Itämerensuomalaisten kielten kaakkoinen kontaktialue nimistöntutkimuksen valossa

Pauli Rahkonen

Väitöksenalkajaisesitelmä Helsingin yliopistossa 11. toukokuuta 2013 Saamme kiit-tää venäläisiä kronikoitsijoita siitä, että tunnemme nimeltä joitain Keski- ja Pohjois-Venäjän kadonneita uralilaisia kieliä puhuneita heimoja. Niin sanottu ”Nestorin kro-nikka” [Povest’ vremennyx let] (PSRL 1965) mainitsee muun muassa kansat nimeltä чудь [tšuudit], весь [?vepsä], меря [merja], мурома [muroma], черемись [mari], мордва [mordva]. Udmurtit (vot(jak)it) ja ehkä komipermjakit tunnettiin nimellä пљрмь [permit] ja komit nimellä печера [pečoralaiset]. Lisäksi mainitaan заво-лочьская чудь  [taipaleentakaiset tšuudit]. Viimeksi mainittu heimo asui lähinnä nykyisen Arkangelin alueen (oblastin) seuduilla. Muun muassa Sofijskaja pervaja letopis’  (MLS) mainitsee  myös heimon nimeltä мещер [meščora]. Kyseisen kroni-kan mukaan tämä heimo asui muromalaisten ja mordvalaisten seassa. Nämä kaikki puhuivat kronikoiden mukaan aomaa kieltään ja luettiin ”vieraskielisten” (tšudojasitšnie, inojasitšnie, HM, kansallisuuksien joukkoon.

Kronikoiden mukaan Venäjän valtakunnan perustajia olivat tšuudit, slaav, krivitšit ja vepsäläiset, mikäli ves’ todella merkitsee nykyisiä vepsäläisiä. Itse asiassa epäi-len,että etelävepsäläiset saivat etnonyyminsä Valkeajärven aluetta ja sen etelänpuo- leisia seutuja asuttaneilta varhaisemmilta asukkailta, joiden kielessä esiintyi esimer- kiksi sana boj < *woja ’oja, joki’ (vrt. Saarikivi 2004: 199–201). Suomensukuisilla heimoilla oli siis vahva edustus Venäjän valtion syntyhistoriassa. Venäläisellä kroni-koitsijalla tuskin oli motiivia paisuttaa heidän merkitystään yli todellisuuden.

Mielenkiintoista on, että tšuudit mainitaan kronikoissa usein heti ensimmäisinä jopa ennen slaaveja ja krivitšejä tai viimeistään toisina. Esimerkiksi: Euroopan väestö: ”Jaafetin alueella asuivat venäläiset,tšuudit ja monet muut kansat, kuten – ”(sitten seuraa luettelo muista kansoista); Venäjälle veroa maksaneet kansat: ”(kansat), jotka puhuvat omaa kieltään, mutta maksavat Venäjälle veroa: tšuudit, merjalaiset – ”; Var-jageille veroa maksaneet kansat: ”– merentakaiset varjagit keräsivät veroa tšuudeilta, slaaveilta – ”; Rus’ien sotajoukon koostumus: ”Oleg (~Helge) lähti sotaretkelle koot-tuaan varjageista, tšuudeista, slaaveista, merjalaisista, vepsäläisistä ja krivitšeistä suuren sotajoukon” (PSRL 1965).

[HM: Varjagi-nimitys on balttia ja sen käyttäjät olivat Itämeren itärannikon "uusviikin-kejä", joiden alue muodostui Tanskan vallattua laänsibalttilaisten skalvien pääkaupungin Jumne ja modostettua siitä oman Jomsburgin viikinkikaupungin. 

http://hameemmias.vuodatus.net/lue/2015/07/untitled-1 ]

Herää myös kysymys, miksi kadonneitten suomalais-ugrilaisten heimojen assimiloi-tuminen tapahtui melko eriaikaisesti. Vepsäläiset, mordvalaiset ja marit ovat säilyttä- neet kielensä tähän päivään saakka. Kuitenkin tšuudit ilmeisesti assimiloituivat jo keskiajalla. Merjalaiset oletettavasti säilyttivät paikoitellen kielensä uuden ajan al-kuun, Tkačenkon (2007) mukaan epäsuorien todisteiden valossa paikoin jopa 1700-luvun alkuun. Kuriositeettina voi mainita lähiaikoina esitetyn Aleksej Fedorčenkon elokuvan ”Ovsjanki” [keltasirkut], suomeksi ”Hiljaiset sielut”. Elokuva kertoo Nejan kaupungista kotoisin olevasta miehestä, joka kertoo tarinaa vielä käytössä olevista merjalaisista tavoista ja suree menetettyä kieltä. Elokuva ei toki ole tieteellinen luon-teeltaan, mutta kuvastaa joidenkin ihmisten suurta kiinnostusta löytää merjalaiset juurensa. Jos merjalaisten assimiloituminen olisi tapahtunut jo varhain keskiajalla, ei tällaista pyrkimystä muistella vanhoja tapoja luultavasti enää esiin-tyisi, vaan kaikki muistojen rippeetkin olisivat vaipuneet unholaan jo kauan sitten.

On tärkeää ymmärtää, että kronikat kertovat näillä heimoilla olleen omat kielensä, mutta ne eivät kerro,minkälaisia kieliä assimiloituneet heimot puhuivat. Nykyaikaan säilyneet kielet mordva ja mari ~ tšeremissi ovat selviä tapauksia. Sen sijaan tšuu- dien, merjalaisten, muromalaisten ja meščoralaisten kielet ovat täysin tuntemattomia – samoin ves’-kansan kieli, mikäli kansan jäsenet eivät olleet nykyisenkaltaisia vepsäläisiä.

Mitään luotettavia kirjallisia lähteitä ei ole näillä kielillä säilynyt. Jotkin kronikoiden heimoista ovat olleet hyvin vaikeita identifioida ja paikallistaa maantieteellisesti. Itse olen esittänyt uudet tulkinnat etnonyymeille ямь [jaami] ja угра [ugra]. Jaamit mie- lestäni sijoittuvat parhaiten nykyisen Kingisepin eli Jaaman alueelle ja olivat näin ollen vatjalaisia.

Ugrat ovat mielestäni Kalugan ja Smolenskin alueilla virtaavan Ugra-joen seutujen balttilaista väestöä,joka tunnetaan kronikoissa myös nimellä голядь (< *galindi). Max Vasmer (1932: 23) esitti jo 1930-luvulla ajatuksen, että Ugralla on Latviassa vastine nimeltä Ogre, joka on keskeinen Daugavan sivujoki. Omassa tutkimuksessani olen keskittynyt ennen kaikkea jo kauan sitten assimiloituneitten tšuudien, merjalaisten, muromalaisten ja meščoralaisten kielelliseen taustaan. Kuten edellä on mainittu, tšuudien assimiloituminen tapahtui varmasti jo keskiajan lopulla, mutta merjaa, muro-maa ja meščoraa puhuttiin useiden arvioiden mukaan paikoitellen ainakin vielä 1500-luvulla, ehkä vielä jopa 1600-luvulla (Tkačenko 2007; Markov 1998).

Kaoottinen tutkimushistoria

Tähänastisessa tutkimuksessa on edellä esitettyjen kadonneitten heimojen etniseksi ja kielelliseksi taustaksi esitetty valtavaa kirjoa erilaisia mahdollisuuksia. Tšuudeiksi on arveltu virolaisia, vatjalaisia ja vepsäläisiä tai jotain näistä erillistä itämerensuo-malaista heimoa (mm. Grünthal 1997: 150–170 ja siinä mainitut lähteet). Epäilemättä jossainhistorian vaiheessa näin onkin ollut. Yhtä varmana pidän kuitenkin, että Nov-gorodin ydinalueella on asunut myös näistä erillinen uralilaista kieltä puhunut heimo, jota kutsuttiin tällä nimellä.

Melko yleisesti katsotaan, että etnonyymi tšuudi palautuu germaaniseen originaa- liin, josta johtuvat nykyäänkin käytössä olevat kansallisuuksien nimet Deutsch ’sak- salainen’ ja Dutch  ’hollantilainen’, vrt.gootin þiuda ’kansa’ (vrt. ESRJ 2003 IV [1950 – 58]: s.v. *Чудь). "

Nuo germaanikansojen nimet EIVÄT TULE GERMAANISTA,vaan teutonien heimon omakielisestä nimestä *teuta, joka ei ole germaania, eikä myöskään kantaindoeu-rooppaa, joskin tulee sietä varmasti hyvin suoraan, vaan kelttiä tai balttia, liettuan tauta = kansa(kunta), lat. natio.


tauta

Straipsnelis:

Indoeuropiečiai turėjo terminą,ekvivalentišką 'genties' ar net 'liaudies', 'nacijos' pava- dinimui. Ypač gerai jį reprezentuoja Vakarų ir Šiaurės Europos kalbos: galų Teuto-, Touto- varduose; s.airių tuath 'gentis, liaudis, šalis',kimrų tud 'liaudis,šalis'; oskų touto 'civitas', umbrų tota-m; go. þiuda, s.ang. þēod, s.v.a. diot(a) (iš jo adj. s. ang. þēodise, s.v.a. diutisc, vok. deutsch 'priklausoantis tautai'), go. þiudans 'karalius'; lie. tautà, la. tauta, pr. tauto 'šalis'; be abejonės ir (kaip skoliniai) r. čužoj 'alien', s.sl. štuždь, etc. Šie faktai rodo buvus ide. *teutā f., iš kurio kilę įvairūs derivatai, plg. *teutonos, *teu-tiskos, *teut(i)yos. [101] Bet ide. *teutā egzistavo ir anatolų bei iranėnų kalbose [p. 107]. Ide. *teutā įprasta kildinti iš veiksmažodinės šaknies *teu-/*tewə- 'augti, brinkti'. Man atrodo, kad tik indų-iranėnų kalbose esama duomenų, leidžiančių rekonstruoti pamatinį veiksmažodį 'būti stipriam, galingam; pajėgti': s.i. tavīti, av. tav-, s. persų tāvayatiy. Senoji ide. *teutā reikšmė buvo 'jėga, galia'; reikšmė sukonkretėjo į 'tautą'. [Buvo nuomonių, kad ide. *teutā pirminė reikšmė – 'aibė, masė, liaudis, šalis', 'pléni-tude', žr., pvz. Pokorny 1959: 1084; Devoto 1962: 319 (: 'l’assemblea'); Benveniste 1969: 366. – W. P. Schmid (BzN 11, 1960, 202) manė, kad 'gens, natio'].

Teutonien ja heidän naapuriensa ja heimoliittolaistensa kimbrien (kymrien?) ja amb-ronien Itämeren ja Pohjameren rannalle jättämät paikannimet eivät ole germaanisia, mutta indoeurooppalaisia ne kyllä ovat.

Kantaindoeuroopan, ja kanbaltin sanavertalo, josta "tauta" tulee (länsibalttilaisttain), on

*ten- = vetää, kasvattaa, ohjata vetämällä, *tenta > *teuta merkitse "kasvateetua, johdettua, (mukaan)vedettyä.

Aivan päinvastaisesta verbijuuresta


*len- = vapauttaa, irrottaa, löysätä, jättää joukosta, josta tulevat mm. liettuan liaudis = väki, rahvas, (vapaa) kamsa,  suomen liuta, lauma, ja venäjän ljudi = ihmiset, latinan populus.

Teuta- sanaan liitetääm myös venäjän tšužoi = on outo vieras(heimoinen).

https://vasmer.lexicography.online/%D1%87/%D1%87%D1%83%D0%B6%D0%BE%D0%B9

Preussi:

http://www.prusistika.flf.vu.lt/zodynas/paieska/

"
tauto

tauto „land (Land) – kraštas“ E 793 nom. sg. fem. = pr. *tautɔ̄, t. y. *tautā; acc. sg. fem. tautan „Lande – kraštą“ III 1714 [194], „Land – t. p.“ III 1721 [1921].

Pr. *tautā „kraštas (Land)“ kildintinas iš „krašto žmonės“ < *„Volk“, kuris bei lie. tautà „Volk“ [LKŽ XV 1032, reikšme „kraštas“ (Bretkūnas) – prūsizmas], la. tàuta „Volk; fremde Leute; ir kt.“ (ME IV 140t., žr. dar Locher Donum Balt. 303tt., Karulis LEV II 380t.) suponuoja subst. balt. *tautā „gentis (bendruomenė, susijusi giminyste, kalba, papročiais, ūkiu ir teritorija)“. Šis – iš ide. *toutā/*teutā „t. p.“ (> go. þiuda „t. p.“ ir kt., žr. Pokorny IEW I 1084t.), kuris yra turbūt iš „(žmonių) gausumas“ – sufikso *-tā vedinys ir verb. ide. *tē̆u-/*tū̆- „pūstis, didėti“ (plg. Pokorny l. c.); žr. dar s. v. taukis.



" Tämä taas voisi liittyä jotenkin goottien Itä-Euroopan alueelle sijoittuneeseen kan-sainvaellusaikaiseen valtakuntaan, jonka vaikutuspiiriin tšuuditkin saattoivat kuulua. Tšuudit tunnetaan monien kansojen muinaistarinoissa, kuten muun muassa saame-laisten legendoissa nimellä čuhti : čuð - vihamielisenä hei-mona (Saarikivi 2006: 11 ja siinä mainitut lähteet), mikä synnyttää sen mielikuvan, että etnonyymiä tšuudi saatettiin käyttää endonyyminä.

Merjan kieltä on yritetty jo melko kauan rekonstruoida nimistön ja venäjän murresa-naston avulla. Tunnetuimpia esityksiä on O.B.Tkačenkon (1985) kirja Мерянский язык ’merjan kieli’. Jo 1930-luvulta asti on väitelty, onko merja lähimpänä maria tai jotain muuta. Marilaishypoteesia ovat kannattaneet muun muassa Vasmer (1932 – 1936) ja Matveev (2006). Vastaan ovat olleet lähinnä suomalaiset tutkijat,kuten muun muassa Ravila (1937,1938),Kalima (1942),Arja Ahlqvist (1997) ja tämän tutkimuksen tekijä (2012). Popov (1974: 24–25) piti ilmeisen perustellusti merjaa itsenäisenä suomalais-ugrilaisena kielenä, jota ei tule lukea itämerensuomen, marin eikä mordvankaan kanssa samaan ryhmään.

Muroman heimoa on yritetty yhdistää mordvalaisiin.Tämä johtunee siitä että heidän materiaalisessa kulttuurissaan on samoja piirteitä kuin mordvalaiskulttuureissakin. Toisaalta esimerkiksi Genning (1967) on katsonut merjan, muroman ja marin muo- dostaneen jonkinlaisen yhteisen kokonaisuuden. Popov (1974: 23–24) taas katsoi, ettei muromalaisia tulisi yhdistää merjalaisiin, mordvalaisiin eikä marilaisiin.

Vasmer (Rjabinin 1997) näyttäisi katsoneen, että nimistö merjalaisalueilla oli melko samanlaista kuin muromalaisalueillakin, mikä tämän tutkimuksen tekijästäkin on lä-hinnä totuutta. Tällöin on huomioitava se kronikoiden mainitsema seikka, että meščo-ralaiset asuivat osin muromalaisten seassa. Näin ollen meščoralainen nimistö täytyy ensin voida erottaa muromalaisesta ja vasta sitten voi analysoida varsinaista muromalaista nimistöä.

Kaikkein hämärimmäksi on jäänyt meščoralaisten kielellinen tausta.Meščoralaiset on usein yhdistetty historiasta tunnettuihin mišääri-tataareihin. Rjabinin (1997) ja Mon-gajt (1961) ovat pitäneet meščoralaisia yhtenä mordvalaisheimona. On myös esitetty, että etnonyymi meščora olisi johdettavissa samasta originaalista kuin unkarilaisia tarkoittava etnonyymi magyar (ks. Rjabinin 1997: 215 ja siinä mainitut lähteet). Alku-peräinen meščoralaisten kieli tuskin oli mordvaa, koskapa 1400-luvulta peräisin ole-va kronikka (Sofijskaja pervaja letopis’) nimenomaan sanoo, että muromalaisten ja mordvalaisten seassa asuneilla meščoralaisilla oli heidän oma kielensä – siis muu kuin mordva. Myös nimistö tukee ajatusta omaperäisestä meščoralaisten kielestä (Rahkonen 2009).

