Palaeolinguistics: The Homeland Problem

The practice of making inferences about the cultures of language users on the evidence of reconstructed languages is called linguistic palaeontology. These inferences may concern the material culture and geographic location of speakers as well as their social relations, mythology, and beliefs – the notion of ‘archaeological culture’ is used to capture both material culture and behaviour (Mallory 2020).

Proto-Uralic Homeland

This is the introductory post for my draft on the Proto-Uralic Homeland, which I have divided into eight pieces according to semantic fields or chronology, or both. During the following week, you will have the opportunity to read every day something more about the reconstructed Proto-Uralic vocabulary and its origin, and what it means for the location of the Urheimat, the date of the split-up relative to neighbouring languages, as well as its relevance for potential dialectal groupings and for the cultural evolution and contacts of its speakers.

The main problem with this task is that, as in Indo-European studies – including linguistics, but also archaeology and now population genomics – there is already an apparently unavoidable baggage of tradition and beliefs among scholars and laymen alike. Most of this baggage is outdated, fully irrelevant for the current state of the question, or just fully wrong, but it constantly drags the evolution of the field and muddles the issue at hand (cf. Laakso 2013).

NOTE. Furthermore, there is even the core issue of whether linguistic paleontology is useful or “scientific” at all (Mallory 2020), which runs parallel to the skepticism surrounding the relevance of migration or population genomics for archaeology or linguistics. I find this unending philosophical questioning usually unnecessary, and more often than not clearly underlying different sociopolitical beliefs, hence I will disregard them completely in this post series.

“Taciturn Uralic Hunter-Gatherers”

Recent chapter Proto-Indo-European, Proto-Uralic and Nostratic: A brief excursus into the comparative study of protolanguages, by J.P. Mallory In: Tracing the Indo-Europeans: New evidence from archaeology and historical linguistics (2019). Interesting excerpts (emphasis mine):

NOTE. The chapter might be read at least partly at Google Books.

The basic semantic fields of Proto-Uralic and Proto-Finno-Ugric have been investigated by Kaisa Häkkinen (2001) who divided their 743 lexical items (284 Uralic and 459 Finno-Ugric) into 37 semantic fields. In order to render her data more comparable, the reconstructed Uralic vocabulary was reclassified into semantic groups broadly similar to those employed above in Proto-Indo-European (…) In order to bring greater sense to the tables, the data were again subjected to the chi-square test to determine whether there were significant statistical differences among the different proto-languages according to semantic field.

An explanation for each of these differences would tax the most inventive of minds and certainly invites an opportunity to display some outrageous speculation. Who could resist?

If we examine Proto-Indo-European, it displays very little evidence of positive loading of its vocabulary compared to the entire data set. The only positive loading occurs for speech, which contrasts with Uralic that shows a negative loading for the same domain summoning up images of chatty Indo-Europeans and taciturn Finns. The negative loading for nature compared with the positive loading for the same domain in Uralic is striking and we see that Uralic is also positively loaded toward terms for flora and fauna as well. The 58 botanical lexemes of Proto-Indo-European are primarily associated with arboreal terms (31), and cultivated plants (21) and only five are possibly associated with wild plant species and consist largely of generic terms (…) We might then be tempted to seek the explanation within the different economic strategies of the speakers of the two proto-languages where Proto-Uralic was a primarily hunting-fishing-gathering society while Proto-Indo-European was associated with a fully developed livestock economy and (perhaps) some cereal agriculture (Mallory 2018). How large should we expect the botanical lexicon in a language to be and do, as the evidence might suggest, agricultural communities, because of their dependence on a restricted range of domestic cultigens and less interest in broad spectrum foraging, have markedly smaller lexical domains for (wild) plant species?

As to the size of the botanical lexicon, linguistic evidence generally suggests that farming societies in areas of very rich and diverse flora may have very sizeable lexical domains. (…) Berlin (1992, 97–98) cites a survey of 29 languages where the average number of plant categories was about 500. Apropos the distinction between hunters and farmers, Berlin’s own survey of 17 languages revealed that the mean number of botanical taxa for seven foraging societies was 197 while that for ten agricultural societies was 520. Berlin (1992, 284–290) emphasises the larger sub generic botanical vocabularies of farmers compared to gatherers and the reasons that have been advanced to explain it, e.g., crop failure among farmers might lead to greater concentration on the exploitation of wild plants than is the case with hunter-gatherers. But what little empirical evidence exists does not confirm this (Berlin 1992, 287) since there does not appear to be much of a correlation between edibility and naming. Moreover, Berlin cautions against accepting the contrasts in lexical size between farmers and foragers at face value since they have not been made between languages occupying the same ecological areas. We should also note that while Uralic has more botanical terms in proportion to its total reconstructed vocabulary, this too is largely comprised of tree names and names for parts of plants rather than specific names for wild plants.

