B.7 Areal effects in prehistoric contacts between Uralic and Indo-European

Tuesday–Wednesday, August 22–23, Department of Slavic Studies (Slawistik), room 1

Organizers: Sampsa Holopainen, Santeri Junttila, Juho Pystynen, Riho Grünthal

(Original symposium call)

Tuesday 23 August

13:30–14:30 (Keynote)

Petri Kallio: 

The Proto-Balto-Slavic Loanword Stratum in the Uralic Languages

As is well-known, there are around 200–300 Early Baltic loanwords in Common Finnic. My paper will focus on whether there are also even earlier Proto-Balto-Slavic borrowings in Finnic and other West Uralic branches. This possibility has been hinted before but only in the case of individual loan etymologies, whereas the Proto-Balto-Slavic loanword stratum in West Uralic has hardly ever occurred to anyone. The reasons for this are twofold. First, most Uralicists are unaware of the Balto-Slavic theory. Second, most Indo-Europeanists have never offered systematic phonological reconstructions of Proto-Balto-Slavic in spite of the fact that the Balto-Slavic unity is now more often accepted than not.

My conclusion is that the Proto-Balto-Slavic loanword stratum in West Uralic can be established even without suggesting any new loan etymologies. True, some of the suggested Proto-Balto-Slavic loanwords could still be considered either Northwest Indo-European or Early Baltic, but at least Northwest Indo-European is a priori less probable than Proto-Balto-Slavic because newer words constantly replace older ones. Then again, Proto-Balto-Slavic was simply the intermediate stage between Northwest Indo-European and Early Baltic, thus confirming permanent contacts between the two language families.

The West Uralic proto-language was probably spoken in the Volga-Oka region during the second millennium BC. Thus, Proto-Balto-Slavic must have been spoken both spatially and temporally nearby, perhaps corresponding to the Trzciniec-Komarov cultural complex (ca. 1800–1200 BC) between the Baltic Sea and the Black Sea. Be that as it may, since West Uralic and Balto-Slavic were contemporaries located in the East European Plain, there should be nothing sensational about the lexical borrowings between them. On the contrary, the absence of such loanwords would be much more surprising.


Luobbal Sámmol Sámmol Ánte (Ante Aikio):

Proto-Indo-European loanwords in Proto-Uralic? A critical analysis of an alleged ancient loanword stratum 

For some reconstructed Proto-Uralic word-stems, Proto-Indo-European loan etymologies have been proposed. Such etymologies have been claimed to demonstrate that the proto-languages of the two families were in contact with each other even prior to their expansion and divergence (e.g., Koivulehto 1991; 2001). Although this interpretation has met with criticism (e.g., Campbell & Garrett 1993; Janhunen 1999; Helimski 2001), so far there has been no attempt to present a detailed critical evaluation of the entire relevant etymological material.

The aim of this presentation is to present such an analysis of the Proto-Indo-European loan etymologies presented for Proto-Uralic and ‘Proto-Finno-Ugric’ word-stems in the works of Jorma Koivulehto (about 40 etymologies in total). First , the individual etymologies will be critically assessed in regard to both the validity of the Proto-Uralic and Proto-Indo-European reconstructions themselves and the alleged phonological and semantic correspondences between the two families. Second, two general methodological problems that may increase the possibilites of finding chance correspondences are discussed : 1) the expansion of the ‘Proto-Indo-European’ etymological corpus to include words with a limited distribution in the language family, and 2) the postulation of many simplifying sound substitutions by which the more complex phonological system of Proto-Indo-European may be represented in Proto-Uralic. The results indicate that a over half of the individual etymologies suffer from problems that render them quite uncertain, and the remaining material is too small to provide clear evidence of contacts between family-level proto-languages, especially considering the two aforementioned general methodological problems. While some of the well-behaved etymologies probably involve genuinely related forms, they may also represent later borrowings from (Proto-)Indo-Iranian to a post-Proto-Uralic dialect continuum.


Campbell, Lyle & Garrett, Andrew 1993. Review: Jorma Koivulehto, Uralische Evidenz für die Laryngaltheorie (1991). — Language 69: 832–836.

Helimski, Eugen 2001. Early Indo-Uralic linguistic relationships: real kinship and imagined contacts. — Christian Carpelan & Asko Parpola & Petteri Koskikallio (eds.): Early Contacts between Uralic and Indo-European: Linguistic and Archaeological Considerations. Mémoires de la Société Finno-Ougrienne 242.