Kaikki tämä suuri mielipiteiden sekava kirjo osoittaa, että luotettavia menetelmiä kie-lellistä taustaa koskevissa kysymyksissä ei ole käytetty. Arkeologit ovat yrittäneet yh-distää tunnettuja kulttuurialueita kronikoiden mainitsemiin heimoihin. Tämä ei tosin ole ollut aivan hedelmätön yritys, sillä varsinkin merjalainen kulttuuri ja nimistö korre-loivat melko hyvin (esim. Leont’ev 1996; Rahkonen 2012). Sen sijaan muromalaisten materiaalisessa kulttuurissa esiintyi joitain samoja piirteitä kuin mordvalaisilla (ks. Rjabinin 1997 ja siinä mainitut lähteet), mutta nimistöllä on sen sijaan läheisiä liitty-mäkohtia merjalaisnimistöön eikä sillä ole juuri mitään tekemistä mordvalaisen ni-mistön kanssa. Pitkien kurgaanien kulttuuria on yritetty liittää muun muassa tšuudei-hin (ks. keskustelu aiheesta Tvauri 2007). Nimistön näkökulmasta tarkastellen näin ei kuitenkaan tulisi tehdä,koska kyseinen kulttuuri näyttää ylittäneen kielirajat. Sen kannattajina on ollut niin tšuudeja,slaaveja kuin balttejakin.

Nimistön analysointi

Riittävän kattavaa nimistöä koskevaa analyysia ei mielestäni ole tässä laajuudessa ennen väitöstutkimustani tehty. Koska arkeologia itsenäisesti useimmissa tapauksis- sa voi luotettavasti kertoa vain materiaalisesta kulttuurista ja geenitutkimus geneetti-sestä perimästä, on keskeisin kielellistä taustaa selvittävä tieteenala juuri nimistöntutkimus.

Jossain määrin voidaan käyttää hyväksi myös venäjän kielessä esiintyviä substraat- tisanoja. Alue, jonka nimistö pitäisi tutkia, on valtavan laaja. Sen vuoksi kannattaa rajata tutkimus vesistönimiin jo siksikin, että vesistönimet säilyvät muutosten pai-neessa paljon paremmin kuin asutusnimet tai mikrotoponyymit (pellot, niityt, purot, lammet, pienemmät suot). Väitöskirjani South Eastern contact area of Finnic lan-guages in the light of onomastics pyrkii olemaan monessa suhteessa uraa uurtava tutkimus. Toki ei voi olla huomioimatta useiden tutkijoiden jo varhemmin tehtyä työtä, mikä vain valitettavasti on jäänyt määrältään vähäiseksi.

Lähteet:

Ahlqvist = Альквист, Арья 1997: Мерянская проблема на фоне многослойности топонимии. – Вопросы языкознания 6/1997 s. 22–36.

ESRJ = Vasmer, Max 2003 [1950–58]: Этимологический словарь русского языка 4 том. Москва: Астрел. Аст.

Genning = Генинг, В. Ф. 1967: Некоторие проблемы этнической истории марийского народа (о мерянской этнической общности). – Происхождение марийского народаs. 52–70. Йошкар-Ола.

Grünthal, Riho 1997: Livvistä liiviin. Itämerensuomalaiset etnonyymit. Castrenianu- min toimitteita 51. Helsinki: Helsingin yliopisto & Suomalais-Ugrilainen Seura.

Kalima, Jalo 1942: Karjalaiset ja merjalaiset. Uusi Suomi 19.07.1942. Helsinki.

Leont’ev, A. E. 1996: The archaeology of the Merya (The early history of North-Eas- tern Russia).

– Gennadij Afanas’ev & Falko Daim (toim.) yhteistyössä Dafydd Kiddin kanssa, Rus-sian monographs in migration period and medieval archaeology. Volume 4 s. 316 – 339. Moscow: Russian Academy of sciences, Institute of Archaeology.

Markov, Aleksej 1998: Мещёра.

http://www.hunmagyar.org/turan/mordvin/meshchera.html.

Matveev = Матвеев, А. К. 2006: Ономатология. Москва: Наука.Mitrofanov, Aleksej – Doukelsky, Vladimir 2003: An old town of the ancient Meshchera.

www.egorievsk.ru.

MLS=Московский летописный свод. Москва. 5.

Mongajt = Монгайт, А. Л. 1961: Рязанская земля. Москва.

Popov = Попов, А. И. 1974: Топонимика древних мерянских и муромских областей.– Географическая среда и географические названия s. 23–25.

Сборник статей. Ленинград. PSRL=Полное собрание русских летописей 9 – 12. Москва 1965.

Rahkonen = Рахконен, П 2012: Границы распространения меряно муромс- ких и древнемордовских гидронимов в верховьях Волги и бассейне Ока. – Вопросы ономастики 1/2012 s. 5–42.

Ravila, Paavo 1937:Max Vasmer Beiträge zur historischen Völkerkunde Osteu-ropas III. Merja und Tscheremissen. – Finnisch-Ugrische Forschungen 24 s. 1 – 76.

Ravila, Paavo 1938: Merja und Tscheremissen. - Finnisch-Ugrische Forschun- gen 26 s. 1–32. Rjabinin = Рябинин, Е. А. 1997: Финно-угорские племена в составе древней Руси. Санкт-Петербург: Издательство Санкт-Петербург- ского университета.

Saarikivi, Janne 2004: Über die saamischen Substratennamen des Nordruss- lands und Finnlands. – Finnisch-Ugrische Forschungen 58 s. 162–234. 2006: On the Uralic substrate toponymy of Arkhangels region. Problems of research methodology and ethohistorical interpretation. – Janne Saarikivi, Substrata Uralica s. 1–64. Tartu: Tartu University Press.

Tkačenko = Ткаченко, О. Б. 1985: Мерянский язык. Киев: АН Украйнской ССР. Наукова думка.2007:Исследования по мерянскому языку. Кострома.

Tvauri, Andres 2007: Migrants or natives? The research history of long barrows in Russia and Estonia in the 5th–10th centuries. - Juhani Nuorluoto (toim.), Topics o n the ethnic, linguistic and cultural making of the Russian North s. 247 – 285. Slavica Helsingensia 32. Helsinki: Department of Slavonic and Baltic Languages and Literatures at Helsinki University.

Vasmer, Max 1932–36: Beiträge zur historischen Völkerkunde Osteuropas – IV. Berlin: Verlag der Akademie der Wissenschaften.

Pauli Rahkonen:South Eastern contact area of Finnic languages in the light of onomastics. Jyväskylä: Bookwell Oy (omakustanne) 2013.

Väitöskirjan tiivistelmä on luettavissa osoitteesssa: http://urn.fi/URN:ISBN:978-952-5866-15-5.
Kirjoittajan yhteystiedot (address):  [email protected]









Pauli Rahkonen


SOUTH-EASTERN CONTACT AREA OF FINNIC LANGUAGES
IN THE LIGHT OF ONOMASTICS


Academic dissertation to be publicly discussed, by due permission of the Faculty of Arts at the University of Helsinki, in the Auditorium XII, Main Building on the 11th of May, 2013 at 10 o’clock.

Contents
Preface…………………………………………………………………………....5
Abstract………………………………………………………………………….. 6
Introduction……………………………………………………...……………….7
1. General remarks…………………………………………...............……… .. 7
2. Research questions and research material ……...........................…….....8
3. Methods and adaptation………………………………...................……. ..12
3.1. General view……………………………………………….......................12
3.2. Methods implemented………………………………................. .............13
3.3. Limits of methods: Problems of reliability and interpretation……………………………………………...........................…....17
3.4. Special case of Russian adaptation……………................…........... ...19
4. Individual articles and their main results…………..........…......................20
4.1. Article 1, The Problem of the Čudes……………....................................21
4.2. Article 2, The Problem of the Meryan-Muroma and ancient Mordvins………………………………….................................                  ........22
4.2.1. Meryan-Muroma hydronyms……………..................................... .........23
4.2.2. Ancient and modern Mordvin hydronyms ....................................24
4.3. Article 3,The problem of hydronyms of unknown origin in Finland..... 25
4.4. Article 4, The Problem of the Meščeras………................................ ......28
5. Special problems…………………………………….…………...........….. 29
5.1. Interpretation of the results of the investigation .....................................29
5.2. Heterogeneous sources…………………….........……...........…....…. ..30
6. The historical Finnic-speaking area………………................…….....…. 30
7. Archaeological contexts…………………………………................…….. 33
7.1. Archaeology and linguistic groups or ethnicity .....................................33
7.2. Principal archaeological cultures between the Baltic–Finland area and Mordovia……….......................................................…….................….. ............37
7.3. Textile Ceramics and Western Uralic…………….............................. ....42
Literature…………………………………………………………...........……. .45


Preface


Already as a teenager I was interested in the Finno-Ugrian past. One of my teachers of the Finnish language,Aarre Rauhala,invited me several times to visit his home and discuss this subject.Later,his daughter Pirjo Uino PhD became a remarkable archae-ologist who has examined especially the past of the Karelian isthmus.However,in my youth I felt a greater attraction to economics and I gradua- ted from Turun Kauppakor-keakoulu in 1978. Soon, I decided that business was not for me and later I went to study Hebrew in Jerusalem in Israel. I graduated with an M.A. in Hebrew with a focus on Bible translation from the American Institute of Holy Land Studies in 1990. Most of the teachers there worked at the Hebrew University of Jerusalem as well. The most prominent among them were professors Chaim Rabin (history of the Hebrew lan-guage), Emmanuel Tov (the Dead Sea scrolls), Gavriel Barkai (archaeology) and Shmuel Safrai (Rabbinic thought). In the 1990s I accidentally became acquainted with toponyms when I worked as a social worker for war veterans in the town of Lahti. A military officer gave me military maps (1939 –45) of the Russian North. When exa-mining those maps I was surprised because most of the hydronyms obviously origi-nated neither from Russian nor any Finnic language. In 1999 I was accepted into the PhD program at the Department of History and Ethnicity of Jyväskylä University. I wrote a manuscript for my doctoral dissertation in 2008 concerning the settlement history of ancient Finno-Ugrian tribes in Central and Northern Russia. However, one of the referees, Prof. Janne Saarikivi, suggested that I complete the work at Helsinki University’s Department of Finno-Ugrian Studies in order to be on a firmer linguistic ground. No doubt, this was good advice. After nearly five years I have finally finished the work.

It is time to express my gratitude to everyone who has contributed to the dissertation and assisted me. I thank my wife Eija-Liisa who has been my main sponsor. Appa-rently due to my age, I never received the kind of financing that would allow me to concentrate full-time on research. I am very grateful to Jyväskylä University and the Suomalais-Ugrilainen Seura (Société Finno-Ougrienne) for their smaller grants. I would like to thank my employer Lahti Church of Baptistmission for their patience as their part-time employee. Many thanks go to the various scholars who encouraged and advised me, especially professors Janne Saarikivi and Riho Grünthal at Helsinki University, and Toivo Nygård and Petri Karonen at Jyväskylä University. Special thanks to archaeologists Christian Carpelan, Pirjo Uino and Mika Lavento for their encouragement and tutoring me in archaeology.

Such linguists or specialists in onomastics as Arja Ahlqvist, Petri Kallio and Jouni Vaahtera have been most important advisers, and I warmly express my appreciation. My friends in Kotus (Institute for the Languages of Finland) cannot be forgotten here. Finally, I want to mention Rivka Bliboim PhD in Haifa and emeritus professor Heikki Kirkinen in Joensuu as important people who supported me at the beginning of my academic career. Finally, I want to mark this dissertation with the same initials S.D.G. as did the great composers Bach and Handel.


Lahti 5.4.2013


Pauli Rahkonen



Abstract


The subject of the present dissertation is the West Uralic past, mainly linguistic and settlement history. It focuses on historically known ancient tribes and their linguistic backgrounds such as the Merya, Muroma, Meščera and Čude as well as on some unknown Uralic tribes and languages. The tools employed are onomastics (mostly hydronyms) and archaeology.

The main results of the study are as follows. The Meščera seem to have been a tribe inhabiting the left bank of the Middle Oka and, surprisingly, they most probably spoke a Permian language.

It seems that linguistically two kinds of Novgorodian Čudes lived in the catchment areas of the Upper Volkhov and Luga. Traces are found of “East Čudes” and, further west, “West Čudes”. Both of these were apparently not Finnic tribes. The language of the East Čudes shows similarities with Meryan. The West Čudian language shows some features of Mordvin and probably Early Proto-Finnic.

The Meryans and Muromas were linguistically close relatives. Their languages may have been only two dialects of the same language. The Meryan language stretched as far as the western parts of Vologda oblast in the north. A kind of Meryan was spoken in the Moscow area as well. The Meryan language had a cognate language in the eastern parts of Novgorod and Tver oblasts which I have called East Čudian. Apparently another related language was spoken in the eastern parts of Leningrad oblast, in the south-western parts of Arkhangelsk oblast and in Karelia in the Lake Onega region probably before the Finnic era. Ancient Mordvin-type toponyms are found in Kaluga and Moscow oblasts. There seem to have been two extreme edges of ancient Mordvin hydronyms, the first in the environs of the town of Tver and another on the left bank of the Volga between the river Kostroma and the estuary of the Unža.


It is possible that an unknown Uralic x-language (or languages) was spoken in Fin- land, Karelia and in the North Russian lakeland. In my opinion, this language pro- bably cannot be derived from Proto-Finnic or Proto-Saami. I have presented a hypo- thesis that this language was spoken by the population (and their descendants) of the early Textile Ceramics culture. In any event, the lexicon shows similarities with the Meryan language as defined by hydronyms.


INTRODUCTION


1. General remarks


This study is an academic dissertation for the degree of PhD in Finno-Ugrian linguis- tics. It consists of four articles which were previously published in various peer- re-viewed journals. The articles focus on hydronyms located between Finland and the present-day Republic of Mordovia in Russia, these regions included. The purpose of the study is to describe,as far as possible,the character of the languages,the bounda- ries of settlement and the linguistic relations of the groups investigated in this disser-tation with other Finno-Ugrian groups. Many of these languages are presently extinct.


The articles in this dissertation are the following:


Article 1: Finno-Ugric Hydronyms of the River Volkhov and Luga catchment areas.
Journal de la Société Finno-Ougrienne 93. Helsinki. 2011. 205–266.


Article 2: Границы распространения меряно-муромских и древнемордов-ских гидронимов в верховьях Волги и бассейне реки Оки. Вопросы Ономастики 1/2012. Екатеринбург. 2012. 5–42.


Article 3: Suomen etymologisesti hämärää vesistönimistöä. Virittäjä 1/2013.


Article 4: The Linguistic Background of the Ancient Meshchera Tribe and Principal Areas of Settlement. Finnisch-Ugrische Forschungen 60. 2009.

The character of these articles is more empirical than theoretical.As already stated above, the main goal has been to determine the linguistic background and areal distribution of hydronyms in territories under investigation that supposedly were populated by Finno-Ugrian peoples in the past.

The old Russian chronicles mention such tribes as the Čudes, Meryans, Muromas and Meščeras, offering some useful information on their principal areas of settlement. Because the languages of these tribes are now extinct without any direct descen-dants, it is very challenging to reliably determine toponymic etymologies based on the languages of these vanished tribes.

Consequently, the methods employed must be well grounded. Therefore, I introduce the methods and criteria used in the articles in detail in this introduction.


In what follows, the Scientific transliteration of Cyrillic is used with the exception of some generally accepted names (e.g. Moscow, Merya, Arkhangelsk, Ryazan, Yaroslavl, Volkhov, Sukhona, Erzya).


2. Research questions and research material


It has long been assumed that the vast area between the modern territories of the speakers of Finnic and Mordvin was once populated by Finno-Ugrians.

This hypothesis was first presented based on toponymic evidence in the 19th century by e.g. Europaeus (1868–70). Later, the same assumption was put forward by many toponymists (Popov 1965; Matveev 2001–2007 etc.), archaeologists (e.g. Rjabinin 1997; Sedov 1979;Tret’jakov 1966) and linguists (Vasmer 1932–35; Kalima 1944).

This area between the Finnic-speaking territory and Mordovia was populated, accor- ding to the old Russian chronicles (PSRL 1965), by tribes which later have been as-sumed to be Finno-Ugrian (e.g.Rjabinin 1997 and attached literature). It is important to note that the chronicles themselves say nothing of the linguistic background of these tribes.They only mention that some of the groups under consideration had their own language (ibid. 1997: 214). The most important source for a reliable examination of these ancient languages is thus historical onomastics.