Faced with such a clear disparity between the reconstructed proto-language and living languages of traditional populations, we might recall the statistical conundrum of the comparative method. There is certainly wide acceptance of Kaisa Häkkinen’s rule of thumb: ‘The narrower the distribution of a particular word, the younger it can usually be presumed to be’ (2001, 170). But then what are the odds that a very substantial proportion of the Proto-Indo-European lexicon relating to plants (on the order of 60–90%) should have disappeared from all twelve major Indo-European groups without surviving in at least one? This also poses the very real question of whether we can validly compare semantic categories between proto-languages or are we merely tabulating comparative attrition rates of once much larger lexicons?

The Language Contact Conundrum

The recent Journal of Historical Sociolinguistics (JHSL) Volume 6 (2020): Issue 2, Comparative sociolinguistic perspectives on the rate of linguistic change, holds some interesting papers on language change and language contact (some of them open access). The main summary of the whole special issue is found in Comparative sociolinguistic perspectives on the rate of linguistic change, by Nevalainen, Säily, and Vartiainen JHSL (2020). Relevant excerpts (emphasis mine):

The comparative historical perspective highlights variability in the rate of change across dialects and languages. Peter Trudgill puts this in a nutshell in his contribution to this issue: “[i]t is clearly true that different language varieties may change at different speeds”.

Although phonetic change may be gradual, at the phonological and lexical levels change can also be implemented abruptly (McMahon 1994: 47–68; Toon 1987: 276–291). (…)

In sociolinguistics, gradual change is typically traced in recorded performance data, where the diffusion of changes can be followed in and correlated with linguistic and nonlinguistic contexts across the speech community. Gradual change enables intergenerational transmission, and its rate can be measured at any given time. However, even a gradual change does not diffuse at a uniform rate but is usually expected to follow an S-curve model, which has a slow beginning, a longer but rapid mid-phase, and a slow tailing off (Blythe and Croft 2012; Denison 2003; Labov 1994: 65–66).

These approaches present three perspectives from which linguists have approached the rate of change in a given language. (…)

1. The focus can be on the chronological states of the language at large, such as Middle English compared to Old and Early Modern English. (…)
2. Different levels of linguistic organization can evolve at different paces, with syntactic processes such as word-order changes being generally slower than morphological changes.(…)
3. Descriptive empirical research shows the extent to which the trajectories of individual changes can alternate between phases of rapid, abrupt change and more or less stable frequency plateaus.(…)

Accounting for the mode of language change, extensive lists of drivers for diversity have been proposed and include factors such as population and community size, technological innovations, geography and ecology (e.g. resource availability and barriers to human contact), and a variety of social factors, among them, political complexity, society type, language maintenance and shift, and language and social identity (e.g. Campbell and Poser 2008: 332–363; Greenhill 2014: 561–572; Trudgill 2011: 21–26, 146–148). The various contingent factors listed by Dixon (1997: 76–85) range from natural causes, material innovations, forms of communication and trade to conquests and religious expansionism, occupations of previously occupied territory and expansion of populations into uninhabited territory.

A major theme running through the contributions in this issue is indeed language contact. It is considered in more detail by Peter Trudgill (2011 and this issue), who presents a typological framework of the sociolinguistic determinants which he shows correlate with the nature of linguistic changes over time. When it comes to the rate of language change, it is argued that languages that are spoken in relatively isolated communities tend to change gradually via intergenerational transmission (i.e. mostly through L1 acquisition), whereas a high proportion of adult L2 speakers in contact situations leads to rapid change (Operstein 2015: 11–13; Trudgill 2011). That is to say, the kinds of change that take place in a given language, and the rate at which these changes progress, crucially depend on the demographic and social history of its speakers. This “contact-genetic” view of language change takes into account the role of adult L2 learners as the driving force of (rapid) change (Thurston 1987: 36), while acknowledging the role of children in the stabilization process of the language variety that emerges from extended contact between two linguistic populations (Trudgill 2004: 28).