Janhunen, Juha 1999. Critical remarks on current Indo-Uralic comparisons. — Hasselblatt, Cornelius & Paula Jääsalmi-Krüger (eds.), Europa et Sibiria. Beitrdge zur Sprache und Kultur der kleineren finnougrischen, samojedischen und paleosibirischen Völker. Gedenkband für Wolfgang Veenker. Veröffentlichungen der Societas Uralo-Altaica 51. Wiesbaden: Harrassowitz. Pp. 211–215.

Koivulehto, Jorma 1991. Uralische Evidenz für die Laryngaltheorie. Wien: Verlag der österreichischen Akademie der Wissenschaften.

Koivulehto, Jorma 2001. The Earliest Contacts between Indo-European and Uralic Speakers in the Light of Lexical Loans. — Christian Carpelan & Asko Parpola & Petteri Koskikallio (eds.): Early Contacts between Uralic and Indo-European: Linguistic and Archaeological Considerations. Mémoires de la Société Finno-Ougrienne 242: 235–263.


Christopher Culver:

Iranian evidence for the reintroduction of *ä to the pre-Proto-Mari vowel system 

Proto-Uralic first-syllable *ä was raised in pre-Proto-Mari and therefore any word reconstructible for Proto-Mari that features first-syllable *ä (> Hill Mari and Northwestern Mari ä, Meadow and Eastern Mari a, see Aikio 2014) must be regarded as a borrowing from elsewhere. Though the bulk of such Mari words are known to be loans from Chuvash or Permian, Gábor Bereczki (2005: 193) has noted in passing that Mari *ä (*a in his reconstruction) is found in words for which Iranian etymologies have been proposed over the 20th century, for example the very ethnonym *märi of the Mari people.

The present contribution gathers all proposed Iranian loan etymologies for Mari vocabulary featuring Proto-Mari first-syllable *ä, drawing on the work of such scholars as Joki (1973), Gordeev (1967, 1979–1983), and Kazantsev (1980). As great strides have been made in recent years in both Iranian historical linguistics and our understanding of Mari diachrony, these proposals are re-examined in the light of present knowledge. Then new etymologies are adduced, e.g. PMari *äpsät ‘smith’ from Proto-Iranian *abi-štata- ‘master craftsman’ (with a usage ‘smith’ attested in the Eastern Iranian branch).

While some earlier etymologies must be discarded, on balance the evidence suggests that *ä did indeed re-enter the Mari vowel system in the era of Mari–Iranian contacts. Additionally, because we find among this Iranian material sound shifts typical of the ancestor of the modern Ossetic language, the Alanic language of the South Russian steppes is proposed as the source.


Aikio, Ante 2014: On the reconstruction of Proto-Mari vocalism. – Journal of Language Relationship 11: 125–157.
Bereczki = Берецки, Габор 2005: Пермско-марийские лексические совпадения – заимствования или общие субстратные элементы?. – Linguistica Uralica 41(3): 187–200.
Gordeev 1967 = Гордеев Ф. И. 1967: Балтийские и иранские заимствования в марийском языке. In: Происхождение марийского народа. Йошкар-Ола. 180–202.
Gordeev 1979–1983: Гордеев Ф. И. 1979–1983: Этимологический словарь марийского языка 1–2. Йошкар-Ола.
Joki, Aulis 1973: Uralier und Indogermanen. Die älteren Berührungen zwischen den uralischen und indogermanischen Sprachen. Mémoires de la Société Finno-ougrienne 151. Helsinki.
Kazantsev = Казанцев, Д. Е. 1980: К вопросу о месте и времени прониковения иранских слов в древнемарийский язык. – Вопросы грамматики и лексикологии. Йошкар-Ола. 90–119.


