3.2. North-West Indo-European

3.2.1. North-West Indo-European community

The North-West Indo-European (NWIE) proto-language is the reconstructible ancestor of Italic, Celtic, Germanic, and Balto-Slavic, and probably other fragmentary European languages like Venetic, Lusitanian, Messapic, Ligurian, or Elymian (Oettinger 1997, 2003; Adrados 1998; Mallory and Adams 2007; Mallory 2013; Beekes 2011). NWIE refers to a long-lasting linguistic community spanning from an early or Pre-NWIE stage, coinciding with a common western development within Early Yamna after the separation of Tocharian ca. 3500–3300 BC, to a post-NWIE Sprachbund, to be identified probably with European Early Bronze Age cultures, in close contact through the Pan-European Bronze Age trade networks including Únětice, until ca. 1600 BC.

Genetic research indicates that there was a patrilineally related community in close contact in the Carpathian Basin, formed first by Yamna settlers in the early 3rd millennium BC, and then by the Classical or East Bell Beaker group from the mid–3rd millennium BC on (Olalde et al. 2018; Wang et al. 2018). A reconstructible Classical NWIE language is then to be associated with this central European population in the centuries before and after 2500 BC (Harrison and Heyd 2007; Mallory 2013; Quiles 2017).

Patrilineally related East Bell Beakers expanded successfully in a short period into wide territories of western, northern, and eastern Europe, areas whose languages later evolved into Celtic, Italic, Germanic, and Balto-Slavic, allowing for certain innovations to spread between these languages. The spread of Bell Beakers ca. 2400–2300 BC is to some extent coincident with the areas of Old European hydronymy (Krahe 1964, 1949; Nicolaisen 1957), a quasi-uniform name-giving system for water courses that shows Indo-European water-words and suffixes following rules of Late Proto-Indo-European word formation (Adrados 1998).

3.2.2. North-West Indo-European evolution

The North-West Indo-European proto-language can be reconstructed based on phonological, lexical, and also morphosyntactical traits[ix]. These are some common features:


·       Final process of laryngeal loss (see §II.2. Laryngeal evolution).

·       Maintenance of a conservative consonant system[x], in contrast with coeval Proto-Indo-Iranian (satemisation and palatalisation) and Balkan languages (devoicing of voiced stops).

Nominal system:  

·       Full development of the known case system, with stable dative-locative-instrumental endings, probably with an origin in PIE adverbial *-bhi- (see above Late PIE): ins. sg. *-bhi, ins. pl. -bhis, dat.pl. *-bhos, expanded to the thematic declension. Compare for example for dat. pl. *-bhos Lat. matribus ‘for the mothers’, Gaul. atrebo ‘for the fathers’; Lusitanian Lugubo Arqueienobo, Venetic louderobos ‘for the children’, Messapic logetibas ‘for the logetis’. Germanic and Balto-Slavic show a characteristic “Northern European” bilabial substitution for *-m- e.g. dat. pl. *-(o)mos, cf. Gmc. *-(a)mz in Goth. -am, ON -m; OLith. sunúmus ‘to/for sons’, výrams ‘to/for men’; OCS kostĭmŭ ‘to/for bones’, gradomŭ ‘to/for cities’ (see below §4.5.3. Northern European).

·       The dual continues a process of loss, remaining as an archaic feature linked to objects naturally occurring in pairs.

·       Morphophonological developments affect ‘unstable’ declensions (viz. proterodynamic, hysterodynamic, etc.) simplified into stable vocalic paradigms and known lexical isoglosses.

Verbal system:

·       Further reduction of the use of athematic stems, with development of semithematic ones, with preterite forms in *-ē-, *-ā-.

·       Expansion of thematic inflection (including new root verbs), i.e. in *-e-, *-e- (particularly in *-ēe-).

·       Further development of the middle and passive systems[xi].

·       Further expansion of the tenseaspect system for modal stems.

·       Simplification with progressive fusion of aorist or perfect stems in an ancient preterite.

·       (Late) trend to develop compound preterites, formed by adding an auxiliary stem to the main stem. The second stem is found made from root *-bheu- ‘become’ in Italic and Baltic; and more controversial *-dhē- or *-dō- ‘do’ in Sabellic (Piwowarczyk 2011) and Germanic, as in Greek; as well as *-es- ‘be’ in Slavic, and possibly in the Latin perfect (Yoshida 1988). These variable endings suggest a common ancestral innovation of the European migrants with alternating formations initially, i.e. not integrated as grammatical desinences into the verbal system.

o   The dual progressively loses its limited scope, accompanying the nominal declension.

Vocabulary and culture (Benveniste 1969):

·       NWIE has the richest set of shared lexical isoglosses connecting any Late Proto-Indo-European branch (see below §3.2.7. Statistics of lexical isoglosses), and many innovative stems are closely intertwined with its morphophonological innovations. The introduction of agricultural nouns from non-Indo-European languages points to a shared linguistic community, until its eventual separation into cultures already incorporating limited agriculture, in contrast with the previous herding-based economy.

·       Instruments in the field of nomina agentis in *-lo-, cf. OHG scūvala ‘shovel’, Lat. capulus ‘handle’, Lith. bar̃škalas ‘rattle’, etc. See below for example NWIE *tekt-lā ‘axe’.

·       The term *pekū supports the maintenance of (or specialisation into) a livestock economy based mainly on cattle, unlike Graeco-Aryan dialects, which include sheep and goats within the term: cf. Lat. pecū, Gmc. fihu. The appearance of farmers in Greek, geōrgói, and the Iguvine Tablets castruo frif (Lat. castra fructus) point to the late adaptation of the tripartite function of society to the incorporation of agriculture in the subsistence economy. The late relevance of agriculture may be possibly inferred from the increased relevance of the stormgod, *perkwunos, built from the root for ‘oak’.

·       The trifunctional sacrifice includes the pig, instead of the Graeco-Aryan goat, probably (at least initially) as the less valuable animal of the traditional three; so e.g. in the Roman suovetaurilia, in the Lusitanian inscription from Cabeço das Fráguas, in East Slavic fairytales, in a stone art of Bilbilis in Celtiberia, and in archaeological remains of sheep, oxen and pigs hoarded together in Scandinavian sites (Prósper 1999). The new economy including swineherding was probably adopted in contact with south-east European cultures, since it is also found in the Greek trittoíai.

·       Expansion of the terms (and concepts of) *ghostis ‘guest’, and *ghosti-potis ‘guest-host’ i.e. ‘host’ (see below § Remade Late PIE stems), closely linked to the use of verbal root *mei-t- ‘exchange’, noun *moinos ‘common’, and *keios ‘household’, probably due to the increased relevance of guest relationships and gift exchange in the NWIE society, compatible with the incorporation of Bell Beaker traditions to the classic IE traditions of exchange and reciprocity.

·       The chiefs of the political and military groups become increasingly formed in *-nos rather than *-potis: cf. Lat. dominus, ‘chief of the house’, tribunus ‘chief of the tribe’; Goth. kindins < *genti-nos ‘chief of the gens’; Goth. druhtins, OHG truhtin ‘chief of the escort’; Goth. þiudans < *teuta-nos ‘chief of the people, king’.

·       The patrilineal society continues, in contrast with Indo-Iranian, with the use of *nets as ‘nephew (usually the son of the sister)’, cf. Lat. nepōs, Cel. *nefot, Gmc. *neφan-, OLith. nepuotis, Sla. *netĭĭ (< *neptios). The presence of this meaning in Greek a-neptios, and conservation of the meaning ‘grandson’ up to the Latin period points to the survival of the custom of marriage between cross-cousins at least until the separation of the different branches.

·       A shared ancestral folk tale in NWIE (da Silva and Tehrani 2016) is “The Grateful Animals” (MFTD 554), whereby a youth earns the thanks of several animals (ants, fish, etc.) and with their help wins the princess by performing three tasks imposed upon him (brings a ring from the bottom of the sea, etc.). More shared tales appear in West Indo-European languages.

Close contacts with Uralic languages in terms of shared vocabulary, especially in the Pre-NWIE and Post-NWIE stages (but apparently not during its classical stage) further contribute to locate the community in space and time.

3.2.3. Early European isoglosses

Early lexical isoglosses shared with Palaeo-Balkan languages include the following:

·       NWIE *aik-tlo- ‘point of a spear, arrow’, from PIE *aik- ‘barb’, in Swe. egel, äjel, OPru. ayculo, Russ. iglá; compare also in *-smo- Gk. aikhmḗ, OPru. aysmis, Lith. (j)iešmas.

·       NWIE *ank-ro- ‘valley’, cf. Lat. ancrae f.pl. ‘valley, gorge’, Gmc. *angra- ‘lowland, meadow, enclosed piece of land’, and Gk. ángos n. ‘valley’.

·       NWIE *ar-ie- ‘plough’, cf. Lat. arāre, OIr. airid, Goth. arjan, ON erja, OHG erien, Lith. árti (ariù), Ltv. at, OCS orati; here also Gk. aróō ‘plough, plant’. It is assumed that Hitt. arrai - arranzi ‘grind, splinter up, crush’ is related, showing the original meaning of the PIE root (Kloekhorst 2008).

·       NWIE *as- ‘ash-tree’, basis for stems in *-n- (cf. Lat. ornus, MIr. onn, MW onn, also extended OIr. uinnius, Russ. jásen’), in *-k- (cf. Gmc. *aska, Alb. ah, Arm. hac’i), in *-i- cf. OPru. woasis, Lith. úosis.

·       NWIE *bheidh-éie- ‘to force’, cf. Gmc. causative *bīdan- ‘wait’, and for its use in mediopassive ‘be persuaded’ > ‘to confide in, trust’, cf. Lat. fīdere, Gk. peithomai.

·       NWIE *bhāgo- ‘beech’ in Lat. fāgus, Gaul. *bāgos, Gmc. bōk(j)ō; cf. Gk. phāgós, Dor. phagós ‘oak’. Maybe here also Russ. boz ‘elder’.

·       NWIE *bhi-lo- ‘(one’s) equal’, cf. OIr. bil ‘good’, Gmc. bila- ‘equal, even’, also Gk. phílos ‘friendly, dear, related, own’.

·       NWIE *bh-no- m. ‘ball, sack; member, penis’ (from *bhel- ‘to swell up’), cf. Lat. follis ‘bag, sack; ball, testicle’, OIr. ball ‘member, penis’, W balleg ‘sack, purse’, ON bollr, E ball. Here also Gk. phallós ‘penis’.

