3.5. Disintegrating Uralic

3.5.1. Disintegrating Uralic evolution

Common Finno-Ugric traits which were probably shared by Samoyedic before its separation include the following:

·       General OV order.

·       Further development of copula support, including copular verb, indefinite/definite terms, third/non-third persons, present/past and indicative/non-indicative oppositions.

·       Cases:

o   Collective marker *-k.

o   Ablative or separative *-ta/-tä.

o   Innovation trend with further distinction of the three local cases, although precise details for a common stage are obscured by later developments.

o   Locative adverb in *-t(t).

·       Coaffix *-s- from western languages is probably to be traced back to the expanding lative of this period.

·       Verbal developments:

o   Past tense marker *-i/i̯, apart from the common in *-ś.

o   Past perfect *-ma/-mä and present in *-pa/-pä may be traced back to this stage, too.

o   Development of a common passive construction can be attributed to this stage, although the innovation continues differently in Proto-Finno-Permic and Proto-Ugric.

·       Development of subordinating conjunctions to combine sentences (apart from the use of nonfinite constructions), probably under the influence of neighbouring Indo-European languages.

·       Reflexive probably formed at this stage from demonstrative pronouns *e- + *čV, but also possibly from a noun meaning ‘(shadow) soul’ (compare the reflexive developed in Proto-Samoyedic).

·       Regular phonetic changes include:

o   In the first syllable, LPU *äx PFU *ē, LPU *ax → PFU *ō, as well as *VV → *V in a closed syllable, and *o → *u in open syllables before a second syllable *-i.

o   In the second syllable, the opposition *i vs. *ë is probably already neutralised, and only *i is retained. This leads eventually (in a process that continued in the different dialects) to the abolition of *ëand *̅in the first syllable, and *i may then be found with *a or *ä in the second syllable.

o   The consonantal system undergoes little change, with only one systematic evolution of *Vx → *VV before a consonant.

3.5.2. Dialectal division

Jaakko Häkkinen (2009) has suggested that Proto-Uralic was early on split into three dialectal groups: West Uralic (later split into Samic, Balto-Finnic, and Mordvinic); Central Uralic (split into Mari and Permic, or both separate from the beginning); and East Uralic (later split into Samoyed, Ob-Ugric, and Hungarian). The Western and Central Uralic groups may have remained united for some time after the separation of the eastern group, since several innovations may be singled out. This phylogenetic tree (Figure 8) is gaining traction among a varied group of Uralicists, including Juha Janhunen (personal communication, 2012), Asko Parpola (2013), or Petri Kallio (2015).

Figure 8. The revised family tree of the Uralic branches (after Häkkinen 2007, 2009). Image modified from Kallio (2015).

The virtual identity of Proto-West-Uralic and Proto-Uralic suggests that Proto-Uralic spread fast (Parpola 2013), likely around the mid-third millennium BC (Kallio 2015), in all likelihood related to the expansion and evolution of Eastern Corded Ware groups, including Battle Axe, Fatyanovo-Balanovo and Abashevo, all of them particularly linked through a shared Corded Ware ancestry and dominated by closely related paternal lineages. This supports a late group of clans (A-Horizon?) which expanded with Corded Ware groups from the Vistula to the east, with Fennoscandian groups maintaining close contacts with populations up to the Urals through the Upper Volga.

All nine well-attested subgroups (Balto-Finnic, Samic, Mordvinic, Mari, Permic, Hungarian, Mansi, Khanty, and Samoyed) are believed to have split soon within the next millennium, because their shared phonological and morphosyntactic isoglosses are rather limited (Kallio 2015). The early expansion of Ugric-Samoyedic (Häkkinen 2012) with Abashevo-related groups into the Andronovo-like cultural horizon through the Seima-Turbino phenomenon seems straightforward (Parpola 2013). However, such a simplistic division of the western group cannot be so easily done, due to convergence and divergence phenomena among Uralic dialects in north-eastern Europe. The possibility of a Finno-Permic-speaking Fatyanovo-Balanovo seems to be preferred in the current literature, with Fatyanovo/Netted Ware expanding West Uralic to the north-west (Parpola 2018), and Balanovo/Chirkovo-Kazan expanding Central Uralic to the north-east.