Linguists have discussed the paradox that Finnic and Mordvin seem to be closely re-lated languages both on the basis of the lexicon as well as the grammatical structure, although geographically they are spoken far away from each other (e.g. Grünthal 2007:115–135). In recent studies it has been suggested that Finnic, Saami and Mord-vin seem to form a mutually more closely connected westernmost Uralic branch (ibid. 2007: 116; Jaakko Häkkinen 2009: 46;

Saarikivi 2011: 88–95,106–117). In earlier studies,many scholars have used the con- cept of a Finno-Volgaic protolanguage (e.g. UEW,Terho Itkonen 1997: 259–262; Bar-tens 1999: 13) and even of a Proto-Finno-Volgaic community (von Hertzen 1973: 88). The latter theory has been widely abandoned among most Finno-Ugrists.

The assumed Finnic-Saami-Mordvin protolanguage or “Western Uralic” (see e.g. Häkkinen 2009) is phonologically more archaic than, for instance,the Permic or Ugric languages.Terho Itkonen (1997: 236) claims that there existed only a small difference between “Finno-Volgaic” and “Finno-Permic” (or even “Finno-Ugrian”). Many resear-chers, e.g.Mikko Korhonen (1981),Pekka Sammallahti (1988) and Petri Kallio (2007:  229) note that the consonant system of Early Proto-Finnic is still very close to that of Proto-Uralic.The vowel system of Proto-Permic,however, differs considerably from the system reconstructed for the Finno-Ugric or Finno-Permic protolanguages (see e. g.PS = Sammallahti 1988: 524; Bartens 2000: 58–59), while the vowels of the Finno-Saami-Mordvin protolanguage show a much greater similarity to earlier stages of the Uralic languages. It is natural to assume that the languages spoken in the area bet-ween the Baltic Sea region and the Republic of Mordovia may have represented in-termediating forms of Uralic languages between Mordvin and Finnic. One of the main goals of my articles was to reconcile this hypothesis.

Basically, hydronyms give only very limited information on extinct languages. How- ever, morphological, phonological and some other regularities in toponyms/ geogra-phical names make it possible to draw conclusions on some features of these languages (cf. Matveev 2001: 73–75).

The research methodology is full of difficulties and presents a number of questions. The methods should resolve certain problems I set out below. The most important questions investigated in the articles of this study are the following:


1) What happened with toponyms when Slavicization took place in the area under study? Were the old names usually adopted or were they translated into Russian? To what extent did Slavic names replace Finno-Ugrian toponyms?


2) How reliably can the original forms of names be defined? What was the role of Slavic adaptation? Is it possible to establish phonological characteristics of ancient substrate languages on the grounds of toponyms?


3) How well have ancient hydronyms and other toponyms been preserved until modern times, or have most of them disappeared?

4) What kind of material can be considered reliable enough for ethnolinguistic research? How should a researcher react to the fact that the quality of the different ways to collect toponyms and the different toponymic materials available in the ar-chives and publications are not homogeneous? Some of the toponyms are collected from field research, some from vast and reliable archives or old literary documents and others from maps.

It is obvious that toponyms collected by interviewing informants provide better data, because in such cases different variants of the same name can be utilized in the research.

Nevertheless,informants have a tendency to create folk etymologies which may da- mage the original form of the toponyms. Old names collected from ancient documents are a risky material as well. Ancient scribes usually represented governmental autho-rities and were speakers of the languages of administrative centers. They may have lacked proficiency in the languages of the autochthonous populations, and conse-quently names in old official documents may contain many errors and misspellings. People who draw maps – ancient or modern – have not always been concerned with accuracy in names. For this reason one should be critical when using maps as research material. The most correct way to carry out the research is to compare the different sources with each other whenever possible.

5) Were the naming motivations similar everywhere in the research area? Are certain toponymic types more likely to occur and is it possible to utilize preconceived models when searching for toponymic etymologies of extinct languages?

6) How can toponyms of two closely related languages be distinguished from each other? In which cases do two different but closely related languages stand behind toponyms and in which cases only two dialects of the same language?

7) What do toponyms reveal of the characteristics of the languages behind the names? For example, the topographical and phonological evidence of hydronyms show in the Meryan language the shifts (word-initial) *a > *vo (*vol(o)- < *ala- ‘lower’) and *a > *o (*kol(o) <*kala ‘fish’) in the first syllable after initial consonants (see below). As for the vocabulary, there apparently existed such words as Meryan *jäγra/ä, East Čudian *jädra/ä (roughly east of the River Msta) and West Čudian *järi (west of the River Msta) ‘lake’.


1 The languages of the Merya and Muroma exhibit many similarities, but there exist minor phonetic differences, e.g. Meryan *joga ~ Muroma *juga ‘river’, Meryan *veksa ~ Muroma *vi̮ksa ‘river connec- ting two bodies of water’ and Meryan *uht(V) ~ Muroma *voht(V) ‘neck of land between two bodies of water’ (see below).


One of the main problems is the quality and reliability of the toponymic material. There are some areas where the material is widely and well documented. Particular-ly good examples are the Oka catchment area whose hydronyms have been collec-ted by G.P. Smolitskaja (GBO 1976) with several variations of names, as well as the catchment area of the River Svir whose hydronyms have been collected by Mullonen –Azarova–Gerd (MAG 1997). In the present study I have chosen to extract the names from regional atlases, usually with a scale of 1:100 000 or 1:200 000. This decision was made for practical reasons, because in many cases this was the only possibility. This means that many small rivers or brooks as well as the variants of the hydronyms are ignored in this research. This is most regrettable, but the material collected from maps is in many cases the only way to proceed, because the research area is too wide for a single scholar to collect the names by interviewing local informants. Toponyms in Finland are well documented and are available in the archives of Kotus (Institute for the Languages of Finland).

One should note that among these ignored microtoponyms,the quantity of renamed Russian material increases and the number of Finno-Ugrian substrate names is smaller. For instance, in the Kenozero region located in the southwestern corner of Arkhangelsk oblast, the percentage of Russian names of brooks,according to the ma- terial collected by Derjagin–Derjagina–Manuxin (1987), is approximately 60% (Rah-konen: unpublished manuscript).Russification obviously took place between the 14th and 18th centuries (cf. Loginov 1999:107–108). In Novgorod oblast, Russification began and was complete much earlier in the Middle Ages, while in Central Russia it occurred approximately between the 10th and 17th centuries (Markov 1998; Rjabinin 1997: 149–244; Tkačenko 2007: 307). Thus the percentage of Russian names of brooks and riverlets in the Novgorod area and Central Russia is sup- posedly higher than in Kenozero. The estimated number might be 70–90% (see also Matveev 2001: 51; Simina 1980 concerning the Russian North).


1 East Čudian *jädra/ä may originate through old Russian influence from the form *jägra/ä, but the shift *g > d may represent an independent Čudian development as well (Rahkonen 2011: 241)


Saarikivi (2006b: 13) states that among the oldest layer of toponyms, most of the sub-strate toponyms are macrotoponyms. Thus, because of the meager amount of collec-ted microtoponyms, something is lost in this study, but not so much as to prevent quite an accurate overall picture. In some cases, the collections of Max Vasmer’s Wörterbuch der russischen Gewässernamen (1961–69) [WRG] are also helpful.


It is good to take into account the fact that the casualties of this loss can be minimized to some extent by focusing the research on what is present instead of what is absent. Thus, one should examine mainly the most common toponymic types as I have mostly done. It should be kept in mind that even if a researcher has the best modern sources, a huge number of original toponyms are lost anyway, and in every case a researcher must be content only with fragments of names.


3. Methods and adaptation


3.1. General view


In Central and Northern Russia the number of substrate names is very high. Accor- ding to Matveev, there exist in Arkhangelsk and Vologda oblasts more than 100 000 substrate names originating from Finno-Ugric languages (see Saarikivi 2000: 405). For the most part, the languages behind those names are now extinct. Generally speaking, searching for the etymology of substrate names is challenging (see below). Some of them are impossible, some very difficult and others easier to etymologize. In this respect there are different categories of substrate names: 2


1) Toponym based on a totally unknown language. example: In Lapland there exist names that are derived from an unknown non-Uralic language (Ante Aikio 2004).


2) Toponym based on an extinct language that did not leave behind direct successors, but having modern cognates.

example: Names of Novgorodian Čude, Meryan-Muroma or Meščera origin in North-western or Central Russia e.g. *Ilmeŕ, Uχtoma, Ule|nka (Ahlqvist 1998; 2001; Rahkonen 2009; 2011; 2012).


3) Toponym based on a vanished language that has direct modern successors. Usually names of this category are linked with protolanguages.

example: Proto-Saami names (e.g. Kuukasjärvi, Njuxca) in Northern Russia or in Southern Finland (e.g. Aikio 2007; Saarikivi 2004b; Mullonen 2002).


2. See also the classification of Saarikivi (2006a: 12–13).



4) Toponym based on a modern language that is no longer spoken in the area under study, but is still spoken somewhere else.

example: Finnish substrate names in the modern Swedish-speaking area in sout- hern Finland and in Ostrobothnia (e.g. Malax, Pedalax, Terjärv) or Finnish names in the modern Russian speaking area that once belonged to Finland (1917–1939) and subsequently to the Soviet Union/Russia since 1944 (Pitkänen 1985; Nissilä 1975).

Each of these categories has an approach of its own.The etymologies of toponyms in category 1 are almost impossible to determine. In my study, especially category 2 is relevant. The methods implemented in this study are described below in some detail. Ante Aikio (2003; 2007) has dealt with category 3 when examining names derived from Proto-Saami in Finland, Mullonen (2002;2008) in Karelia and in the River Svir catchment area and Saarikivi (2000; 2006b) and Matveev (2001; 2004) in the Rus-sian North.3 The Finnish substrate names in Swedish-speaking islands of the Turku area in Pitkänen’s (1985) study fall into category 4.


As pointed out by Saarikivi (2006b: 9–10), researchers have too often tried to find the correct etymology only by researching the lexicon of attested languages. Practically their sole criterion for  the toponymic etymology has been that the words in stems should look similar to the toponyms they are investigating. For example the etymo-logy of the Vuoht-hydronyms are usually derived from Nsa vuohčču ‘narrow bog’ (SPK s.v. Vuohtajärvi; Sammallahti, Pekka 1989 s.v. vuohčču). A more exact ana-lysis shows that in many cases of Vuoht-hydronyms in these environments, no bog is found at all. It is true that many transparent toponyms can be etymologized in this way. However, there remains a danger of folk etymology and other possibilities for wrong conclusions.


3.2. Methods implemented


I have endeavored to follow methodological rules as simple, unambiguous and reli- able as possible. I have not invented any new methods that have not been presented earlier by other scholars, e.g. by Ageeva (1989), Matveev (2001), Mullonen (2002), Saarikivi (2006a/b), Salmons (1992) and others.


3. Aikio has used Proto-Saami as a tool of research, not as a realistic spoken language.



The following principles have been essential in my work in order to find the most probable etymologies:


1) It is safer to focus the study on naming motivations occurring frequently. In that case the etymology must be one of the most common. Because the motifs of naming in Uralic languages everywhere are rather similar to each other, it is possible to track down the meanings of the words behind the toponyms by comparing them with cognate languages or protolanguages.

2) There may be only a few dozens of motifs in frequently repeated stems of hydro- nyms (see Kiviniemi 1990: 188–189; Saarikivi 2004b: 181–187; 2006b:18–21). If the number of toponyms belonging to the same type is sufficient, it is possible to observe regularities or irregularities in the structure, phonology and morphology (see Matveev 2001: 73–75 ).

3) In Uralic naming systems certain opposite pairs are usual in toponyms. Such pairs include ‘big–little’, ‘white–black’, ‘upper–lower’, etc. Even if the language behind the hydronyms is unknown and extinct but suspected to have been Uralic, it is possible to examine these pairs by comparing the hydronyms with living Uralic languages and with layers of protolanguages. For instance, in the Meryan-Muroma territory the word *il(e) can be compared with Finnic *ülä ‘upper’<PU *üli (PS536). 4 There occur such il(e)-toponyms in Meryan and Muroma as Ile|me|nka (GBO95), Ile|m|kaIli|nda (GBO193), Ile|nda (GBO228), Ili|mdina (GBO269), Il’|d (AJO58/59) and Ile|zem (AKO84).These are regularly located in the upper course of their water system.

Correspondingly, it is possible to discern the word meaning ‘lower’ in the substrate languages by comparing hydronyms with Mordvin alo/ala,Mari ül-, ProtoSaami *vōlē (Lehtiranta 2001: 152) and Proto-Finno-Ugric *e̮ la (PS536). In the Meryan-Muroma area there occur such vol(o)-hydronyms as Vole|š|ka (GBO207), Volo|kša (GBO196), Volo|ga (GBO84), Volo|š|ka (GBO195, 196, 262, 270, AJO 60), Vol|ovskoe oz.(AJO69), Vol|inka (AJO106/107), Vol (AKO121), Vol|ma (AKO58/84), Vol|manga (AKO48,Vol|myš (AKO43),Volo|mša (AKO155) and Vol’|ma (AKO31). The pairs ‘big–little’ and ‘white–black’ can be compared with assumed modern cognate languages and protolanguages as well. Of course, there is a danger of folk etymology and it is impossible to reach full certainty in all details.


4. In the Meryan area there exists also another stem for ‘upper’ vere, which can be compared with Mordvin veŕ ‘upper’; cf. such rivers as Vere|ksa (AJO57), Vere|me|evka (AKO173), Ver|ženka (AKO184), Vere|na (AKO197) and Ver|ža (GBO210).


4) There are lakes, rivers, brooks, etc., which are named according to a regular se- mantic criterion, in many cases on the grounds of specialized geographical terms. I have utilized especially names with the meaning ‘isthmus, route for dragging a boat over a neck of land’ < PU *ukti̮ and ‘river connecting two bodies of water’ < PFP *veksi or *viksi. 5  Their topography is usually easy to verify. However, there must be a sufficient number of this type of material and their topographical position must be regular.

Especially lakes, more than rivers, may be obviously long, narrow, round or curved which are typical motifs for naming. In Southern and Central Finland, quite a common stem for lake names is kukk(V)-, kuuk(V)-. If the shape of the lake is long, the name can be derivedfrom Proto-Saami *kukkē ‘long’. But if the lake is not long in shape, it might be derived instead from Finnish kukka ‘flower’ – in this case probably ‘water lily’ – or in some cases from kukko ‘male bird’ or kukku- ‘hill’; cf. Finnish kukkula ‘hill’ (see Mullonen 1994). The rule of thumb is that the etymologies of lake names are usually easier to determine than those of rivers when using topography as grounds for finding the correct etymology.

5) Examining ethnonyms may give hints for placing hydronyms in the context of the correct extinct languages. The number of ethnonyms must be large enough. It is im-portant to also take into account the density of ethnonyms that occur in toponyms. Stray names do not tell much about the boundaries of settlement. There are some dangers when using ethnonyms as evidence. Firstly, it is useful to know whether the ethnonyms were used as an exonym or endonym. Secondly, it is well known that the ancient Russians called at various times several different Finno-Ugrian tribes чудь [čude] and the Old Russian ethnonym немец [nemec] could mean ‘Scandinavian’ or ‘German’. In my research most of the toponyms derived from ethnonyms are of Russian origin. However, one should note that, for instance, Karelians called Finns ruottši ‘Swede’ and Finns called, at least occasionally, Ka- relians ryssä ‘Russian’ according to the their statehood (as opposed to ethnicity).

It is not always easy to determine the origin of hydronyms derived from closely rela-ted languages such as Finnish and Karelian because of the proximity of these langu-ages. Even surrounding ethnonyms may not necessarily offer the correct solution. Saarikivi (2000: 406–407; see also Saarikivi 2006b: 55) has presented the possibility that the Korela toponyms in the Arkhangelsk and Pinega territories possibly refer to an old Karelian population in the northern Dvina basin.However,this does not neces- sarily mean that all the Finnic hydronyms are of Karelian origin. The Finnic names may originate from an earlier or later era than the ethnonyms.


5 In this case the form from the PFP level is defined on the grounds of Komi vis ‘river connecting two bodies of water’ and the name material in Central and Northern Russia and in Finland.


6) The study of formants is very useful (Matveev 2001; 2004; Mullonen 2002: 191 – 228; Saarikivi 2006b: 29–34). Saarikivi (2006b: 18) has defined formants as phono-tactic types of single-morpheme opaque toponyms having a characteristic feature that makes it possible to understand the word as a place name.Formants are word- final elements in toponyms often based on derivational suffixes or original generics of toponyms that have been modified and rendered opaque.For example, Hima|nka is constructed out of an old Finnish anthroponym Hima and the formant -nka (the origin of this formant has been debated in detail Räisänen 2003). To set formants within the context of different languages is sometimes very difficult.

