4.5. Supradialectal communities

4.5.1. West Indo-European

Close contacts within a West Indo-European group (formed by Italic, Celtic, and Germanic), supposed to derive from a dialect continuum spanning Pan-European trade contacts during the Bronze Age, may be found in certain common traits. Most likely, because there were no closer contacts between Germanic and Celtic, or between Germanic and Italic, these common traits are in fact North-West Indo-European features that did not survive in Balto-Slavic, the most divergent dialect of the group, probably partly isolated early in close contact with Palaeo-Balkan and Uralic languages (see §4.13. Balto-Slavic).

One common feature of the three dialects is the merge of *k kw, usually considered a parallel development in ‘centum’ languages, because it is also found e.g. in Gk. hippos. Nevertheless, since it is not found in Lusitanian, it may be posited as a common development, or one due to areal influence.

Shared folk tales among the three dialects (da Silva and Tehrani 2016) include “The Treasures of the Giant” (MFTD 328), whereby a boy or hero sets out to steal from a giant (a variant is Jack and the Beanstalk); and “The Animal Bride” (MTFD 402), where the youngest of three brothers succeeds best in the quests set by his father, bringing the best cloth, the most beautiful bride, etc., and the mouse (cat) who has helped him changes herself into a beautiful maiden. Substrate words

Non-IE substrate words shared only by West Indo-European include e.g. *eregw-o-, *gnid-, or *knuʔ- (see above § Substratum common to NWIE and Palaeo-Balkan), and also:

·       WIE *akwā ‘water’, found in Gmc. *axwō- and Ita. akwā-, may be a loan from a non-IE language (de Vaan 2008). Traditionally claimed to be a variant from PIE *ap- ‘water’.

·       WIE *aru- ‘ore’ as basis for Lat. raud-, Pre-Gmc. *arud (cf. ODu. arut, OHG aruz, ariz), comparable to Sumerian urud(u) ‘copper’.

·       WIE *bholik- ‘coot’, cf. Lat. fulica, Gmc. *balika/ōn-.

·       WIE *ghazt-o-, *ghazdh-o- ‘spear’, cf. Lat. hasta, OIr. gat, gas, Gmc. *gazda- (de Vaan 2008)

·       WIE *kagh- ‘enclosure’ as basis for Lat. caulae < *kagh-el-ā-, W cae, OBret. cai ‘pen, enclosure’ < *kagh-io-, ON hagi, OE haga, OHG hag < *kagho(n)- ‘enclosure, pasture’.

·       WIE *kapro- ‘goat’, cf. Ita. *kap-ro-, Cel. *gabro-, Gmc. *kapro-. The consonantal vacillation and a-vocalism support its non-IE origin.

·       WIE *kar-(gu-) ‘sacrificial mound’, cf. OIr. carrac, carn, OW carrac, creic, carn, ON hǫrgr, also in Finn. karkko, karkku ‘pile, stone wall, cliff’.

·       WIE *kragr- ‘heron’, cf. MW crehyr < *krəxar-, Gmc. *xraigran.

·       WIE *kwe-kwer- ‘gourd’, cf. Lat. cucurbita ‘gourd’, OE hwerhwette ‘cucumber’.

·       WIE *-msl- ‘blackbird’, in Ita-Cel. *mesal- (cf. Lat. merula, W mwyalch, Bret. moualc’h), Gmc. *a-msl- (cf. E ousle, OHG amsala); the reduction to *-msl- and the addition of a non-IE prefix *a- points to a non-IE origin (Kroonen 2013).

·       WIE *marko- ‘horse, stead’, found in Gaul. acc. markan, MIr. marc, MW march, ON marr, OE mearh, OHG marh, marah, has been connected to OInd. mára-, but a-vocalism probably points to a non-IE origin, possibly a Eurasiatic wanderword through steppe nomads (Matasović 2009).

·       WIE *sem/b- ‘rush’, cf. OIr. simin, sibin(n), sibhean(d) ‘rush, reed; corn-stalk’, OS semith, MLG sem(e)de, OHG semida, MHG semede, sebede.

·       WIE **smmér-/*sémmr- ‘clover’, possibly behind OIr. seamar, Gmc. *smērōn- (restricted to Scandinavian).

