of methodological issues, please refer to my original article in JIES. 4 for an in-depth discussion on the concept of basic lexicon, see Gianfranco Forni - Maximizing the reliability of macro-comparisons. General criteria and a detailed proposal for the semantic aspects. In: Atti del Sodalizio Glottologico Milanese, vol. IV n.s., Edizioni dell'Orso, Milano 2011, ISBN 978-88-6274-267-2. 6 Historical Phonology Historical Phonology This dictionary uses Pre-Basque (as reconstructed by MICH, T and T-E) as its starting point. Therefore, this section will first provide a brief outline of Pre-Basque, followed by the sound laws that can derive Pre-Basque from Proto-Indo-European. For the sake of legibility, in the present section reconstructed protophonemes are not preceded by an asterisk; for example, protophoneme *p is simply written p. Pre-Basque: an internal reconstruction Based on dialect variations, other internal evidence, and phonological treatment of loans, Michelena and Trask (MICH, T, T-E) reconstructed so-called Pre-Basque, i.e. the phonology of Basque around two thousand years before present. Such work, combined with an analysis of known loans, led to the creation of a list of native Basque terms and the reconstruction of their PreBasque phonological shape. Pre-Basque is an excellent starting point for any historical linguist investigating the affiliation of Basque to other language families, because it:      filters out any terms that have already been recognized as loans, and only reconstructs native Basque terms; provides an accurate segmentation of Basque lexical items, based on internal evidence; identifies the common sources of current dialectal variations; provides a standardized inventory of phonemes and phonotactic rules; is not based on any assumption about external affiliations of Basque. The phonemic inventory reconstructed for Pre-Basque is the following (T, chapter 3): 7 Historical Phonology Fortes (-p-) -t- -k- -tz-tz -ts-ts -nn-nn -ll-ll -rr-rr Lenes b-b- (d-) -d- g-g- z-z- s-s- n-n- l-l- -r- Vowels i e a o u (marginal phonemes are within brackets) Lenes never occurred word-finally. Fortes5 never occurred wordinitially (and plosive fortes never occurred word-finally, either). Pre-Basque (like Basque) has two lenes and two fortes sibilants: <s> and <ts> are apico-alveolar, whereas <z> and <tz> are laminoalveolar. Pre-Basque had no */m/ phoneme at all6. Trask (T 128, 133-136, 172-176, 180-183) reconstructs that no word in Pre-Basque could begin with any of /p, t, k, d/ or /m/, and illustrates Martinet’s hypothesis that, before Pre-Basque, words could begin with /ph, th, kh/, which later weakened to /h/ and were eventually lost. Also note that that *h was not phonemic in Pre-Basque, and is therefore conventionally represented as *[h]. Finally, Pre-Basque had no word-initial consonant clusters (T 127). As Pre-Basque is largely accepted as a valid reconstruction by Vasconists, this dictionary builds upon Pre-Basque, as reconstructed by Michelena and Trask in MICH, T and T-E: with few, minor exceptions, Pre-Basque is the starting point for the Indo-European etymologies of Basque basic, native lexicon proposed in this dictionary. 5 Fortes consonants can be conventionally represented either by capital letters or by geminates. This dictionary will adopt the latter convention. 