Sampsa Holopainen & Niklas Metsäranta:

Indo-Iranian loanwords in Permic

In this paper we aim to present a revised stratigraphy of Indo-Iranian loanwords in the Permic languages. For the study of Indo-Iranian loans, Permic is an interesting branch, as the number of loans in Permic is higher than in any other branch of the Uralic family (with the possible exception of Hungarian). However, in earlier research, such as Joki (1973) or Rédei (1986: 61–82), the distribution of the loanwords in Permic has been poorly discussed, and the loans in Permic have been presented as a monolithic layer. Koivulehto (1999: 209) assumed that some loans confined to Permic are quite early borrowings, but he never dealt with the whole material of Indo-Iranian loans in the Permic languages. It has been argued (Koivulehto 1999: 208–209) that distribution is a problematic criterion when the stratigraphy of loanwords is studied, and when one takes a closer look at the loanwords in Permic, it is rather easy to see that they represent different layers.

In our presentation we will investigate the following questions: 1) How many different chronological layers are found in Permic? 2) What can be said of the donor language(s) – which loans show sound-changes that are typically Alanian (cf. Cheung 2002), and which are ambiguous or point to other varieties of (Indo-)Iranian? 3) Are the loans that are found in Permic and the neighbouring branches parallel loans or inheritance?

Additionally, we are firmly of the opinion that previous treatments haven’t exhausted Permic lexicon of Indo-Iranian loans and new loan words can still be found. Some hitherto undiscovered Indo-Iranian loan words in Permic include: Udm gur ’oven, stove’, gureź ’mountain, hill’, Komi gor(j) ’sauna stove’, goruv ’foot of a mountain’ < PP *gɔr ’mountain; stone; oven’ ← PI *gari- ’mountain, rock, cliff’, Udm merta-, Komi murtal- ’to measure’ < PP *murt(-) ’to measure; measurement’ < Pre-PP *marta(-) ← PII *maH-tra- ’measurement’. Also convincing alternatives to some earlier etymologies can be found: PP *pad (> Komi, Udm pad in pad-vež, pad-vož ‘cross-road’) is not from PI *pántHā- ‘path’ (cf. Rédei 1986: 74) but from a reflex of PI *padá- (> Avestan paδa- ‘steap, tread’, Ossetic fæd ‘footprint’) – the earlier etymology requires anachronic phonological developments, with *a retained but *nt > *d taking place after borrowing, which cannot be the case.


Cheung, Johnny 2002: Studies in the historical development of the Ossetic vocalism. Wiesbaden: Ludwig Reichert.

Joki, Aulis J. 1973: Uralier und Indogermanen. Helsinki: Suomalais-Ugrilainen Seura.

Koivulehto, Jorma 1999: Varhaiset indoeurooppalaiskontaktit: aika ja paikka lainasanojen valossa. – Pohjan poluilla. Suomalaisten juuret nykytutkimuksen mukaan. Ed. Paul Fogelberg. Helsinki: Suomen Tiedeseura. 207–236.

Rédei, Károly 1986: Zu den indogermanisch-uralischen Sprachkontakten, Wien: Verlag der Österreichischen Akademie der Wissenschaften.

Wednesday 24 August


Anthony Jakob:

Symmetry, sound change, and the Baltic loanwords in Finnic

Vowel changes should not be viewed in isolation, but as part of a larger system. In this talk, I discuss some cases of front-back symmetry in Finnic vowel changes that are revealed through the study of Baltc loanwords.

The substitution of Baltic *ō as Finnic *ū, *ou (1a), supports the idea that Finnic derives from an earlier *ā (1b), and also suggests that the Baltic loanwords postdated the Early Proto-Finnic change *ow > *uw (1c). The examination of corresponding changes to the front vowels brings us to a discussion of the "sporadic change" *ai > *ei usually assumed to have occurred in loanwords such as Finnish heinä 'hay'. As the principle of regular sound change demands a reconstruction other than *ai, it is suggested that this group rather refect PF *e̮i–a. In this context, it is suggested that PF *ī and *e̮i are regular substitutes of East Baltic *ẹ̄ (2a), that Finnic derives from an earlier *ä (2b) and that the Baltic loanwords postdated the Early Proto-Finnic change *ej > *ij (2c). Postulate 2a can be viewed as an argument for a specifcally East Baltic origin of the Baltic loanwords, for which other arguments are presented. Postulate 2c demands a reappraisal of Finnic *tīneh 'pregnant (of animals)'. It is argued that this word cannot derive from *tejnəš, as usually assumed, but must rather refect *tüjnəš, with a late change *üj > *ij (paralleling the established sound law *iw > *üw). In this context, I ask whether *tīneh really can be considered a Baltic loanword.