In Uralic languages there are formants which are common in several languages.The most common may be the formant -m(V) that occurs all over Central and Northern Russia and in Finland (Mullonen 2002: 222–228; Saarikivi 2006b: 31). I have dealt with the problems related to the formant -lja (-ля) in article 1 of this volume and the formant -ks(V)/-kš(V) in article 2. It must be noted that these particular formants as well as some others can be attached to stems originating from many different languages.

7) The principle presented by Salmons (1992: 267), Ageeva (1989) and Saarikivi (2006b: 15) is essential in my work when determining whether a toponym is a sub- strate name. That is, irregular phonetic correspondences occur between the langua-ges or dialects which reflect similar toponyms or loanwords originating in a substrate language. For example, the stems Koitere and Koitter- occurring in Finland show a phonetic irregularity,but are most probably derived from the same original (Rahkonen 2013: 25–27).

8) The entire geographical distribution of each toponymic type is very important when tracking down core areas of hydronyms. In my work I have defined the total distri-bution of several stems and formants of hydronyms. For instance,the distribution map of the formant -lja (Rahkonen 2011: 235–238) proves that this formant is in most ca-ses probably not of Slavic origin,because the distribution does not match with the ter- ritories of old Russian dialects (Vaahtera: personal information; DARJA 1986; 1989).

In some cases it is possible to determine also the directions of spread. Several Nov-gorodian Finno-Ugrian toponymic types can be connected with similar types in the Upper Volga area, and at the same time it is possible to notice the discontinuity bet-ween Estonia and the Lake Ilmen area (Rahkonen 2011: 252–255).This means that Novgorodian types of Finno-Ugrian names most probably did not spread from Esto-nia or vice versa,but instead they are linked with the Upper Volga area. In Finland, such stems as vieks- (viiks-, vääks-), vuoht- (voht-, oht-, uht-), suont- (sont-), kem(V) have clear and mutually similar continuity with the corresponding hydronyms between Inner Finland and the North Russian lakeland and the Rybinskij–Kostroma Volga area (Rahkonen 2013:19–27).6 In Finland these stems do not occur in the archaeological area of coastal cultures that is traditio- nally understood to have been inhabited by Proto-Finns (e.g. Terho Itkonen 1983; Kallio 2006: 16–18; 2009: 41).


9) A sufficient number of toponyms is needed when attempting to define phonetic characteristics and sound shifts that occurred in an extinct language. It is highly desi-rable to find as reliable a definition as possible for the historical development of the examined substrate languages. If the sound laws are known,it is easier to find the ori- ginal form of hydronyms and to decide how to place toponyms in the context of diffe-rent languages. This helps to determine the correct etymologies as well. It is regret-table that the number of types of many substrate toponyms in my research areas is not as sufficient as one would wish.

However, I have dared to define the system of vowels in the Meryan-Muroma language to some extent in article 2.


3.3. Limits of methods: Problems of reliability and interpretation


The methods presented above acceptably solve most of the problems that are con- nected with toponymic types that occur sufficiently frequently. For several toponymic types, a relatively reliable  etymology can be given because of the regular topogra-phic characteristics of the objects they denote. In many cases it is possible to verify the ‘upper’ and ‘lower’ bodies of water, special semantic geographical characteris-tics, principal areas of ethnonyms and the distribution of hydronyms under investi-gation. The linguistic criteria, such as various phonetic and morphological  reliable etymological solutions and defined the connections of the now extinct languages within the Uralic language family.




6 The North Russian lakeland consists of the large lakes Onega, Lača, Vože, Belozero and Kubena. characteristics, offer good possibilities for comparing names with living modern languages or with well-defined protolanguages. Altogether in this thesis I have presented on these grounds several



In some cases, the narrowness of the research material weakens the etymology of rare stems of hydronyms. In my thesis this applies especially to the toponyms of Nov-gorod and Tver oblasts. It seems that linguistically there existed two types of Čudes. 7 One group was close to the Meryans whereas the other group remains obscure.

In any case, even the latter one was most probably not a Finnic tribe.

It appears that in Finland, a Uralic language was spoken that may have been close to languages spoken in the North Russian lakeland and in Rybinskij Volga region (Rahkonen 2013, map 23). This layer of hydronyms may be very old, dating back even to the early Middle Ages.

Sometimes the closeness of two different languages poses problems for the identifi-cation of the substrate . It may be difficult to distinguish e.g. Finnish and Karelian hydronyms from each other. I faced the same problem when trying to define the distribution and boundaries of Meryan and Muroma hydronyms (Rahkonen 2013).

These two languages seem to have been close relatives,but there are some phone- tic differences as well. Sometimes a small detail may provide the solution. For ins-tance, in those Finnish Häme dialects in which > l, the generic ‘bay, gulf’ is fre-quently represented by the word pohja (NA; GT2000; Rahkonen: unpublished mate-rial); e.g. Häme|pohja, Mylly|pohja.Traditionally it has been thought that one of the markers of the Häme dialect (> l) are the toponyms with the generic salin-< *sadin; e.g. Salin|korpi, Salin|niemi, Salin|suo, Salin|Kallio (NA).


In my latest article (2013) I faced the problem of interpretation of hydronyms with the stem vieks-, viiks-, vääks- ‘connecting river between two bodies of water’ and uht-, oht-, v(u)oht- ‘neck of land, route over land between two bodies of water’. The motif of the toponyms is easy to verify.

However, the linguistic origin of the words behind these stems is very difficult to de- termine, because those words or their descendants do not occur in modern langua- ges,neither Finnic nor Saami.Are they words that have been lost from modern langu- ages or already from their protolanguages? Or did these words belong to some ex-tinct Uralic language that was spoken in reliable etymological solutions and defined the connections of the now extinct languages within the Uralic language family.


7 This claim is based on two different words for ‘lake’: *jädra/ä mostly on the right bank of the river Msta and *järi on the left bank (Rahkonen 2011: 242, map 8). This corresponds with the distribution of the words in which *a > *vo-, o [right bank] and *a ~ *a (Introduction, maps 1 and 2).


3.4. Special case of Russian adaptation


Here I am obliged to deal with the problem of Russian phonological and morphologi- cal adaptations of originally Uralic toponyms. This is the case especially concerning my claim that in Meryan, the sound shift *a- > vo- occurred in absolute initial position, while in other positions in the first syllable there occurred the sound shift *a > o. The development corresponds with the linguistic history of Saami (cf. Kor- honen 1981; Sammallahti 1998). In principle, this phenomenon could be explained by Russian adaptation.In Old Russian adaptations of some toponyms there existed a sporadic prosthetic v in front of initial o as well; e.g.Oresa ~ Voresa, Orga ~ Vorga in the Svir’ region (see Mullonen 2002: 56). However, the prosthetic v was not regular in different Russian dialects and therefore there should be variants with and without prosthetic v in widely occurring toponyms. As pointed out below, in the Meryan type of toponyms prosthetic v is regular everywhere.


map 1. Territories of *a ~ a



map. 2. Territories of *a- > vo- word-initially and *a > o in other positions in the first syllable.


I have examined the possibility of Russian influence on original Finno-Ugrian *a utili- zing five different and rather widely distributed variants of stems that seem to vary between *a- ~ vo- and *-a- ~ -o-. They all occur in Central and Northern Russia. They are variants of the following type: al- ~ vol-, and- ~ vond-, par- ~ por-, paž- ~ pož-, kal- ~kol-.

The result can be seen on maps 1 and 2.Obviously in most cases the shift *a>vo-, -o- does not depend on Old Russian influence, but corresponds with the areas of Meryan-Muroma and Novgorodian East Čudic hydronyms in general (see Rahkonen 2011: 241, 254–255; 2012a: 19 – 27). As expected, the known historical territories of the ancient Mordvins and Veps show the conservative form *a ~ a (map 1). The his-torical areas of the Merya and Muroma defined by Leont’ev (1996) clearly show the shift *a > vo-, o (map 2).


4. Individual articles and their main results


The articles included in this dissertation have a thematic focus. Most of the area bet-ween the historical Finnic and Mordvinic-speaking territories, those included, is exa-mined. Each of the articles examines place names (mostly hydronyms) within one or more different linguistic groups.


Research of this particular geographical area is highly important for several reasons.


1) The historical development of Western Uralic languages is not exhaustively described yet.

map 1. Territories of *a ~ a

This concerns especially the mutual relations between Finnic, Saami and Mordvin.

Toponyms that are inherited from extinct Western Uralic languages might shed some new light on this problem.


2) The settlement history of the Uralic tribes that have vanished can be properly examined only by means of onomastic studies, because we do not have a sufficient amount of literary documents. The boundaries of different ancient tribes who lived between the modern Uralicspeaking areas in Northern, Central and Northwestern Russia have been defined in these articles by utilizing toponymic types and phonological features in hydronyms.


4.1. Article 1, The Problem of the Čudes


Article 1 [published in SUSA 93, 2011], written in English, examines the problem of the Novgorodian Čudes. There has long been the idea that the Čudes are linguisti- cally a Finnic group (see Grünthal 1997: 150–171 and referred literature). However, no systematic, comprehensive linguistic study of the characteristics of the relevant substrate has been carried out. My approach to solving this problem was the following:


1) Determining the boundaries of transparently Finnic hydronyms.


2) Carrying out a linguistic analysis of Uralic formants and their geographic connections or correspondences.


3) Searching for possible phonetic Finnic features,i.e. such late Proto-Finnic sound shifts as> h, > t and the initial consonant h that should occur frequently if the language of the Čudes was Finnic as it has been repeatedly assumed. 8.


4) Examining the distribution of ethnonyms, especially Čude and Mer(e)/Ner(e).

5) Determining whether there exist hydronyms between Estonia and the Lake Ilmen area linking the ancient Estonians with the Novgorodian Čudes.


There do exist some hydronyms in Novgorod and Tver oblasts that are definitely Fin-nic and, presumably, of a fairly late origin. For phonetic reasons they seem to be of Karelian origin from the 17th century or later when a considerable amount of Ortho-dox Karelians moved to these regions from Käkisalmi county.Examples of the phone- tically Karelian type of hydronyms include Hobol’ka < *Hoaba|la, Huba < *Huaba (from Karelian dial. hoaba and huaba ‘aspen = haapa’), Kagra or Kägrä ‘oats = kaura’ or ‘curved = [käyrä, liett. kebras’, HM] and Mušto < *Mušta [liett. muštas = lyöty, mustelma, HM] ‘black’.



8 The consonant h did not occur in Uralic protolanguages.


V. L. Vasil’ev has published articles concerning archaic toponyms in the ancient No-vgorodian Land. He has presented some suggestions of old Baltic toponyms (2001: 8–13) and extensive research into toponyms based on Old Slavic anthroponyms (2005). His conclusions often seem tenable, but he has not dealt much with Finno-Ugrian toponyms in more detail.


The main problem of this article was the source material. The names are collected from local atlases and WRG (see section 3). The Finnic names of Ingria could have been collected more accurately from the archives of the Institute for the Languages of Finland (NA). However, the hydronyms mentioned in local atlases are sufficient for determining the boundary of Finnic hydronyms.

This article is a fundamental part of my thesis, because of the rather versatile presen-tation of the methods I have used. The total stock of hydronyms does not reveal many names that could be interpreted as Finnic. Of course, the research material is rather narrow and the assimilation of the Novgorodian Čudes took place early. This may influence the results to some extent. In any event, it is likely that these Čudes were not a Finnic-speaking group.


4.2. Article 2, The Problem of the Meryan-Muroma and ancient Mordvins


In Article 2 [published in Voprosy Onomastiki 1/2012], written in Russian, I concent-rated on defining the boundaries of Meryan-Muroma and ancient Mordvin hydronyms. Arja Ahlqvist (1998;2001;2006 etc.), O.B.Tkačenko (1985) and A.K. Matveev (e.g. 2006) have written on Meryan toponyms on several occasions. However, Tkačenko was not critical enough when using Russian dialectal words from Kostroma oblast as his research material. Matveev did not take into account the fact that there have histo-rically been Mari populations in Kostroma oblast and in Vologda oblast as well. This led him to erroneously consider many Mari stems of hydronyms to be Meryan.

Arja Ahlqvist has criticized these Matveev’s views as well (1998; 2001). However, Ahlqvist, in turn, did not extend the scope of her research to the Russian North or to Tver and Novgorod oblasts. I used the following criteria when defining what actual Meryan hydronyms are:


1) The demand that an actual Meryan hydronym must at least partially occur in the traditional Meryan core area in Yaroslavl and/or Vladimir oblasts.

2) Meryan hydronyms must represent steady and recurring phonemic characteristics from wellestablished words of earlier protolanguages. For example, the hydronym Andoma in Kostroma oblast cannot be of Meryan origin, because hydronyms with the stem and- do not occur at all in the traditional core areas of the Meryans in Yaroslavl and Vladimir oblast and because of the Meryan sound shift initial *a- > vo-. The latter criterion is well attested by several Vond-hydronyms (protolanguage *and- > Meryan *vond-) in Vladimir (Vondega ~ Vondoga GBO202, Vondina GBO200, Vonduxa GBO221), Yaroslavl (Vondel’AJO91) and Kostroma (Vond AKO175, Vondanka AKO73) oblasts. 9


The source material is not fully homogeneous in this research. It was possible to make use of the well-collected material of Smolickaja’s GBO to some degree, but partly I had to be content with local atlases (e.g. AJO, AKO, AVO, ATO) and Vasmer’s WRG. Regrettably, other sources were not available.


The phonology of Meryan is presented cautiously. The lack of different variants of hydronyms casts uncertainty on the result. However, there is no other reliable way to determine what the characteristics of the proper Meryan toponyms are. I believe that I have presented relevant criteria for evaluating toponyms in the light of phonetic regularities.


4.2.1.Meryan-Muroma hydronyms


When defining the boundaries of Meryan hydronyms I used three different methods in order to ensure the results. If all three methods show approximately the same picture, the result is supposedly on a firm footing. These three methods are the distributions of:


1) Common topographical terms and their dialectal variants occurring in the Meryan core area: *veksa ‘river connecting two bodies of water’, *uχta ‘isthmus, neck of land (for dragging boats over)’, *jäγrä ‘lake’ and molo- ‘move from one catchment area to another’.


2) Semantically opposite words *il(e)- ‘upper-‘, *vol(o)- ‘lower’, *väz(ä) ‘little’ (see Rahkonen 2009: 170–176) and the ethnonym *mer(e)/*ner(e) ‘Meryan’.


9 In the Belozero area several names with the stem And- are found. These names probably originate from an ancient tribe called Ves’ in the old Russian chronicles. The language was likely related to Upper Volgaic languages (personal comment of Irma Mullonen). The stem might be derived from *e̮mta, which had the meaning ‘give (food)’ (PS 541). In Finland, corresponding hydronyms Antaman- järvi [Ruokolahti] and Antolampi [Suomussalmi] are found (NA). The motif is ‘lake that gives food’.


3) Typically Meryan formants -hta/-gda and -pol, -bol (see e.g. Ahlqvist 2001: 446– 447, 451–453, Matveev 2001: 206–211; 2004: 29) and in addition, the not only Meryan but more widely occurring formant -ksa/-kša.10


All of these hydronyms show rather identical results: Meryan and Muroma were close relatives. In Moscow oblast occur similar hydronyms as in the core area of the Me-ryans (Yaroslavl and Vladimir oblasts). Meryans also lived in the western parts of Vo-logda oblast. Hydronyms in the Svir and Karelia areas may reflect a Meryan dialect or a closely related language.


4.2.2. Ancient and modern Mordvin hydronyms


When studying the boundaries of the ancient Mordvins I used as my method the distributions of:

1) Opposite pairs based on the Mordvin words iń(V) – vešk-, višk-, vež- ‘big–little’, veľ al(a/o) ‘upper–lower’ and aš(o), akšačem() ‘white–black’.11

2) Generics of hydronyms: Erzya Mordvin eŕke ‘lake’, Erzya Mordvin lej and Moksha Mordvin läj.

3) A specific based on Mordvin piče, pičä ‘pine’ compared with Meryan-Muroma *peč(ä).

Their distributions were compared with one another.

The referees noted a couple of matters. Firstly, what is the semantic motif of Vond- hydronyms? The stem can be derived from the PFP-level root *e̮mta ‘give’ (PS541). In Mordvin the corresponding word ando|ms has the meaning ‘give food’ (MW 1 s.v. andoms). Fish was one of the basic foods for ancient Finno-Ugrian tribes.

