On Magyar migration

Jan 13, 2011 by

In yesteday’s posting I mentioned that from the linguistic point of view the closest relatives of Hungarian are the Khanty and Mansi languages spoken about 2,500 miles to the northeast, on the eastern slopes of the Urals. The affinity of Hungarian with Khanty and Mansi can be illustrated by such cognates as the following numerals: the Mansi χūrəm, the Khanty xutəm and the Hungarian három for ‘three’; or the Mansi χōt, the Khanty xut and the Hungarian hat for ‘six’. But unlike Hungarian which is spoken in Central Europe, Khanty and Mansi are the language spoken by the indigenous inhabitants of the Khanty-Mansi Autonomous Okrug in the Russian Federation (its capital is Khanty-Mansiysk, an oil boom town and a centre of alpine skiing). Approximately 13,600 people speak Khanty and under 3,000 speak Mansi (generally, Mansi have been more assimilated by the Russians than the Khanty). So how come Hungarian is spoken so far from its linguistic brethren? The answer is migration.

The area around and just south of the Ural Mountains is probably the location of the original Finno-Ugric homeland. Historical linguists argue as to whether Finno-Ugric languages were originally spoken on the eastern or the western slopes of the Ural Mountains, but it is fairly clear that it was somewhere close to the Urals. The original homeland of the Hungarians (or Magyars, as they call themselves) -– known to historians as “Magna Hungaria” –- was located in the Kama basin in the upper Volga region.

But by the turn of the 9th century CE, the Magyars moved to the Pontic steppes, north of the Black Sea. In the next century, they moved further west, pushed by another nomadic group from the Eurasian flatlands: the Pechenegs. These were speakers of a Turkic language, which has heavily influenced the vocabulary of the Russian language.

Under the pressure from the Pechenegs, the Magyars crossed the Carpathian Mountains and by the end of the 9th century settled in the Great Hungarian Plain, the one place in Europe that looks like the great Eurasian flatlands and where pastoralist lifestyle can be easily maintained.

Here, the Magyars encountered a population speaking a Romance language, which was probably quite similar to an older form of Romanian. However, the Magyars managed to imposed their Ugric language on the local Romance-speaking population.

Interestingly, scholars now agree that the number of the Magyar conquerors was relatively small, at most about 30% of the resulting population. Moreover, their genes were further diluted by subsequent exchanges with their neighbors. As a result, the genetic effect of the Magyar conquest was modest at best: Hungary does not stand out on the genetic map of Europe, such as the map showing the distribution of PC2 (the second principle component of genetic variation among Europeans; the reader is referred to Cavalli-Sforza and Cavalli-Sforza 1995, Cavalli-Sforza 2000 for a more detailed discussion of how these principle components are calculated):

In contrast, Finns, Estonians, Karelians and especially the Saami are highlighted on the map of PC2. This genetic uniqueness of Finns (and related peoples) has been recently supported by the findings of the study conducted by the Institute for Molecular Medicine Finland (FIMM), which compiled the Finnish Gene Atlas based on DNA from more than 40,000 Finns. Unlike the Hungarians, speakers of Finnic languages managed to retain their special genetic profile because they descended from a more extensively Finno-Ugric founding population.

Related Posts

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below:

  • Erik the Reader

    "Here, the Magyars encountered a population speaking a Romance language, which was probably quite similar to an older form of Romanian. However, the Magyar s managed to imposed their Ugric language on the local Romance-speaking population. "

    Totally false, with due respect that is a pretty far fetched presumption on the Hungarian conquest!
    If you would read about the conquest of Pannonia in any history book you will not find it written that Pannonia was romance speaking.
    What about Svatopluk? 🙂 Was he romance?
    And what about the Moravians, Slovenes, Karantans, Franks, and Bavarians?

    A Hungarian

  • Asya Pereltsvaig

    @Erik the Reader: Thank you for your comment! Are you suggesting that the pre-Magyar population in the area was Slavic-speaking? This is a distinct possibility too, I agree.