Santeri Junttila:

Finnic loanword studies in the 2010's

Our knowledge of the Finnic past owes much to loanword studies, since most neighbouring Indo-European languages have entered the literary era centuries earlier. Prehistoric layers of lexical borrowings have thus been a favoured subject of academic research since Vilhelm Thomsen.

The research history of Finnic etymology was investigated/analysed by Santeri Junttila (2015), with an emphasis on the borrowings from Baltic to Proto-Finnic. Junttila’s overview covers the period from 1869 to 2009. A further decade of active loanword research has passed since then, concluding at the 150th anniversary of Thomsen’s dissertation in 2019. New information has been gained, new interpretations have been presented and new questions have arisen. My presentation will shortly evaluate the results from the latest decade of research.


Ilona Kivinen:

Adjectives behaving badly – The adjectives with suffix -is in North Saami

The adjective system in the Saami languages has interested many scholars, but its complexity has been reducing the real interest to study the adjectives in Saami more closely. Mostly the interest has concerned the attributive marker -s and its origin (Nielsen 1933; Rießler 2006).

Both derived and non-derived adjectives may or may not have separate attributive form. The attributive forms also vary between the Saami languages and dialects. The attributive is not always marked, instead, sometimes the marked form is predicative, such as North Saami ođas ’new (pred.)’, which has attributive form ođđa ’new (attr.)’. The predicative form is marked with -s whereas the attributive form is unmarked.

North Saami is an interesting language for this type of study because it lies in between the eastern and western group of Saami languages. Apparently it seems that partly the adjective marking system in Saami has emerged during the time of division of Proto-Saami, which can partly explain the complexity of the system (Rauhala 2018). The eastern and western Saami languages have developed slightly different systems and contacts between the dialects have increased the complexity especially in North Saami.

This paper takes into closer observation the adjectives ending with -is in North Saami. It is a suffix that can be divided into different inflectional types. At least three types are seen: the contracted type njálggis: njálgá, attributive form njálgga, three syllabled type boaris: boarrása, attributive form boares, and another three syllabled type stuoris: stuorrása, attributive form stuorra.

This presentation concentrates on the etymological background of the adjectives with suffix -is. The material for this study has been collected from Álgu – etymological database for Saami. I will try to find out whether etymological background affects the inflectional type of these adjectives. How did such word types emerge? How old loanwords there are? Have some types attracted certain loanwords? Such questions I will try to answer. My current hypothesis is that especially Scandinavian loanwords have been taken to the type stuoris: stuorrása, with attributive stuorra, but the question remains if the whole type has emerged together with these loanwords.


Nielsen, Konrad 1933: A note on the origin of attributive forms in Lapp.– Liber semisaecularis Societatis Fenno-Ugricae. Mémoirés de la Société Finno-Ougrienne 67. Helsinki: Société Finno-Ougrienne. 296–307.

Rauhala, Ilona 2018: Suorggitmeahttun adjektiivvaid ráhkadanvuogit vuođđosámegielas. – Dutkansearvvi dieđalaš áigečala 2/2018: 85–128.

Rießler, Michael 2006: Om samiskans attributiva adjektivform. – Andrea Amft & Mikael Svonni (eds.): Sápmi Y1K. Livet i samernas bosättningsområde för ett tusen år sedan. Sámi dutkan 3. Umeå: Umeå universitet. 135–150.


Martin Joachim Kümmel:

Early satem loanwords in Uralic: Indo-Iranic or Balto-Slavic or what?

Most very early apparent loanwords from Indo-European into Uralic show features of an early satem stage of IE, i.e. affricate or sibilant reflexes of IE palatals and no evidence for labiovelars. Traditionally these have been mostly assigned to Indo-Iranic which is known to have had long-lasting contact with Uralic in later times. Other early borrowings are ambiguous with regard to satem reflexes but otherwise fit well and have therefore also assigned to the same stratum. However, the vocalism of some of these words does not fit Indo-Iranic well, since they appear to show mid vowels in Uralic, and Proto-Indo-Iranic (PIIr) merged all non-high vowels in phonologically low */a/, */ā/. Especially Uralic *e seems to correlate with original PIE *e, e.g., in *rećmä 'rope', *sejti 'bridge', *pejmä 'milk', *mertä 'man', *kečrä 'spindle', *jewä 'grain', *tejniš 'pregnant animal', while for *o there is more evidence that it may continue PIIr *a as well. Hence the question arises whether these items can represent a prestage of PIIr, possibly even late PIE itself, or whether we might instead assume a stratum of borrowing from a different satem branch, e.g., Balto-Slavic which probably had areal contact with Indo-Iranic. The aim of this contribution is to critically discuss the criteria for distinguishing sources of such early loans against the most recent reconstructions of Indo-Iranic, Balto-Slavic and early stages of Uralic. Some cases show unmistakable features of Indo-Iranic, viz. West Uralic *kekrä 'circle' with *r for PIE *l, while the near-synonym *keklä (only in Saamic) rather looks its prestage or rather a borrowing from a different IE language. But in most cases, this is less clear, and a more thorough investigation of this question might be useful. An important point here is if there are any cases that require a Balto-Slavic source and can thus be used to establish this possibility beyond doubt.