Nevertheless, it seems certain that the Battle Axe culture must have spokenif not the actual Finno-Samic dialectsa Para-West Uralic dialect in continuous contact with the West Uralic area, based on 1) the early separation of Samic from the core West Uralic area (Parpola 2013); 2) the lack of non-Uralic substrates in Balto-Finnic (Kallio 2015); 3) the scarce non-Uralic topo-hydronymy in the East Baltic and around the Gulf of Finland (Saarikivi 2004), comparable to that on the Upper Volga region; 4) the strong influence of a Balto-Finnic-like substrate on Pre-Germanic (see §4.12.2. Finno-Samic influence on Pre-Germanic) and Proto-Balto-Slavic (see §4.13.2. Uralic influence on Balto-Slavic); 5) the Palaeo-Germanic and late Balto-Slavic / early Proto-Baltic superstrate on Balto-Finnic (see below §4.18. West Uralic); and 6) the easy replacement of that hypothetic Para-West Uralic dialect by incoming Balto-Finnic peoples. Finno-Permic

The following is the likely phonetic evolution of Disintegrating Uralic to the Proto-Finno-Permic stage (Sammallahti 1988):

·       In stressed syllables *ë *, *̅ *ō (through an intermediate *̅?); *ë *ŭ, *a *u, although PFU *i had already started lowering towards *a.

·       The consonantal paradigm remains largely the same, although * *v (with secondary * developed word-initially in Proto-Permic).

·       The phonotactic system is somewhat different from the previous one: there are several PFP stems with word initial *r; geminate consonants may have also contained *čč.

3.5.3. Contacts with Indo-Iranian

Contacts between Disintegrating Uralic and Proto-Indo-Iranian were intense and long-lasting, as revealed by the different loanwords proposed to have been acquired in different stages (Koivulehto 1991; Carpelan and Parpola 2001; Katz et al. 2003; Blažek 2005)even if some are disputed (Aikio and Kallio 2005)which have survived in spite of strong posterior Indo-European influences, such as that of Palaeo-Germanic on Finno-Samic (see below § Palaeo-Germanic borrowings).

Pre-Proto-Indo-Iranian loanwords in Disintegrating Uralic[xvi] include the following:

·       PFU *aiša ‘shaft’< Pre-PIIr. *aīšā́ ‘shaft’ (< CIE *ħihseħ/ʕwihseħ ‘shaft’)‚ cf. OInd. īšā́, Av. aēsa-.

·       PFU *en-ti < Pre-PIIr. *gjen-ti <*genh- ‘be born’.

·       PU *inš-mi *išmi ‘wonder, sign’ < *gjn̥-né/n-ˀ- ‘know, recognise’ (Koivulehto 1991). Similarly, PFU *inši ‘man’ < PIIr. *gjˀ-(i)e ‘generate’ (Kümmel et al. 2001) hence ‘offshoot; creation, being; kin, family’. The substitution of *gjn̥- by *in is explained by the impossibility of the consonant group **i̯n- in Uralic (Koivulehto 1991), while the presence of a laryngeal may be explained by the late survival in specific groups (see below §II.2.5.2. *CR̥HC). Compare for the adoption from a palatalised velar PFU *sei- ‘eat’ < (Pre-)PIIr. *e- <*gi̯eh- ‘chew’ (Koivulehto 2003).

·       PFU *kekrä ‘cycle’ < Pre-PIIr. *kekro-‚ cf. Skt. cakra- ‘wheel, cycle’.

·       PFU *kesträ ‘spindle, spin’ < Pre-PIIr. *kētstro-, cf. Skt. cāttra-m ‘spindle’.