·       WIE *taks-, cf. Cel. *tazg-, Gmc. *ϑaxs-, also in Lat. taxo (beside taxus), probably a loanword (Kroonen 2013).

·       NWIE *u̯--t- ‘dyer’s rocket’ cf. Pre-Ita. *ulout-o-, Pre-Gmc. *olt-ā-. West Indo-European lexicon

Common West Indo-European lexicon likely of North-West Indo-European origin includes:

·       WIE *al-e- ‘to feed’, in Lat. alere, OIr. alid, W alu, Goth. alan, ON ala. Strong verb to PIE root *ħel-.

·       WIE *bhlē-u̯o- adj. ‘blond’, cf. Lat. flāvus ‘blond’, OIr. blá ‘yellow’, W blaw ‘grey’, Gmc. *blēu̯a- ‘blue’.

·       WIE *bhlō- ‘flower’, cf. Lat. blō-os, Cel. blo-tu-, Gmc. *blōan, blōman-.

·       WIE *gus-tus m. ‘taste’, cf. identical cognates Lat. gustus, Gmc. *kustuz ‘trial’, OIr. guss ‘excellence, force’ < *gus-tu-. Derived from root *geus.

·       WIE *kaput- ‘head’, cf. Lat. caput, Gmc *xa(u)buda-; further (in *-k-) OIr. cúach, W cawg.

·       WIE *kent-no- ‘skin’, cf. Lat. centō ‘blanket, patched cloth’, OIr. ceinn ‘head’, W cen ‘skin’, ON hinna ‘thin skin, membrane’, OE hion ‘meninx?’.

·       WIE *ker-u- ‘deer’, cf. Lat. cervus, OW caru, MW carw-, Gmc. *xeruta- (with animal sufifix -ut-a-), from PIE root *ker- ‘horn’.

·       WIE *koldo- ‘destruction’, cf. OIr. coll, Gmc. *xalta-, and probably here irregular Lat. clādēs ‘destruction’ (reconstructed as from *k-dh-, but with no cognates outside Italic).

·       WIE *krispo- ‘curl’, cf. Lat. crispus, Gmc. f. *xrispō-, W crych < *kripso-.

·       WIE *ks-e- ‘run’ cf. Lat. currō ‘run’, OIr. carr, W car ‘vehicle’, Gmc. *xurzōn ‘to rush’.

·       WIE *lut- ‘mud’, cf. Lat. lutum, OIr. loth, Ger. Lotz ‘deep spot in a creek (to scoop water)’.

·       WIE *mto- ‘mouth’, cf. Lat. mentum, MW mant, Gmc. *munϑa.

·       WIE *nad- ‘bind’, as basis for *nad-to- in Lat. nassa ‘wicker-work basket’, OIr. nassae ‘bound’; as o-grade *nōd-o- in Lat. nōdus ‘node’, ON nót, pl. nœtr ‘net’; as *nad-i- in Gmc. *nata ‘net’, OIr. nenaid- ‘nettle’; as verb *nad-ske- in OIr. nascaid, MBret. naska ‘bind’; etc. Compare Skr. niṣká- ‘golden necklace, golden ornament’ and Thrac. acc. pl. neu. NHSKOA ‘adornaments’.

·       WIE *na-tr- ‘snake’, cf. Lat. natrix, OIr. nathir, gen. natrach, Goth. nadrs, ON naðr.

·       WIE *preus- ‘freeze’, cf. Ita. pruīna ‘frost’, W rhew ‘frost, rime’, Gmc. *φreusan- ‘to freeze’.

·       WIE *psei-so- onomatopoetic verb ‘blow; hiss; whisper; fart’, cf. Lat. spirāre, W ffûn, Gmc. *φīsan-.

·       WIE *rēgs ‘king, ruler’, cf. Ita. *rēks, Cel. *rīgs, Gmc. *rekaz (cf. ON -rekr, -reki), also in derivative fem. *rēgnā ‘queen’, cf. Ita. *rēginā, Cel. *rīganī, and *rēg(i)o-, ‘royal, mighty, cf. Cel. *rīgo (and as a loan in Gmc. *rīks, *rīkia-). They continue a Late PIE concept, being evident in PIIr. f. *rā́ȷ́niā; PIIr. m. *rā́ȷ́ān (<*rēgon-, also found in Bret.  rīgon) is usually explained as a back-formation from the feminine form in *-n, because the West IE form is a root noun and thus believed to be ancient. However, this form is not found outside West Indo-European.