  • Erik the Reader

    @Asya Pereltsvaig
    The historic record tells that Pannonia itself at the time of conquest was mostly slavic but with also avars and franks andt it belonged mostly to East Francia, Moravia and some regions also was of the Bulgars.

  • Erik the Reader

    Also one should remember that the Hungarians were already pretty mixed up at the time of the conquest.
    Many of the tribes of the Union were Turkic. Like the three Kabar tribes. If in Hungarians there ever were any Finno-Ugric genes, it must have been very low even at the conquest. And even after settling down in the new homeland the Hungarians enriched their ranks with other Turkic tribes too like the Cumanians (they settled them down in the Kunság) and Petchenegs.

    It's no surprise that the Hungarians have considered tradionionaly their origin as Turkic and in the 19 century dismissed the Finno-ugric theory regarding the Hungarian language.

  • Asya Pereltsvaig

    @Erik the Reader: It is quite possible that the Magyars picked up some Turkic genes and even linguistic features during their trek from the historic homeland into the present-day Hungary. I did mention the Petchenegs. Do you know of any specific examples of Turkic words in Hungarians? I've seen a few, such as kávé 'coffee', but it doesn't tell us much (the word has been borrowed into many languages that had meager interaction with Turkic speaking tribes). As far as I know, there are good cognates between Hungarian and Khanty/Mansi (also with Finnish, Estonian etc.), but what about basic vocabulary cognates with Turkic languages? Without this sort of evidence, any theory of Turkic origin for Hungarian is mere politics/ideology.

  • Erik the Reader

    @Asya Pereltsvaig
    At least 10% of Hungarian vocabulary is of Turkic origin, most of it comes before the
    Hungarian conquest. I list some words.
    An example:

    Cebimde çok küçük elma var. Zsebemben sok kicsi alma van. I have many little apples in my pocket.

    Some Word list:

    anne = anya = mother
    ata = atya = father (archaic)
    iki = kettő, két= two
    ikiz iker twean
    sayı szám number
    Kim = Ki = Who
    çok = sok = many
    küçük = kicsi = little
    var = van = there is
    ben = én = I
    o = ő = he/she
    Kiminle = kivel with whom
    kaç = hány = how many?
    sarı sárga yellow
    gök (sky) kék blue
    iyi = jó = good
    yeni új new
    eski old ócska old
    yaka nyak neck
    sakal = szakál beard
    ölmek (to die) = öl (to kill)
    çalmak (to steal)= csal (to cheat)
    elma alma apple
    kapı (door) kapu (gate)
    arpa árpa barley
    arslan oroszlán lion
    keçi kecske goat
    koç kos ram
    kapmak (take)= kap (get)
    buğday = búza wheat
    boğa = bika = bull
    yas gyász mourning
    yel = szél wind
    deniz tenger sea
    biti betű letter
    eke (let him sow) eke plough
    bil know bir have the power
    çarp slap, csap slap,
    dol get filled tel fill, full

    also of turkish origin are word denoting family like
    gyermek (child), kölyök (kid), iker (twin)
    body parts like térd (knee), boka (ankle), gyomor(stomach), köldök(umbilical), tar (bold), csipő (hip), szeplő (freckle)

  • Erik the Reader

    @Asya Pereltsvaig I have not asserted that the Hungarian languages is of Turkic origin, but I have stressed that many of the Hungarian tribes were ethnically Turkic which is a historical fact.

  • Asya Pereltsvaig

    @Erik the Reader: I have no argument about Hungarians being heavily mixed ethnically/genetically (Magyars, Slavs, Turkic groups etc.)

  • Asya Pereltsvaig

    @Erik the Reader: and thank you for the comment on Turkic words in Hungarian — very interesting!

  • Erik the Reader

    One thing I still have to mention that the region conquered by Hungarians was scarcely populated at that time. Hungarian historiography estimates the number of conquering Hungarians to 500,000 based on the fighting force they could yield. After settling down they ravaged Europe with campaigns for at least 70 years. The issue of genetic studies are interesting and in fact there are many. Some even support the historical or traditional view of the Hungarian population groups regarding their origin.