6 as we will see below, it actually had a marginal *m phoneme, deriving from primary or secondary *mm. 8 Historical Phonology Introduction to sound laws The PIE and Pre-Basque sound systems are readily available (e.g. in M-A and T respectively). An intermediate stage, called ProtoBasque in the present research, has been tentatively reconstructed, and can be derived from a first set of sound laws listed below. It should be borne in mind, though, that any reconstructed protolanguage is somewhat arbitrary, since it tries to depict a prehistoric language at some given point in time, whereas the evolution of PIE into Pre-Basque was a continuous sequence of diachronic changes. It is therefore more useful to focus on sound laws (i.e. the sequence of changes) than on any given sound system at some intermediate stage. Sound laws will be listed in chronological order. Only well-attested sound laws are listed below, whereas scarcely attested sound laws are omitted. For a full discussion of sound laws, please refer to JIES. The following conventions will be used for PIE phonemes:    three laryngeals7 will be used, namely h1, h2, h3, while hx will represent a laryngeal of unknown type (sometimes written as H by other authors); semivowels will be written as w and y, while some authors use u̯ (or u) and i̯ (or i); aspirates will be represented as Cʰ (e.g. dʰ), labials as Cʷ (e.g. kʷ) and palatals as g̑ , k̑ , etc. 7 so, for the sake of uniformity, M-A’s h2, h4 and ha (=h2/4) will all be “normalized” to h2 9 Historical Phonology Main sound laws linking PIE to Proto-Basque Laryngeals display an unremarkable behavior: h1(e) > e; -Vh1(-) > -V(-) h2(e) > a; eh2 > a; exception: h2eu > eu h3(e/o) > o m is generally fricativized to /v/ (probably via [ṽ]), while rare mm (whether primary of from nm) is preserved as m: m>v nm, mm > m o usually yields u (with some exceptions): o, ō > u (/o) Dental stops assibilate before /u/: tu, du, dʰu > tsu Initial consonant clusters are not permissible: CC- > CVC- (anaptyxis) (exception: st- > s-). 10 Historical Phonology Non-labial word-initial plosives are fricativized8 (a phenomenon closely resembling initial consonant mutations found in Celtic languages, such as Welsh, Breton and Scottish Gaelic); wellattested cases include: p- > f- (possibly [φ]) t- > θk-, k̑ - > xbʰ- > v- (possibly [ ]) d-, dʰ- > ðg̑ ʰ- > - (exception: g̑ ʰC- > gC-) while labialized initial consonants generally yield b: kʷ-, gʷ-, dw- > bsw- > b- (exception: swVr- > sVrw-) This set of sound laws does not apply to verb roots: any consonant can occur in root-initial position in verbs (T 178-179), because both finite and non-finite verb forms always have one or more prefixes before the root. Since root-initial plosives were never word-initial in verbs, they follow the next set of sound laws. Word-medial plosives display different behaviors; well-attested cases include: -bʰ- > -b-VpV- > -VfV-d- > -ð- near e, i; -d- near o, u -dʰ- > -θ- near i; -d- elsewhere -g- > -g-gʷʰ- > -g-g̑ - > -g-g̑ ʰ- > -k-kʷ- > -b8 based on areal similarities, this trajectory to eventual loss is more likely than aspiration, contra JIES. 11 Historical Phonology Final -s is lost after vowels: -Vs > -V Word-medial consonant clusters undergo several simplifications and changes; a notable one is the following: -st- > -s- in pre-tonic position, -st- in post-tonic position Vowels are generally preserved, except: er, ēr > ar in closed syllables en > an (-)el- > (-)al- (exception: ell > ill) ei > e Proto-Basque - the proto-language obtained by applying the above sound laws - is still recognizably Indo-European. Its evolution into Pre-Basque, on the other hand, involves dramatic sound changes that will obscure its origins. 