·       WIE *āti- ‘prophet’, cf. Lat. tēs, Cel. *āti-, Gmc. *ātí-, *āt-ó-; contrasting with *kou̯i- / *kou̯hēi- of Palaeo-Balkan and Indo-Iranian groups.

·       WIE *āstu- ‘desert, waste, empty’, cf. Lat. stus, OIr. fás < *āsto-, Gmc. *u̯ōstu-.

·       WIE *u̯iros ‘man, cf. Lat. vĭr, OIr. fer, Goth. wair; against long vowel in all other languages (Matasović 2009). Possibly related originally to Dybo’s law (Kroonen 2013).

·       WIE *-tu- ‘seen’, cf. Lat. vultus, OIr. fili, filed ‘seer’, Goth. wulþus. Celtic-Germanic isoglosses

Common Celto-Germanic vocabulary (before their respective sound shifts) includes a reflection of an ancient shared cultural sphere:

·       WIE *ako-lo- ‘awl’, in Weslh ebill, ON -all, OE awul.

·       WIE *baiso- or *baso- ‘boar’, cf. Proto-Brit. *basio, Gmc. *baiza-.

·       WIE *bhodh-a ‘war; battle’, in MIr. bodb, badb m./f. ‘war-god(dess); scald-crow’, ON bǫð, OE beado, OHG batu-.

·       WIE *bhors-o- ‘bass’, cf. Gael. barsch, Gmc. *barsa, possibly from an original meaning ‘pine needle’; cf. here Lat. fastīgium ‘roof’, Skr. bhr̥stí- y OCS borshchŭ.

·       WIE *bhrozdh-o- ‘edge’, cf. OIr. brot, Ice. bradd, OE breard, OHG brart.

·       WIE *dhreibh-e- ‘hurry’, cf. Cel. *drippi- (<dhribh-ni?) ‘hurry’, Gmc. *drīban- ‘to drive’.

·       WIE adj. *drou-sd-o- ‘trustworthy’, cf. OIr. druit, Gmc. *trausta-. Probably from IE roots *dreu- ‘trust’ + *sed- ‘sit’.

·       WIE *dhrūto- ‘joker’, cf. identical MIr. drúth ‘professional jester, fool’, MW drut ‘dear, foolish, foolhardy’, ON trúðr ‘juggler, fool’, OE trūð ‘trumpeter, actor, buffoon’.

·       WIE *elko- ‘evil, mean’, cf OIr. elc, ON illr, borrowed into Finn. elkiä.

·       WIE *gheislo- ‘hostage’, cf. Cel. *gēs(t)lo-, Gmc. *gīslaz ‘hostage’, from *gheidh- ‘desire, wait for’, possibly through ‘one who is waiting (to be released)’ (de Vaan 2008).

·       WIE *gel-o- ‘charcoal’, cf. OIr. gúal < *goulo-, Gmc. *kula- < *gul-o-. An innovation from a root meaning ‘burn, shine’; cf. Toch. B śolie ‘oven’  Skr. jvālā ‘light, torch’, Russ. zolá ‘ash’.

·       WIE *keng-e- ‘to limp’, cf. OIr. cingid, MW ry-gyng, OE hincian, OHG hinkan; here probably also Gmc. *skank-, O.Ind. khañjati, Gk. skázdō.

·       WIE *koito- ‘forest’, cf. OW coit, Bret koat ‘wood’, OCo. cuit ‘silva’, Goth. hai ‘open field’, ON heiðr ‘heath, moor’, OE hǣð, OHG heida.

·       WIE *kork- ‘oats’, cf. OIr. corca, coirce, W ceirch, Bret kerc’h, ON hargr, Ice. -hagra, NW hagre, OSwe. hagri.

·       WIE *lok- ‘fault, offence’, cf. OIr. locht ‘fault, shortcoming, vice; offence; (physical) blemish’, Gmc. *laxan- ‘to blame, reproach’.

·       WIE *lugh- ‘to bind by oath’, cf. OIr. luige ‘oath’, Goth. liuga ‘marriage’.