One way forward is to consider a wider range of simultaneous factors behind alternating rates of linguistic change as proposed, for example, by Natalie Operstein (2015: 8), who argues that the alternation between periods of abrupt and gradual change “is likely to be correlated with a specific type of societal structure, language transmission, typical pathways of linguistic change, sources of new linguistic structures, agents of change, as well as its outcomes”, in other words, a variety of drivers for change. In practice, the evidence for influences on language variation and the rate of change is gathered from various sources and on different levels of linguistic organization.

One example of a major cultural disruption in the history of Britain is the Norman Conquest (1066), which had a colossal impact on medieval English. Providing systematic corpus-based evidence for variation in the rate of language change over an extended period of time, Nevalainen et al. show that multiple interacting factors follow this punctuating event at some temporal remove from it. Although the Norman Conquest served as a catalyst for major social and linguistic processes, concomitant factors like variation in population size, for example, cannot be separated from the social consequences of epidemics and prolonged military activities that simultaneously increased social and geographic mobility and effected changes in social network structures in the language community.

What is “Uralic” and what “non-Uralic”?

The most striking aspect about the paper referred to by Mallory, written by Kaisa Häkkinen (2001: 169-178), is that it dedicated around half (!) its discussion to reject the value of loanwords and Neolithic-related words in assessing the potential Proto-Uralic homeland: ‘flour’, ‘porridge’, ‘bee’, ‘honey’, ‘sheep’, ‘copper/bronze’, ‘tin’, ‘domesticated animal’…all appeared to be reconstructible for the parent language (which we will see is not exactly the case for some of them), but meant little if anything for the “true” Uralic homeland, because they (a) are not found in all dialects, or (b) are loanwords, or (c) they might reflect something that Uralic speakers knew but didn’t practice, or (d) they might have been unrelated to their reconstructed meaning, or (e) are traditionally (e.g. UEW) “believed to be…”

Different ad hoc arguments which Häkkinen implicitly did not find applicable to the many other compiled reconstructs that the reader must assume are “naturally” Uralic. Believe it or not, this trend of rejecting loanwords and protoforms referring to the Neolithic package in Proto-Uralic as “non-Uralic” using any and all available dialectical resources remains as alive in Uralic studies as it was decades ago. Today, this trend of overanalyzing the same words focuses on a “more objective” linguistic perspective, to disembowel them and expose any irregularity, or to cast doubts about their ultimate meaning. Nevertheless, many words with similar or greater irregularities are assumed to follow some sort of unknown irregular sound laws that are yet to be found, or their far-fetched reconstructed meanings from some 50-year-old dictionary are taken at face value. The difference, indeed, lies in what the researcher would like to consider naturally Uralic.

The most evident reason for this unscientific behavior in Uralic (as in Indo-European) studies is putting unsubstantiated anthropological preconceptions above formal linguistic assessments: no matter the differing homeland preferences – be it the Urals, Siberia, the Arctic, or East Asia – the common aim is trying to negate a ‘Neolithic’ economy of Proto-Uralic speakers, by any means. I am sure that the personal and professional reasons for this behavior are manifold, but I honestly doubt anyone should be interested in assessing them: that bias is fairly easy to trace, and poses no problem at all for the interested reader. In fact, the need to continue obsessing about excluding these same set after more than a century of research is witness enough of the futility of that effort.

The less evident aim is the common desire to reconstruct the “core” Proto-Uralic vocabulary, “clean” of “foreign” influences. This objective – not rarely seen shamelessly expressed with exactly those same words – is as fallacious as it is attractive; to you, to me, and to everyone else. Because of its natural appeal, it is important to always take it into account as the most dangerous bias when assessing the homeland problem. Language contact and language change are intertwined (see above), and their effects are essential to understand language evolution in general, and most certainly Proto-Uralic in particular. Loanwords, of course, are an important aspect to assess such contacts.

Humor me for a moment, and imagine a futuristic post-apocalyptic world some two or three thousand years from now, where mostly Future American and Future Romance languages would be spoken, and only Middle English and Old/Middle French and some Proto-Romance could be reconstructed (see a similar example in this blog from three years ago).