Secondly, was the original orthography of Rha (~ Volga) known in old maps really correct? It is very important to note that any other orthographic variations do not oc-cur.The oldest mention of the name is known from Greek sources in the form ΡΑ (Ra) with uncials.Vasmer (ESRJ 1 s.v.Волга) has presented the Greek quotation of the name with minuscules ͑ Ρα (Hra or Rha). We do not know where he found this Greek form, but if it reflects the original,it is necessary to observe the spiritus asper, too. The sound h in the middle of the word would be very unnatural for Russian speakers and there is no reason to suspect that the Russians added it to the original name. The early cartographers were Flemish (Mercator, de Jode, Ortelius) and it is possible that they wrote *ra > rha by analogy with the name Rhein. An original Rha < *reha is pos-sible as well. Thirdly, it should be noted that there is a possibi- lity of misspelling in the variants Vojmiga ~ Vojmira. However, the names Nev|ra and Nev|lej point to the strong possibility of ra = lej ‘river’. Further, there is an ety- mological possibility for -bal, -bol, -pal,-pol to derive them from *poole (cf. Finnish puoli ‘side’).


10 An asterisk (*) here means that the word is a reconstructed Meryan word, not proto-language.

11 Interestingly enough, the modern Mordvin word raužo ‘black’ occurs nowhere in hydronyms (GBO; Inževatov). It seems obvious that the word appeared in use only recently,while in earlier times the word čeme- ‘black’ (orig. ‘rusty’) was used; cf. the rivers Čeme|r|itsa ? < *Čeme|ra (GBO113), Čeme|s|koj (GBO266), Čemi|s|lej (GBO256), Čem|rav|skoj (GBO117).


The evidence of the hydronyms suggests that the ancient Mordvins used to live re-markably further west of their present area and the Republic of Mordovia. Today the Mordvins live east of Ryazan,while people who spoke a Mordvin-type language lived in Moscow oblast, in the southern parts of Tver oblast and in the Volga area between Kostroma and Nizhny Novgorod.


4.3. Article 3, The problem of hydronyms of unknown origin in Finland


Article 3 (published in Virittäjä 1/2013) is written in Finnish.The amount of large lakes in Finland without any clear etymology is high. Among the names of the 85 biggest lakes, at least 25% are of unknown origin according to the writers of Suomalainen Paikannimikirja (in English The Book of Toponyms in Finland). In my opinion the number is even higher. In theory this can be explained in two ways:

1) Words behind those hydronyms have disappeared from the known Finnic and Saamic languages or already from their earlier layers, Proto-Finnic and Proto-Saamic.

2) The opaque hydronyms are substrate names originating from some unknown language(s).

They may originate from extinct Uralic languages or from some Paleo-European language(s). If the language was Uralic, it would be possible to compare the words behind the names with modern languages or with Uralic protolanguages.

[HM: They are not unknown origin:

http://hameemmias.vuodatus.net/lue/2013/12/suomen-balttilaisia-vesistonnimia-imatra-keitele-simpele-inkere-kymi-vanaja-saimaa-paijanne-jne
]


It is very important to determine the whole distribution of these names, and outside of Finland as well in the Russian North or in Estonia. Mainly I have used the same me-thod as in article 2, i.e. names connected with necks of land (routes used for dragging boats over the neck) and rivers connecting two bodies of water. In order to determine whether the words related to the substrate language notion from which the names under consideration were derived have disappeared from Finnish or Saami, one should prove that there are systematic sound correspondences with the reconstruc-ted protolanguage lexical stock. If the word behind a hydronym cannot be derived accordingly by means of Finnic or Saami sound correlations from the protolanguage, the word cannot have disappeared from these languages.

In addition, I have examined the distribution and etymological backgrounds of certain stems, such as kem(V)- ?? ‘stream’ and sont/d-, suont- ?? ‘curved’, because they are located in the same area as the above-mentioned stems uhta-, ohta-, v(u)ohta < PU *ukti̮ ‘track (over a neck of land)’, veks-, vieks-, vi(i)ks-, vääks- < PFP *viksi ‘river connecting two bodies of water’. 12 .

The quality of the sources is better than in the case of those studies dealing with the Russian toponyms, being based on the wide collections of the archive of Kotus [Institute for the Languages of Finland] (NA).

One may hope more information in variants of the stems based on *ukti and *viksi. It is not easy to explain the reasons for such variants as Uht-, V(u)oht- and Oht- or Vieks-, Viiks- and Vääks-. I have assumed that these are dialectal differences that existed already in the original substrate language. Because we do not know any- thing of the phonemic characteristics of this language or its dialects, it is impossible to say anything more detailed.

One of the referees noted the relatively long distance (7 and 15 km) over the necks of land. It is worth of remembering that the original meaning of the word PU *ukti̮ was ‘track’ according to Pekka Sammallahti (PS 536);cf.modern Hungarian út ‘way, road’. The point is that there has been a road or track in common use from one catchment area of a water system to another. The distance across the isthmus is not important, but rather one must ask, was there a road or pathway in common use already when these sites were named? In all these cases I have presented (Rahkonen 2013: 12 – 13; maps 1–4) vuoht(V)-names are located close to such ancient tracks, which necessarily were in use already since the times people travelled by boat.


Because of the criticism of one referee it should be explained why I assume that vuoht-hydronyms are not derived from Saami *vuoččo < NwGerm *watjō- ‘wet bog’ (Aikio 2006: 12).

At first, as I have pointed out in the article, in some cases there is no bog at all in the vicinity of the vuohthydronyms Vuohtajärvi in Reisjärvi and Vuohtojärvi in Pihtipudas.

Secondly, all the vuoht- hydronyms are located at the last point in the upper courses of their catchment area both in the Suomenselkä area and in Karelia.

Thirdly, it is interesting that in the core area of the historical sound shift *čč > ht in the Savo region the original Saami toponym *vuoččō is represented as vuotso and not vuohto; cf.Vuotsinsuo [Joroinen] (NA).If the vuoht(V)-toponyms originate from Saa-mi, then they may have been adopted into Finnish only after the sound shift *ts > ht in the Savo dialect. This may be a rather late event. However, this sound shift does not explain the vuoht(V)-hydronyms in the Republic of Karelia in Russia


12 The etymological background of the stems kem(V)- and s(u)ont/t- is very obscure. Here the meanings ‘river’ and ‘curved’ are presented only as my suggestion, not as a solid fact.

[HM: Kemi (khem) om iranilainen joki-sana. Myös suomen sana ori (oriin, oriiden) merkityksessä uroshevonen on iranilainen sana. Sillä ei ole mitään telemistä liettuan arklys,latgallin orklys -sanojen kanssa,jatka tulevat kyntää-verbistä *arti, joka taas tulee kantabaltin ilma-sanasta *aaras; sen sijaan ori on kyllä samaa kantaa kuin liettuan vyras = aviomies.]


It may be hard to accept my explanation of the variant Indiager ~ Inari due to the lack of a reliable phonetic explanation. I have assumed that there existed these two vari-ants of the same name for a long time side by side. If groups of people speaking Pro-to-Saami came to Lapland in the early Iron Age,there is a good reason to believe that the Saami-speaking population lived side by side with the original inhabitants for a relatively long time. It is known that the variant *ind with d in this particular hydronym was documented already in the year 1593. The first variant without d is documented approximately at the same time in the year 1567 (SPK s.v. Inari) and is not necessa-rily older than the variant with d. This means that the x-language and the variant *Indjäγ(ə)r could have been in use for a long time after the beginning of the linguistic saamification in Lapland. One should remember that the biggest lake on the Kola peninsula has two variants of names as well, the Saami name  Avvir and the official name Imandra (EKI; Sammallahti 1989: 512) The hyd- ronym Imandra as well as probably its corresponding name Imatra are of unknown origin (SPK s.v. Imatra).

[HM: IMATRA on aivan ehdottomasti balttisana. Sen vanhin kantabalttilainen muoto on ollut joko

a) *en-mat- eli "katsoa sisään, joukkoon, itselle/omille kuuluvaksi", pistää silmään, olla erikoinen, jsota en-mat-r- = "laatuaan silmiinpistävä", liettuan imatra(s) = huomattava erikoinen silmiinpistävä, tai sitten

b) *en-man- "ajatella, lukea joukkoon, passiivin partsiipin preteriti *en-man-t-, päätteellä -r- = kaltainen.

Liettuassa luullaan, että erkoinen on suomeksi "**imatra", ja se on päässyt luikahtamaan ainakin yhteen sanakirjaankin!

Kanta-IE:ssä ja -baltissa on kelvannut muoto *enmanras, mutta vasarakirveskieles-sä, kuurissa tai kiettuassa -n-:n ja -r-:n väliin tukee -t- tai -d-, joissakin balttikielissä myös -k- tai -g-.

(Jos sana olisi liettuaa, se voisi tulla myös sanats *ishmantras eli "poiskatsottava", muodon **ihmantra kautta. Molemmat kelpaavat khden reviirin rajamerkiksi, sellainen Imatrankoski on todennälöisesti ollut,ja toinen näistä osapuolosta on ollut ilmeisimmin saamelainen, toinen kuurilainen,kuten muuten joskus Tampereellakin.)

http://hameemmias.vuodatus.net/lue/2013/12/suomen-balttilaisia-vesistonnimia-imatra-keitele-simpele-inkere-kymi-vanaja-saimaa-paijanne-jne

6. įmatyti...įmato...įmatė = katsoa jo(ta)kin joksikin, johonkin kuuluvaksi

įmat|yti (o, ė) усматривать = valvoa, katsoa perään, nähdä (jokin jonakin), pitää (jotakin jonakin); различать = erottaa; разбирать = poimia (erilleen), laatia (valikoima ym.), rajata (mukaan)

Yksi sanakirja antaa sanan "įmatrus"suomalaiseksi käännökseksi ”imatra”:

http://www.zodynas.lt/zodynai/lietuviu-suomiu

” Žodžio "įmatrus" vertimas:

Imatra ”

Verbin "įmatyti" se kääntää ”laatia”, englanniksi ”make out”, joten kysen esimerkiksi kokoelman, näyttelyn tai kirjallisen esityksen laatimisesta:

” Žodžio "įmatyti" vertimas:

make out ”

Mutta se ei ole sanan kirjaimellinen perumerkiktys joka taas on ”johonkin kokonaisuuteen mukaan katsominen”. "

Imatra-nimestä ei kannata yhtäänenenemppää vääntää. Se on selvä kuin pläkki.

Inari ja Enare sopivat erinomaisesti saman vanha balttilainan piiriin.]


It should not be surprising to assume that the name Indiager originates from the Svir and Sukhona area, because as I have pointed out (Rahkonen 2013: 26), many other hydronyms in Lapland have correspondences in the Svir region.

The distribution of the above-mentioned stems give proof of linguistic connections between Finland, Karelia and the North Russian Lakeland and even the Rybinskij–Kostroma-Volga region. The range follows the distribution of the so-called Textile Ware (1900–800 BC) (see the discussion below in section 5.3). These types of hydronyms occur neither in the so-called area of the Coastal Cultures in Finland nor in Estonia. The main problem is that in my research, I did not have the possibility of examining a larger amount of corresponding hydronyms outside Finland.

Article 3 must therefore be understood as only a preliminary enterprise pointing to- wards a good possibility for an unknown Uralic x-language.A larger and more accu- rate study is needed. One basic problem is that the only reliable means for finding the correct etymologies for the stems treated here lies in the investigation of regularly occurring geographical qualities related with the name types under consideration, otherwise the etymological reliability is questionable. For this reason, the problem is how to find enough required stems of hydronyms in order to verify the regularity.

There exist in Finland names of lakes with the formant -(V)ri. This formant may origi-nate from a Western Uralic word derived from the proto-form *jäkrä ~ *järkä ‘lake’ < Proto-Indo-European *ieuHr (see Rahkonen 2011: 239–241 and 2012: 30; Mallory & Adams 1997: 636). It is very important to examine which kind of stems these formants are connected with. Do the stems originate from the known Uralic languages Finnic and Saami or are they of some unknown origin?

[HM: Ja (kaikki) balttilainen kuten vasarakirvesalkuperä on siis tässä "tutkimuksessa" (ainakin vesintönnimiltä ja kaikilta paikanniltä...) MÄÄRITELMÄLLISESTI SULJETTU POIS!!!]


4.4. Article 4, The Problem of the Meščeras


In article 4 [published in FUF 60, 2009], written in English, I have examined the lin- guistic background and the principal settlement area of the Meščeras. Most of the research material has been adequately collected and published by Smolickaja (GBO 1976). This tribe is known from some old Russian chronicles, but not the Primary Chronicle [PVL]. According to a chronicle Pervaja sofiskaja letopis’, the Meščeras lived in the midst of the Mordvins and Muromas (Rjabinin 1997: 214, PSRL 1965). This makes the research of hydronyms especially demanding. I solved this problem by studying especially such opposite pairs as white–black, big–little and upper–lo-wer, antonyms that are characteristic of the toponymy of northwestern Uralic areas. In this way it was possible to distinguish toponyms based on different languages from each other. I compared the results with the available archaeological data, the ethno-nyms and the correspondences with Udmurt  toponyms, because it maybe alleged that the language of the Meščeras was a Permian one.

The weaknesses of this particular article might be my, at that time (2009), insufficient knowledge of Old Russian anthroponyms. Consequently, it is possible that I have interpreted a few hydronyms to be of Uralic origin instead of Old Russian,such as Vele|gošča (GBO111) and Vele|goži (GBO109), for instance (see Vasil’ev 2005). I interpreted the stem vele- in these names as comparable to Mordvin veľ ‘upper’ instead of the Russian anthroponym *Velegostь > Velegošč.

However, in the Oka catchment area there exists a discrepancy of formants in hydro- nyms, e.g. Tor|goša ~ Tor|gaš ~Tor|goš ~ Tor|gošča ~ Tor|gušča (GBO121). This may offer support for the possibility of a Uralic origin (cf.Mordvin kuža < *kuša ‘mea-dow, glade’) of the above-mentioned toponyms with a possible Russian folk etymo-logy. My knowledge of Old Russian anthroponyms increased by the time I wrote the later articles (2011 and 2012) and similar points of view were taken into consideration when analyzing the material relevant for these publications.

One may recall the possibility that vil- ‘upper’ could belong to any Finno-Ugrian lan-guage where Mordvinic *i > *e did not take place. In principle this is correct. However, one should remember that the form vil- is found especially in the same areas with names containing the Permian-looking un- ‘big’, ič- ‘little’ and ul ‘lower’. Udmurt vi̮l(i) ‘upper’ is substituted by Russian vili even in Udmurtia; cf. Vili|šur ‘upper river’ (Rahkonen 2009: 171).

The list of 31 hydronyms which have correspondences in Udmurtia should have been explained better and in greater detail. Unfortunately, there was no space for a longer article in the publication.

Today I would have shortened the article, leaving this section out. One of the prob- lems is that the names are of the type CVC-. Thus the stems of the names consist of one syllable only due to the rather regular loss of second-syllable vowels in Udmurt (Bartens 2000: 61–63).However,I have not divided the stems of the hydronyms with-out reason. In most cases I have followed Atamanov’s presentation of the Udmurt formants such as Čep|ur|inskaja (GBO141),cf.the formant -ur/-i̮r, Kad|ym|ov (GBO 167), cf.t he formant -um/i̮m (Atamanov 1988: 79) and Ir|mes~Ir|mez (GBO 216), cf. the formant -mez, -mes, -mis (ibid. 1988: 60–61). I admit that I should have written it down more clearly.

Altogether, this my earliest article was partly a trial run in scholarly writing, and there-fore I admit that its level is not as high as I hoped. However, I suppose that the basic result that the Meščera were a tribe speaking a Permian language is justified well enough.


5. Special problems


5.1. Interpretation of the results of the investigation


Some problems are related with the presentation of the results of the investigation. I admit that not all of etymologies or the interpretations included in the articles of this study are reliable to the same degree: some are fairly evident, whereas others are hypotheses offered as a direction for future research. In any event, in several cases I tried to take the uncertainties into consideration in my conclusions. How- ever, I am of the opinion that it is useful to present not only what is a watertight case but also what is probable or possible, if it is clearly stated in the text and if there is even some good reasons lending support to the hypotheses.

As I have written above in section 3 and especially in my article 1 (Rahkonen 2011: 214–222), there are many methods and criteria to employ when endeavoring to prove a hypothesis on past languages with the help of toponyms. The certainty grows if there are several different pieces of evidence pointing in the same direction. These criteria are e.g. linguistic characteristics, topography and distribution of the toponyms. If all the evidence agrees, the interpretation of the results rests on a firm foundation. But if only one piece of evidence suits the interpretation and nothing is arguing against it, then the result is somewhat uncertain but could be possible. If there is nothing that points in any clear direction, neither linguistic characteristics, topography nor distribution, then any results presented are speculative.