  • Asya Pereltsvaig

    @Erik the Reader: Thank you for your latest comment! There are plenty of genetic studies of Hungarians but from I've seen they strongly suggest that the "invaders" (or "settlers", if you prefer) could not have constituted that large a proportion of the resulting population, so the area could not have been "sparsely populated" as you suggest.

  • Erik the Reader

    "so the area could not have been "sparsely populated" as you suggest. "

    The problem is more complex then that. You can't sweep aside later events which deluted or altered at least the original geen pool. I can try to scatch some of it.

    You have to be aware that after the Hungarian conquest in the following centuries there were new population influx in Hungary and also great population loss occured like the Mongol onslaught in the 13th century. Cumans were settled down in the 13th century. A great number of German settlers were invited to Hungary during the Middle Ages. And with the advent of ottoman turks new problems arose. People running away from Turks sought refuge in Hungary. During the conquest of one third of Hungary by Ottomans and during the turkish wars Hungary experienced depopulation. After Ottoman turks left many parts of Hungary were repopulated by a new wave of immigrants. A new wave of Germans arrived Slovaks, Serbs were settled down in Hungarian regions.

  • Asya Pereltsvaig

    @Erik the Reader: Thank you for your comment and the information about Hungary's historic demography.

  • guest

    Several of Erik the Reader’s Turkic-Hungarian connections are questionable and rely heavily on visual or phonological similarity in line with the spirit of the mid-19th century when comparative linguistics was less rigorous and still very much “in development” with its methods. He ignores careful analysis of potential sound changes or even an examination of newer research from macro-comparative linguistics. In addition most of the other Turkic items in Erik’s list have been explained better as borrowings (e.g. alma ~ elma, gyermek ~ śarmъk, iker ~ ikiz) rather than reflexes of a supposed and exclusive Hungaro-Turkic or Hungaro-Altaic root.

    5 of the more interesting (mis)identifications by Erik are below.

    1) iki – két/kettõ

    There MAY instead be a link between ‘két/kettõ’ (and by extension Finnish ‘kaksi ~ ‘kahden’, Mansi ‘kit/kitig’ etc.) with Turkish ‘kat’ meaning “layer”, Mongolian ‘gagc’. meaning “single” and/or Manchu ‘Gaqda’ meaning “one of a pair”. If ‘iki’ and ‘két/kettõ’ were related as Erik insinuates, then he would fail to explain why the Turkic words lack a final ‘t’ (or some other dental consonant) whereas the Hungarian (and Uralic ones) words do (cf. Finnish ‘kaksi ~ kahden’, Mansi ‘kit/kitig’ etc.)

    2) var – van

    There MAY instead be a link between Hungarian ‘van’ (from an earlier ‘val-‘ ~ ‘ol-‘) and the Mongolian ‘ol-‘ “to obtain” and/or the Manchu ‘o-‘ “to make, become”). On the other hand, Turkish ‘var’ has been reconstructed as descending from an earlier noun ‘bār’ meaning something like “existence”. A derivative of this old word is the Old Turkic ‘barim’ meaning “riches” which in turn probably entered Hungarian as a loanword ‘barom’ meaning “idiot” or “beast” (in semi-nomadic societies, wealth was positively related to quantity of animals owned).

    3) o – ő

    This relies too much on visual similarity. However there MAY instead be a link between Turkish ‘o’ and Hungarian ‘a(z)’ “that”. Demonstratives in one language can turn up as personal pronouns in other
    languages (e.g. Finnish ‘tämä’ meaning “this” being cognate with
    Estonian ‘tema’ meaning “he/she/it”). The Hungarian pronoun ‘ő’ is most likely a cognate of Sami ‘son’ meaning “he/she” among other Finno-Ugric words as there’s a pattern with certain words in Finnic or Sami that begin with ‘s-‘ lack it in Hungarian and the Hungarian word tends to have a long vowel arising from having lost a consonant cluster or final consonant still existing in the Finnic or Sami word (e.g. Finnish ‘syli’ and Hungarian ‘öl’ both meaning ‘lap’; Finnish ‘syksy’ and Hungarian ‘ősz’ both meaning “autumn”).