·       NWIE *deuk-e- ‘pull’, in Lat. dūcere, Osc. duc-/doc-, and Gmc. *teuhan-, as well as MW dwc < *duk-e, Toch. A. śuk, tskāt, and also Alb. n-duk-. Compare in zero-grade with suffix *-ie- mediopassive Gk. da-dússomai ‘is ripped’. Its wide distribution in Tocharian, Italic, Celtic, and Germanic makes it most likely an ancient western isogloss.

·       NWIE *gerbh-e- ‘carve’, cf. OIr. cerbaid ‘hack; lacerate’, Gmc. *kerban-, OCS žrěbŭ, žrêbij, Russ. zérebej; also zero-grade Gk. gráphō, ‘to scratch, carve, write’.

·       NWIE *gleubh-e- ‘carve, cut out’, in Lat. glūbō, Gmc. *kleuban; also Gk. glúphō, OCS glǫbokŭ, Ru. glubókij ‘deep’.

·       NWIE *glā /glinā ‘glutinous substance, clay’, cf. Gmc. klaia-, Lith. gléinė, Russ. glej, glína; also Gk. glía, glínee, MArm. kałǰin.

·       NWIE *gōb-o- ‘barley’, cf. Lat. gōbius, Gk. kōbiós ‘goby, gudgeon’, SCr. gûb, Sln. gúba ‘barbel’.

·       NWIE gwel- ‘wound’, cf. OIr. at-baill, W a-ballu ‘die’, Gmc. *kelan-, Lith. gélti ‘sting, hurt’, Ltv. dzelt ‘to sting’; also Arm. keł ‘wound, sore, ulcer’.

·       NWIE *kan-e- ‘sing’, cf. Lat. canō, OIr. canaid, OW canam, preserved in nouns in *kan-o- ‘singer’, cf. Gmc. *xanan- ‘rooster, singer’, Gk. ēi-kanós ‘rooster’, MDu. -hane ‘singer’.

·       NWIE *kiker- ‘chick pea’, cf. Lat. cicer, OPru. keckers, also Gk. kíkerroi, Arm. sisen, Alb. thjer(r), thíerr.

·       NWIE *kna-e- ‘scratch’, cf. OIr. -cná, W cnoi, OHG nōen, Lith. knója, Gk. knaíō.

·       NWIE *klep-e- ‘steal’ cf. Lat. clepere, Gmc. *xlefan-, in *-t- in Gk. kleptō. Derived from European root *klep-, compare derivatives in *-ni- OIr. cluain ‘deceit’, agentive in *-tu- Goth. hliftus, p. part. in *-to- OPru. au-klipts ‘concealed’, and noun in *-ā- Cz. klopa ‘lapel, flap’.

·       NWIE *kwap-o- ‘smoke, steam’, cf. Lat. vapor, Goth. -hvapjan, MHG -wepfen, Lith. kûpêt; also Gk. kapnós.

·       NWIE *lap-na- ‘lick’, cf. Lat. lambere, Gmc. *lappōn, *labbōn, Lith. lapènti, Russ. lópat’ ‘gobble up’; further Gk. láptō, Alb. lap ‘lick up water’.

·       NWIE *leudhis ‘(free, common) people’, cf. ON ðr, OHG liut, Lith. liáudis, Ltv. ļaudis; and *leudh-eros ‘free’, cf. Lat. līber (<Ita. *louðeros), Gk. eleútheros.

·       NWIE *leug- ‘bend’, cf. OIr. -loing, Goth. -lūkan, as well as Lat. luctāre, Lith. lùgnas, Gk. lúgos (see below also NWIE *leugh- ‘lie’).

·       NWIE *loubh- ‘rind(?)’, cf. Lat. liber ‘bark, rind’, Gmc. *lauba- ‘leaf, foliage’, Russ. lub ‘bast’, Lith. lubà ‘blank’, OPru. lubbo ‘bast, plank, shelf’; further Alb. labē ‘rind, cork’.

·       NWIE *lokus ‘lake, pond, pool’, cf. Lat. lacus, OIr. loch, Gmc. *lagu-, OCS loky; further Gk. lákkos ‘pond, cistern, pit, reservoir’ < *k-o-.

·       NWIE *lu(n)k- ‘lynx’, cf. MIr. lug, OHG luhs, Lith. lū́šis, dial. lųnšis, OPru. luysis, Russ. rýsь; also Gk. lúnks, gen. lunkós, Arm. *lusan-n.

·       NWIE *mori-/mrei- ‘sea’, cf. Ita. *mari, Cel. *mori, Gmc. *mari, BSl. *marja, and maybe Arm. mawr (possibly in combination with *meuH- ‘make wet, wash’). Compare with potential eastern derivative PIIr. marā́dā ‘shore, limit, boundary’.

·       NWIE *musa- ‘fly’, cf. Gmc. *musī-, OPru. muso, Lith mùsė, Ltv. muša, OCS muxa, Russ. múxa, or Gk. muia; also Lat. musca < *mus-kā-.

·       WIE *not-o- ‘back’ cf. Lat. natis ‘rump, buttocks’, Goth. *nota ‘rear of a ship’, cognate with Gk. nōton n. ‘back’.

·       NWIE *oi-no- ‘one’, in Ita. *oino-, Gel. *oino-, Gmc. *aina-, Bal. *oino-; cf. also Gk. oinḗ, OCS inŭ ‘some(one), other’. Derived from root *oi-, compare cognates Skt. éka- < *oi-ko-, Av. aēuua-, OPru. aiva < *oi-o-.

·       NWIE *oia ‘type of tree’, cf. Lat. ūva ‘grapes’, OIr. ‘yew’, OHG iwa ‘yew’, OPru. iuwis ‘’yew’, Lith. ‘bird-cherry’, Russ. íva ‘willow’, Cz. jíva ‘willow’; compare also Gk. ē ‘service-tree’, Arm. aygi ‘vineyard; vine’.

·       NWIE *pā-no- ‘cloth’ cf. Lat. pannus, Gmc. *φanan-, Gk. pḗnē; maybe also here OCS o-pona ‘curtain’.

·       NWIE *pā-u̯o- ‘small, few’ cf. Lat. parvus, paucus, Gmc. *φaa-, Gk. pauros.

·       NWIE *pel-en- ‘skin’, cf. Lat. pellis, Gmc. *φella-, OCS pelena, and Gk. pello-ráphos ‘sewing skins together’.

·       NWIE *plṓ-e- ‘swim’, cf. Gmc. *φlōan ‘flow’, OCS plaviti ‘to let swim’, PGk. *plṓu-e- ‘to swim’.

·       NWIE *sal-ik- f. ‘willow’, cf. Lat. salix ‘willow, osier’, OIr. sail, gen. sailech, MW pl. helyg, MBret. halek, Gaul dat. salico-genne, Gmc. *salixōn- ‘willow, sallow’ (ON selja, OE sealh, OHG salaha). Compare *selik- in Gk. Myc. e-ri-ka /helikā/, Arcadian helíkē ‘willow’ < *sel-ik-ā.

·       NWIE *sēmen ‘seed’, cf. Ita. *sēmen, *sēmō (collective), Gmc. *sēman-, BSl. *sēmen, also found in Gk. hēma (<*sē-m). From common PIE root *seh- ‘press in, introduce → sow’, cf. Hitt sai- ‘press in, sow’.

·       NWIE *skabh-e- ‘scrape’, cf. identical Lat. scabere ‘to scratch’, Gmc. *skaban ‘to shave, scrape’, and full-grade Lith. skóbti ‘hollow out’; further Gk. skáptō ‘to dig (out), work the earh’ < *skabh-ie, and OCS skoblĭ, Russ. skóbelĭ ‘plane’.

·       NWIE *sleim-ak- ‘slug’ cf. Lat. līmāx, Russ. slimák, Gk. leímaks; further Gmc. *slīma- (probably also here *slīu̯a- ‘tench’), Lith. sliēnas m.pl., OCS sliny.

·       NWIE *smerd- ‘hurt’ as basis for Lat. mordēre ‘bite’, Gmc. smertan ‘hurt’, Lith. smirdė́ti, OCS smrŭti ‘smell’; also Gk. smerdnós ‘terrible, dreadful’, Arm. mart ‘fight, battle’.

·       NWIE *(s)nē- ‘spin’, cf. Lat. nēre, OIr. sníid, W nyddu, Gmc. nēan-, Lith. nýtis, Ltv. nĩtis, Russ. nit’; further Gk. neō.

·       NWIE *tpel-u̯ā- ‘willow, elm’ cf. Lat. tilia, Gmc. *φelo, possibly Russ. topol’; also Gk. pteléa, Arm. t’ełi.

·       NWIE *teutā ‘people’, the most common ethnic self-designation in European languages, cf. Ita. *touta, Cel. *toutā, Gmc. *ϑeudō, Ltv. tàuta, Lith. tautà, OPru. tauto ‘country’, or Messapic taot-; also found in Balkan languages, cf. PAlb. *tetan ‘all; people’, Phryg. acc. pl. teutous, or “Illyrian” Teutana ‘queen’. Possibly a PIE word that evolved from a more general meaning, cf. the uncertain cognates Hitt. tuzziš ‘army, camp’, Sogd. twð’k ‘crowd’, Persian tōda ‘heap, pile’.

·       WIE *ogw- ‘wet’, cf. Pre-Ita. *og(w)-edho-, Pre-Cel. *ogw-lo-, Pre-Gmc. *ogw-o-, PreGk. *ug(w)-ro-.

·       NWIE *ogwh-ni- ‘ploughshare’ cf. Lat. vōmer, Gmc. *agnis-an, OPru. wagnis, also Gk. ophnís.

·       NWIE *rad-i- ‘branch, root’ cf. Lat. rādīx, Cel *radi-o-, ON rót, Goth. waurts, also Gk. rhā́dīks Alb. rrënj/ë, -a, maybe Toch. B itsako.

·       NWIE *trozd-u- ‘thrush’, cf. Lat. turdus (<*torzd-o-), Gmc. *ϑrastu-, Bal. *strozd-o, Sla. *drozd-o-. Compare with Arm. tordik ‘thrush’ < *dorzdh-.

3.2.4. Classical North-West Indo-European

Lexical isoglosses not shared with other IE languages include different categories (Oettinger 2003): Remade Late PIE stems

Most likely a term inherited from Late PIE, but remade in form or meaning from the same or other root:

·       NWIE *bhei- ‘hit’, cf. OLat. perfines ‘you shall hit’, OIr. -ben, -benat, perf. , béotar, Gmc. *bi()ēn-, OCS biǫ, bi, bišę. Stem not found in other IE languages, although maybe here Alb. (m-)bin.