5.2. Heterogeneous sources


As I have written on several occasions (Rahkonen 2011: 214; 2012: 7), a conside- rable problem is related to the fact that the toponyms investigated derive from very heterogeneous sources. Usually onomastic studies are carried out in a relatively nar-row area where researchers themselves have collected the material (e.g. Mullonen 2002). I have amassed a considerable collection of local atlases and maps covering almost all of Northern (European) and Central Russia. The archive of Kotus (Institute for the Languages of Finland) offers a fine collection of the toponyms in Finland.

The topography of the objects can be examined on the site

http://kansalaisen.karttapaikka.fi.


6. The historical Finnic-speaking area


Finnic is a label for several different languages descending from Late Proto-Finnic, such as Livonian, Estonian, Võro or South Estonian, Vote, Ingrian, Veps, Karelian (3 main branches) and Finnish. Livonian, Vote and Ingrian, though still spoken in the first part of the 20th century, are now practically extinct but relatively well documen-ted languages.Veps and all Karelian branches are in great danger of disappearing in the near future. Estonian and Finnish are strong dominant languages in the areas in which they are spoken. According to the evidence of toponyms, there has been at least one but probably two or three Finnic languages spoken in the (Northern) Dvina catchment area and elsewhere in Arkhangelsk oblast (Matveev 2001; 2004; Saarikivi 2006b). It is difficult to distinguish archaic Karelian or Veps dialects from a possible local independent Finnic language.

There are many theories on the homeland of the Finnic languages. Terho Itkonen (1983) held to a hypothesis that the Gulf of Finland was the center of early Finnic. He stated that Late Proto-Finnic was divided into three dialects: a northern dialect locted in the southern coastal area of Finland, a southern dialect located in Estonia and an eastern dialect in the bottom of the Gulf of Finland and south of Lake Ladoga. When defining which words are of Proto-Finnic origin, this model of three quarter of Finnic is usually used as a basis of definition. Any word that is suggested to be derived from Proto-Finnic should occur in the northern, southern and eastern branches of modern Finnic languages. Kallio (2007: 243) divided Finnic into three early dialects: the dia-lect of the Gulf of Finland (whence Western Finnish,Karelian,Veps,Vote and Northern Estonian),the dialect of the Gulf of Riga (whence Livonian) and “Ancient Čudian” (whence Southern Estonian).

Ante Aikio has previously supported the idea that Proto-Finnic was not spoken in the southern coastal area of Finland (Aikio & Aikio 2001: 11;). Jaakko Häkkinen (2010a: 32), in turn, following Saarikivi (2004a: 224–225), places the origin of Proto-Finnic somewhere southeast of Finland.

In any case, Finnic probably had its origin somewhere around the Gulf of Finland. Names of large and central rivers such as Vuoksi (< Finnic vuo ‘stream’) and Neva (< Finnic neva ‘marsh, river’) must be very old and might represent Proto-Finnic hydronyms.13.

In the southern coastal area of Finland, the names Kymi and Nietoo < *Niet|oja (id. later Porvoonjoki) may also be of Finnic origin and derive from, respectively, kymi ‘stream’ (see SSA I s.v. *kymi; see however SPK s.v. Kemijärvi; Rahkonen 2013: 24) and nieto(s) ‘heap of snow’ (SSA II s.v. nietos), in hydronyms probably ‘high (snowy?) banks of a river’.


[HM: Iranilaisella Kemillä (-in) = iso joki, khem) ja preussilaisella Kymillä (en) = Ky-mis = vaatenhka, poron ja lampaan sianpuolukalla parkittu käy hyvin, ei ole mitään tekemistä keskenään, paitsi korkeintaan että ne ovat aika uusia lainoja. Tämä vaate-nahka on sanana samaa juurta kuin mm. gemssi ja säämiskä, mahdollisesti myös kumi, vikka se kreikassa tarkoittakin pihkaa, jota taas myös terva on. Ja ne nahka-vaatteet usein tervattiin. Erityisesti nykyisen mietan puvun näkäinen tervattu nahka-asu oli tämän loskaisen alueen merenkulkijoiden, kuurilaisten, liiviläisten ja hansan kauppiaiden, työasu, jonka myös kauppiaat maissa omaksuivat näyttääkseen mm. turvallisuussyistä merimiehiltä. Kauppiaat rupesivat tekemään mustia asuja muista-kin materiaaleista.Siitä se pukukulttuuri alkoi,ja noista preussilais-piireistä alkoi moni muukin maailmanlaajuinen jokapäiväisen elämän kulttuuri, mm. siideri, salaatti ja pizza. (Salaatti tarkoitti medän "rosollia",mössöksikeitettynä borshtshia,joka oli tehty nauriista ja toutaimesta, liett. salatis, tai muusta viljel- lystä kalasta.) Vaatenahka on erittäin yleinen paikannimien lähde Mokkasta ja Chamonix´sta maailman vanhimpaan pääkaupunkiin Damaskokseen.


Mustion|joki is clearly a Finnish name < *must|oja ‘black river’. [HM: Liettuaa: mushtas  = "lyöty" = musta]

The river name Vantaa remains somewhat obscure, although Nissilä (see SPK s.v. Vantaanjoki) has derived it from the Finnic word vana ‘water route’. In western Fin-land the names of large rivers, such as Aura and Eura, are supposedly of Germanic origin (Koivulehto 1987).

[HM: Balttia ne ovat kaikki, Vanduo = liettuan vesi, Aura on vasrakirveskilen "ilma" (se on myös rajuilma ja tulva balttikielissä), ja Eurakin on samaa taustaa, mahdol-lisuuksia on monia, vaikkpa *Edra, "Ruokajoki"!]


In Estonia the names of many of the most important rivers might be of Finnic origin: e.g. Ema|jõgi Est. ema ‘mother’ [Tartu district] (?? cf. the Lake Piiga|ndi < Est. piiga ‘maiden’), Pärnu [Pärnu district] < Est. pärn ‘linden’, Valge|jõgi [Loksa dis- trict] < Est.valge ‘white’, Must|jõgi [Võru district] < Est. must ‘black’ (balttia). It is possible that Emajogi and especially Piigandi are the result of later folk etymolo-gizing of a name with some unknown origin. However, as a naming motif there exist in Finland numerous toponyms with the stems Finnic *emä (e.g.3 Emäjoki), *neit(V)- ‘maiden’ (e.g. Neitijärvi, Neittävänjoki, Neittävänjärvi) and Saami stems that can be derived from Proto Saami *nejte̮ ‘id’ (GT2000; NA).

[HM: Myös Suomessa on vironkilisiä vesitönnimiä, ainakin Näsi (-in) = viron näsiä (Daphne mezereum), tärkeä viljelty myrkkykasvi rautakaudella. Myös tampere voisi olla vrios: Tamiheimo tai "tammiperä" (tampera, tammaen pohjoisin luonnovarainen esiintymipaikka), kuurin Tampere = venevinssi on kuitenkin todennäköisempi. Tam-pereen esitetyt germaanietymologiat palautuvat nekin kuuriin : *Dambere = pdottu, kesytetty, valjastettu).


13 It seems that the word *neva occurred in some language(s) in the Oka catchment area as well,most probably in Ancient Mordvin; cf. the names of rivers Nev|lej GBO242,264, Nev|ra GBO223, Neva GBO185.




map 3. The historical southern boundary of Finnic hydronyms, excluding
hydronyms produced by the Karelian refugees of the 17th century.


These seemingly very old names of relatively large rivers in southern Finland, modern Leningrad oblast and Estonia support the hypothesis that Proto-Finnic was spoken for a long time on both sides of the Gulf of Finland and it thus basically cor- responds to the hypothesis of Terho Itkonen (see above). In the Novgorod, Tver or Vologda oblasts of Russia, Finnic names for large rivers cannot be found (Rahkonen 2011: 229). For this reason, it is likely that the Late Proto-Finnic homeland was the area around the Gulf of Finland.


Beyond the southeastern boundary of the modern or historically known Finnic-spea-king area, there exists a toponymic layer belonging to the supposedly non-Finnic Novgorodian Čudes (seeRahkonen 2011). In theory it is possible that Proto-Finnic and Proto-Čudian separated from each other at an early stage or it is even possible that Proto-Čudian was identical with Proto-Finnic. However, this cannot be proven, because there is not enough material available describing what Novgorodian Čudic was like exactly.


7. Archaeological contexts


7.1. Archaeology and linguistic groups or ethnicity


Because in most of the articles I have presented archaeological data, it is reasonable to discuss the problems and restrictions of archaeology in resolving linguistic or ethnic matters. Lately, there has been a great deal of discussion on the correlation between archaeological data and ethnic or linguistic groups (e.g. Tvauri 2007 and attached literature). The subject is essential in resolving, for instance, the problems of “linguistic homelands” and directions of linguistic spread.

Mallory (1997: 94, 106–117) has written on the possibility of using archaeology and linguistics to solve the problem of the “homeland” of the Indo-Europeans. He is skep-tical about the matter and sees no purely linguistic method for resolving the problem of IE origins. He also presents several limitations in resolving the problem by means of archaeology, but he cannot ignore the results of archaeology entirely.

Malcolm Ross (1997: 141,158,162) criticizes the accuracy of glottochronology and recommends that linguists compare archaeologically discernable events with linguis- tic events. He feels that the comparative method in linguistics is unable to absolutely date linguistic events. He emphasizes the need to understand that the methods of these two disciplines are very different and stresses the importance of finding sequ-ences of linguistic events comparable with sequences of material-cultural events. He speaks of reconstructed linguistic and material-cultural events as manifestations of change in human societies. Relationships should be sought between them. Without going into further detail here, it is worth mentioning that he showed with the help of Central Papuan how to match the spread of Oceanic languages across the Pacific with the spread of the Lapita culture.

Johanna Nichols (1998: 259) presents a linguistic relative chronology on the Eura- sian steppe and their periphery.She admits that in order to define an absolute chro- nology,one must utilize both comparative linguistics and the results of archaeology. Some Finnish linguists present similar opinions (Koivulehto 2006: 153–154; Kallio 2006: 15–16).

In short, there is a wide consensus among linguists that cooperation is needed bet- ween linguistics and archaeology Saarikivi & Lavento (2012: 177–216 ) have criti- cized the so-called traditional way of combining archaeologically defined cultures with linguistic groups. In their conclusions they present three key points against the traditional view:


1) Ethnicity and linguistic identity do not always correlate.


2) Linguistic diversity was supposedly more significant in the past than today.Modern examples show that different linguistic groups may have a rather similar material culture even though the density of languages is high.


3) Languages spread not only by migration, but also by language shift. Language shift easily spreads over culturally significant boundaries.


First of all, the concept of ethnicity should be defined. For instance,in Finland there are many populations who have lived in Finland for a very long time.All of them could be considered different ethnic groups. There are Finnish-speaking, Swedish-spea-king and Saami-speaking populations. However, in Finland’s sport teams one might find representatives of all these linguistic groups and yet the spectators (and the players themselves) would refer to all of them as Finns.

Barth (2009) writes that ethnic identities can be considered as boundary identities defined by the members and the neighbors of the group. The group may be linguistically or culturally heterogeneous.

On the other hand, even though the Roma in Finland are native speakers of the Fin- nish language, they are often distinguished in the minds of other peoples of Finland as an independent ethnic group. It is thus possible to speak of “ethnicities” and “eth-nic groups”. Ethnicity might occur at many different levels. Of course,several scholars have presented their own definitions, but in the following I do it in the way most relevant for my research.


1) Representative identity: armies, sport teams, citizenship or everything else that is permanently somehow differing and distinguishes them from other nationalities. In this case it might be useful to speak of an ethnic macrogroup or of a nationality and their ethnic subgroups.

2) Linguistic identity.If the ethnicity is determined only according to the language, this may cause serious difficulties. Are the Finnish Muslim Tatars or Jews in Finland con-sidered to be ethnic Finns,because they usually speak Finnish as their first language (or in some cases Swedish)? In Russian Karelia the usual situation is that grandpa-rents speak Karelian as their native language. The next generation is bilingual, but Russian is the more natural language.

Young adults and children may hardly understand Karelian; Russian is obviously their native language. The question is, do people living in the same village, or even in the same house as a family, belong to different ethnic groups? According to the Russian authorities all of them were some years ago registered as Karelians.


3) Sociocultural identity. Especially in imperial or/and authoritarian states,such as the former Soviet Union or China and to some degree even the United States (as histori-cally an imperial state), there is a large number of ethnic groups who share quite a si-milar “imperial” culture because the official culture has a superior prestige or it might even be dangerous to profess a local identity. In the United States there are ethnic Native Americans who speak English as their native language but still retain a high sense of their own ethnicity. There are many speaking Spanish as their first langu-age, but they are identified as “white Americans”. African-Americans may have their own subculture, partially their own type of English language or dialect and they have an African genealogy. However, they are deeply American.


4) Religious identity. A good example are the Jews. If a Jewish person receives Christian baptism, he or she is no longer regarded as a Jew according to the Rabbi-nate or the immigration laws of the state of Israel. However,so-called Messianic Jews who believe in Jesus as their Messiah identify themselves clearly as Jews. For the anti-semitic Nazi regime, the definition of Jewishness was not dependent on religion. On the other hand, there exist religious evangelical Christians all over the world who wish to identify themselves as Jews, because some of their ancestors had a Jewish-looking name, but they cannot adequately demonstrate Jewish roots.


Archaeology may reveal something of all of the factors mentioned above.

Archaeological findings may include such objects as:


1) characteristic equipment of armies – representative identity
2) letters, stones, houses or utensils with ancient inscriptions – linguistic identity
3) cultural finds such as pottery, remnants of housing systems, etc. – cultural identity
4) ethnically characteristic religious ornamentation and/or cultic objects, such as the Jewish menorah, Saami drums, etc. – religious identity

The result is that on an archaeological basis, ethnicity can in most cases be defined only partially, not absolutely, i.e. only specific aspects of past ethnicities are discer-nible in the archaeological material from a particular period and region. For the rea-sons presented above,in any scientific research each of these definitions – represen- tative, linguistic, cultural and religious – should be examined separately. It is possible to excavate these findings, but the interpretation of the material is not always easy in the correct context. The more ancient the findings are, the more difficult their interpre-tation is.If there is already some historical context,such as literary sources in the case of archaeology in Egypt, the Levant or Mesopotamia, then the interpretation is much more reliable.

Because ancient prehistoric inscriptions in Uralic languages do not exist, archaeolo- gy cannot independently provide absolute evidence of Uralic linguistic groups in the prehistoric era.14 This does not mean that the results of archaeology are useless. Ar-chaeological data may in certain cases confirm conclusions based on toponymic or linguistic (cf. loanwords) material or help in dating them (Koivulehto 2006: 153– 154). However, archaeological results cannot usually be the basis for conclusions with the exception of found inscriptions or certain national religious objects, such as ancient Scandinavian runic stones, stars of David, etc.

There are some rules of thumb that can be useful and lend extra significance to ar-chaeological findings.From the point of view of archaeology,it is very important to dis- tinguish among all the research material objects that may be relevant for defining eth-nicity but not necessarily the linguistic background. Saarikivi & Lavento (2012: 197, 200) admit that indicators for ethnicity may exist, but they state that they are not easy to identify.

There are object finds that apparently were trading material.Saarikivi&Lavento (2012: 195) point to the “fruit knives” found in Lapland as an example. It is evident that these knives can only tell about contacts between Lapland and more southern areas. How do we know? Because the distribution of those knives is geographically much wider than most of the other findings, which seem to be more local. Thus everything that might be traded between different communities must be ruled out when researching linguistic or ethnic groups by means of archaeology.


14 The oldest writings in any Uralic language are the Old Hungarian texts (10th century) and Novgorodian birch bark documents from the late Middle Ages.


There may be material that is typically “imperial” or “representative” in use by many linguistic groups. It is not always easy to determine which material is of this kind. If a certain non-commercial object is at first typical only in a restricted area that is suspec-ted to have been culturally and politically strong and central, and afterwards its distri-bution increases remarkably elsewhere, it is possible that the character of this object is “representative” or “imperial”. Along the periphery of distribution a former culture is often mixed with the new one, especially in the beginning of the  expansion. This is the case during the early stages of Russification of former Uralic-speaking communi-ties (Rjabinin 1997). This kind of situation means that the users did not necessarily all speak the same language, but probably belonged to the same political alliance.