    4) yeni – új

    This relies excessively on a vague similarity between Turkish’s initial ‘y-‘ and the Hungarian word’s final ‘-j’. The Hungarian word has been linked most convincingly with Finnic words (e.g. Finnish ‘uusi’, Komi ‘víl’, Sami ‘ođđa’) with the Hungarian word’s long vowel and ‘-j’ arising from the loss of earlier consonants. On the other hand, the Turkish word descends from an earlier ‘jaŋï’ or ‘jeŋi’ which interestingly might be linked to Finnish ‘hyvä’ (from earlier ‘*šüvä’) and Mordvin ‘čiv’ both meaning “good”.

    5) Cebimde çok küçük elma var ~ Zsebemben sok kis alma van.

    This is a tired example used by proponents of a close genetic linguistic link between Turkish and Hungarian including Hungarians who feel a deeper emotional connection to Turks over all other peoples for whatever reason. The problem is that it’s cherry-picking and all of the words can be explained as borrowings rather than cognates (e.g. alma ~ elma) or are actually unrelated despite the superficial similarity (i.e. van ~ var). The only suffix that has any plausible connection is the Turkish ‘-im’ and the Hungarian ‘-em’, and even then these could be just independent developments of a process where the suffixes originated from the personal pronouns. The pronouns themselves MAY however be related (Hungarian ‘én’ may be a cognate of Turkish ‘ben’. By the way, this connection has been tied to descent from some common Eurasiatic root for the 1st person singular (cf. ‘mine’, ‘mon/ma’, ‘mein’, Estonian ‘minu’, Mongolian ‘miniy’ with the last two meaning “mine”). There’s nothing earth-shattering about this structure, and is merely a good demonstration of heavy agglutination typical of many Eurasiatic languages – not just Hungarian and Turkish. For the sake of inclusiveness, the Finnish equivalent of this sentence could be (ab)used as demonstrating a relationship because of the typological similarity.

    Cebimde çok küçük elma var
    Zsebemben sok kis alma van.
    Taskussani on monta pientä omenaa.

    N.B. Apart from the words, Finnish requires a different word order here (verb in 2nd position, rather than final in the Hungarian and Turkish example), and the suffixes differ again from Hungarian. (inessive suffixes are ‘-ssa’, ‘-ben’ and ‘-de’, possessive suffixes are ‘-ni’, ‘-em’, ‘-im’).

    An example of cherry picking with the express purpose of trying to maximize the Finnish-Hungarian connection would be to show the tired example used in an article in The Economist.

    Elävä kala ui veden alla. (Finnish) ~ Eleven hal úszik a víz alatt. (Hungarian)
    “The living fish swims under the water”

    I could go on and break down more links that Erik has mentioned but I think that you get the picture. The focus on emphasizing a close Turkic-Hungarian connection is odd and rests on several dubious etymologies. At the same time, you can see that I’m not diametrically opposed to the spirit of Erik’s arguments since I’ve examined  research from other comparative linguists (including the “Nostratic” school) and find it plausible because of the more rigorous techniques used. There are indeed proposed etymologies showing how Hungarian words that are underlined as or thought of as “Finno-Ugric” may indeed extend further to other Eurasiatic language groups, which Erik hasn’t even considered or examined judging by my brief analysis.

    In general, it is true that of all of the languages out there Hungarian shows most similarity to certain Uralic languages – not only in vocabulary, but also in grammar (right down to suffixes or even concepts of verb conjugation. E.g. Hungarian conjugation into two classes “subjective” and “objective” has parallels with the patterns used in some Uralic languages including Mansi. There’s no such division of conjugation in Turkic languages). I sense that the unsaid problem is that many people ranging from Hungarian nationalists/”patriots” to the comparative linguists of the past have not questioned the equation of language to the ethnic group. By this equation, if Hungarians speak a Finno-Ugric language, then they can only be ethnically Finno-Ugrians. You can see that some Hungarian nationalists could find this upsetting because Hungarians then wouldn’t belong with a desired group of people – in this case, the “glorious” Turks.