·       WIE *maghus ‘young boy(?)’, cf. OIr. mug ‘servant’, Corn. maw, Goth. magus ‘boy’, ON mǫgr, OE magu, etc.

·       WIE *mon-go- ‘mane’, cf. OIr. mong, Gmc. *mankan.

·       WIE *nent-e- ‘fight’, cf. OIr. néit ‘battle, combat, fighting’, OHG gi-nindan ‘to dare’.

·       WIE *oitos m. ‘oath’, cf. Cel. *oito- behind OIr. oeth, and Gmc. *aiϑa. Given the close formal and semantic agreement, it is unlikely that the formation goes back to PIE only to surface in these two neighbouring branches independently, so it is likely to have arisen in a shared cultural zone with similar legal traditions (Kroonen 2013).

·       WIE *orbhom n. ‘inheritance’, cf. OIr. orbe, ON arfr, OE ierfe, OHG arbi, erbi. Maybe developed in a shared European cultural zone from a root meaning ‘pass over’, cf. Hitt. arp- ‘to associate (someone) with, to combine, to join together’.

·       WIE *pleid-e- ‘to strive’, cf. MW llwydaw ‘to succeed’, Gmc. *φlītan- ‘to strive, fight’.

·       WIE *roi-no- ‘hill, border’, cf. Bret rūn ‘hill’, ON rein ‘marge, strip of land’, MHG rein ‘marge, elevation’.

·       WIE *slak- ‘hit’, probably onomatopoetic; cf. MIr. slachta ‘hit’, also MIr. slacc ‘sword’, Gmc. *slaxan ‘to beat, strike, slay’.

·       WIE *sret- ‘swirl’, cf. MCo. streyth ‘river’, MIr. srithit ‘stream of milk or blood’, OHG stredan ‘to seethe, swirl’.

·       WIE *streib-o- ‘stripe’, cf. OIr. sríab ‘stripe, line’, Gmc. *strīpan- ‘stripe’.

·       WIE *t(o)nˀr̥os ‘thunder; god of thunder’, in Cel. *toranos- (metathesised *torˀnos), Gmc. *ϑunraz. Compare Lat. tonitrus, O.Ind. tanā.

·       WIE *tegu- adj. ‘fat’, cf. Cel. *tegu-, Gmc. *ϑeku.

·       WIE *egh-no- / *ogh-no-  ‘wagon’, cf. Cel. *egnos, Gmc. *agnaz, probably reveals an ancient trend to replace the common noun *oghos (cf. Sla. *vȏzъ) with one formed in *-no.

·       WIE *elt-i- ‘wild’, cf. Cel. *elt-i- in MIr. geilt ‘lunatic, panic-striken fugitive’ Gmc. *elϑa-. Maybe from *gwhel-ti- (Matasović 2009).  

More Celtic–Germanic isoglosses can be found in Lane (1933) and Hyllested (2010). Italic-Germanic isoglosses

The shared Italo-Germanic lexicon before their characteristic sound shifts includes words related to nature and to the divine, among others:

·       WIE *ankr-o- ‘lowland’, in Lat. ancrae f.pl. ‘valley’, Gmc. *angra- ‘lowland, meadow’, cf. Gk. ankos ‘valley’.

·       WIE *ark-u- ‘arrow; bow’, cf. Lat. m. arcus < *arkus ‘bow, arch’, Gmc. *arxu̯ō- < *arku̯ā f. ‘arrow’.

·       WIE *at-no- ‘year’, cf. Ita *atno-, Gmc. *aϑna-.

·       WIE *aus-e- ‘to scoop’, cf. Lat. haurīre, Gmc. *ausan-.

·       WIE *bhlē- ‘blow’, cf. Lat. flāre, Gmc. *blēan-.

·       WIE *bhg-ne- ‘break’, cf. Lat. frangō, Gmc. *bruk(k)ōn-.

·       WIE *bhrā-ie- ‘to bore’, cf. Lat. forāre ‘to bore through, pierce’, Gmc. *burōan- ‘to bore’.

·       WIE *bhrod-n- ‘to bud’, cf. Lat. frondis ‘foliage, leaves’, Gmc. *brut(t)ōn- ‘to bud’.