·       NWIE *bhor-e- ‘fight’, in Lat. ferīre ‘strike’, Gmc. *baran- ‘fight’, OCS brati ‘fight’, Russ. borót’ (borjú) ‘overpower, throw to the ground’; here also Lith. bárti (barù), Ltv. bãrt ‘scold’ (in *-e-), and OIr. barae ‘anger, hostility’ (in *-n-). Probably also related derivatives in *-ā- Lat. forāre, Gmc. *burōan-.

·       NWIE *dhghmōn ‘man, human’, usually reduced to *ghmōn, cf. Ita. *hemō (OLat. hemō, Osc. humuns, Umbr. homonus), Gmc. *gum-an-, Bal. *ǯmō̃ (OLith. žmuõ, OPru. smoy). A different reduction is found in OIr. duine < *don-i̯os, probably from metathesised form *gdon-i̯os from extended **ghdhmon- + *-i̯os. Comparable with LPIE *deios ‘god’, from *deus ‘sky’, hence “celestial”, as opposed to this term “earthly”, from *dhghem-, ‘earth’.

·       NWIE *ghóstis ‘guest’, cf. Ita. *hostis, Cel. *gostis, Gmc. *gastiz, Sla. *gostь, probably from *ghes- ‘eat’. Also found in compound *ghost(i)potis, ‘host’, cf. Ita. *hostipotis, Sla. *gospodь. Comparable with parallel Skt. átithi-pati-.

·       NWIE *gwhor-mós ‘warm’, cf. Ita. *formos, Gmc. *warmaz, OPru. gorme (and gormapada), Lith. gorme, Ltv. gar̂me. In contrast, compare Palaeo-Balkan *gwher-mós in Gk. thermós, Phryg. Gérmee, Arm. ǰerm, PAlb. *dźerm-. PIIr. gharmá- is probably a parallel development of an o-stem, pointing to an original LPIE paradigm *gwhór-mo- ‘heat’, *gwhr-mós ‘warm’.

·       NWIE *ghórdhos ‘enclosure, fenced circle’, cf. Gmc. gardaz, BSl. *gardas, also in Toch. B. kerccī < *gʰórdʰijo-; dubious is the origin of Alb. gardh ‘fence’. The change is the development of an o-stem from *gʰr̥dʰós, cf. Hitt. gurtas, PIIr. *gʰr̥dʰás.

·       NWIE *kéios ‘household’, cf. Ita. i-stem *keiis, Gmc. hīu̯on ‘married couple, household’, Ltv. sive ‘wife (< household)’. Shift in meaning to the extended family group, including personal relationships such as the spouse and ‘friends’i.e. those with reciprocal (moral) obligationscomparable to the Greek philos (Benveniste 1969). Compare for the original meaning Skr. śéva- (<ĉáias) ‘dear, precious, friendly’. Compare also *koimos ‘home’, cf. OIr. cóim ‘dear’, Gmc. *xaimas (probably borrowed in Baltic, cf. Lith. káimas, Ltv. cìems, OPru. caymis), Gk. *koimo- ‘bed’.

·       NWIE *-er-os ‘northern (wind?)’, from CIE *k(e)hu-er-o-, cf. Lat. caurus, Lucr. cōrus (< Ita. *kaero-) ‘north wind; the north-west wind’, Lith. šiàurė ‘north’, šiū́ras, šiūrùs ‘cold, northern’, OCS sĕverŭ ‘north’; from the same root MW cawad ‘gust, shower’, OBret. couatou ‘showers’ (<Cel. *ke-Vt-), which confirms that -r- belongs to the suffix. Goth. skūra windis ‘storm (of wind)’, OHG skūr ‘rain shower’ and Arm cՙowrt ‘cold’ are probably not related.

·       NWIE adj. *mói-nos ‘common, shared’, from *mei- ‘exchange’, cf. Ita. *moi-ni-, *moinos ‘duty, obligation, task’, Gmc. *mainas ‘common’, Lith. maĩnas ‘exchange’, OCS mĕna ‘exchange, substitution’. Also here *kom-moinos ‘common, general’, cf. Lat. commūnis, Goth. ga-mains, OE ge-mǣne, OHG gi-mein, etc. Possibly here also verb *moi-t-éie- ‘exchange’, cf. Lat. mutāre, Goth. maidjan.

·       NWIE *pérkwus ‘oak-tree’, cf. Ita. *kwerkus, Cel. *φerkus, Gmc. *φerxuz; with possible derivative *pérkwūnos, the stormgod (with wife pérkwūnā), a main god of the European pantheon, possibly linked to the stronger agricultural character of the new population; cf. Cel. Ercunia (< fem. Φerkuniā, probably a forest goddess), ON. Fjörgyn (‘Earth’, moder of Thor), probably also here Gmc. *φerkuna-  ‘mountain’, Lith. Perkūnas (borrowed in Finn. Perkele, Mordv. Pur’gine-paz), and Thrac. Perkōn/Perkos. Possibly remade are Sla. Perun, Alb. Perëndi. For a comparison probably here belongs (with the same obscured phonological developments of mythological names, usually caused by taboos), OInd parźána ‘rain god’; interesting also Gk. keraunós ‘thunderbolt’. See below for the god’s magical hammer *meldh-n-, lightning, and for the WIE epithet *tr̥ˀnos, ‘thunder’.

·       NWIE *pol-o- ‘pale’, cf. Lat. pullus, Gmc. *φala-, Lith. pal̃vas, OCS plavŭ. The uo-suffix for colour names is used in this European stem, instead of that found in Pre-PIIr. *pelu-so-.

·       NWIE *réidh-e- ‘ride, hurry’, in Cel. *rēd-o- (cf. OIr. réidid, W rhwydd-hau ‘to hurry’), Gmc. *rīdan-, Lith. riedėti. Also found as ‘ride’ with sexual connotation, cf. Gmc. *ridra- ‘penis’ with instrumental ending (Kroonen 2013). Classified as of unknown origin in Oettinger (2003), it seemsat least phonologically and morphologicallya remade Indo-European root (Kümmel et al. 2001). The stem seems related to root *rei- ‘arrange’, whose semantic evolution includes the meaning ‘ready, quick’, then to ‘easy, simple’ cf. NWIE *reidh-i- e.g. in OHG bi-reiti ‘ready’, OIr. réid, ‘even, light, easy’, OW ruid, Lith. raidùs ‘ready, quick’ (Matasović 2009).

·       NWIE *rótos (<DIE *Hró-tos) ‘wheel’, cf. Ita. *rotā (collective) Cel. *rotos, Gmc. *raϑaz, Lith. rãtas, Ltv. rats. The Proto-Balto-Finnic loan may have been Germanic or Baltic. Other languages (including NWIE) show other derivatives from *Hret- ‘roll’.

·       NWIE *eik- ‘defeat, conquer’, cf. Lat. vincō, OIr. -fich, -fechad ‘fight’, ON vega, Goth. -waih ‘fight’, Lith. veikiù; contrasting with Graeco-Aryan root, cf. Skt. źáati ‘defeat’, Gk. ā ‘violence’. Peculiar semantic or sound developments

New terms with peculiar semantic or sound developments:

·       NWIE *dreu-o- ‘certain’, from ‘tree, wood’, hence ‘strong’, cf. OIr. derb ‘certain’, Gmc. *treu̯u̯u- ‘loyal, trustworthy’, trūēn- ‘trust’, OPru. druwit ‘to believe’. For a similar shift of meaning, compare Lat. rōbustus adj. ‘made of oak; strong’, from rōbur ‘oak, strength’.

·       NWIE *gˀnóm ‘grain’, cf. Ita-Cel. *grānom, Gmc. *kurną, BSl *źirˀna-.

·       NWIE *londh- ‘open land, waste’, cf. Cel. *landā (<*ldh-), Gmc. *landa- (*londhom), OPru. lindan (<*ldh-), maybe here Russ. ljadá. Probably originally from an ablauting paradigm nom. **lōndh-s, gen. ldh-os, Acc. londh- (Matasović 2009).

·       NWIE *selbh-o- ‘self’, in Gmc. *selba(n), Venetic sselboi-sselboi ‘to oneself’ (reduplication similar to Lat. ipsipse). The lack of relation of both languages probably points to a common North-West Indo-European origin. It has been suggested that this stem is derived from *s(u)e- ‘self’.

·       NWIE *(s)kel- etc. ‘commit a crime, be guilty’, in Lat. scelus ‘crime’, Goth. skal ‘be guilty, must’, Lith. skelù ‘be guilty’.

·       NWIE *stabhos m. ‘beam’ in identical roots behind OIr. sab ‘shaft, pole’ Goth. stabos ‘letters’, ON stafr ‘staff; stave’, OE stæf ‘staff, stick; letter’, OHG stap, LIth stābas ‘post’. The original Cel.-Bal. correspondance in *-a points to it as the original vowel; further in *-ro- cf. Swe. staver ‘fencepost’, ODa. stavær, OCS stoborŭ ‘pillar’.

·       NWIE *steup- ‘to bend’, in W ystum ‘bend, turn’, Bret. stumm, Gmc. *stūpēn ‘to stoop, to bend’; Ltv. staūpe ‘horse track’ (cf. Nor. staup ‘puddle; deep track; sharp turn’).

·       NWIE *odh- etc. ‘bail, surety’, cf. Lat. vas, Goth. wadi, Lith. vãdas, etc. Combinations from roots and affixes

New combinations from roots and affixes:

·       NWIE *al-no- ‘all, whole’, cf. Osc. allo, OIr. uile, Goth. alls, ON allr, Lith aliaĩ. New is the formation in -no- from LPIE *al-, ‘other’, with controversial original laryngeal (*ħ or *ʕw) and vocalism (*a or *o).

·       NWIE *áksis ‘axis’, cf. Lat. axis, W echel (< Cel. *aksi-lā), BSl. *aśis (Lith. ašis, Sla. *osĭ). Common is the formation in *-i.

·       NWIE *bhā-e- ‘speak’, cf. Lat. fārī, OE bōjan, OCS bajati, bajǫ. Common is the suffix *-e.

·       NWIE *bhlē-e- ‘to bleat’, cf. Lat. flēre, Gmc. *blēan, Ltv. blêju, ORuss. blějati.

·       NWIE *ghórnos ‘gut’, cf. Lat hernia, Gmc. *garnō, Lith. žárna. Common is the noun in *-no- (extended in Lat. *n-o-).