·       PFU *mertä ‘man, person’ < Pre-PIIr. *m-- ‘death; mortal’ cf. OInd. mr̥tá, OAv. mərəta ‘dead’, also in other LPIE dialects ‘mortal, person’. For o-grade Pre-PIIr. *mor-to- ‘mortal, man’, cf. OInd. márta, Av. masa-, hence PFP *marta, ‘dry(cow), farrow’ < PIIr. *márta-.

·       Maybe from this period LPU *oča ‘see; beware, guard; wait’ ~ Av. axša- < Pre-PIIr. *ok-se- ‘watch’ < *Hokw-se- ‘see; watch’, although it is difficult to explain the apparent Pre-PIIr. *-ks- PU *-č-, so possibly a later, PIIr. loanword that diffused also to Samoyedic. This word has also been explained as from the same root as PU *att- (*ott-) ‘see, look’, also ‘watch, guard, etc.’, although the medial *-t(t)- *-č- is equally difficult to explain.

·       Ob-Ugric *peečəɣ ‘cattle’ ~ Pre-PIIr. obl. *pečeu- points probably to an early, Pre-PIIr. loanword, before the evolution into PIIr. *pačau-. The PFP equivalent, *poča(), may point thus to a slightly later PIIr. stage, which may in turn suggest more continued contact of Finno-Permic with Indo-Iranian languages after the separation of the Ugric community.

·       Difficult to pinpoint is the origin of PFP porćas ‘piglet’ ~ PIr. *párĉah (cf. Av. pərəsa), because of the adoption with o-vocalism, which suggest an ancestral (palatalised) *porĉos.

·       PFU *-teksä ‘ten’ < Pre-PIIr. *dek-s-, cf. Skt. daśa- ‘ten’.

Proto-Indo-Iranian loanwords include:

·       PFU *ara ‘value, price’ < PIIr. *argha- < *algwha- ‘value, price’, also applied to the value of a slave (see above § Economy and technology).

·       PFU *asëra ‘lord, leader’ < PIIr. *asura- ‘god; leader, lord’, cf. OInd. Ásura-, Av. Ahura- ‘lord’.

·       PFU *ćata ‘one hundred’ < PIIr. atá- (<*któm), cf. Skt. śatám.

·       PFU *ama < PIIr. *ama- ‘twin’.

·       PFU *kurë ‘dig’ < PIIr. *kar-, enlarged *karš- ‘pull; plough’, cf. PIIr. *kší- ‘ploughing, furrow’, PIr. *kāraa ‘to sow, plant, plough’, etc.

·       PFU *mëkšë ‘bee’ < PIIr. *makš- ‘fly, bee’(Lubotsky 2001).

·       PFU *ora ‘awl’ ~ OInd. *ā́rā- (<*ēlā), after the merging of liquids, probably PIIr. and not Pre-Indo-Aryan as proposed by Koivulehto (1991), since it is found in other LPIE dialects with the same meaning. Also, Gmc. *ala(n)- <*HoH-ló-? remains unsuccessfully explained, and a borrowing from a Pre-PIIr. cognate is thus possible, although not warranted with the current data.

·       PFU *ora ‘slave’ < Pre-PIIr. *ara-, the self-denomination of Indo-Iranians, hence ‘Aryan taken as a war-captive, prisoner’, with a semantic shift mirroring Medieval Latin sclāvus ‘slave’, from Late Latin Sclāvus ‘Slav’, because Slavs were often forced into slavery in the Middle Ages.

·       PFU *šistV (*šikśtV) ~ OInd. siktha- ‘beeswax’.

·       PFU *sosra ‘one thousand’ < PIIr. *sa-hasra- / Pre-PIIr. *s-gjhesro- (<*s-ghéslo-), cf. OInd. sahásram, Av. hazanram.