·       WIE *bhrug-i̯e-, in Lat. fruor, Goth. brukjan, etc.

·       WIE *gentis, *gˀti- ‘clan, kin, race’, cf. Ita. *gentis, *gnā-ti-, Gmc. *kindiz, *kundiz, evolved from ‘people of the same descent’. Compare for the basic meaning ‘child, birth, offspring’ Sla. *zę̀tь ‘son-in-law’ (< BSl. **źéntis), Gk. génetis ‘origin, source’, PIIr. *ĵā́tiš ‘birth, production’.

·       WIE *kar-n- ‘flesh’, cf. Lat. carō, Gmc. *xarund-; probably here also Lith. karnà ‘Lindenblast’.

·       WIE *kat- ‘goat’, in Lat. catulus ‘young animal’ (cf. MHG hatele ‘young goat’), Gmc. *xattu ‘hat’, ON haðna f. ‘young goat’.

·       WIE *kneigwh-e- ‘bow (down)’, cf. Lat. cōnīveō ‘to be tightly closed, close (of the eye), Goth. hneiwan, ON hníga, OE hnīgan, OHG hnīgan.

·       WIE *kolso- ‘neck’, cf. Lat. collum ‘neck; hill’, Gmc. *xalsa-, with common delabialisation; cf. Gk. polos < *kwolos, Lith. kaklas, Ltv. kakls < *kwokwlos.

·       WIE *spar- ‘spear’ cf. Lat. sparus, Gmc. *spar(r)an, *speru-. Compare maybe from the same root Alb. spardh(ë) ‘kind of oak’, potentially with a meaning closer to the original.

·       WIE *takē- ‘to be silent’, cf. Lat. tacēre, Gmc. *ϑagēn-, also in *taknā- ‘to silence’ (Kroonen 2013).

·       WIE *tong-éie- ‘think, know’, cf. Gmc. *ϑankan-, Lat. tongēre (the root is also found in Tocharian, see above § Northern Indo-European).

·       WIE *ā̆de- ‘to wade’, cf. Ita. *u̯āde, Gmc. *adan. Possibly from older *gweħdh-/gwħedh- (Witczak 2012).

·       WIE *éik- ‘holy’, cf. Ita. *ik-tm-ā-, Gmc. *u̯īxa, *u̯īx-nā.

4.5.2. Italo-Celtic

The main reason for a proposed Italo-Celtic subgroup is the shared innovations of Proto-Celtic and Proto-Italic, which are unusual enough not to have been parallel developments, but rather reflect an early linguistic unity.

Common morphological developments include, from more to less likely to be dated to a common period of genetic relationship (Zair 2018):

·       Morphological innovation:

o   Superlative in *-is-mo-, apart from the inherited *-t-mo.

o   Reinterpretation of *-ī as an o-stem genitive, even though it alternates with other gen. sg. ending (viz. *-oso), is supported to stem from a common trunk by its presence in Venetic and Messapic.

o   Passive ending 1pl. *-mor, 3.pl. *-ntro.

o   The so-called “-ā- subjunctives”, which have been reinterpreted (and thus their origin obscured) later in the different dialects.

o   i-stemisation of the suffix *-stħo- (Weiss 2017).

·       Shared lexemes, of which those unique to both are few. The most significant are:

o   The prepositions with ablatival rather than directive function (Lat. , OIr. di- ‘from’, Lat. in-de ‘thence’, OIr. de ‘from him’).

o   *trāns (cf. Lat. trans ‘across’, MW. tra ‘beyond’).

o   Shared lexicon with cultural concepts, for example:

§  *tersā ‘earth’ (etymologically ‘dry land’, from *ters- ‘dry’), cf. Lat. terra, Osc. teerúm, OIr. tír, W tir;

§  *sodom ‘seat, throne’ (from *sed- ‘sit’), in Lat. solium ‘throne, OIr. suide ‘seat’.

o   Substrate words unique to both, for example:

§  *bodo- ‘yellow’, cf. Lat. badius, OIr. buide;

§  *krbh- ‘basket’, cf. Lat. corbis ‘basket’, OIr. corb < Cel. *karbanto ‘chariot’, with a semantic evolution similar to non-IE *kistā- ‘basket’ in Gk. kístē Lat. cista, ‘basket’, and also cissium ‘a kind of car with two wheels’, probably from Gaulish, in turn from Cel. *kistā.