Niklas Metsäranta, Terhi Honkola & Veronika Milanova:

Borrowability of kinship terms in Uralic

The aim of our research is to study the kinship terms and their borrowability in Uralic languages, both from non-Uralic languages to Uralic and within Uralic. We collected kinship terms from twenty Uralic languages and used a list of in total 115 kin terms as the basis of our data collection. We collected kinship terms from each main branch of Uralic: Saami, Finnic, Mordvin, Mari, Permic, Khanty, Mansi, Hungarian and Samoyed. The research material is collected from individual dictionaries.The main research questions we seek to answer based on our material are:

1.Which kinship terms have been borrowed in Uralic languages?

2.When have the borrowings taken place?

We have attempted to answer the first research question by going through etymological materials written on Uralic languages, etymological dictionaries, articles etc. to see which kinship terms in our material have a loan etymology. The reliability of loan etymologies was evaluated and the most convincing ones (classified as probably or clearly borrowed) are considered reliable enough to base any further assumptions on. In our presentation we will be reporting our results, i.e. which kinship terms are generally loanwords in Uralic languages we surveyed and compare our findings with those observed in a global study of kin term borrowings (Honkola & Jordan in revision).In regards to the second question, there are different layers of loanwords, the main bulk of them are either older loans borrowed into the proto-language of an individual branch, e.g. North Saami gállis ‘husband, old man’ < Proto-Saami *kāllēs ← Proto-Scandinavian *karlaz ‘man, old man’ (Aikio 2009: 250) or rather recent loans to individual languages, e.g. Skolt Saami zeeʹtt ‘son-in-law’ ← Russian зять ‘id.’ (Itkonen 1958 [2011]: 540), Udmurt kart ‘husband’ ← Tatar kart ‘old man’ (Csúcs 1990: 207). There are only a handful of borrowed kinship terms which based on their distribution could be more ancient than the proto-language for each individual branch, Proto-Saami etc. The word for ‘younger sister’, for example, is found in Mordvin, Mari and Permic, Erzya Mordvin sazor, Meadow Mari šüžar, Udmurt suzer, which are borrowed from Proto-Indo-Iranian *swasar ‘sister’ or later from Proto-Iranian (Holopainen 2019: 222–224). However, as the words don’t exhibit vowel correspondences typical of inherited Proto-Uralic vocabulary, it seems safe to assume that despite being ultimately from the same source, they have been borrowed at a time when Proto-Uralic had already started to disperse into different dialects.Kinship terms are universal and part of the basic vocabulary, they are also closely connected to various systems of social organisation such as marriage and inheritance patterns. Hereby, studying the borrowing of kin terms allows us to detect places and times of supposedly intense contacts between speakers of different Uralic and non-Uralic languages, both past and more recent.


Janne Saarikivi:


Proto-Germanic, Proto-Finnic and Baltic Sea substrates

It is a well-established fact of the Uralic linguistic research that there is a notably Germanic vocabulary layer in Finnic and Saami. The amount of Germanic borrowings in Finnic may be approx. 550 (e.g. LÄGLW) of the 2200 or so Proto-Finnic lexemes. While the exact locating and dating of the Germanic - Finnic contacts continues to be an open question (e.g. Aikio 2006, Koivulehto 2014 for some of the recent accounts on this problematics), it is clear that the Proto-Germanic contacts were a central areal effect dividing Finnic and Saami apart of the rest of the Uralic.