• What some scholars in American Studies would be pretending to do when getting rid of (1) all French-like loanwords, of (2) the irregular substratum words [i.e. Celtic], and (3) everything that according to their preconceptions looks “non-English”, is to get to the “true” Old English homeland.
• However, what they would be desperately trying to do is to link the ancestor of Future American to some territory they believe is the “true” American homeland, using their preconceptions to guide their research.
• In the end, what they would actually get is merely a “Middle English without Anglo-Norman and Celtic”; not Old English, not Proto-West-Germanic, not Proto-Germanic, and certainly not their respective homelands…Just Middle English without the most relevant pieces of information to reconstruct its culture and to locate it in time and space. Furthermore, they wouldn’t be able to understand that there are other intermediate contact layers to be taken into account (such as the influence of Norse), which might be further distorting the situation.

The reality of the language does not allow us to distinguish “pure” or “core” vocabulary: Middle English must have been spoken in the most logical territory where it could have received continuous “Old/Middle French”-related influence (i.e. from Latin to Anglo-Norman and Middle French), and an earlier but also somehow continuous Celtic influence; in other words, Western Europe. It’s as simple as that. Everything that we would reconstruct from the language – its phonology, its dialectal evolution, the culture reflected by the meaning of words… – all would show what you would expect from those intense exchanges.

The next series of posts follows that simple example: I try to trace the Proto-Uralic homeland with all available Proto-Uralic reconstructions; just like reconstructing Middle English would be useful to trace the Middle English homeland. For a “clean” or “pure” Uralic homeland, I believe the fantasy genre would be much more appropriate.

Featured image: Erdal-Bilderreihe Nr. 126 Bild 3, by Gerhard Beuthner (1930?).

Bibliography

Please note:

Many words are referenced only with the UEW Nº and one or more recent texts which sum up all (relevant) previous references. This is, after all, a text about the Uralic homeland, not specifically about etymology.