·       Shared phonological features (although parallel developments cannot be discarded):

o   NWIE *CR̥ˀC → Ita.-Cel. *CRāC.

o   Distant assimilation *p…kw- *kw…kw-.

The dozens of shared words between Latin and Celtic come usually from late loanssimilar to the shared vocabulary between Germanic and Baltic, and Germanic and Slavic (see below)dating to the time of the Roman expansion. The best description of the group is thus probably still that of a “drowned” subgroup, sharing “a rather short period of common development followed by a long period of divergence” (Cowgill 1970). The early estimates for Proto-Italic or Proto-Italo-Venetic and Proto-Celtic languages (see below) put this community most likely in the centuries around the turn of the 3rd–2nd millennium BC.

4.5.3. Northern European

Based on shared vocabulary of Indo-European and non-Indo-European origin (and on the lack of closer genetic relationship within the NWIE group), it has been proposed that Germanic and Balto-Slavic may have shared a common Indo-European substratum with strong non-Indo-European influence (Kortlandt 2016).

This hypothetic Indo-European language without known descendants, based on the phonetic and morphological similarities, may be identified with a North-West Indo-European branch influencing both Pre-Germanic and Pre-Balto-Slavic rather early, before the early satemisation trend in Balto-Slavic, because no traces of the long proposed Centum IE substrate[xx] of Balto-Slavic can be found (Hyllested 2007). A Northern European language could thus be tentatively identified with the language of early vangard Yamna migrants from Hungary who settled into the Saxony-Anhalt region, by then dominated by the Corded Ware culture.

These pioneer settlers of the Northern European Lowlands were later replaced by East Bell Beakers migrating northwards, as the most likely source of both, a Pre-Germanic (in Scandinavia) and a Pre-Balto-Slavic community (in central-east Europe). Given the close contacts through the Northern Plains during the Neolithic, and the Pan-European influence of Únětice in the north during its classic periodapart from close contacts of cultures around the Baltic during the Bronze Ageit could be proposed that it was a third, neighbouring language from Únětice which influenced both.

Whichever the actual nature of the Northern European substratum, it could have been the source of:

·       Common lexica of Indo-European origin found in Germanic and Baltic, and to some extent in Slavic, limited to social phenomena, and especially to technical terms for wooden tools and utensils, as described by Stang (1972). His 68 compiled isoglosses were reduced to 25 by Nepokupnij (2000)[xxi]. Well-known examples of Germanic–Balto-Slavic correspondances include (Dini 2018):

o   Lith. alùs ‘beer’, Ltv. alus ~ OCS olŭ, ~ OIce. ol ‘beer’, Engl. ale.

o   Lith. draũgas ‘friend’, Ltv. draugs ~ OCS drugŭ ~ Goth. ga-draúhts ‘soldier’, OIce. drótt ‘army’.

o   Lith. kliẽpas ‘loaf of bread’, Ltv. klaips ~ OCS chlěbŭ ~ Goth. hlaifs, OIce. hleifr,

o   Lith. rugys ‘rye’ ~ Sl.: Russ. rožĭ ~ OHG roggo. Compare here also Iranian Pamir (e.g. Shughni royz ‘ear of rye’) and dubious Thracian bríza ‘emmer-wheat, rye’.

o   Lith. valdýti, Ltv. valdīt ~ OCS vladǫ / vlasti ~ Goth. waldan ‘rule a household’; particular is the extension of root *uel-, cf. Lat. valeō and other IE cognates.

·       Northern European subsitution of *-bh- endings for *-m-. Due to their consistent vocalism, it is likely that the original NWIE endings in *-bh- (see above §3.2.2. North-West Indo-European evolution) underwent a reinterpretation to *-m- in Germanic and Balto-Slavic.

o   It is usually interpreted[xxii] as a substitution based on the adverbial suffix in *-m- (cf. Lat. ill-im ‘from there’; HLuw. abl-instr. pron zin ‘from/with this’), therefore replacing an old adverbial ending (in *-bhi-) for another (Lundquist 2018); compare also Lith. raĩbas/ raĩmas, ‘varied’.