    This is juvenile in my view but the equation’s strength persists still (cf. Romanian insistence that they must be descendants of the glorious Roman legions above all just because their native language (Romanian) is more similar to Latin than neighbouring languages). At the same time, most modern linguists think that the equation of language to ethnicity is wrong-headed because a group of people can go through “language shift” (i.e. ethnic group X which initially speaks X can speak language Y later but still imagine themselves to be part of ethnic group X.). An example of this are the Bulgarians. In the Dark Ages the Bulgars were nomads who spoke a Turkic language. Yet when they conquered part of the Balkans, they stuck around and became assimilated to the Slavs and Thracians who outnumbered them. Today we call them Bulgarians who speak a Slavic language, consider them Slavs and often Orthodox Christians. Yet Bulgarians acknowledge that they have still some vague connection to the modern Chuvash and Volga Tatars, who in one way or another are descendants from the Bulgars of old.

    For the Hungarians, we basically have a people of Eurasiatic heritage but who speak a Uralic language. Big, fat hairy deal. It’s not different from the Normans who spoke Old French and felt no connection to the Anglo-Saxons and Celts they conquered but eventually became assimilated and their descendents conducting themselves as “English” speaking only “English”.

  • Pingback: Do Languages Spread Solely By Diffusion? « Cultural Geography « GeoCurrents()

  • Ricko9

    Your information seems to be at odds with other genetic information I have found. Not being knowledgeable on the subject I find it confusing.





    Paivi Lahermo at alii (“Y
    chromosomal polymorphisms reveal founding lineages in the Finns and the Saami”,
    Eur J Hum Genet. 1999 May-Jun;7(4):447-58.) have already done the job for you:
    they have screened 502 males from 16 Eurasian ethnic groups for Tat C and they have found out that all the Finno-Ugric populations
    do have Tat C (up to 63.2 %!);  only the
    Hungarians DO NOT carry Tat C ! The highest incidence of Tat C is in the
    Ob-Ugrian Khanty! Thus, the Magyars cannot have come from there.



    Racially the
    Finno-Ugric language group is just about as diverse as humanly possible. The
    small tribes living east of the Ural Mountain are
    Mongoloids, the Finns are of Northern European stock, and the Hungarians are
    typical, Central-Europeans. Research in the 1940’s indicated that among King Árpád’s people (those that conquered the Carpathian Basin eleven hundred years ago in 895 A.D.) the Finno-Ugric stock
    totaled just 12.5%. This accounting for only a small percentage of the total
    population of the Carpathian Basin, other
    possibilities seemingly have more to offer regarding the origin of the
    Hungarians and their language. Let’s investigate those, along with a short
    recap of the official version of events.  Before we begin and as a
    reminder, Hungarians call themselves Magyar – a name which appears often in the text.

    Let us start with the results of the latest
    genetic research. Between 1984 and 1989 the Hungarian and German
    Academies of Science jointly conducted a genetic research project that
    resulted in the following findings:

    have evaluated the deletion of the so called inter-genetic 9-bp, of which the
    presence or absence is a determining factor in establishing racial
    relationships. The Asiatic origin of 9-bp is completely missing from the
    Hungarian population. We have found the Asiatic M haplo-group in the Finns, the
    Ezras and the Lapps, but we did not find it in a single Hungarian individual
    tested.” (The
    three-page summary of this joint study appeared in the weekly publication Élet és Tudomány (Life and
    Science) as the  article „Népességünk Genetikai Rokonsága” (“Genetic
    Relations of our Population”); written by Dr. Judit Béres, the leading Hungarian scientist
    in the group,  it appeared in the September 21, 2001

    Thus, the latest scientific
    research refutes the claim that Hungarians are genetically related to the
    Finn-Ugric peoples.
    Logically, this fresh information should call for a new review and revision of
    where exactly the Hungarians originated from.