• Aikio, Ante. “New and old Samoyed etymologies.” Finnisch-Ugrische Forschungen 57 (2002): 9-57. Link at Academia.edu
• Aikio, Ante. Studies in Uralic Etymology I: Saami Etymologies – LU XLIX (2013), 161-174. http://dx.doi.org/10.3176/lu.2013.3.01
• Aikio, Ante. “On the reconstruction of Proto-Mari vocalism.” Вопросы языкового родства 5 (127) (2014).
• Aikio, Ante. “The Finnic ‘secondary e-stems’ and Proto-Uralic vocalism.” Suomalais-Ugrilaisen Seuran Aikakauskirja 2015.95 (2015): 25-66.
• Aikio, Ante. “Studies in Uralic Etymology IV: Ob-Ugric Etymologies.” Linguistica Uralica 51.1 (2015): 1-20. Link at Academia.edu.
• Aikio, Ante. “Proto-Uralic.” (2019) — To appear in: Marianne Bakró-Nagy, Johanna Laakso & Elena Skribnik (eds.), The Oxford Guide to the Uralic Languages. Oxford University Press.
• Aikio, Ante. Uralic Etymological Dictionary (draft version of entries A–Ć) (opened for comments and discussion on , 1/17/2020)
• Beekes, Robert. Etymological dictionary of Greek (2 vols.). Brill, 2009.
• Bjørn, Rasmus Gudmundsen. “Pronouns and Particles: Indo-Uralic Heritage and Convergence.” The Precursors of Proto-Indo-European. Brill Rodopi, 2019. 30-49.
• Carpelan, Christian & Asko Parpola. “Emergence, Contacts and Dispersal of Proto-Indo-European, Proto-Uralic and Proto-Aryan in Archaeological Perspective.” Early Contacts between Uralic and Indo-European: Linguistic and Archaeological Considerations. Helsinki: Memoires de la Société Finno-Ougrienne 242 (2001): 55–150.
• Czentnár, András 2015: Egy finnugor etimológia török kapcsolatai. – Keletkutatás 2015. tavasz: 9–19.
• de Smit, Merlijn. “Contacts between Finnic and Celtic?.” (2019). Link at Academia.edu.
• De Vaan, Michiel. Etymological dictionary of Latin and the other Italic languages. Vol. 7. Leiden · Boston, 2008.
• Derksen, Rick. Etymological dictionary of the Slavic inherited lexicon. Brill, 2007.
• Derksen, Rick. Etymological dictionary of the Baltic inherited lexicon. Brill, 2015.
• Grünthal, Riho. “Finno-Ugric ’Dog’ and ’Wolf’.” Etymologie, Entlehnungen und Entwicklung: Festschrift für Jorma Koivulehto zum 70. Geburtstag. Mémoires de la Société Néophilologique de Helsinki LXIII (2004): 83-96.
• Grünthal, Riho. “Baltic loanwords in Mordvin.” Riho Grünthal–Petri Kallio szerk.: A linguistic map of prehistoric Northern Europe (2012): 297-343.
• Häkkinen, Kaisa. “Prehistoric Finno-Ugric culture in the light of historical lexicology.” Early Contacts between Uralic and Indo-European: Linguistic and Archaeological Considerations Papers presented at an international symposium held at the Tvä rminne Research Station of the University of Helsinki. 1999.
• Häkkinen, Jaakko. “Kantauralin ajoitus ja paikannus: perustelut puntarissa.” Suomalais-Ugrilaisen Seuran Aikakauskirja 2009.92 (2009): 9-56.
• Holopainen, Sampsa, Santeri Junttila, and Juha Kuokkala. “Indoeurópai jövevényszavak és a második szótagi labiális magánhangzók fejlődése az uráli nyelvekben.” A nyelvtörténeti kutatások újabb eredményei IX. (2017).
• Holopainen, Sampsa. “Indo-Iranian borrowings in Uralic. Critical overview of sound substitutions and distribution criterion.” PhD Thesis (2019).
• Holst, Jan Henrik. “On the theory of a Uralic substratum in Baltic.” Contacts between the Baltic and Finnic languages 151 (2015): 173.
• Hyllested, Adam. “Internal reconstruction vs. external comparison: the case of the Indo-Uralic larnygeals.” Internal Reconstruction in Indo-European: Methods, Results and Problems. Copenhagen: Museum Tusculanum (2009): 111-136. PDF at Academia.edu.
• Hyllested, Adam. “Again on pigs in ancient Europe: The Fennic connection.” Etymology and the European lexicon. Proceedings of the 14th Fachtagung der Indogermanischen Gesellschaft, 17–22 September 2012, Copenhagen (2017): 183-196.
• Klemenčič, Simona. “Bojan Čop’s Indo-Uralic Hypothesis and Its Plausibility.” The Precursors of Proto-Indo-European. Brill Rodopi, 2019. 88-101.
• Kloekhorst, Alwin. Etymological dictionary of the Hittite inherited lexicon. Brill, 2007.
• Kroonen, Guus. “The Proto-Indo-European Mediae, Proto-Uralic Nasals from a Glottalic Perspective.” The Precursors of Proto-Indo-European. Brill Rodopi, 2019. 111-114.
• Iversen, Rune, and Guus Kroonen. “Talking Neolithic: Linguistic and archaeological perspectives on how Indo-European was implemented in southern Scandinavia.” American Journal of Archaeology 121.4 (2017): 511-525.. Paper available at Leiden.nl.