·       NWIE *gwhou̯ēe- ‘watch; be considerate of; worship’, cf. Lat. faveō, ON , OCS govĕjo, govĕti. Common is the combination of suffix *-ēe-.

·       NWIE *gwhréndh-e- ‘grind’ cf. Lat. frendere, OE grindan, Lith. gréndžiu <*-ie- ‘to plane, scour’.

·       NWIE *dhul-(n)o- ‘blind; fool’ cf. Cel. *du̯allo- ‘blind’, Gmc. *dula- ‘foolish, crazy’, Ltv. duls ‘furious’.

·       NWIE *dhus-e- ‘to lose one’s senses’, in Lat. furere ‘be mad, rave’, Gmc. *dusēn- ‘to slumber’, Lith. dūstù ‘suffocate’, Ltv. dust ‘to gasp’, OCS. duxŭ ‘breath’, Lith. dùsas ‘short breath, asthma’; cf. Cel. *du̯allo- ‘blind’, Gmc. *dula- ‘foolish, crazy’, Ltv. duls ‘furious’. Here also *dhus-k- ‘obscure’, cf. Lat. fuscus, Gmc. duska- ‘dark’, and also with different suffix *dhus-no- ‘brown’, cf.  OIr. donn, OHG tusin, OE dosen.

·       NWIE *ēdskā ‘food, feed’, cf. Lat. ēsca, Lith. ėskà, Ltv. êska.

·       NWIE *k(V)l-ni- ‘mountain path’, cf. Lat. callis, Sla. *kol-ni-ki (cf. Serb. klánac, Czech klanec).

·       NWIE *k-men- ‘hill’, cf. Lat. culmen, columen, ON holmr, holmi, OE holm, Lith. kálnas.

·       NWIE *kā́ros ‘dear; love’, cf. Lat. cārus, OHG huor, Ltv. kãrs.

·       NWIE *ker-n- ‘wild boar’, from an original ‘horn’, cf. OIr. craín ‘sow’, W cranan ‘wild sow’, Lith. šer̃nas, Old Late Frankish chranni-chaltia ‘pig’s den’.

·       NWIE *kwegh-ne- ‘crouch down, flinch, be startled’, cf. Lat. conquiscō (perf. -quēx- < *kwēgh-s-), Gmc. *xekkan-, OCS čeznǫti, is-čeznǫti.

·       NWIE *kŕ̥sro- ‘hornet’, cf. Lat. crābrō, OE hyrnet, Lith. šìršė, ORuss. s(t)rŭšenĭ etc.

·       NWIE *krōpo- m. ‘shed’, cf. OIr. cró, MW creu, Gmc. *xrōφa-, OCS stropŭ.

·       NWIE *kūtis f. ‘skin’, cf. Lat. cutis, Gmc. diz; further OPru. keuto, Lith. kiáutas, and also MIr. codal, Lith. kiáuklas.

·       NWIE *lāmā ‘bog, hollow’, cf. Lat. lāma, Ltv. lâma, OCS lam.

·       NWIE *lentos ‘soft’, cf. Lat. lentus ‘pliant, flexible; tough; sticky; slow’, Gmc. *linϑa ‘flexible, soft’, Lith. leñtas ‘quiet, calm’.

·       NWIE *meldh-n- ‘lightning; thunder weapon of the stormgod’, cf. Gmc. *meldunaz, W mellt, Ltv. milna, OCS mlŭnĭĭ.

·       NWIE *mto- ‘dead’, in Lat. mortuus, Sla. mrŭtvŭ.

·       NWIE *oinoko- ‘unique’, in Lat. ūnicus, OS ēnag, etc. OCS inokъ.

·       NWIE *oketā ‘harrow’, cf. Lat. occa, OW ocet, Gmc. *agiϑō- Lith. akěčios, OPru. aketes.

·       NWIE *pelen-, pn- (ablauting) ‘fine flour, milldust’, cf. Lat. pollen, OPru. pellane, Lith. pelenaî, Ltv. pęlni.

·       NWIE *plek-te- ‘to plait, braid’, in Lat. plectere, Gmc. *φlextan-, OCS pletǫ, plesti. Extension of the root *plek-, compare Gk. plékō, Skt praśna-.

·       NWIE *porkelo- ‘piglet, hoglet’, cf. Lat. porculus, OHG *φarxeli, Lith. parselis. Common is the newly formed diminutive, which turns NWIE *porkos into a more general meaning ‘pig’, evolved from the older LPIE ‘piglet’ found in PIIr. *párĉas. Similarly, from LPIE sūs ‘pig’, diminutive *sīnos is formed, cf. Lat. suīnos, Gmc. *sīnaz, Ltv. sivēns, Slav. *svinŭ, *svinĭa.

·       NWIE *pˀ- ‘first’, cf. Lat. *prāmo- (in prāndium), Faliscan pramo, Gmc. *φurmaz, Gaul. ramus, Ligurian pramion, Lith. pìrmas. Common is the suffix -mo-, which sets it appart from other LPIE developments in *-o- (Indo-Iranian), *-to- (Greek), etc.

·       NWIE *reudh-o- ‘red’, cf. Lat. dial. rūfus, rōbus, Umb. rofu (acc. pl.), OIr. rúad, W rhudd, Goth rauþs, Lith. raũdas, OCS rudŭ; contrasting with eastern DIE dialects forming it in *-r-, cf. Toch. B ratre, Skt. rudhirá- ‘bloody’, Gk. eruthrós.

·       NWIE *rudhēe- ‘be red’ (‘become red’), cf. Lat. rubeō, OIr. ruidi, OHG rotēn, ORuss. rŭdeti se, Lith. ruděti.

·       NWIE *salo- ‘drab, dull-brown or -gray’ cf. W salw, ON solr, ORuss. slavo-očie.

·       NWIE *souk-nā- ‘suck’, in identical Lat. sūgere (cf. sūcus ‘juice’), Gmc. *sukk/gōn; further Ltv. sùkt (notice lack of satemisation), OCS sŭsǫ, Russ. sosú.

·       NWIE *stengw-e- ‘to push back’, cf. Lat. re-stinguere ‘to push back, suppress’, Gmc. *stinkan ‘to thrust, clash; to stink’, Lith. sténgiu ‘to exert oneself’. Nasal infix present from root *stegw- (Kroonen 2013).

·       NWIE *stor-on- ‘starling’, cf. Lat. sturnus, Gmc. *staran-, also Cz. sternad ‘bunting’. Comparable with zero-grade root in *-lo- in Gk. astralós ‘starling’.

·       NWIE *strenghe- ‘be stiff, tighten’, cf. Lat. stringere, Gmc. adj. *strang()a, *strangi, *strunka-, *strunga- (which imply an underlying verb *strengh-e-), Lith. stringù ‘to stick’, Ltv. stringt ‘to stiffen; to wither’, adj. strangs ‘brave’; Pol. za-strząc ‘to come to a halt’.

·       NWIE *strou-eie- ‘to strew’, cf. Lat. struere ‘to arrange, construct, compose, build’, OIr. sruïd ‘to throw’, OBret. strouis ‘to strew’, Gmc. *strauan ‘to strew’ (cf. Goth. straujan, ON strá, OE strēowian, OHG strewen, etc.), OCS o-struǫ ‘to destroy’. NWIE causative or intensive formation from LPIE *str-neu- ‘to strew’, compare Lat. sternere, Gk. stórnūmi, Skt. stnā́ti.

·       NWIE *tekt-lā ‘axe’, cf. Lat. tēlum, OIr. tál, OHG dehsala, ORuss. tesla. Common is the formation in *-lo- in the field of nomina agentis, see above.

·       NWIE *tlokw-e- ‘talk’, cf. Lat. loquor, OIr. ad-tluchedar ‘gives thanks’, do-tluchedar ‘demands’, Russ. tolk ‘sense, meaning’.

·       NWIE *erb-es- ‘twig, rod’, cf. Lat. verbera pl. ‘(rod for) punishment’, verbēnae pl. ‘twigs of the laurel’, Lith. vir̃bas ‘twig, rod’, Ltv. vir̃bs ‘thin stick’, Pru. arwarbs ‘langwyt’, Sla. vĭr̃ba ‘willow / Salix’. From Balto-Slavic borrowed into Proto-Balto-Finnic, cf. Finn. virpa, varpa, Est. viŕb, varb. From PIE root *χwerb-, *χwb- ‘willow’, cf. Hitt. urpasta(n-), urpusta-, ‘leaf, peel’, Gk. rhábdos (*rhabos) ‘twig, rod, staff’. Possibly related to an Afroasiatic root, cf. Semitic *ɣurab (cf. Akk. urbatu-, Hebrew ʕărābā ‘willow, Euphrates poplar’), Egyptian ʕзb “a kind of tree” (Blažek 2018).

·       NWIE *u̯ēros ‘true’, cf. Lat. rus, Cel. *īros, Gmc. *eraz, OCS vĕra. Likely cognates Gk. ē͂ra ‘please’ and Hitt. uarr- ‘help’ show a different, probably original meaning.

·       NWIE *mis ‘worm’ cf. Lat. vermis, W gwraint, Gmc. *urmiz, Lith. var̃mas, OPru. wormyan, warmun, Sla. vĭrmĭe; probably here Alb. rrime. Common the formation in *-is, compare Gk. rhómos, OArm. ordn (uncertain). Cognates in Indo-Iranian and Armenian include a proto-form *kwmis. Different function from LPIE

New words substituting common PIE terms for stems with a different function in Late PIE:

·       NWIE *bhars, *bharsinā ‘barley’, cf. Lat. far, farinā ‘spelt, flour’, Faliscan far, Cel. *bargos, *barginā ‘cake, bread’, Gmc. *barizaz, *bariz-īnaz, Sla. *barsina-. Instead of LPIE *eo- ‘barley, cereal’.

·       NWIE *bhardhā ‘beard’, cf. Lat. barba (<Ita. *farβā), Gmc. *bardaz, BSl. *bardā́. Instead of LPIE *smokru-.

·       NWIE *bhergh-e- ‘protect’, cf. OIr. commairge < *kom-bhorgh-iā-, Gmc. *bergan, OCS brě ‘care’, Lith. birginti ‘save’. Probably connected to *bhergh- ‘mountain’ hence ‘to take high ground’, whence ‘to keep safe’. Balto-Slavic words are sometimes rejected because of the lack of satemisation in these forms (and thus derived from late dialectal innovation from *bhergh- ‘safegard’), although satemisation is late and irregular in Balto-Slavic (see below §4.13.1. Balto-Slavic evolution).