In the Germanic studies, there is a growing consensus that the Germanic languages emerged on the basis of a Pre-Germanic substrate that is preserved in some parts of the Proto-Germanic lexicon not having parallels in the other Indo-European languages (Kroonen 2012, Kroonen 2013, and the literature mentioned in these sources). In my paper I proceed from the observation that there are also words considered as Germanic borrowings in Finnic that do not have any well-established etymology outside Germanic. Many of them represent words of local character, confined to the phenomena of the Baltic Sea area (Fi. hylje(h) < Proto-Finnic *šülkeš ~ English seal < Proto-Germanic *selxaz, Fi. siika < PF *siika ~ PG *sīka-z). I analyze this type of vocabulary pointing to a possibility that some of the words in question may represent not borrowings from Germanic to Finnic but vice versa, and even more likely, a kind of Baltic Sea substrate from extinct Palaeo-European sources that is common for both the Proto-Germanic and the Proto-Finnic.

Aikio, Ante 2006: On Germanic-Saami contacts and Saami prehistory. JSFOu 91. 9-55.
Koivulehto 2014. Die Urheimat der Germanen. MSFOu 274. 433-455.Kroonen, Guus 2012: Non-Indo-European root nouns in Germanic: evidence in support of the Agricultural Substrate Hypothesis. MSFOU 266. 239-260.
Kroonen, Guus 2013: The Proto-Germanic etymological dictionary. Leiden: Brill.


Abel Warries:

An investigation of possible Tocharian loanwords in Samoyed

Loanwords from Indo-European languages are well-known and quite numerous in most Uralic languages, and in some cases they go back to a very early time period. In particular, borrowings from (pre-)Proto-Indo-Iranian are so early that they can look similar to Proto-Uralic vocabulary; and some are found with a wide distribution in both western and eastern Uralic languages. Far fewer Indo-Iranian loanwords have been identified for the Samoyed languages on the eastern periphery of the Uralic family, although at least a few do exist (e.g. Janhunen 1983). In addition to these early Indo-Iranian borrowings, there are also a handful of Samoyed words proposed to derive from another branch of Indo-European, namely from Tocharian. However, there is stlll a lot of uncertainty and disagreement on that point. For instance, Kallio (2004) proposes one new loanword from pre-Proto-Tocharian, but he does not accept those proposed by Janhunen (1983) and Napoľskikh (2001).

In this presentation I will adduce and discuss some additional Samoyed words that may be derived from pre-Proto-Tocharian. These new comparisons are based on Proto-Samoyed reconstructions found in Samojedischer Wortschatz (Janhunen 1977) and on words that are only known from the northern Samoyed languages. Some regular patterns of sound substitution between the pre-Proto-Tocharian and the Samoyed forms may be tentatively established. I will furthermore assess whether these words are likely to be early or late borrowings into pre-Proto-Samoyed, and whether pre-Proto-Tocharian (and not rather Indo-Iranian) can indeed be considered the donor language for some or all of these words with the use of any diagnostic features available in the material.

Janhunen, Juha. 1977. Samojedischer Wortschatz: gemeinsamojedische Etymologien. Castrenianumin toimitteita 17. Helsinki: University of Helsinki.

Janhunen, Juha. 1983. “On Early Indo-European-Samoyed Contacts”. In: Juha Janhunen et al. (eds.), Symposium saeculare Societats Fenno-Ugricae. Helsinki: Société Finno-Ougrienne. 23-42

Kallio, Petri. 2004. “Tocharian Loanwords in Samoyed?” In: Irma Hyvärinen et al. (eds.), Etymologie, Entlehnungen und Entwicklungen: Festschrif für Jorma Koivulehto zum 70. Geburtstag. Helsinki: Société Néophilologique. 129-137

Napoľskikh, Vladimir. 2001. “Tocharisch-uralische Berührungen: Sprache und Archäologie” In: Christian Carpelan et al. (eds.), Early Contacts between Uralic and Indo-European: linguistc and archaeological considerations. Mémoires de la Société Finno-Ougrienne 242. Helsinki: Société Finno-Ougrienne. 367-383.


Krzysztof Witczak:


Are there Samoyed borrowings in the Indo-Aryan Languages?