12 Historical Phonology Main sound laws linking Proto-Basque to Pre-Basque The main phenomenon is the loss of all fricatives (apparently except medial -θ- which later evolves to /z/): f>∅ θ- > ∅- x- > ∅v- > ∅ð>∅ γ- > ∅- This set of sound laws is probably the main reason why Basque is so hard to recognize as an IE language: while the loss of wordmedial fricatives is a widespread phenomenon (e.g. in several Romance languages, notably in French), the loss of word-initial fricatives is much less frequent in IE languages – the most notable exception being the loss of PIE *p- (via *f-) in Celtic. Sibilants undergo palatalization: some simplifications and tsu- > zu-(t)su(-) > -(t)zu(-) /-(t)zi(-) s > s before e and after i, e; z elsewhere ts > ts after i, e; tz elsewhere w (whether primary, or secondary from uV) becomes g: w>g Word-medial θ becomes z: -θ- > -z- 13 conditioned Historical Phonology Vowels undergo several further changes: -V > -∅ in words with more than two syllables - and sporadically in disyllabic words, too Vu > V a- > i- in pre-tonic position e- > ∅- in pre-tonic position Consonant clusters undergo several simplifications and changes; notable ones include: -Vktz- > -Vitz-rs- > -s- (see T 163) -rst(-) > -rtz(-) / -st(-) -rw- > -rr(-) -ly- > -rr-lC- > -ll- The above sound laws lead to a slight difference in the reconstruction of Pre-Basque (vs. MICH and T): while primary PIE *m is lost (via *v), secondary *m (from *mm, *nm) is preserved in Pre-Basque. Main sound laws linking Pre-Basque to Basque For a detailed description, see T (chapter 3) and T-E (pages 25 to 38). The most notable sound changes from Pre-Basque to Basque are the following:      loss of intervocalic *n: *VnV > VV rhotacism of intervocalic *l: *VlV > VrV fluctuation of intervocalic d and r ([ɾ]) *bVn > mVn *bo- > o14 The Dictionary [EXCERPT] The Dictionary [EXCERPT] Structure of each entry All dictionary entries have the same structure: modern Basque term; dialect variants (when listed) are followed by the acronym of the dialect (see acronym list); English translation of the Basque term; Pre-Basque reconstruction, when available (source: T and/or T-E); in the rare cases where my sound laws lead to a PreBasque reconstruction different from Trask’s, Trask’s reconstruction is presented first, followed by my own reconstruction (preceded by “recte”); Proto-Basque reconstruction; Proto-Indo-European (“PIE”) reconstruction (preceded by a question mark if the PIE etymology is doubtful); Proto-Celtic reconstruction (if applicable); Celtic cognates (if applicable); other IE cognates; optional comments about the etymology, incl. intermediate forms between PIE and Pre-Basque (when this is useful to make the phonetic evolution explicit); sources of the Basque term, its Pre-Basque reconstruction, PIE reconstruction and IE cognates, followed by a reference to etymology number in JIES (see source acronyms in the bibliography section). As is usual in Basque linguistic work, the letter <h> is ignored in the alphabetical ordering. 15 The Dictionary [EXCERPT] Transcription conventions All Greek words are transliterated. For Hittite, the following conventions are used: What is usually written as follows by Hittitologists … … will be written as follows is this dictionary: i̯ y u̯ w ḫ h š s For PIE, see “Introduction to sound laws” sub-section in the “Historical Phonology” section. A conjecture on perfective participles in –n Contra Trask (T 212-214, T-E 46), I believe perfective participles in -n are best explained as deriving from a circumfix *e-…-an (possibly from an older *e-…-en: see izen ‘name’). I will adhere to this view throughout the dictionary. Intriguingly, this circumfix resembles a similar, widespread Germanic circumfix for perfective participles, appearing e.g. as ga-…-an in Gothic, as gi-…-an in Old Saxon and Old