• Junttila, Jussi Santeri. “Proto-Finnic loanwords in the Baltic languages?: An old hypothesis revisited.” Contacts between the Baltic and Finnic languages. Suomalais-Ugrilainen Seura, 2015. 12-38.
• Kallio, P. (2004). Tocharian loanwords in Samoyed. Etymologie, Entlehnungen und Entwicklungen. Festschrift für Jorma Koivulehto zum 70. Geburtstag, 129-137. Link at Academia.edu.
• Kallio, Petri. “Germanic ‘maggot’ and ‘moth’.” Amsterdamer Beiträge zur älteren Germanistik 54.1 (2000): 117-122. Link at Academia.edu
• Kallio, Petri. “Suomen kantakielten absoluuttista kronologiaa.” Virittäjä 110.1 (2006): 2-2.
• Kallio, Petri. “The non-initial-syllable vowel reductions from Proto-Uralic to Proto-Finnic.” Per Urales ad Orientem: Iter polyphonicum multilingue. Festskrift tillägnad Juha Janhunen på hans sextioårsdag den 12 (2012): 163-175. Download PDF.
• Kallio, Petri. “The Indo-Europeans and the Non-Indo-Europeans in Prehistoric Northern Europe.” Language and Prehistory of the Indo-European Peoples: A Cross-Disciplinary Perspective (2017): 187-203. Link at Academia.edu.
• Keresztes, László. “13. Mansi.” The Uralic languages. Routledge, 2015.
• Koivulehto, Jorma. Uralische Evidenz für die Laryngaltheorie. Vol. 566. Viena: Österreichische Akademie der Wissenschaften (1991).
• Koivulehto, Jorma, and Theo Vennemann. 1996. Der finnische Stufenwechsel und das Vernersche Gesetz. Beiträge zur Geschichte der deutschen Sprache und Literatur 118:163-182.
• Koivulehto, Jorma. “The earliest contacts between Indo-European and Uralic speakers in the light of lexical loans.” Early contacts between Uralic and Indo-European: Linguistic and archaeological considerations (2001): 235-263.
• Koivulehto, Jorma. “Frühe Kontakte zwischen Uralisch und Indogermanisch im nordwestindogermanischen Raum.” A. Bammesberger & T. Vennemann: Languages in Prehistoric Europe. Heidelberg (2003): 279-317.
• Koivulehto, Jorma. “Wie alt sind die Kontakte zwischen Finnisch-Ugrisch und Balto-Slavisch?.” Slavica Helsingiensia 27 (2006): 179-196.
• Kroonen, Guus. Etymological Dictionary of Proto-Germanic. Brill, 2013.
• Kümmel, Martin Joachim. “Thoughts about Pre-Indo-European Stop Systems.” The Precursors of Proto-Indo-European. Brill Rodopi, 2019. 115-130. https://doi.org/10.1163/9789004409354_010 (an old version is available at Academia.edu and ResearchGate.net)
• LIV². Rix, Helmut, and Martin Kümmel. LIV, Lexikon der indogermanischen Verben: die Wurzeln und ihre Primärstammbildungen. Vol. 54. L. Reichert, 1998.
• Lubotsky, Alexander. “The Indo-European suffix*-ens-and its Indo-Uralic origin.” The Precursors of Proto-Indo-European. Brill Rodopi, 2019. 151-162.
• Palmér, A. I., Jakob, A., Thorsø, R., van Sluis, P., Swanenvleugel, C., & Kroonen, G. (2021). Proto-Indo-European ‘fox’and the reconstruction of an athematic ḱ-stem. Indo-European Linguistics, 1(aop), 1-30.
• Parpola, Asko. “The problem of Samoyed origins in the light of archaeology: On the formation and dispersal of East Uralic (Proto-Ugro-Samoyed).” Per Urales ad Orientem: Iter polyphonicum multilinguae (2012): 287-98.
• Parpola, Asko. “Finnish vatsa~ Sanskrit vatsá and the formation of Indo-Iranian and Uralic languages.” Suomalais-Ugrilaisen Seuran Aikakauskirja 2017.96 (2017): 245-286. DOI: https://doi.org/10.33340/susa.70229
• Pystynen, Juho. “The State of the Finno-Permic Lexicon.” (2017).
• Saarikivi, Janne. “The divergence of Proto-Uralic and its offspring. A descendent reconstruction.” – To appear in: The Oxford Guide to the Uralic languages, Marianne Bakró-Nagy & Jelena Skribnik & Johanna Laakso (eds.)
• Schrijver, Peter. “Lost languages in northern Europe.” Early Contacts between Uralic and Indo-European: Linguistic and Archaeological Considerations. Mémoires de la Société Finno-Ougrienne 242 (2001). Link at Academia.edu.
• Schrijver, Peter. “Varia I. More on Non-Indo-European surviving in Ireland in the first millennium AD.” Ériu 55 (2005): 137-144. Link at Academia.edu.
• Witczak, Krzysztof Tomasz. “Are there traces of a Finno-Ugric substratum in Proto-Slavic?.” Slavisticna Revija 68.1 (2020): 73-89.
• Zhivlov, Mikhail. “Indo-Uralic and the Origin of Indo-European Ablaut.” The Precursors of Proto-Indo-European. Brill Rodopi, 2019. 219-235.
• Живлов, М.А. Неиндоевропейский субстрат в финно-волжских языках. X Чтения памяти С.А. Старостина ~ 27 марта 2015