There exist also findings that are typical for microareas. According to Saarikivi & La-vento (2012:197) this is true, for instance, in women’s costumes and traditional orna-ments. Women who spoke close dialects of the same language may have had distin-guishing features in their costumes;cf.the oval tortoise brooches of the Häme and Sa-vo/Karelia type (Taavitsainen 1990: 77-112) or the different kinds of “horse-shaped” pendants found in Central Russia (Makarov 2006:277,Fig.4;Rjabinin 1997:180–181).

Therefore, this kind of material is not relevant for defining larger linguistic areas, such as the ancient total areas where the Meryan-Muroma language was spoken (see Rahkonen 2012: 19–27) or the total Finnic-speaking area. They may show regional and dialectal peculiarities. As a regional ethnic marker in Finland one can point to rowboats of the Häme and Savo types, which are clearly different from each other. As mentioned by Saarikivi&Lavento (2012:210),the connection between material and ethnicity becomes more difficult to solve the larger archaeological entities are, as in e.g. the distribution areas of the Textile Ware or the Ananino Culture in northern and central European Russia.


If the distribution of many different objects of a local nature (non-commercial and non- “representative”) creates an uniform areal entity and matches the toponymic evidence, there is good reason to suspect that in this area culture and language belonged to the same linguistic group. This may offer some possibilities for dating the toponyms of the area. It is important to highlight that,in most cases, archaeological evidence alone cannot testify to an ethnos or linguistic group.


7.2. Principal archaeological cultures between Finnic and Mordvin


Of course it is true that so-called archaeological cultures are often defined by simpli-fied means. An archaeological culture cannot be defined only as a territory where a certain type of ceramics was used. Any material culture consists of several factors.

A great number of archaeologically defined cultures existed between the territories of historical Finnic and Mordvin speakers already for a long time before these langua-ges developed. The first one of note was the early Neolithic Ljalovo culture (circa 5000–3650 BC) [map 3]. This culture originated in the territory between the Upper Volga and the Middle Oka (Carpelan 1999: 257). It spread from there northwest towards the Lake Onega region (Carpelan & Parpola 2001: 79).

After the Ljalovo culture came the late Neolithic Volosovo culture (3650–1900 BC) [map 4] almost in the same region in the Volga–Oka area, stretching to the River Volkhov and the lakeland of the Russian North. This culture influenced also cultures in Finland, leaving traces in the Kierikki Ware and Pöljä-Jysmä Ware cultures (ibid. 2001: 83).

The third archaeological culture of note created in Central Russia was the Textile or Netted Ware culture launching the Bronze Age (1900–800 BC) [map 5]. Textile Ceramics spread from the Upper Volga-Oka region via the southern waterways (the Volga and Mologa routes) towards the River Volkhov and via the more northern way (the Belozero route) to the district of the Northern Russian lakeland, Karelia and the inlands of Finland. This culture had several subcultures even in Finland (Lavento 2001: 166–168). The differences between the subcultures in Russia have regrettably been poorly examined.

In the Upper Volga-Oka region,the original Textile Ceramics culture split in the early Iron Age into the western Djakovo and eastern Gorodec cultures.The Djakovo cul- ture reached the Lake Ilmen area in the west (Patrushev 2000:90;Fig. 31) bordering on the Kalmistonmäki-Volxov culture south of Lake Ladoga. The Kalmistonmäki-Volxov culture was another daughter culture of the Textile Ceramics.

It has been assumed that the language of this population was Proto-Finnic (Uino 2006: 363), but this is very hard to prove.



[HM: Kantasuomi on muodostunut kampakeraamisen kulttuurin SU-kielen pohjalta vasrakirveskilen vaikutuksesta.]


The feature common to all these Central Russian cultures is the southwestern boun-dary (see the SW-boundaries in maps 5–7) that runs roughly from Kaluga oblast over the watersheds between the Volga and Dnieper and between the Lovat’ and Za-padnaja Dvina (Daugava) to the (north)west. As already mentioned in this section, there has been a fierce debate about the correlation between archaeological cultural and linguistic areas. The main question is, does any archaeological cultural area correlate with historical linguistic areas? It is obvious that at least such long-lasting archaeological boundaries as the one mentioned above in central and northwestern Russia (lasting over 5000 years!) or the boundary between the coastal and inland cultures of Finland (which lasted around 4000 years) no doubt correlate with topo-nyms (Rahkonen 2011: 211, map 2; 2013, map 23; Rjabinin 1997: 4, map 1 according to Sedov; Kriiska & Tvauri 2007: 148, map).

In the following, three maps are presented. At first, it is necessary to state that the his-torical Finnic and Mordvin-speaking areas mentioned on the maps illustrate the situ-ation in the historically known era, and not the era of the presented archaeological cultures. However, if the population behind the above-mentioned cultures spoke Pre- and Proto-Uralic and the later Finno-Mordvin protolanguage (West Uralic?) as assumed by many scholars, the location of these cultures might explain something of the development of the prehistoric Uralic entity and the reason why Finnic and Mordvin are relatively close to each other.


The Ljalovo culture (circa 5000–3650 BC) [map 5] developed in the Upper Volga re- gion. For example, Carpelan has considered the Ljalovo population to have been speakers of Proto-Uralic (Carpelan 2006: 85). Asko Parpola (2012: 288) suggests that it was the population of the Volosovo culture (3650-1900 BC) [map 6] who were the speakers of Proto-Uralic. Petri Kallio (2006: 16) has presented similar opinions. Apparently due to the influence of Corded Ware, a new type of ceramics appeared after the Volosovian era. This was called Textile Ceramics or Netted Ware (1900 – 800 BC) [map 7]. Parpola (2012: 288) reminds us that it corresponds to the distri-bution of West Uralic (see also Rahkonen 2013, map 23). Von Hertzen (1973: 90) has stated that there is a Volgaic loanword stratum in Finnic that,according to him, might have been borrowed through the population of Textile Ceramics.

Lately a doctrine has arisen that protolanguages were spoken in relatively narrow areas called their “homeland” (see e.g. Aikio 2006:42–43; Mallory 1997:93; Nichols 1998: 224). I would like to briefly comment on this doctrine:

1) No protolanguages were ever spoken in reality in the forms of the different stages presented and defined by linguists. This means that there never existed a Proto-Uralic [PU] or Proto-Finno-Permic [PFP] language as a monolithic language. Instead, they are theoretical stages that correspond to the real linguistic situation only partial-ly. The phonetic shifts probably happened inside of a certain close, but not exactly similar, linguistic community step by step over a long time. The recon-struction of protolanguages usually endeavors to illustrate only the last stage.

2) Every language, ancient and modern, has always had different dialects. Some of them develop towards new languages and some of them never do. Dialects interact and exchange borrowings from each other over a long time, causing “linguistic waves” proceeding in many directions and revising the concrete language all the time.

3) If we compare the European situation (two basic language families:Indo-European and Uralic) with the native languages of North America, we notice that there are huge areas covered by relatively few different language families. In the western part of Canada (including Alaska as well) the Athabaskan languages have been widely spoken, while in the eastern part Algonquin languages have been spoken. In the northeastern coastal area of the USA were speakers of Iroquois languages, in the southeastern areas speakers of Muskogean languages and in the Central Plains speakers of Siouan languages.(The Native Americans 1991) [TNA]. However, it is true that in Siberia there have been perhaps 10 language families. Thus, it is difficult to tell whether the northern European linguistic situation should be compared with North America or Siberia.

Is there any basis to assume that,for example,in Canada those two basic language families of American Indians were originally small groups living in a narrow restricted area and then spread over very wide territories,assimilating all the other languages? For me it is more probable that these linguistic groups already existed in some ancient form (Pre-Athabaskan and Pre-Algonquin) when the population crossed the Bering Strait after the Ice Age.A similar situation is possible in the area between the Baltic Sea and the Urals. Proto-Uralic must be a descendant of Pre-Uralic, which is beyond the capability of comparative linguistics to reveal. But when talking about the ethnic settlement history, the population of the Pre-Uralic language must be taken in-to account. One can assume that a few newcomers speaking a very early Pre-Uralic language migrated to the place which became the narrow “homeland” of the Pre-Ura-lic-speaking population and then began to slowly spread across the surrounding ter-ritory. Therefore, it is possible to hold that the speakers of Proto-Uralic could have al-ready lived in a relatively large area. Of course, this cannot be proven, but it remains a possibility.



map 5. The Ljalovo culture (circa 5000–3650 BC) between the much later traditional areas of Finnic and Mordvin speakers (? >0 AD), as featured in Huurre (2001: 25).




map 6. The Volosovo culture (circa 3650–1900 BC) between the much later traditio-nal areas of Finnic and Mordvin speakers, as featured in Carpelan (1999: 263).





map 7. The Textile Ceramics culture (circa 1900–800 BC) between the later traditio-nal areas of Finnic and Mordvin speakers, as featured in Carpelan (1999: 269).


7.3. Textile Ceramics and Western Uralic


It should be observed that the territory between the historical Finnic and Mordvin-speaking areas matches quite well with the area of the so-called Textile Ceramics [circa 1900–800 BC] (cf. Parpola 2012: 288). The culture of Textile Ceramics could function as a bridge between these two extreme points. Languages that were spoken later in this vast territory between Finland–Estonia and Mordovia seem to derive from Western Uralic (WU) as well.I have called those languages Meryan-Muroma, Eastern and Western Čudian and an unknown “x” language spoken in inland Finland, Karelia and the Lake Region of the Russian North (Rahkonen 2011;241;2012a:19–27; 2013: 5–43).This might mean that the territory of the Early Textile Ceramics reflects to some extent the area of late Western Uralic.15.


15 However, it is possible that also non-Uralic languages were spoken in the area of the Textile Ware Culture.


The archaeologically problematic area is Estonia, Livonia and Coastal Finland – the area traditionally assumed to have been populated by the late Proto-Finns.The Tex- tile Ceramics culture was absent there. It is very difficult to believe that the Textile Ware population in inland Finland migrated or was even the main factor bringing the Pre- or Early Proto-Finnic language to Estonia or Livonia. There are no archaeological or toponymic signs of it.

Therefore, I am forced to believe thatTextile Ceramics did not bring Uralic-speaking people to those regions. This makes it possible, but not absolutely proven, to assume that some type of Uralic language was spoken in the region of the Gulf of Finland already before Textile Ceramics spread to the northwest (circa 1900 BC).

The Corded Ware population in Finland is thought to have been NW Indo-European by many scholars (e.g. Koivulehto 2006: 154–155; Carpelan & Parpola 2001: 84). At least, it is probable that the Corded Ware culture was brought to Finland by waves of migration,because the representatives of the former Late Comb Ceramics partially lived at the same time side by side with the Corded Ware population. However, it is possible that the immigrants were a population that spoke Proto-Ura- lic, who had adopted the Corded Ware culture from their Indo-European neighbors, possibly from the population of the Fatyanovo culture, e.g. in the Valdai region.

[HM: Suomessa oli itäbalttilaista vasarakirveskansaa, eikä keitä tahansa nuorake-raamikkoja, esimerkiksi latvialaisten ja liettualaisten ja preussilaisten esi.isiä. Suo- mesta vasarakirveskansaa siirtyi myös Ruotsiin, joka näin sai ilmeisimmin ensim-mäiset indoeuoppalaista kieltä puhuvat asukkansa.]

This was suggested by Terho Itkonen (1997: 251) as well. In that case the popu- lation of the Typical and Late Comb Ceramics may have spoken some Paleo-European language (see Saarikivi 2004a).

[HM: Suomi on muodostunut aikasemman SU-kielen ja pohjalta vasrakirveskielen vaikutuksesta.]

In the Early Bronze Age, the Baltic Pre-Finnic language that I have suggested must have been very close to late WU and therefore no substantial linguistic differences existed between the Baltic Pre-Finns and the population of Textile Ceramics in in-land Finland. I admit that this model is difficult to prove, but I have presented it prima-rily in order to offer new models of thinking.16 At least, there is no archaeological or linguistic reason against this idea.

[HM: Täytyy sanoa, että en täysin ymmätänyt ideaa...

Pangermanisen esittämät "paleosuomalsiet sanat" ovat tavallisesti osoittautneet balttilainoiksi (jänis), liiviksi (ilves) jne.
]


A few remarks should be made on the history of settlements in Estonia as well. This might be important when thinking of common Finnic roots. The early settlements in the Bronze Age in mainland Estonia were concentrated in three parts of the country: the northern coastal area, the Pärnu valley area and the area between the lakes Võrtsjärv and Peipus (Kriiska & Tvauri 2007: 95). In the late Bronze Age, the popu-lation in the area of the Pärnu region seems to decrease remarkably (ibid. 2007: 97).

In the Roman Iron Age (50-450 AD) there seem to have been three principal centers of population: 1) the Northern Coastal region,2) the Midland region (Põltsamaa) and 3) the southern area (Tartu-Võru region).


16 Of course the population of the Textile Ware culture was not linguistically totally homogenous, but supposedly spoke many Western Uralic dialects and possibly different Paleo-European languages as well.


In the western coastal area,especially in the Pärnu region,only a few archaeological sites are found. In the Haapsalu region there is a small concentration of graveyards (ibid.2007:131).In the Age of Migrations the situation is similar to previous eras(ibid. 2007:149). In the Viking Age (800–1050 AD) all of the country seems to have been almost equally populated, with the exception of the Pärnu region and the area north of the Lake Peipus as the situation had been already during the entire Iron Age.The population seems to have been divided into northern and southeastern parts with the Pärnu valley as the borderline (ibid 2007: 173). This may be reflected in the formation of northern and southern Estonian languages in accord with the hypothesis of Kallio (2007, see above).




map 4. Principal settlements in Estonia according to the archaeological evidence in the period 0–1050 AD, as featured in Kriiska & Tvauri 2007.


Literature


AGEEVA = Агеева Р.А.1989:Гидронимия Русского Северо-Запада как источник
культурно-исторической информации. Москва.
AHLQVIST,ARJA 1998:Merjalaiset - Suurten järvien kansaa.Virittäjä 102:1:24–55. ― 2006: Ancient Lakes in the Former Finno-Ugrian Territories of Central Russia: An Experimental Onomastic-Paleogeographical Study.The Slavicization of the Russian North.Juhani Nuorluoto (ed.). Slavica Helsingesia 27. Helsinki. 11–49.
AHLQVIST = Альквист А.2001:Субстратная топонимия Ярославского поволжья. Очерки Исторической Географии.Северо-Запад России славяне и фиинны.Под. ред. А.С.Герда и Г.С.Лебедева. Издательство С.-Петербургского Университета.
AIKIO, ANTE 2003: Suomen saamelaisperäisistä paikannimistä. Virittäjä 1/2003.

― 2004: An essay on substrate studies and the origin of Saami. Etymologie, Ent-leh-nungen und Entwicklungen Festschrift für Jorma Koivulehto zum 70. Geburtstag. Irma Hyvärinen, Petri Kallio & Jarmo Korhonen (eds.).Mémoires de la Société Néo- philologique de Helsinki 63. Helsinki: Société Néophilologique. 5–34.