High German, as ge-…-en in Modern German, and as ge-…-an in Old English (where ge- was pronounced /je/, then became i- in Middle English, and eventually disappeared in Modern English). Whether this is just a chance resemblance or a result of early contacts with Germanic populations, it is too early to tell. 16 A haur, aur 'child' PRE-BASQUE: *[h]aur 'child' PROTO-BASQUE: *faurPIE: *peh2u-ro- / *ph2eu-ro- 'little, small' OTHER IE LANGUAGES: Latin puer 'child, boy' (< *ph2u-ero- ‘smaller’), Latin parvus ‘small’, Greek paûros ‘small’, Faroese faur ‘few’, English few COMMENT: *peh2uro- > *pauro- > *faur- > aur SOURCES: T-E 20, 118, 386, 391; M-A 211, 320; V 448, 496; B 1158; KR 132; IEW 842f; JIES #88 17 B B bizi 'alive, living; life' PRE-BASQUE: *bizi 'alive, living' PROTO-BASQUE: *biutsu PIE: *gʷ(e)ih3- 'to live' > *gʷih3-etoPROTO-CELTIC: *biwo- 'alive' CELTIC: Middle Welsh byw ‘living, alive; life’; Old Cornish biu ‘life’ OTHER IE LANGUAGES: Greek bíotos ‘life’ (< *gʷih3-eto-); Avestan jiyā-tu- ‘life’; Sanskrit jīvitá- ‘living’; Lithuanian gyvatà ‘life’; Old Prussian giwato ‘life’, Gothic qius ‘alive’, English quick, Latin vīvus ‘alive’, Latin vīta ‘life’ (< *gʷiH-wo-t-), Oscan biítam ‘life’ (< *gʷiH-t-) COMMENT: *gʷih3eto- > *gʷioto- > *gʷiutsu- > *biutsu > *biuzi > bizi. SOURCES: T 129, 312; T-E 145, 385, 387, 392; M-A 188; M 67; B 216; V 685; KR 320; DERK 563; IEW 467f; LIV 215f; JIES #17 buru 'head' PRE-BASQUE: *buru 'head' (recte *bulu) PROTO-BASQUE: *γebulu PIE: *gʰebʰōl / *gʰebʰ-(e)l- / *gʰebʰ-l-o- 'head' OTHER IE LANGUAGES: Greek kephálē ‘head’, Tocharian A śpāl ‘head’, Gothic gibla ‘pinnacle’ COMMENT: *gʰebʰol >*gʰebʰul- > * ebul-u- > *ebulu > *bulu, followed by regular rhotacism of intervocalic *-l- after Pre-Basque. Also see berarri ‘ear’. SOURCES: T 129, 173, 285; T-E 19, 148, 387, 392; M-A 174; AD 642; KR 173; B 682; JIES #120 18 E E egun 'day; today' PRE-BASQUE: *egu(-n) 'day; sun' PROTO-BASQUE: *ðewuPIE: *di-eu- > *dei-w-o- 'day, clear sky, heaven, sky-god' PROTO-CELTIC: *dīy(w)o- 'day', *dēwo- ‘god’ (< PIE *deiwo-) CELTIC: Old Irish dïe ‘day’, Old Welsh did ‘day’, Cornish dew ‘god’ OTHER IE LANGUAGES: Latin dies ‘day’, Latin diu ‘by day’, Latin deus ‘god’ (< PIE *deiwo-), Armenian tiw ‘day’, Sanskrit diáuḥ ‘heaven, sky god, day’, Greek Zeus ‘Zeus’ COMMENT: *deiwo- > *deiwu- > *ðeiwu- > *ðewu- > *ewu- > *egu-. The original Pre-Basque word was *egu 'day', whence egu-n 'today' with -n locative ending, later generalized to nominal status. Eki 'sun' (< *egu-ki) and egu-zki 'sun' also derive from *egu. SOURCES: T 175, 211, 310; T-E 19, 41, 165, 386, 392; M-A 301; M 96, 101; V 167, 170, 172-173, 315; MART 616; B 498; IEW 184f.; JIES #110 19 G G gizon (c excl. Z), gixon (B G), gízun (Z LN) 'man, person' PRE-BASQUE: *gizon 'man' PROTO-BASQUE: *giθonV PIE: *dʰg̑ ʰom-yo- / *dʰg̑ ʰm-on- 'human, earthling’ PROTO-CELTIC: *gdonyo- 'human, person' CELTIC: Gaulish -xtonio /gdonio/, Old Breton don OTHER IE LANGUAGES: Latin homō ‘man, human being’, Gothic guma ‘man’ COMMENT: a similar metathesis (*dʰg̑ ʰ- > *gd-) can be seen in ProtoCeltic. In Pre-Proto-Basque, the process must have been *dʰg̑ ʰ- > *g̑ ʰdʰ- (metathesis) > *g̑ dʰ- (loss of first aspiration) > *gdʰ- (loss of palatalization) > *gidʰ- (anaptyxis in consonant cluster). SOURCES: T 142, 173, 271; T-E 19, 37, 206, 387, 393; M-A 206; M 156; V 287; KR 195; IEW 414; JIES #40 20 I I ibai (B G HN L LN Z), hibai (L LN) 'river' PRE-BASQUE: *ibai (recte *ibani) 'river' PROTO-BASQUE: *abanPIE: *h2ep-h3on- / *h2ebʰ-n̥ - 'river' PROTO-CELTIC: *abon- 'river' CELTIC: Old Irish ab 'river', Middle Welsh afon 'river' OTHER IE LANGUAGES: Palaic