·       NWIE *bhlad- ‘invocation’, Lith. blódeti, Ltv. blãdêt, also in loanword Fi. luote ‘enchantment’; impossible to know the true source of Lat. flāmen ‘priest’, Gmc. *blōtan- ‘to sacrifice’, if *bhlad- or *bhlag- (see above § Ritual and religion).

·       NWIE *mais- ‘more’ cf. Osk. mais (adv.), W mwy, Goth. maiza, OPru. muisieson (adv.). Instead of Late PIE *pleh-is-.

·       NWIE *piskos, cf. Ita. *piskis, Cel. *ɸēskos, Gmc. *φiskaz, Sla. *piskor’, *piskar’. Contrasting with eastern isogloss *dhghu- (see below §4.13.3. Contacts with Palaeo-Balkan languages). Root variant or parallel root

New formations with a root variant or parallel root, usually further obscured with dialectal innovations on the newly created, unstable forms:

·       NWIE *ag-r-o- ‘tree fruit’ cf. Cel. *agn-io-, agr-on-a, Gmc. *akrana- (see above § Northern Indo-European for root *ag-).

·       NWIE *bhel- ‘henbane’ cf. Cel *belesa, Gmc. *bel-uon-, Sla. *bel-(e)n-o-.

·       NWIE *bhei- ‘bee’ as basis of forms in *-k-o- (cf. Lat. fūcus, OIr. bech, W begegyr, OCS bicela), in *-n- (Gmc. *bion-), or *-ti- (Lith. bìtė, OPru. bitte).

·       NWIE *bhreus- ‘breast, chest’, in *-t- (cf. OIr. brúasach, Gmc. *breusta-), in *-o- (cf. Russ. brjúxo), in *-n- (cf. OIr. brú, gen. bronn, MW bru, bron; OIr. bruinne).

·       NWIE *dhlgh-e- ‘owe’, *dhlgh-o- ‘debt’, cf. Cel. dligid, dliged, Gmc. *dulga, OCS. dlŭgŭ.

·       NWIE *el-o- ‘auk, swan’, as basis for Lat. olor < Ita. *elōr; OIr. elu < *el-jā? W alarch < *el-r-sko; Swe. al-, Ice. -ella, ON alka, OE ealce, apart from those in *el(bh)-, cf. Gmc. *albut, Sla. *elbedь, *olbądь, etc. The alternation *e/*a may point to the influence of *albhos ‘white’, on the root.

·       NWIE *geldh-e- ‘payment, compensation’ cf. OIr. gell < *geldo- ‘pledge surety, promise’, Gmc. *geldan < **gelde- ‘pay, be worth something’, OCS. žlěsti (žlědǫ) < *gelde ‘to pay, compensate’.

·       NWIE *gebh- ‘gnaw’, as basis for OIr. gop ‘muzzle, snout, beak’, Gmc. *kebra(n)- ‘gnawer’ in OE ceafor, cefer, OHG kevar(o) ‘beetle’, ODu kevera ‘grasshoper’, OCS o-zobati ‘to spoil’, Russ. zobát’ ‘eat, peck’, etc.

·       NWIE *g(e)r-s-e- ‘make sound’, basis for Lat. garrio ‘to chatter’, Gmc. *kerzan- ‘to creak, to cry (of birds)’, iter. *kurzōn-, Lith. gasas ‘sound, rumour, glory’.

·       NWIE *glebh- ‘round object (?)’, as basis for Lat. globus ‘round object’, glēba ‘lump of earth, clod’, Gmc. *kulba(n) ‘round object’, Lith. glébiu ‘to embrace, clasp’, Ltv. glêbt ‘to guard, protect’.

·       NWIE *ghlend-e- ‘shine, look’, cf. OIr. as-gleinn ‘to examine’, Gmc. *glintan ‘to shine, look’, Ltv. glen̂st ‘to (scarcely) perceive’, OCS ględati ‘to look at, see’ Russ. gliadét’ ‘to look at’.

·       NWIE *grem- ‘push’, cf. Lat. gremium ‘lap, bosom, armful’, Gmc. *krimman-, ‘crumble’, *krummōn- ‘squeeze’, Lith. grùmti ‘to push, shove, cram’.

·       NWIE *gos-dh-o/i- ‘nail, wheel hub’ cf. W both, Gaul. bottos, MIr. bot, OCS gvozdĭ.

·       NWIE *ī-lo- ‘empty land, clearing’, as basis for W ial, OE idel, OHG ital, Ltv. jêls, Russ. jályj, jálovyj.

·       NWIE *kank- ‘jump (related to horses)’, with derivative *kankisto- ‘stallion’, cf. W caseg, Runic /hangist-/, Lith. šankùs, šankinti.

·       NWIE *kerd/t- ‘gird’ cf. OIr. cris, crys, Russ. čeres, etc.

·       NWIE *kostā, *kost-i ‘rib, bone’ cf. Lat. costa, ORuss. kostŭ.

·       NWIE *krengho- ‘ring’, cf. Umbr. críngatro, ON hringr, OCS krǫgŭ.

·       NWIE *kors-e- ‘to card’, cf. Lat. carro <*krs-e-, Gmc. *xarza- <*kors-o- ‘flax’, Lith.  karsiù, Ltv. kārst < *kors-ie-.

·       NWIE *kout-no- ‘testicle’, cf. Lat. cunnus ‘vagina’, W cwd ‘bag, sack’, Gmc. *xauϑan- ‘testicle’, Lith. kutỹs ‘pouch’.

·       NWIE *kreis-, cf. Lat. crīnis ‘hair’, ON hrís ‘brushwood’, OE hrīs ‘branch, brush’, OHG hris ‘twig, branch; shrubs’, OPru. craysi ‘talk’, crays ‘hay’.

·       NWIE *kret- ‘tremble’ as basis for Ir. crothaim, ON hrata, OE hratian, MHG razzen, Lith. kretù, krintù, Ltv. krist, etc.

·       NWIE *k-ti- ‘door’, cf. Lat. crātis ‘construction of wickerwork, hurdle’, Gmc. *xurdi, from *ker- ‘hinge’, cf. Gmc. *xer(r)an-; here possibly OPru. corto ‘fence’.

·       NWIE *kendh-r- a type of plant, cf. Lat. combrētum ‘kind of rush’ < *kendh-r-, Ir. contran ‘water horehound’ < *kundh-r-, Lith. šveñdrai ‘cattails’ < *kendh-ro-, also here ON hvǫnn < *kondh-nā-.

·       NWIE *leugh-e- ‘lie’, cf. OIr. fol-lugaid, Gmc. *leugan-, OCS lŭgati. Possibly originally ‘bend (oneself)’ (Kümmel et al. 2001), see above *leug- .

·       NWIE *leig- ‘bind’, cf. Lat. ligāre ‘fasten, bind’, Gmc. *līka- ‘leech-line, bolt-rope’, Russ. dial. po-lyhaty-s’a ‘to connect’; maybe here Alb. lidh ‘bind, tie’.

·       NWIE *mazd-o-, *mazd-to- ‘wooden stick’ cf. Lat. mālus, OIr mátan, ON mastr, OCS mostŭ.

·       NWIE *mein- ‘think’, as basis for OIr. mían ‘desire, inclination’, Gmc. *mainan-, OCS měniti, probably also OLat. meinom.

·       NWIE *mus-o- ‘moss’, cf. Lat. muscus (< *mus-ko-), Gmc. *musa(n)-, Lith. mùsos, Russ. mox.

·       NWIE *oldh- ‘hollowed out tree trunk; boat’, basis for Gmc. *aldo(n)-, Sla. oldh-ia-. Probably an Indo-European root for boat, retained in Northern languages.

·       NWIE *(s)pen-d- ‘pull, draw’ cf. Lat. pendeō, OLith. spándau, OCS pęndĭ.

·       NWIE *pen-o- ‘swamp’ as basis for MIr. an, en, Gaul. anam, Gmc. *φana-, OPru. pannean.

·       NWIE *pe-pel- ‘butterfly’ cf. Lat. papilion-, Gmc. *φīφaldra, OPru. pepelis, penpalo, Lith. píepala, Ltv. paîpala, Russ. pérepel.

·       NWIE *per- ‘to be with young’ as basis for Lat. pariō, Gmc. *φarz- Lith. peréti; cf. Gk. poreĩn, OInd. pūrdhí ‘give’.

·       NWIE *pleu-d- ‘flow, drift’, cf. OIr. luaid-, ON fljóta, OE flēotan, Lith. pláusti, pláudžiu.

·       NWIE *pleu-s- ‘(wool) flock’ cf. Lat. plūma, OIr. , MHG vlies, Lith. pláuzdinis.

·       NWIE *plout-o- ‘transverse board’ cf. Lat. pluteus ‘movable screen of wood or wickerwork, parapet’, ON fleiðr ‘cross-beam’, Lith. plaũtas, Ltv. plàuts ‘shelf, sideboard’, SCr. plúto n. ‘flotsam’.

·       NWIE *pol-kā ‘arable land’, cf. Gaul. olca, Gmc. *φalgō-, also behind Pre-Bal. *plek-ie- ‘plough’ cf. Lith. plešiu, Ltv. plest.

·       NWIE *preg- ‘desire’, in OBret. pl. rogedou ‘orgies’, W rhewydd ‘lascivious’, Gmc. *φreka- ‘avaricious’, Pol. pragnąć ‘yearn for’; cf. Pol. prażyć ‘stew’, Slov. prážiti ‘stew’.

·       NWIE *pk- ‘(smouldering) embers’ as basis for OIr. riches, W rhys-yn, Lith. pirksnys, Ltv. pìrkstis.

·       NWIE *segh- as basis for *seg(h)-lo- ‘sail’, in Ir. séol, W hwyl, Gmc. *segla- ‘sail, canvas’; cf. also Lat. sagum ‘coarse woollen cloak’, Lith. obs. sãgė ‘shawl, warp’.

·       NWIE *sent- ‘feel’, in Lat. sentiō ‘sense, feel’, Gmc. *sinnan- ‘head for, long for’; cf. Lith. sintė́ti ‘think’, OCS sęštĭ ‘sensible, wise’.

·       NWIE *skerdh- ‘cut’, cf. OIr. scerdaid, Gmc. *skertan, Lith. skerdž.