    Dr. László Marácz – Professor of Linguistics, Amsterdam University

     The Untenability of the
    Finno-Ugric Theory from a Linguistic Point of View.

    I. The turning point of research in the field of
    Hungarian language relationships.

    book entitled Magyar fordulat – Politikai vélemények közép-Európáról (Hungarian
    Turning Point – Political opinions concerning Central Europe)
    was published in Dutch in November 1995. The goal of the Hungarian Turning
    Point was to break through several rings of taboos created by news black-outs,
    to reveal the oppression against the Hungarians and their culture in the Carpathian
    Basin and to speak the truth about the injustice of the
    Dictated Peace Treaty of Trianon.  As a linguist, I raised the question of
    the Finno-Ugric relationship of the Hungarian language. I came to the
    conclusion that this theory was untenable from a linguistic point of view. In
    the Hungarian Turning Point, I presented ten points summarizing my criticism of
    the Finno-Ugric/ Magyar language relationship.


    the Magyar (Hungarian) language is classed as belonging to the
    Uralic/Finno-Ugric/Ob-Ugric language group. The (linguistic) ‘relatives’ of the
    Magyar language are to be found amongst the Uralic/Finno-Ugric/Ob-Ugric
    languages which include Khanty, Mansi, Ostyak and Vogul. Magyar is a distant
    (linguistic) cousin of Finnish, Estonian and so on, since Finnish and Magyar
    are supposedly separated by 4000 years and share no more than a few
    recognisably similar words.

    theories concerning the origins of the Magyar are many and varied. However,
    there is often a confusion between racial and linguistic kinship. It was not
    that long ago that the dogmatists insisted that the Magyar were related to the
    racially disparate Uralic Finno-Ugric peoples, based entirely on very
    narrow linguistic similarities alone. The theory has its roots in the work
    of Joannis Sajnovics published back in 1770 when linguistics had not as yet
    evolved into the ‘exact’ science it is today. Interestingly, such racist
    ‘beliefs’ never arose in the case of equally, distant, linguistic relatives
    such as English and Hindi, for example.

    The winds appear to have changed.
    We can now read such comments from [jla] as “What has
    been falsified is probably the antiquated idea of Finno-Ugrian cultural or
    even racial relationship.” Doesn’t seem to be too sure, though. And from
    [msz] “The linguistic affinity which ties the
    Hungarians to the Ugric language family does NOT mean ethnic relationship or
    common origin”.

    enlightened view was not always the norm even among professionals. As recently
    as 1982 [sh] defined
    ‘Magyar’ as ‘One of the Mongoloid Race (entering Europe in 884), dominant in Hungary’. The Finno-Ugric people somehow turned into Mongols!
    Shades of Masaryk [als]

    On the link between
    Magyar and the Finn-Urlick language family – How is it possible that a
    community of scientists can base its work, for over 100 years, on a fundamental
    belief in the U-node, which is simply not supported by the evidence, and indeed
    is contradicted by a good body of evidence?” (Marcantonio. 2003: 277) 
    Marcantonio finds the explanation in the influence of Darwinism, but it is now
    sure that we have to break away from the paradigm of the Ural-Finno-Ugric
    linguistic relationship. “I believe that a shift in the paradigm can no longer
    be delayed.” (Marcantonio. 2003: 278)  However  “it is in the interest
    of those who are sitting in the stronghold of the  Finno-Ugric theory that
    their ‘truth’ remains unchanged.”  (Péter Király: Magyar Nyelv.
    96/2000/67.) [from Marácz, Bakay]

    What does all this mean?

    • Indeed there’s much controversial or conflicting genetic information out there. In part it is because the field of population genetics is developing quickly, both in terms of methodology and in terms of better theories. Good luck trying to make sense of it all!

  • Pingback: Traditionalism vs. Assimilation Among Indigenous Peoples of Siberia - Languages Of The World | Languages Of The World()

  • Pingback: Meeting Darwin’s Last Challenge: Matching Genes, Languages, and Geography (Part 3) - Languages Of The World()