·       PFP *śuka ‘awn, chaff’ < PIIr. / Pre-PIAr. *śuka- ‘needle’, cf. OInd. *śuka- ‘insect’s sting, ear of corn’, Av. *śuka- ‘needle, pin’.

·       PP *sur ‘beer’ < PIIr. *surā- ‘alcohol’, PFU borrowing due to the *s- (Lubotsky 2001).

·       PFP *taš ‘star’ < PIIr. (or PIr.) *tištria- ‘Sirius’.

It is difficult to distinguish any Pre-Proto-Indo-Aryan or Proto-Indo-Aryan loans from those usually proposed:

·       PFU *anta ‘grass’ < PIAr. ándhas- ‘sprout of the soma plant’. The Old Indian word has been connected to Gk. ánthos ‘flower’, less likely to Alb. endë ‘flour’ and Arm. and ‘field’; if so, then any cognate from DIE *andho- ‘sprout’ to Proto-Indo-Iranian would be as good a candidate for the loanword as the Pre-Indo-Aryan stage.

·       PFU *uχë- ‘drink’ ~ OInd. źuhṓti ‘pour in fire, sacrifice’; the initial * points to an earlier stage, cf. PIIr. *j́ʰu-j́ʰeu- <*gu-gheu- ‘pour’.

·       PFU *kuŋe ‘moon; month’ ~ OInd. Gugū́ ‘lunar Goddess’.

·       PFU *reś ‘rope’ < PIAr. *raśmi ‘rein’, is probably from a previous PIIr. *rać- < Pre-PIIr. *reć- given its vocalism, from IE *rek- ‘bind’ (Lubotsky 2001).

Loans closer to Proto-Iranian, identified by their meaning or phonology, include the following (Lubotsky 2001):

·       Ob-Ugric *koras ‘god; heavens’ < Middle Iranian *xar- ‘bright sun’.

·       PP *maäk/maäg ‘stake’ ~ PIIr. *maūkha-.

·       PP *ńań ‘bread’ ~ PIIr. *nagna-.

·       FV *oraśe ‘(castrated) boar’ ~ PIIr. *arā́ha ‘wild boar’, a non-IE word (see §3.4.3. Asian agricultural substratum).

·       PFP *śaka ‘goat’ ~ PIIr. *sćāga-/sćaga.

·       PFU *šorń⁠i or *šar(a)ńa ‘gold’ ~ PIIr. j́ʰaranya (<*gholʕw-) ‘gold’, cf. Av. zaranya-.

·       FP *śuka ‘chaff, awn’, only found in Iranian, cf. YAv. śuka- ‘needle’.

·       Vog. tas ‘stranger’ < Pre-PIr. *dasu- ‘foreigner’ (meaning shift from PIIr. ‘people’, see § Society and laws); an early borrowing due to the *s- (Lubotsky 2001).

·       PP *vork ‘kidney’ ~ PIIr. *vtka-.

For more on Indo-Iranian loans in Uralic, see Kümmel (2019).

3.5.4. Schleicher’s fable in Disintegrating Uralic

uči – šepät

uči, ńarana olima,

šepäi̯ näki;

će küsä u̯iχim u̯iχitä,

će enäm kantam,

će koi̯im suχim kantata.

uči šepäi̯ moni:

śüðjämä ćärkë

uräm šepäi̯ ai̯atam äntitä.”

šepät monit: “kuntal, uči!

śüðjämät ćärkë äntitä:

urä, asëra, učin śäχrätä

eči päu̯im erčam teki,

učin aptë epä olik.”

e kulimä uči ńurmik kulkiśa.


·       For ‘not having wool’, the more specific PFU word *ńarV ‘hairless skin’ is found in the first sentence in the essive case (in *-na), with the use of a copulative verb, and both terms in the nominative, with the dependent construction in the past perfect (or participle?).

·       For the negative verb, a system similar to Proto-Finno-Samic is used, conjugating it with the third person singular marked by dialectal LPU present *-pa/-.