hāpna 'river', Latin amnis 'river' COMMENT: *h2ebʰn̥ - > *aben- > *aban- > *iban-; Pre-Basque should be reconstructed as *ibani, with subsequent regular loss of intervocalic *n. No need to derive ibai ‘river’ from ibar ‘valley’ (contra T-E 217): given this compelling IE etymology for ibai, the opposite is more likely to be true, i.e. ibar may be a derivative of ibai. SOURCES: T 313; T-E 217, 388, 393; M-A 126; M 23-24; V 39; K 295; IEW 1; JIES #48 itz (G HN L LN), hitz (L LN Z) 'word' PRE-BASQUE: *[h]itz 'word, speech, language' PROTO-BASQUE: *yektsu PIE: *yek- 'to say, speak' + *-toPROTO-CELTIC: *yextV- 'speech, language' CELTIC: Middle Welsh ieith 'language', Middle Breton yez 'language' OTHER IE LANGUAGES: Old High German jiht ‘utterance’, Umbrian iuk- ‘word’ 21 I COMMENT: *yekto- > *yektsu- > *yeitz > itz (with *yVi- > i- as in ihi ‘reed, rush’) - or possibly straight from zero grade *ik-to- > *iktsu > *iitzu > itz. SOURCES: T 174, 178; T-E 233, 390, 393; M 435-436; V 308; KR 272; IEW 503f; LIV 311; JIES #55 izar 'star' PRE-BASQUE: *izar 'star' PROTO-BASQUE: *asar PIE: *h2s-tḗr 'star' PROTO-CELTIC: *sterā 'star' CELTIC: Old Irish ser, Middle Welsh ser OTHER IE LANGUAGES: Greek astḗr 'star', Hittite haster(a)- ‘star’, English star COMMENT: *h2stḗr > *astér > *asér > *azár > izar. SOURCES: T 174, 313; T-E 19, 235, 389, 393; M-A 129; M 355; B 156; K 326; KR 478; IEW 1028; JIES #58 22 N N neska 'girl' PRE-BASQUE: *neska 'girl' PROTO-BASQUE: *γ(e)n-et-ska (diminutive) PIE: *g̑ enh1- 'to bear, engender, generate' > *g̑ (e)nh1-(e)toPROTO-CELTIC: *genetā 'girl' CELTIC: Gaulish gnata ‘girl’, Welsh geneth ‘girl’ OTHER IE LANGUAGES: Oscan genetaí 'to the daughter', German Kind ‘child’ SOURCES: T 141, 174, 175, 271; T-E 7, 19, 296, 387, 393; M 157; V 260; B 272-273; KR 288; IEW 373f; LIV 163; JIES #68 23 O O otz, hotz 'cold (n., adj.)' PRE-BASQUE: *[h]otz 'cold' PROTO-BASQUE: *ougtsu PIE: *h3eug- 'cold' + *-to- / *-tuPROTO-CELTIC: *owg-ro- 'cold (adj.)', Proto-Celtic *owxtu- 'cold (n.)' CELTIC: Old Irish ócht 'cold' (< PIE *h3oug-tu-) OTHER IE LANGUAGES: Latvian aũksts (< PIE *h3oug-s-to-), Armenian oyc ‘cold’ COMMENT: *h3eug-to/u- > *ougtu- > *ougtsu > *ogtsu > *otzu > otz (for sporadic loss of final vowel in disyllabic words, compare e.g. or ‘dog’, zur ‘wood’). SOURCES: T 137, 174, 178; T-E 20, 47, 322, 386, 394; M-A 348; M 301, 304; MART 569; IEW 783; JIES #74 24 S S sehi (L LN), sei (G L), seĩ (old B), sein (unified Basque), segi (Sout) 'boy, child' PRE-BASQUE: *seni 'boy, child' PROTO-BASQUE: *seunPIE: *suhx-nu- 'son' < PIE *seuhx- 'to beget' OTHER IE LANGUAGES: Old High German sun, English son, Old Prussian souns, Sanskrit sūnú-, Russian syn 'son' COMMENT: *suhx-nV- > *sūn- > *seun- > *sen- + -i adj. suffix. For *uhx > *ū > *eu compare euli ‘fly’. Basque seme 'son' < Pre-Basque *sen-be has the same origin. Also see suin ‘son-in-law’. SOURCES: T 133, 140, 177, 188, 269; T-E 123, 338, 339, 385, 386, 394; M-A 211; DERK 483; KR 492; IEW 913f; LIV 538; JIES #75 seme ‘son’  see sehi ‘boy, child’ 25 Z Z zur (c excl. R), zũr (R), zul (B) 'wood' PRE-BASQUE: *zur 'wood' PROTO-BASQUE: *tsuru PIE: *doru 'tree, wood' PROTO-CELTIC: *daru- 'oak' CELTIC: Old Irish daur, Middle Welsh dar ‘oak’, Old Breton dar 'oak' OTHER IE LANGUAGES: Greek dóru ‘wood, tree’, Hittite tāru ‘wood’, English tree COMMENT: *doru > *duru > *tsuru > *zuru > zur, with loss of *-u as in Celtic reflexes9. SOURCES: T 144, 173, 178, 310; T-E 20, 381, 390, 395; M-A 156; M 91; B 349; K 849; KR 522; IEW 214f; JIES #86 9 for other examples of sporadic loss of final vowel in disyllabic words see e.g. otz ‘cold’, or ‘dog’. 26 27