― 2006:On Germanic-Saami contacts and Saami prehistory. Suomalais-Ugrilaisen Seuran Aikakauskirja 91. Helsinki: Suomalais-Ugrilainen Seura. 9–55.
― 2007: The Study of Saami substrate Toponyms in Finland. Onomastica Uralica 4. Helsinki: Debrecen. 157–197.
AIKIO, ANTE & AIKIO, ASLAK 2001: Heimovaelluksista asutusjatkuvuuteen: suoma- laisen väestöhistorian tutkimuksen pirstoutuminen. Muinaistutkija 4/2001. 2–21.
AJO = Атлас Ярославской области (1:100 000) 2002: Москва: Роскартография.
AKO = Атлас Костромской области (1:100 000) 2009: Москва: Роскартография.
ATO = Атлас Тверской области (1:100 000) 2005: Москва: Роскартография.
AVO = Атлас – Вологодская область (1:200 000) 2008: Санкт Петербург: ФГУП Аэрогеодезия.
BARTENS, RAIJA 1999: Mordvalaiskielten rakenne ja kehitys.Suomalais-Ugrilaisen Seuran toimituksia 232. Helsinki: Suomalais-Ugrilainen Seura.
― 2000: Permiläiskielten rakenne ja kehitys. Suomalais-Ugrilaisen Seuran Toimituksia 238. Helsinki: Suomalais-Ugrilainen Seura.
BARTH, F.1969: Ethnic Groups and Boundaries. The Social Organization of Culture Differences (Ed.by F.Barth). Scandinavian University Books, Bergen, Oslo, London.
CARPELAN, CHRISTIAN 1999: Käännekohtia Suomen esihistoriassa aikavälillä 5100...1000 eKr. Pohjan poluilla. Suomalaisten juuret nykytutkimuksen mukaan. Paul Fogelberg (ed.). Bidrag till kännedom av Finlands natur och folk 153. Societas Scientiarum Fennica. 249–280.
― 2006: On Archaeological Aspects of Uralic, Finno-Ugric and Finnic Societies be- fore AD 800. The Slavicization of the Russian North. Juhani Nuorluoto (ed.). Slavica Helsingensia 27. Helsinki: Department of Slavonic and Baltic Languages and Literatures at Helsinki University. 78–92
CARPELAN, CHRISTIAN & PARPOLA,ASKO 2001: Emergence, contacts and dis- persal of Proto-Indo European, Proto-Uralic and Proto-Aryan in archaeological pers- pective. Early contacts between Uralic and Indo-European: Linguistic and Archaeo-logical considerations. Mémoires de la Société Finno-Ougrienne 242. Helsinki: Suomalais-Ugrilainen Seura. 55–150.
DARJA = ДАРЯ 1986: Диалектологический атлас русского языка. Центр Евро- пейской цасти СССР. Выпуск 1: фонетика, под. ред. Т.Ю. Строгановой и С.В. Бромлей. Москва: Наука.
DARJA = ДАРЯ 1989: Диалектологический атлас русского языка. Центр Евро- пейской цасти СССР. Выпуск 2: морфология, под. ред. С.В. Бромлей. Москва: Наука.
DERJAGIN, DERJAGINA, MANUXIN = Дерягин В. Я., Дерягина З. С., Манухин Г. И. 1987: Топонимика Кенозера 1-2. Архангельск. 1987.
EKI = Eesti Keele Instituut
ESRJ = ФАСМЕР, МАКС 2003: Этимологический словаь русского языка 1-4. Москва: Астрель. Аст.
EUROPAEUS, D. E. D. 1868–1870: Tietoja suomalais-ungarilaisten kansojen muinaisista olopaikoista. Suomi II. 7:1–190, 8: 27–106.
GBO = Смолицкая,Г.П. 1976: Гидронимия бассейна Оки. Москва: Издательство Наука.
GT2000 = 2000: GT2000 Tiekartasto. Vantaa: Karttakeskus Oy.
GRÜNTHAL, RIHO 1997: Livvistä liiviin: Itämerensuomalaiset etnonyymit. Castre-nianumin toimitteita 51. Helsinki: Helsingin yliopisto. Suomalais-Ugrilainen Seura.
― The Mordvin languages Between Bush and Tree: A Historical Reappraisal. Sámit, sánit, sátnehámit. Riepmočála Pekka Sammallahtii miessemánu 21.beaivve 2007. Jussi Ylikoski & Ante Aikio (ed.). Mémoires de la Société Finno-Ougrienne 253. Helsinki: Société Finno-Ougrienne. 115–135.
HERTZEN, ERIK VON 1973: Itämerensuomen lainakerrostumien ikäämisestä. Suo- malais-Ugrilaisen Seuran Aikakauskirja 72: Helsinki. Suomalais-Ugrilainen Seura.
HÄKKINEN, JAAKKO 2009: Kantauralin ajoitus ja paikannus:perustelut puntarissa. Suomalais-Ugrilaisen Seuran Aikakauskirja 92. 9–56.
― 2010a: Jatkuvuusperustelut ja saamelaisen kielen leviäminen (osa 1).Muinaistut- kija 1/2010. Helsinki: Suomen Arkeologinen Seura. 19–36.
INŽEVATOV = Инжеватов И.К.1987 Топонимический словарь Мордовской СССР. Саранск.
ITKONEN, TERHO 1983: Välikatsaus suomen kielen juuriin. Virittäjä 87: 190–229, 349–386.
― 1997: Reflections on Pre-Uralic and the “Saami-Finnic protolanguage”. Finnisch-Ugrische Forschungen 54, Heft 3. Helsinki. 229–266.
KALIMA, JALO 1942: Karjalaiset ja merjalaiset. Uusi Suomi-lehti 19.7.1942
― 1944: Einige Russische Ortsnamentypen. Finnisch-Ugrische Forschungen 28: 99–150.
KALLIO, PETRI 2006: Suomen kantakielten absoluuttista kronologiaa. Virittäjä 1/2006. 2–25.
― 2007: Kantasuomen konsonanttihistoriaa. Sámit, sánit, sátnehámit. Riepmočála
Pekka Sammallahtii miessemánu 21. beaivve 2007. Jussi Ylikoski ja Ante Aikio (eds.). Suomalais-Ugrilaisen Seuran Toimituksia 253. Helsinki: Suomalais-Ugrilai- nen Seura. 229–249.
― 2009: Stratigraphy of Indo-European Loanwords in Saami. The Roots of Saami Ethnicities, Societies and Space/Places. Tiina Äikäs (ed.). Publications of the Giellagas Institute. Oulu. 30–45.
KIVINIEMI, EERO 1990: Perustietoa paikannimistä. Helsinki: Suomalaisen Kirjallisuuden Seura.
KOIVULEHTO, JORMA 1987: Namn som kan tolkas urgermanskt. Klassiska problem inom finlandssvensk ortnamnsforskning. (Ed.) Lars Huldén. SSLS. Studier i nordisk filologi. 67. Helsingfors: SLS.
― 2006: Arkeologia,kielihistoria ja jatkuvuusteoria.Arkeologian lumoa synkkyyteen: Artikkeleita Christian Carpelanin juhlapäiväksi. Helsinki: Yliopistopaino. 153–165.
KRIISKA, AIVAR & TVAURI, ANDRES 2007: Viron esihistoria. – Hannu Oittinen & Andres Tvauri (suom.). Helsinki: Suomalaisen Kirjallisuuden Seura.
LAVENTO, MIKA 2001: Textile Ceramics in Finland and on the Karelian isthmus. Helsinki: Suomen Muinaismuistoyhdistyksen Aikakauskirja 109.
LEONT’EV, A. E. 1996: The Archaeology of the Merya (the Early History of North-Eastern Russia). Russian monographs in Migration period and Medieval Archaeolo- gy. Gennadii Afanas’ev and Falco Daim in collaboration with Dafydd Kidd (eds.). Volume 4. Moscow: Russian Academy of Sciences. Institute of Archaeology.
LOGINOV = Логинов, К.К.1999: Еще раз о «вепском» прошлом Водлозерья. Вепсы: История, культура и межэтнические контакты.Карельский научный центр РАН институт языка, литературы и истории. Петрозаводск: Издатеьство Петрозаводского государственного университета. 104–111 MAG = Муллонен, И. И. & Азарова, И. В. & Герд, И. В. 1997: Словарь гидронимов юго восточного Прила- дожья бассейн реки Свирь.Санкт-Петербург:Издательство Санкт Петербургского университета.
MAKAROV, N.A.2006: Cultural Identity of the Russian North Settlers in the 10th – 13th Centuries: Archaeological Evidence and Written Sources. The Slavicization of the Russian North. Juhani Nuorluoto (ed.).Slavica Helsingensia 27.Helsinki: Depart- ment of Slavonic and Baltic Languages and Literatures at Helsinki University.
MALLORY, JAMES P.1997:The Homelands of the Indo-Europeans.Archaeology and Language 1. Ed. Blench Roger and Matthew Spriggs. London: Routledge.
MALLORY & ADAMS 1997: Encyclopedia of Indo-European Culture. J.P. Mallory and D.Q. Adams (eds.). London: Fitzroy Dearborn Publishers.
MARKOV, ALEKSEJ 1998: Мещёра. www.dlc.fi/~Kokov/mesher/meshchera. Meshchera. www.dlc.fi/~kokov/mesher/index.html
MATVEEV = Матвеев, А. К. 2001: Субстратная топонимия Русского Севера I. Екатеринбург: Издательство Уральского университета.
― 2004:Субстратная топонимия Русского Севера II.Екатеринбург:Издательство
Уральского университета.
― 2006: Ономатология.Москва: Российская академия, отделение историко филологических наук. «Наук».
MULLONEN = Муллонен,И.И.1994:Очерки вепской топонимии.Санкт Петербург.
― 2002: Топонимия Присвирья. Проблемы этноязыкого контактирования.Петро- заводск: Российская академия наук, карельский научный центр: Институт языка, литературы и истории.
― 2008:Топонимия Заонежья. Словарь с историко-культурным комментариями. Петрозаводск: Карельский научный центр РАН: Институт языка, литературы и истории.
NA = Kotimaisten kielten tutkimuskeskuksen nimiarkisto. Institute for the Languages of Finland.
NICHOLS, JOHANNA 1998: The Eurasian spread zone and the Indo-European dispersal. Archaeology and Language 2. Ed. Blench Roger and Matthew Spriggs. London: Routledge.
NISSILÄ, VILJO 1975: Suomen Karjalan nimistö. Joensuu: Karjalaisen Kulttuurin Edistämissäätiön julkaisuja.
PARPOLA, ASKO 2012: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 multilingue. Mémoires de la Société Finno-Ougrienne 264.Helsinki: Suomalais-Ugrilainen Seura. 287–298
PATRUSHEV, VALERY 2000: The Early History of the Finno-Ugric Peoples of European Russia. Societas Historiae Fenno-Ugricae. Oulu. 2000.
PITKÄNEN, RITVA LIISA 1985: Turunmaan saariston suomalainen lainanimistö. Suomalaisen Kirjallisuuden Seuran Toimituksia 418. Helsinki: Suomalaisen Kirjalli-suuden Seura.
POPOV = Попов, А.И.1965:Географические названия (Введение в топонимику). АН СССР,
Географическое общество СССР. Москва–Ленинград: Наука.
PS = Sammallahti, Pekka 1988:Historical phonology of Uralic languages. With spe- cial reference to Samoyed, Ugric and Permic. Uralic Languages. Denis Sinor (ed.). Description, history and foreign influences. Leiden – New York – København –Köln.: E.J. Brill. 478–554.
PSRL 1965 = Полное собрание русских летописей 9–12. Москва.
RAHKONEN,PAULI 2009:The linguistic background of the Ancient Meshchera tribe and principal areas of Settlement. Finnisch-Ugrische Forschungen 60:162–202.
― 2011: Finno-Ugric Hydronyms of the River Volkhov and Luga catchment areas. Journal de la Société Finno-Ougrienne 93. Helsinki. 205–266.
― 2012a: Границы распространения меряно-муромских и древнемордовских гидронимов в верховьях Волги и бассейне реки Оки. Вопросы Ономастики 2/2011. Екатеринбург. 5–42.
― 2013: Suomen etymologisesti hämärä vesistönimistö. Virittäjä 1/2013.
RJABININ = Рябинин, Е. А. 1997: Финно-угорские племена в составе древней Руси. Санкт Петербург: Издательство Санкт Петербургского университа.
ROSS, MALCOLM 1998: Sequencing and dating linguistic events in Oceania: the linguistics/ archaeology interface. Archaeology and Language 2. Ed. Blench Roger and Matthew Spriggs. London: Routledge.
RÄISÄNEN, ALPO 2003:Niemet mieltä kiehtovat: etymologista nimistöntutkimusta. Suomalaisen Kirjallisuuden Seuran toimituksia.Helsinki:Suomalaisen Kirjallisuuden Seura.
SAARIKIVI, JANNE 2000: Kontaktilähtöinen kielenmuutos, substraatti ja substraat-tinimistö. Virittäjä 3/2000. 393–415.
― 2004a: Is there Paleo-European substratum interference in western branches of Uralic? Journal de la Société Finno-Ougrienne 90. Helsinki. 187–214.
― 2004b: Über die saamischen Substratennamen des Nordrusslands und Finn- lands. Finnisch-Ugrische Forschungen 58: 162–234.
― 2006a: Introduction. Substrata Uralica. Studies on Finno-Ugric Substrate in Northern Russian Dialects. 7–50.
― 2006b: On the Uralic Substrate Toponymy of Arkhangelsk Region: Problems of Research Methodology and Ethnohistorical Interpretation. Onomastica Uralica 4. Debrecen. 1–64.
― 2011: Saamelaiskielet:historiaa ja nykypäivää.Irja Seurujärvi-Kari & Petri Halinen & Risto Pulkkinen. Tietolipas 234. Helsinki: SKS. 77–119.
SAARIKIVI, JANNE & LAVENTO,MIKA 2012:Linguistics and archaeology: a critical view of an interdisciplinary approach with reference to the prehistory of Northern Fennoscandia. Damm,Charlotte&Saarikivi, Janne (eds.): Networks, interaction and emerging identities in Fennoscandia and beyond. Mémoires de la Société Finno-Ougrienne 265. Helsinki: Société Finno-Ougrienne. 177–216.
SALMONS, JOE 1992: Northwest Indo-European vocabulary and substrate phono- logy. Perspectives on Indo-European language,culture and religion: studies in honor of Edgar C. Polomé. Vol. 2. Roger Pearson (ed.). Journal of Indo-European Studies. Monograph Series. 265–279.
SAMMALLAHTI, PEKKA 1989: Sámi‒Suoma sátnegirji. Saamelais-suomalainen sanakirja. Ohcejohka: Jorgaleaddji Oy.
― 1998: The Saami languages: an introduction. Kárásjohka: Davvi Girji
SEDOV = Седов, В.В.1979:Этнический состав населения Новгородской земли. Финно- угры и славяне. Ленинград. 74–80.
SIMINA = Симина, Г. Я. 1980: Географические названия (по материалам пись- менных памятников и современной топонимики Пинежья). АН СССР. Географи-ческое обшество СССР. Ленинград. Наука.
SMJ = Словарь марийского языка 1-10. Йошкар-Ола. 1990–2005.
SPK = Suomalainen paikannimikirja 2007: Sirkka Paikkala (ed.). Jyväskylä: Kartta-keskus, Kotimaisten kielten tutkimuskeskus.
SSA = Suomen sanojen alkuperä I-III. 1992–2000: Itkonen, Erkki – Kulonen, Ulla- Maija (ed.). Suomalaisen Kirjallisuuden Seuran Toimituksia 556,Kotimaisten kielten tutkimuskeskuksen julkaisuja 62. Helsinki: Kotimaisten kielten tutkimuskeskus. Suomalaisen Kirjallisuuden Seura.
TAAVITSAINEN, JUSSI-PEKKA 1990: Ancient hillforts of Finland: problems of ana- lysis, chronology and interpretation with special reference to the hillfort of Kuhmoi- nen. Helsinki: Suomen Muinaismuistoyhdistyksen aikakauskirja.
TKAČENKO =Tкаченко, О.В. 1985: M ерянский язык. Киев:АН Украйнской ССР. Наукова думка.
― 2007: Исследования по мерянскому языку. Кострома.

TNA = Richard Collins (ed.) 1991: The Native Americans. Salamander Books Ltd.
TRET’JAKOV = Третьяков П. Н. 1966: Финно-угры, балты и славяне на Днепре и Волге. Москва. Лениниград.
TVAURI,ANDRES 2007:Migrants or Natives?The Research History of Long Barrows in Russia and Estonia in the 5th–10th Centuries. – Topics on the ethnic, linguistic and cultural making of the Russian North. Slavica Helsingensia 32.Helsinki: Depart- ment of Slavonic and Baltic Languages and Literatures at Helsinki University.
UEW = Rédei, Károly (ed.) 1986–91: Uralisches etymologisches Wörterbuch. Budapest.
UINO, PIRJO 2006: The Background of the Early Medieval Finnic Population in the Region of the Volkhov River: Archaeological Aspects. The Slavicization of the Rus- sian North. Juhani Nuorluoto (ed.). Slavica Helsingensia 27. Helsinki: Department of Slavonic and Baltic Languages and Literatures at Helsinki University. 355–373.
VASIL’EV = Васильев, В.Л. 2001: Метонимическое калькирование архайских гидронимов в Приильменье. Топонимия и диалектная лексика новгородской земли. Велики Новгород: Министерство образования Российской Федерации. Новгородский государственный университет имени Ярослава Мудрого. 6–13.
― 2005: Архаическая топонимия новгородской земли. Велики Новгород: Нов- городский Межрегиональный Институт Общественных Наук.
VASMER, MAX 1932–1936: Beiträge zur historischen Völkerkunde Osteuropas I-IV. Verlag der Akademie der Wissenschaften. Berlin.
WRG = Vasmer, Max 1961–69: Wörterbuch der russischen Gewässernamen I-V. Berlin. 1961–1969.

https://www.facebook.com/pauli.rahkonen.52


Henkilön Pauli Rahkonen kuva.