469 29 16MB
English Pages 424 [425] Year 2023
Romanian Folklore and its Archaic Heritage A Cultural and Linguistic Comparative Study
Ana R. Chelariu
Romanian Folklore and its Archaic Heritage
Ana R. Chelariu
Romanian Folklore and its Archaic Heritage A cultural and Linguistic Comparative Study
Ana R. Chelariu Emerson, NJ, USA
ISBN 978-3-031-04050-4 ISBN 978-3-031-04051-1 (eBook) https://doi.org/10.1007/978-3-031-04051-1 © The Editor(s) (if applicable) and The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023 This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cover illustration: Friptuleac Roman / Alamy Stock Photo This Palgrave Macmillan imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
“Aion is a child at play, playing draughts; the kingship is a child’s.” —Heraclitus, Fr. 52
To my mother
Acknowledgements
The completion of my work took many years of research and a lot of effort to overcome the language barrier, and which could not have been realised without the encouragement and support of family, friends and distinguished professors, colleagues from the International Association of Comparative Mythology, the Institute of Archaeonythology, American Romanian Academy, and Romanian Universities. I will always cherrish the memory of Professor Dean Miller, a true friend of Romanian studies, and Professor Nick Allen, whose Hindu studies were generously made available to me. Special thanks and gratitude are due to Professor John Colarusso from MacMaster University, for his encouragement and guidance over personal correspondence. I owe a great deal to Professor Allen Ahsby for helping me in my struggle with the English language. Professor Stefan Stoenescu’s advise for the Romanian version of this work is much appreciated. Professor Adrian Poruciuc’s interests in Indo- European linguistics and Romanian folklore encouraged me in the beginning, and many thanks to Professor Mircea Diaconu from Stefan Cel Mare University, Suceava, Romania for his interest and reviewings of my articles. Appreciation goes to Professor Joshua James Pennington from the Peerwith, for his help in editing and language corrections. ix
x Acknowledgements
The many Romanian friends from diaspora, among whom renown folklorist Coca Eretescu, the poet and literary critic Mirela Roznoveanu, and the linguist Mihai Vinereanu, deserve special thanks. Last but not least, this work could not have been completed without the patience and understanding of my talented husband Serban, painter and poet in his own right, and my beautiful exceptional daughter, Andrea, my first and most understanding editor; and her wonderful family, Scott and my three grandkids, Sean, Claire and Thea, always the joys of my life, who I hope one day may be able to read this work.
Contents
Part I Getae-Dacians History and Religion in Ancient Sources 1 1 I ntroduction 3 2 Brief Overview of Recent Theories on the Indo-Europeans’ Homeland 9 3 Cucuteni-Tripolye: The Indo-European Homeland? 17 4 Some Recent Genetic and Archaeological Conclusions 25 5 Daco-Romanian Language: An Indo-European Branch 29 Part II Comparative Method: Myth, Fairy Tale, Folk Tale 39 6 The Development of the Comparative Method 41 7 The Concept of Myth 45
xi
xii Contents
8 Myth Between Symbol and Metaphor 53 9 M yth and Fairy Tales 63 10 Mythic Time Versus Fairy Tale Time 69 Part III Traits of Indo-European Mythic Motifs in Romanian Folk Stories 73 11 The Mythic Motif of Man’s Creation 75 12 C osmogony: Fârtat and Nefârtat, the Romanian Twins among the Indo-European Divine Twins 83 13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains in Romanian Beliefs 97 14 Th e Romanian Goddess Ileana Simziana: The Sun and the Moon Marriage109 15 The Sun and a Mortal Girl Marriage: The Romanian Song of Cicoarea ‘The Chicory’121 16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony125 17 R omanian Feminine Spirits iele, rusalii, șoimane141 18 Th e Romanian Păcală among the Indo-European Tricksters145 19 Th e Hero Slaying the Dragon in the Romanian Song Iovan Iorgovan153
Contents
xiii
20 Youth Rivalry Fights: From Nart Sagas to the Romanian Song Mioriţa169 21 Metamorphoses in Myth and Fairy Tales181 22 Metamorphoses in Youth Initiation Rites191 23 The Romanian Folk Story “The Enchanted Pig”: The Motif ‘Beauty and the Beast’199 24 Indo-European Social Structures and Youth Initiation Rites in Romanian Folk Customs205 25 F ather Christmas: Romanian Moş Crăciun—A Solar Myth215 26 C onclusions229 Part IV Daco-Romanian Language Position Among the Indo-European Languages 235 27 Th e Daco-Romanian Cultural Vocabulary237 A ppendix257 B ibliography389 I ndex411
About the Author
Ana R. Chelariu was born in Bucharest, Romania, on November 19, 1946. Early on, while studying Romanian language and literature at the University of Bucharest, she began her career writing original fairy tales for the Radio-TV Romania, and was awarded First Prize for an original fairy tale, transmitted on the radio broadcasting for children “Inşir-te mărgăritar.” The interest in fairy tales, particularly the relationship between myth and fairy tales, was the subject of her Master of Arts graduation thesis at the University of Bucharest, entitled Nemesis in the folktale type A-Th 325, the Wizard and his Apprentice. She has published articles, book reviews, and stories for children in magazines such as Neue Literatur and Cutezătorii; she worked as a freelance editorial advisor for the publishing house ‘Cartea Românească’. In 1978, she published her first book, a collection of original fairy tales The Secret of Happiness, Ion Creangă Publishing House, Bucharest. In January 1979, Ana immigrated to the United States. She continued her studies, graduating in 1981 from Rutgers University, with a Master in Library Science. While working as the director of a public library in Northern Jersey, she continued her studies in mythology, folklore, and language history. As a member of the Society of Romanian Studies, she presents her research studies on the relation between myth and folktale at various xv