·       NWIE *skok-e- ‘tremble’, cf. OIr. scuichid ‘to move, start, go’, MW ysgogi ‘to move, stir, tremble’, Gmc. *skakk/gon- ‘shake’, OCS skiti, SCr. skòčiti ‘to jump, leap’.

·       NWIE *skt-e- ‘jump; gush forth’ (for the semantic connection of both meanings, compare with English ‘spring’), cf. Lat. scatere, Gmc. *skut(t)ōn-, Lith. skàsti (skantù, skataũ).

·       NWIE *soit- ‘charm, spell’, cf. OCo. hudol ‘magus’, W hudol ‘charming, illusory’, ON seiðr, Lith. saĩtas, seĩtas ‘magic’.

·       NWIE *streig- ‘to stroke’, cf. Gmc. *strīkan ‘to stroke’, OCS strišti ‘to cut, slip’; with nasal infix Lat. stringere ‘to skim, scratch’.

·       NWIE *tek-(i)e- ‘ask, request’, cf. Gmc. *ϑegan-, cf. Bret. tizaff ‘receive’, Lith. tèkti (tenkù) ‘to reach (for)’; originally ‘reach out the hand’.

·       NWIE *tk-e- ‘thrive, prosper’, cf. Cel. *tnk-o-, *tonk-eto-, Gmc. *tenk-e-, Bal. *tk-e-, Sla. *tk-neu-. For potential cognates of *temk-, see (Kloekhorst 2008).

·       NWIE *tenk-s- as basis for an expansion of the meaning ‘pole’, cf. Lat. tēmō (<*tenk-s-mon-), Gmc. *ϑinxs-lo-, OPru. teansis < *tenk-s-i. An original root **tengh- could be deduced from Slavic and Iranian cognates.

·       NWIE *teuk-o- ‘thigh, hip’, cf. Cel. *tuk-nā, Gmc. *teuk-o, BSl. *touk-o-.

·       NWIE *tb-o- ‘crowd, village’, cf. Cel. *trebā, Pre-Gmc. *tb-o-, Pre-Bal. *trobā. Maybe here also Ita. *turbā, PGk. *turbā.

·       NWIE *(s)tronk-o- ‘dirty’, cf. Cel. *(s)tronko-, Gmc. *ϑranxa-, Bal. *tronk-ā-. Maybe here also Gk. truks ‘wine residue’.

·       NWIE *u̯idh-o- ‘forest, tree’, OIr. fid, OW guid < *idh-u-, Gmc. *idu, possibly also Lat. dīvidēre, Lith. vidùs n. ‘middle’ (< “forested area between two centres of habitation”), O. Pru. widdewū.

·       NWIE *u̯ol-t- ‘(cut) tuft of hair, curl’, as basis for OIr. folt, falt (in *-o-), OE weald (in *-u-), Lith. váltis, Ukr. vólotĭ (in *-i-).

·       NWIE *u̯ors- ‘callus, wart’, cf. Lat. verrūca < Ita. *u̯e/ors-, Gmc. *arza; maybe here OPru. warsus ‘lip’ < ors-u- (cf. Gmc. *erilas ‘lip’).

3.2.5. European agricultural substratum

The presence of words of non-Indo-European origin is particularly interesting to assess the routes of expansion of European languages. Their ‘Neolithic’ and ‘Chalcolithic’ naturerelated to agriculture and to metallurgy, respectivelyconnects them to the Neolithisation wave that brought Middle East farmers from Anatolia mainly into South-East Europe, to the south of the loess belt of the European plain, where the first metallurgic centres developed close to the steppe. The finding of a NWIE substrate common to Palaeo-Balkan languages is still more indicative of the origin of the substrate language, which should be located in the north Pontic and north-west Pontic area around the lower Danube.

The following lists contain 24 stems shared with Palaeo-Balkan languages, which may be attributed to a common period of expansion of west Yamna settlers ca. 3300–2800 BC, and 45 stems only in North-West Indo-European which should be added to the ca. 25 stems in West Indo-European and Northern Europeanwhich may be attributed to the period of isolation of Yamna settlers in Hungary, and to the formation and expansion of East Bell Beakers, i.e. ca. 2800–2300 BC. Substratum common to NWIE and Palaeo-Balkan

Pan-European substratum words include:

·       NWIE *aig- ‘oak’, in Lat. aesculus ‘winter oak’, Gmc. *aik-, OPru. ansonis, Lith. áuol-, áužuolas, Ltv. uôzuōls, and also Gk. aigílōps ‘kind of oak’.

·       NWIE *agws-i- ‘axe’, in Lat. ascia < *ask-iā-, Gmc. *akesī< *agwis-ī, Gk. aksínē <*agws-i-. Formal incongruences suggest a non-IE origin, maybe comparable with Akkadian aṣṣinu ‘axe’ and Aram. aṣṣīnā. Here possibly Myc. a-qi-ja, too.

·       NWIE *ak-, adj. *ak-no- ‘maple’, cf. Lat. n. acer, adj. acernus, Gmc. *axurna-, adopted in Sla. *avor-ovŭ ‘made of maple’, Lith. aornas; possibly here Gk. ákastos, ákarna.

·       NWIE *ar- ‘(a tree with) nuts or cones’, cf. Gaul. *arua & *araua, Pru. *reisas, Lith. ríešas, dial. ruošutỹs Sla. *orexŭ, *orĭxŭ; Gk. árua, auará (<*araa?), Alb. árrë. Compare Proto-Basque *hurr “hazelnut” (Blažek 2018).

·       NWIE *ard- (<*Hd-) ‘heron’, cf. Lat. ardea, OIce. árta ‘a kind of duck’; from *rod- cf. Sla. *roda, Gk. erōdiós, rhōdiós.

·       NWIE *bak-(t)lo- ‘stik’, in Lat. baculum, Gmc. *pagila-, also OIr. bacc, MW bach ‘hook’ < *bakko-, cf. Gk. báktron.

·       Pre-Gmc. *digh-ā- ‘goat’, also in Balkan IE *dig-ā- cf. Gk. dizda, Alb. dhi. The incongruence of both forms points to a non-IE substrate word.

·       West IE *eregw-o-, *eregw-indh-o- ‘pea’, cf. Lat. ervum, OIr. orbaind pl. ‘kinds of grain’, Gmc. *aru̯īt-; cf.Gk. erébinthos, órobos ‘(chick)pea’ (Kroonen 2012).

·       NWIE *gnid- ‘nit’, cf. Ita. *gnind-ā, Gmc. *knid-, Pre-Bal. *gnind-ā, Pre-BSl. *knid-ā; also Gk. konís, konídos, Alb. thëní, both from *k(o)n-id-, and Arm. anic ‘louse’. All forms point to an original source **c~ʕ(o)n-ĩd (Kroonen 2012).

·       NWIE *ghersd- ‘barley’, cf. Lat. hordeum, Gmc. *gerstō; also here Palaeo-Balkan *grisdh-, cf. Arm. gari, Gk. krithḗ, Alb. drithë. Irregular forms point to a non-IE agricultural substratum (Kroonen 2013).

·       NWIE *i(o)rk- ‘goat’, cf. Lat. (h)ircus, W iwrch, Bret. yorch, OHG irah, ireh, irh; further Gk. íorkes, Arm. ors. Unexplained consonantism and vocalism may point to a non-IE source.

·       NWIE *kailo- ‘the whole’, cf. Ita. kailo- ‘sky’, W coel ‘presage, omen’ OBret. coel ‘priest’, Gmc. *xaila- ‘whole, sound’, *xail-sōn- ‘give oaths, interpret omens’, OCS cělŭ ‘whole’, OPru. kailūstiskan ‘health’. Possibly a term in the augural sphere, in contrast with temple ‘the part’ (de Vaan 2008).

·       NWIE *kanab- ‘hemp’, cf. Gmc. *hanapa ‘hemp’, Russ. konopĭ ‘rope’ (> Bal. *kanapi-), Gk. kánnabis. Probably an old Wanderwort, maybe of oriental origin, cf. Sumerian kunibu ‘hemp’.

·       NWIE *kˀkw-(n-) ‘heel; ankle, tarsal joint’, cf. Lat. calx, calcis, BSl. *kulˀk (de Vaan 2008).

·       NWIE *krem-u- ‘wild garlic, ramson’, cf. Cel. *krm-u- ‘garlic’ Gmc. *kre/om-us-(o-), Slavic cognates from **kerm-(o)us-a, Baltic from **kerm-us-a; cf. Greek from **kre/om-us-o-. Kroonen (2013) assumes that all forms are remade from an underlying amphidynamic paradigm *krem-uos, *km-us-ós, acc. *krom-ués-m.

·       NWIE *mk-/bk- ‘wild carrot’, cf. Gmc. *murxōn-, Russ. borkan (into Fi. porkkana), Ru. morkóv’, Gk. n. pl. brákana ‘wild vegetables’.

·       NWIE *ploudh- ‘lead’, cf. Lat. plumbum (<*plumdh-), MIr. lúaide, Gmc. *lauda-, also *blīu̯a, and same word behind Gk. mólubdos, etc. Possibly through the same source as Lyd. marida- ‘the dark metal’, cf. CLuv. marai- ‘black, dark-coloured (?)’.

·       NWIE *rāp- ‘turnip’, Lat. rāpum, OHG ruoba, ruoppa, Lith. rópė; compare Sla. *rěpa (<*rēp-), Gk. rhápus, rháphus (<*rap/*raph-). Alternating vocalism and vacillation in Gk. p/ph suggests a loanword (de Vaan 2008).

·       NWIE *sam(n̥)dh-o ‘sand’, cf. Lat. sabulum < *sadh-lo-, Gmc. *samda-, Gk. psámmos < *sam-ndh-o-.

·       NWIE *(s)gwhongh-o- ‘sponge, mushroom’, cf. Lat. fungus < *gwhong(h)-, Gmc. *sa(m)b/ppan-, OCS gǫba <*g(h)umb(h)-; further Gk. sphóngos < *skw(h)ong-, Arm. sunk < *suongwh- (Kroonen 2013).

·       NWIE *upno- ‘oven’ cf. Gmc. *uφna, OPru. wumpnis < *u(m)p-no-? Myc. i-po-no-, Gk. ipnós. Dubious are Lat. aulla, Skr. ukhá. Maybe here also Hitt. appen-, apn- ‘baking film, fire-pit, broiler (oven)’.

·       NWIE root *k- ‘heather’, cf. Cel. *roiko-, Gk. ereíkē <*er-eik-o-, PSl. *versŭ, *verskŭ (<*erk-), Lith. vìržis, Ltv. vìrsis (<*k-).