o   However, it seems likely that this change of an already grammaticalised case ending in two different dialects would have been helped by certain regional features. Northern European, as the source of this common trait, could have undergone the change *-bh- *-m- due to a simple sporadic phonetic change, similar to the commonly assumed for Mongolian or Tungusic *b- → *m (Street 1983). This change may be also supported by:

§  a phonetic substrate (assuming e.g. an underlying nonobstruent stop that alternated with other nasal nonobstruents);

§  the influence of a substrate language with similar oblique cases;

§  or both.

o   Uralic languages (from Early Proto-Uralic to West Uralic dialects) are known for their lack of voiced and aspirated bilabial and velar stops, which would compel their speakers to adopt *-bh- as a different but similar (i.e. bilabial) phoneme; and Uralic oblique cases and most dialectally innovated paradigms were usually made in nasals (cf. LPU acc. *-m, gen. *-n, loc. *-nV, dat-lat. in *-n or *-ń⁠). All this could suggest that Northern European was heavily influenced initially by Uralic or a closely related language. Interesting in this respect may also be the example of Livonian dative in -n, only partially stemming from the Uralic genitive in *-n, and which has strong links to the Latvian dative in *-m- (Seržant 2015).

·       Generalisation of mediopassive endings in *-i and specialisation of the mediopassive system, in contrast with the the original alternation of endings in *-i (middle voice) and *-r (impersonal–passive), the latter surviving in frozen remains possibly up to Proto-Slavic, e.g. in the suffix -žĭdo ‘each, every, everyone’ (cf. Russ. ždyj, Pol. kazdy), which seems to reflect an archaic impersonal / middle-passive ending from PBSl. *-ghido-r (Majer 2012). Supporting this adoption of a substrate proto-middle, later diverging in both branches, it has been suggested that n-infix athematic intransitive change-of-state verbs shared by Germanic and Balto-Slavic developed from the proto-middle through the 3sg. ending *-e (Watkins 1969; Jasanoff 1978; Darden 1996, 2018).

·       Witness to this intermediate substrate may also be other typological features common to Germanic, Balto-Slavic, Uralic, and Northern Eurasian languages (Klesment et al. 2003), although many are constrained to Balto-Slavic and Uralic, which developed in neighbouring territories.

·       The bear taboo seems to affect Indo-European languages with a close relationship with Uralic peoples and culture, making them replace the inherited *ŕ̥tkos ‘bear’ with epithets: so e.g. Germanic *beran- ‘the brown one’ (cf. Lith. bė́ras, Ltv. bērs ‘brown’); Slavic medv-ědĭ ‘honey-eater’, which has a parallel in OInd. madhv-ád (but cf. common PIIr. ŕ̥ĉšas ‘bear’, absent in Slavic); and OPru. clokis, Lith. lokỹs, Ltv. lākis, from Proto-Baltic *talk-, *tlkak- hence ‘trampler, stomper, pounder’ (the IE root is conserved in Lith. irštvà ‘bear-den’). This taboo has a parallell in Finno-Ugric languages: compare PF *karhu, from *karheda ‘rough, coarse’; PFP from PU *oča ‘dam net, enclosure’, possibly by association to the asterismand mythical origin fromBig Dipper (cf. PF *otava, Saami oahci, oahtse, Permian ); in Ob-Ugric languages the bear is the most sacred and the most feared animal, and is called ‘the old one of the forest’, ‘the little idol’, ‘the holy beast’, etc. (Cushing 1977).

·       Shared folk tales among Eastern Baltic cultures across the Baltic Sea, and also through East European cultures in the Forest Zone, also point to strong interaction between Germanic, Balto-Slavic, and Finno-Permic populations (Bortolini et al. 2017).

All these innovative traits and cultural similarities could have also been aquired through intense Bronze Age contacts between Palaeo-Germanic peoples from Scandinavia and Pre-Balto-Slavic peoples from Central-East Europe through the Northern European Lowlands and the Baltic Sea, without a need to propose a third, intermediate language. This is also supported by the closer lexical and morphological influence between Germanic and Baltic languages, product of later contacts in the same regions. Interesting in this regard are a certain number of borrowings from late Palaeo-Germanic into Proto-Balto-Finnic and Proto-Slavic (see below § Proto-Germanic loanwords in Slavic and Finnic).