The paper discusses the origin and etymology of the Indo-Aryan terms for ‘bile’, ‘yellow’ and ‘brass’. The lexemes have no known cognates either in Iranian or in other Indo-European languages; they are commonly treated as borrowings from an unknown source („aus unarischer Quelle gekommen”; Mayrhofer 1963: 292). In fact, a scrupulous analysis of the available Indo-European and Uralic lexical material shows that at least two of the Indo-Aryan items in question are probably Samoyed borrowings:

1. OInd. (AV) pittá- n. ‘bile’, Pa., Pk. pitta- n. ‘id.’ etc. (Turner 1966: 462) ← Samoyed dial. *pitta ‘bile’ (< Ur. *piša ‘bile’), cf. Selk. Ke patte, patta, Ta pătɜ, Ty. pāt, pāte ‘bile, gall’; Yur. O paďe ‘bile, gall’; Yen. Ch. fóreʔ, B fódeʔ ‘id.’; Twg. fáta ‘id.’; Kam. pada ‘bile, gall’; Koib. poda ‘id.’; Mot. hadìide ‘bile’ (Collinder 1977: 68; Rédei 1986: 384–385). The secondary meanings ‘yellow, green’ (see No. 2) and ‘brass’ (see No. 3) are also documented in the Uralic languages (including Finno-Permic ones). The Finno-Ugric languages use a different term for ‘bile, gall’, FU. *säppä, cf. Fi. sappi ‘bile, gall’, Est. sapp, Lap. N sap'pe ‘gall’, Mordv. M säpä, E sepe ‘bile’, Vog. tääp ‘gall of the bear’; Hu. epe ‘bile, gall’ (Collinder 1977: 126; Rédei 1986: 435–436). Thus, the Samoyed origin of the Indo-Aryan words seems certain.

2. Skt. pīta-, also pītaka-, pītala- adj. ‘yellow’, whence Pa. pīta-, pītaka- adj. ‘yellow’; Pk. pīa-, pīala- adj. ‘yellow; P. pī̃  m. ‘yellow colour of saffron’; G. pīyɔ m. ‘mucus or gum in the eyes’ (Turner 1966: 466) ← Sam. *pita ‘yellow, green, bile-coloured’ (< Ur. *piša ‘yellow, green’, originally ‘bile-coloured’), cf. Yur. O paďe jilāχā ‘green’, paďerāχā ‘bile-like; bile-coloured; yellow / gallenartig; gallenfarbig, gelb’ (← Yur. O paďe ‘bile, gall’); Selk. padә ‘green’, Ty. patelča- ‘to colour, dye / färben’; Taig. паттанге ‘green’ (Rédei 1986: 384–385). Cf. also Mordv. M. piža ‘green, young, small’ also ‘copper, brass’, E piže ‘green, blue’, also ‘copper, brass’. The Samoyed origin of the Indo-Aryan words seems highly probable for phonological reasons (Ur. * develops into Sam. *t).

The article also discusses the Indo-Aryan and Samoyed words denoting ‘brass’:
(3) Skt. pittala- n. ‘brass, bells / Messing, Glockengut’, also pītala- n. ‘brass’, whence Pk. pittala- n. ‘brass’; P. pittal m. ‘brass’, pitlī, pitlīā adj. ‘brazen’ etc. (Turner 1966: 462) vs. Ur. *piša ‘brass’, originally ‘a metal having a yellowish colour’, cf. Selk. Ty. patәrkä ‘brass / Messing’ (cf. Selk. Ty. pāt, pāte ‘bile’, patelča- ‘to dye / färben’). Note also Mordv. M. piža ‘copper, brass / Kupfer, Messing’, E piže ‘copper, brass’ (< ‘green, blue’).
It is suggested that the Indo-Aryan terms for ‘brass’ should be treated as a separate innovation, created on the basis of the two borrowings indicated earlier (see No. 1 and No. 2), rather than a third item of Samoyed origin.

Björn Collinder, 1977: Fenno-Ugric Vocabulary. An Etymological Dictionary of the Uralic Languages. Hamburg: Helmut Buske Verlag.
Manfred Mayrhofer, 1963: Kurzgefaßtes etymologisches Wörterbuch des Altindischen. B. II. Heidelberg: Carl Winter. Károly Rédei, 1986: Uralisches etymologisches Wörterbuch. Budapest: Akadémiai Kiadó.
Ralph Lilley Turner, 1966: A Comparative Dictionary of the Indo-Aryan Languages. London: Oxford University Press.