Words restricted to the Mediterranean area (probably because of their natural distribution) include:

·       *el(e/a)i- ‘olive, oil’, cf. Lat. oleum, Gk. elaíā, elaios, elaion, Arm. ewł;

·       *thuōiko- or tū(i)ko- ‘fig’, cf. Lat. fīcus, Gk. sũkon (Boeotian tũkon), Arm. tՙuz.

·       *éinom ‘wine’, in Ita. *īnom, Arm. gini; *óinos in Myc. wo-no, Gk. oînos; *ih(o)n- in Hitt. ianis, iniant-, Luw. inia-, ian(i)-. It may be explained as derived from PIE *ehi- ‘turn, bend’, but it could have been a Mediterranean loan; cf. Kartvelian *ɣino-, Semitic *ain-. Substratum words only in North-West Indo-European

Substratum words in NWIE have sometimes the suffix *-is- or *-(e)n-, and include the following (Matasović 2013; Oettinger 2003):

·       NWIE *ab(e)l-u- ‘apple’, in Osc. Abella, Cel. *abul, Gmc. *aplu-, *ap(a)laz, Bal. *ābel-, *ābul-, OCS ablŭko. Dubious is the appurtenance here of Dacian kinoúboila (<*kun(i)-ābulo-) and Thracian dinupula (<*kun-ābulo-), or Gk. abíllion (< *abúlion), as well as the origin in Semitic *ʔabul- (or *ʔubal-?) ‘various kinds of fruits and cultural trees or plants’ (Blažek 2018). It bears resemblance to a South European isogloss *mālom ‘apple’ cf. Lat. mālum, Gk. melon; it contains the marginal IE phoneme *b (which is bilabial like southern *-m-); and contains a non-IE prefix *a-, so both words are possibly related to the same substrate language, with different (later) regional influences.

·       NWIE *als- ‘alder’, cf. Ita. *alsno-, BSl. al(i)snio-, alisā, Gmc. *als-, *alis/zō. The suffix variation *-s- / *-is- and vocalic vacillation in BSl. points to a non-IE origin (de Vaan 2008).

·      NWIE *aeksna‘oats’, cf. Lat avēna, Sla. *ovĭsŭ ‘oats’, Lith. avižà, Ltv. àuza, OPru. wyse; shape of suffix *-eks-/*-iks- (possibly in *-gs-) points to a substratum origin.

·       NWIE *bhabh- ‘bean’, cf. Lat. faba, Faliscan haba, Gmc. *baunō <*bab-nō, Sla *bobŭ. Germanic form in *-no, and the Palaeo-Balkan root in *-k- (cf. Gk. phakós, Alb. bathë) suggest that its origin is a European language belonging to a deeply agricultural culture (Kroonen 2013).

·       NWIE *b(h)at- ‘strike, hit’ as basis for Lat. fatuus, Gaul. (through Latin) battuo, anda-bata, Russ. batŭ.

·       NWIE bhog- ‘(water source)’, cf. Gmc. *baka ‘creek’, Sla. *bagno ‘swamp, marsh’, MIr. búal ‘water; bathing, healing, cure’ (Kroonen 2013).

·       NWIE *edh-lo- ‘conifer’, cf. Lat. ebulus, Lith. glė, Ltv. egle, OPru. addle, OCS ela, etc. Here probably *odh-o-ko- in Gaul. odocos.

·       NWIE *epr- ‘boar’, cf. Pre-Ita. *apr, Pre-Gmc *epr, Pre-Sla. *epr.

·       NWIE *ghaid- ‘goat’, cf. Lat. haedus ‘kid, young goat’, Gmc. *gait, OCS koza; cf. Semitic *gadi̯- ‘goat’, cf. Arab. gadun, Hebr. geδī, Berb. agaid (Kroonen 2012).

·       NWIE *ghladh- ‘even, flat’, cf. Lat. glaber, OHG glat, Lith. glod(n)ús, OCS gladŭkŭ.

·       NWIE *gramma ‘gramiae, rheum in the corner of the eye’, cf. Lat. grāmiae, grammōsus, Sla. *grŭměždĭ, *krŭmelĭ, OIce. kramr ‘damp’ Goth. qrammiþa‘moisture’. Maybe related to Gk. glámōn‘blear-eyed’, then with liquid dissimilation (de Vaan 2008).

·       NWIE *(s)grap/bh- ‘hornbeam’, cf. Lat. carpinus, Lith. skrõblas, Sla. *grabrŭ. Maybe here Umbrian Grabovius (epithet of Jove) and Ancient Macedonian grā́bin ‘a kind of tree’.

·       NWIE *kabu- ‘falcon’, cf. Pre-Ita. *kapu-, Pre-Gmc *kabhug-, Pre-Sla. *kabhuk-, also in Etruscan capu < *kapu. Considered a late Wanderwort because of the impossible IE *p ~ *bh correspondance, it could have been influenced in Italic by capere ‘to take’ (“the snatching bird”).

·       NWIE *kars- ‘willow (wool)’, cf. Lat. carrō, Lith. karš.

·       NWIE *kasni- ‘garlic’, cf. Cel. *kasninā, Sla. *česnъ, possibly Slavic vocalism influenced by root *kes- ‘to comb (?); to peel’.

·       NWIE *kattos ‘cat’, cf. Lat. cattus, Cel. *katto-, Gmc. *kattōn-, Russ. kótka. A non-Indo-European word usually believed to have entered Europe from North Africa (cf. Nubian kadīs ‘cat’, Arabic qiṭṭa) through Latin catta, but this is not proven.

·       West IE *knu- ‘nut’, cf. Lat. nux < *knuk, Cel *knū-s, Gmc. *xnut < Pre-Gmc. *knud-. Because of its nature as root noun, it has been proposed that it is likely to be the result of the vowel-less coda of the word in the source language (Kroonen 2012).

·       NWIE *koilo- ‘lean, naked’, cf. OIr. cóil, cél, W cul, Ltv. kails.

·       NWIE *kolombh- ‘pigeon, dove’, cf. Lat. columba, Sla. *goląbĭ; cf. also cf. Lith. gelumbẽ ‘blue cloth’, OPru. golimban ‘blue’ vs. Russ. golubój ‘pale blue’, OPru. gołęby ‘greyish, blue-grey’ (Derksen 2008).

·       NWIE *kos(u)lo- ‘hazel-tree’, cf. Lat. corulus ‘hazel-tree’ < *kosulo-, OIr. coll, W coll < Cel. *kos-lo- ‘hazel-tree’, OHG hasal, OIce. hasl < Gmc. *xaslaz ‘hazel-tree’, OLith. kasulas ‘hunting spear’. The diverging suffixes, restriction to Europe, and the fact that it concerns a tree-name could point to a non-IE origin (de Vaan 2008).

·       NWIE *kos- ‘basket made of wicker’ as basis of Lat. quālum, cōlum (<*kos-lo-) and OCS košŭ (<*kos-o-).

·       NWIE *lens-/*lent- ‘lentil’, cf. Lat. lēns, OHG linsa, Sla. *lęntĭa. Gk. láthuros ‘pulse, chickling’ probably unrelated.

·       NWIE *leisā ‘furrow, plough furrow’, cf. Lat. līra, OHG -leisa, OCS lĕxa (but cf. Goth. lais).

·       NWIE *ms- ‘taste, savour’, as zero-grade basis for Cel. *mlasso (in *-to-), Russ. molsát, Czech mlsati (in *-o-).

·       NWIE *mokinā‘belly’, cf. Cel *mok-inā ‘bellows’, Gmc. *magan (<*mok-on-) ‘stomach’, Sla. *mok-inā ‘pouch, purse, scrotum’. The plain k may point to a non-IE origin (Kroonen 2013).

·       A common non-Indo-European source for ‘many, much’ is behind Cel. *menekki-, Pre-Gmc. *monogh-o-, Pre-Sla. *munogho-.

·       NWIE *morā ‘incubus, nightmare’, cf. OIr. Morigain, OHG mara, Ukr. móra, SCr. mora.

·       NWIE *nāti- ‘nettle’, cf. Cel. *ninati, Gmc. *natōn, *natilōn, Lith. nõtrynė ‘nettle’, OPru. noatis, perhaps also Sla. *natĭ; a non-IE form.

·       NWIE *plout-o-, *plut-e-o- ‘board(-construction)’, cf. Lat. pluteus, ON. fleyðr, Lith. plaũtas.

·       NWIE *polk-ā- ‘ploughed, turned’, cf. Gaul. olca, OHG felga, Russ. polosá.

·       NWIE *poug(h)-o- ‘upright, aloud’, cf. OIr. óg, óge, Czech. pouhý.

·       NWIE *rouk-o- ‘garment’, cf. Celt *ruxtu- < *roukkā-, Gmc. *rukkaz, OCS ruxo ‘garment’ (Matasović 2009).

·       NWIE *ruk-s-o- ‘clothes’ cf. Cel *roukk-, *ruk-, Gmc. *rukkaz, Sla. *ruxo; alternation between geminate and simple *k (and *x in Slavic) looks non-IE.

·       NWIE *silubhr- ‘silver’, cf. Celtib. silabur, Gmc. *silubra-, Sla. *sĭrebro, Lith. sidãbras, sidabras; cf. Basque zilhar (Kroonen 2013).

·       NWIE *sekūr- ‘axe’, cf. Lat. secūris, OCS sekyra, etc.

·       NWIE *skoi-/skui- ‘thorn, needle (from tree)’, cf. Cel. *skiat- ‘thorn bush’, Bal.  *skiā ‘needle from tree’, Sla. *xvoa‘needles of a coniferous tree’; root of non-IE shape.

·       NWIE *skerb(h)- ‘sting with a thorn’ as basis for Corn. Bret. spern (with rn < rbn), Lith. skverbiù, skvebti.

·       NWIE *(s)līu̯o- ‘bluish’ in Lat. līvidus, OIr. , W lliw, OCS sliva; compare also *sloīkwo- in OHG slēha, slēwa, OE slāh.

·       NWIE *slougo- (collective *slougā) ‘attendant’ in OIr. slóg, slúag, W llu, *tego-slougo- in OIr. teglach, OW telu; Lith. slaugà, slaugýti; OCS sluga, služiti.

·       NWIE *sorb- ‘berry’, cf. Lat. sorbus ‘service-tree’, Lith. serbentà ‘blackberry, redcurrant’, dial. sarbentà, Russ. sorobalina ‘rose hip, blackberry’.

·       NWIE *spar- ‘sparrow’, cf. Lat. parra ‘a kind of bird’, Cel. *sfrao- ‘crow’, Gmc. *spara(n)-, OPru. spurglis; also compared to Gk. sparásion ‘sparrow-like bird’.

·       NWIE *terp- ‘be rigid, stiff’, present *tpēe- in Lat. torpeō, Lith. tipti (tirpstù, tirpaũ), OCS u-trŭpěti.

·       NWIE *tlokw-, tl̥kw- ‘interpret’, etc. Lat. loquitur, OIr. ad-tluchedar, do-tluchedar; OCS tlŭkŭ, ORuss. tlŭkŭ.

3.2.6. Old European topo-hydronymy

The latest stage of the classical period, after the expansion of the East Bell Beaker folk over all of Europe, may be identified in linguistics with the expansion the Old European hydronymy (Krahe 1949, 1964; Tovar 1977; Udolph 1994; Kitson 1996), as found in Central Europe, France, Italy, the British Isles, Iberia, Scandinavia, and the Baltic (Figure 5). The language of hydronyms can be classified as of Common Indo-European nature (Tovar 1977; Kitson 1996; Adrados 1998), with the typical a-vocalism evidencing a post-laryngeal stage:

·       Interesting from an etymological point of view are those related to IE *danu- ‘river’, and *u̯eis- ‘flow, flood’. Other common roots found in old river names are *al- (e.g. *Alma), *alb- (e.g. *Alba) *dra-, *kar-, *sal- (e.g. *Sala), *er-; cf. also *Isara.

·       Common affixes include *-l-, *-m-, *-n-, *-r-, *-s-, *-st-, *-k, *-- and *--.


·       To the north of the Alps the most common endings are in *-a, while to the south the most common are in *-os, which may reveal a later vocalisation (i.e. Germanic–Balto-Slavic vs. Celtic–Italic) of previous names.

Figure 5. Old European hydronymic map for the root *sal-, *salm-. Modified image from Krahe (1964).



The following is a simplified schema of possibilities for two common roots:

















































Regarding the stratigraphy of topo-hydronymy in Europe, there are Old European layers of hydronyms in ab-/ub-/ap-/up-/ur- ‘water, river’, as well as toponyms in kerso-, turso-, alau̯o-, lako-, mido-, silo-, tibo-, or basto-, which function as first members of compounds in different regions of Europe, including those historically inhabited by non-Indo-European speakers (such as Iberian and Basque in Spain and France, or Etruscans in Italy). This supports the huge social and linguistic impact of the expansion of Bell Beakers all over Europe, before the later emergence and spread of Indo-European and non-Indo-European speakers over that ancient Indo-European linguistic layer (Villar Liébana et al. 2011; Villar Liébana 2014).

Figure 6. Old European topo-hydronymy, European block. Frequency projection using the Kringing method. Baltic data have been compensated to diminish the regional weight of the Baltic series. The high density of topo-hydronyms in the Baltic area, due to the active use of some terms in ancient and modern Baltic languages, makes their prehistoric distribution around the West Baltic (and its statistically inferred projection to the east and north) dubious. Image modified from Villar Liébana et al. (2011).

3.2.7. Statistics of lexical isoglosses

The reconstructed NWIE vocabulary can be summarised as follows[xii] (Table 1):

Table 1. Shared isoglosses related to the North-West Indo-European language.




Northern IE (IE)



NWIE + Balkan (IE + non-IE)



  NWIE + Balkan (IE)



  NWIE + Balkan (non-IE)



NWIE (IE + non-IE)



  NWIE stems (IE)



  NWIE substrate (non-IE)



  NWIE roots (IE)



West IE



  West IE stems + roots (IE)



   *West IE substrate (non-IE)










Northern IE (IE + non-IE)

25 (~45)


Other European



Rows with obscured background: main layers of NWIE vocabulary. In bold: numbers counted for statistics of the North-West Indo-European vocabulary. In italics: approximate numbers or estimations.

·       Core NWIE lexicon of ca. 172 stems: 67 identical stems and 45 substrate words, unique to the group, all of them connecting at least three non-neihbouring languages. Apart from that, ca. 60 shared roots of Indo-European origin, suggesting that the terms they represent were possibly recent innovations before the disintegration of the NWIE community. Some uncertain cognates have not been included.

·       Core NWIE lexicon shared with Balkan languages of ca. 70 stems: listed above are 46 stems of Indo-European origin, and 24 of non-IE origin, although the number including uncertain roots is bigger. Still more interesting is that the NWIE stem usually shows a consistent form, different from Balkan cognates, which suggests that, even though tthe word was probably adopted at an early time, the NWIE language developed later independently; that is, these adopted substrate words in NWIE either underwent the same innovation, or remained without change, contrasting with what happened in Balkan languages.

·       ‘West Indo-European’ isoglosses shared among Italic, Celtic, and Germanic are ca. 130, exluding the many early and late borrowings, i.e. before any of their respective sound shifts (see below §4.5.1. West Indo-European): 115 are of IE origin, and 15 likely of non-IE origin (Kroonen 2013). Around 25 of them are uncertain, and a third is approximately shared between Celtic and Germanic, and another third between Italic and Germanic. Shared Italo-Celtic (IE and non-IE) stems may be around 15 (de Vaan 2008; Matasović 2009), whichif we presuppose a non-genetic relationship between bothshould be added to the common Western IE isoglosses. Many West IE cognates are identical stems, which further supports its inclusion as ancestral, NWIE vocabulary.

·       Shared vocabulary between Germanic and Balto-Slavic has been recently set to ca. 25 stems (see below §4.5.3. Northern European), which seems the more conservative number of true ancestral cognates. Nevertheless, Kroonen (2013) listed ca. 220 broadly described ‘Northern European isoglosses’ (related stems excluded), with 40 of them listed as uncertain, and approximately one third of them shared only with Baltic, and one third only with Slavic languages, which leaves a general estimation of ca. 60 shared stems or roots. It is often unclear how many of these shared stems or roots are loanwords from later periods, though, because of the later attestation of these languages. Also, the number of reconstructed Proto-Slavic words of Germanic origin is bigger (Derksen 2008), as is probably that of Proto-Baltic.

·       ‘Whole European’ isoglosses, adding NWIE and Balkan languages, comprise more  than 300 independent stems, including any North-West Indo-European branch (in combination with other branches, or not): ca. 300 with 50 uncertain cognates shared with Germanic (with or without other NWIE cognates) were included by Kroonen (2013), and a similar number of ca. 300 terms shared with Italic are found in de Vaan (2008). Without a closer inspection beyond a summary description of a European isogloss, it is unclear how many of these hundreds of (stem or root) cognates may be shared with other branches, or if they may be later (i.e. Bronze Age) borrowings.

Since there is no stronger genetic relationship between any of the four main branches of the group, all reconstructed stems pointing to an ancestral Indo-European form should suggest shared vocabulary within the parent NWIE proto-language. Furthermore, earlier isoglosses shared with Balkan languages and with Tocharian are often remade stems in NWIE, showing peculiar innovations (or else maintaining a more archaic shape), and most should be therefore also included as part of the NWIE community.

The vocabulary reconstructed for a NWIE proto-language can be thus so described:

·       Basic lexical inventory of ca. 255 NWIE stems or roots, as a conservative estimation.

·       Extended NWIE vocabulary of ca. 380 stems or roots, including West Indo-European (without Italo-Celtic) and Northern European cognates (mainly confirmed stems).

·       With a lesser degree of confidence, the number of Pan-European isoglosses (i.e. excluding at least Indo-Iranian) would lie between 600 stems or roots and 900 potentially ancestral and common shared items, including all uncertain and dialectally quite restricted cognates. In any case, more than half of these reconstructed common terms are found exclusively among NWIE proto-languages, without Balkan cognates.

Figure 7. The levels of Indo-European reconstruction (Mallory and Adams 2007).

Compared with the strict Proto-Indo-European inventory of less than 1,500 units (Mallory and Adams 2007), the shared NWIE lexicon of 255/380 items represents a fairly large proportion, especially when we consider the different stagesIndo-Uralic, Early PIE, Middle PIE, early Late PIEwhere distinct layers of vocabulary were accumulated (Figure 7). For comparison, Italo-Celtic, which is supposed to have shared a short common period of development (probably during the Early Bronze Age) based on morphosyntactic similarities, has a common lexicon of less than 20 items; whereas Balkan languages, probably neighbouring each other during the whole Bronze Age, share less than 50 lexical units.

Languages with strong foreign influence show a marked reduction in Indo-European heritage. Hittite, for example, has an estimated inherited Indo-European lexicon of ca. 422 words (Tischler and del Monte 1978), which may have probably been enlarged in more recent research, although some etymologies have also been rejected (Kloekhorst 2008). This proportion of inherited vs. loaned vocabulary of about 5:3 or 2:1 is not much different from what we see in Greek (Zeilfelder 2017), but it seems larger than what is found in North-West Indo-European or Proto-Indo-Iranian.

3.2.8. Schleicher’s fable in North-West Indo-European

óu̯is éku̯ōs-kwe

óu̯is i̯ósi̯ās u̯ĺ̥ˀnā né est

éku̯ons dedórke;

tom gwˀu̯úm u̯óghom u̯éghontm̥,

tom mégām bhórom,

tom u̯īróm ōkú bhérontm̥.

óu̯is éku̯obhos u̯eukwét:

“kērd ághetor moi,

ghmónm̥ éku̯ons ágontm̥ u̯idtéi.”

éku̯ōs u̯eukwónt: “kl̥néu, óu̯i!

kērd ághetor nos u̯idtbhós:

u̯īrós, pótis, óu̯i̯om u̯ĺ̥ˀnām

sébhei gwhórmām u̯éstim kwr̥néuti.

óu̯i̯om-kwe u̯ĺ̥ˀnā né esti.”

tod kekluu̯ṓs óu̯is ágrom bhugét.

A recitation[xiii] of the text is available at <https://youtu.be/_6ne-xvC0TU>.


·       For *(dh)ghmon-: The other PIE word apart from *u̯īrós (possibly *u̯irós already appearing in the NWIE stage), *ner-, commonly used to translate ‘man’ in the fable, is not used here because of its more specialised use in NWIE exclusively as ‘manly, strong’ and mostly in archaisms, cf. Italo-Celtic *ner- (as Lat. neriōsus, OIr. nert), Gmc. *ner- (OHG Nerthus), Bal. *ner-/nor- (Lith. Nertėti, OPru. nertien).