xvi
About the Author
conferences organized by the Society, studies published in English in the Romanian Civilization magazine. In 2001, she published together with the Romanian writer Nina Ceranu Libertăţile bufniţei [The Owl Freedoms], Ana@West and Nina@ East, a collection of e-mail correspondence between two writers who never met in person. The book was reviewed in a few newspapers: Ildico Achimescu, National Premiere, in Timişoara was published the first epistolary novel on the Internet; Ion Arieşanu: Looking through books; Alex. Stefănescu: Chat on the Internet. In 2003, Ana Chelariu published in Romanian Metafora metaforei; studiu de mitologie comparată (The Metaphor of the Metaphor, a Study of Comparative Mythology), Bucharest, Cartea Românească Publishing House, book presented at the National Book Fair, Bucharest, November 2003. The book was also presented at the ‘Mihai Eminescu’ literary club, New York City, January, 2004, featured in the magazine Lumină Lină, and at the Romanian Cultural Institute, New York City, February 2004. The study was reviewed by Mircea A. Diaconu: “Un teritoriu fascinant şi recuperat, mitologia comparată [A Fascinating and Retrieved Territory, Comparative Mythology]”, [Dacia Literară, XV, nr. 55, 4/2004, Iaşi], Timotei Ursu “Tot despre Crăciun [About Christmas Again]”, in Lumea Liberă [Free World], New York City, Nr. 798, 22 January, 2004; on line http://www.romanianvip.com/2009/03/metafora-metaforei-ana-radu/ “Metafora metaforei” de Ana Radu Chelariu, 26 March 2009— Eugen Evu. The same year, 2003, she published at the Eubeea Publishing House, Timişoara, a bilingual children’s story, Romanian-English, Nea Nae mănâncă luna/Master Nick Eats the Moon. She published in Balkanistica, vol. 16, 2003 (South East European Studies Association, University of Mississippi), a book review of The 3000-Year-Old Hat: New Connections with Old Europe: the Thraco- Phrygian World, by Irina and Nicolas Florov, Vancouver, 2001. Ana continues to participate in conferences and publish articles on various topics, particularly on Romanian mythology in relation with Indo-European language and culture in the Journal of American Romanian
About the Author
xvii
Academy, the magazine Origini [Roots], and the Internet publication of the Romanian language, Conexiuni. In 2013, she published ‘The Two Pennies Pouch’; a Romanian folktale, based on ‘Punguța cu doi bani’, by Ion Creangă, Amazon.com., with illustrations by Serban Chelariu. She participates annually with communications at the conferences organized by the International Association of Comparative Mythology, such as: The Role of Metamorphosis in the Initiation Rites: the Flight of Transformation from Myth to Romanian Folktales, Strasbourg, 2011. Metaphors and the Mythical Language—Examples from Romanian Mythology, Tubingen, 2013. She published Metamorphosis amid Myths, Initiation Rites and Romanian Folk Tales in Nouvelle Mythologie Comparée/New Comparative Mythology, Lingva, nr, 3, p. 49, 2016–2017. She publishes reviews of books for The Journal of Folklore Research Reviews; Indiana University. She is a member of the American Romanian Academy, the Society of Romanian Studies, the South-East European Studies, International Association of Comparative Mythology, The Institute of Arhcaeomythology, New Jersey Library Association, and American Library Association. Ana lives with her family in New Jersey, United States of America.
Abbreviations
adj. adjective adv. adverb Alb Albanian arch archaic Arm Armenian ARom Aromanian Bret Breton, Celtic Bulg Bulgarian Bur Burushaski Corn Cornish, Celtic Czech Czech DEX.RO Dicționarul Explicativ al Limbii Române online DRom Daco-Romanian Dutch modern Dutch f. feminine Gaul Gaulish Goth Gothic Grk Greek Hit Hitite Ice Icelandic IEW Indogermanisches Etymologisches Wörterbuch Illyr Illyrian inf. infinitive xix
xx Abbreviations
Iran Iranian IstrRom Istro-Romanian JIES Journal of Indo-European Studies Lat Latin Latv Latvian Lig Ligurian Lith Lithuanian Luv Luvian m. masculine MA Mallory & Adams Maced Macedonian MDutch West Law Germanic ME Middle English MglRom Megleno-Romanian MHG Middle High German MPers Middle Persi n. noun NE modern English NHG New Hogh German NIce New Icelandic Norw Norwegian NPres New Persian NWels New Welsh OCimmr Old Cimmerian OCS Old Church Slavonic OE Old English OIce Old Icelandic OInd Old Indian OIr Old Irish OIran Old Iranian OLat Old Latin ON Old Norse OPers Old Persian OPrus Old Prussian ORus Old Russian Osc Oscan Oss Ossetic Phryg Phrygian
Abbreviations
PIE Proto-Indo-European Pk Pokorny pl. plural PN personal name Pol Polish PSl Proto Slav RN river name Rus Russian Serb Serbian sg. singular Skt Sanskrit Slovk Slovakian Slovn Slovenian Thrac Thracian Toch A B Tocharian A and B Trk Turkish Ukr Ukrainian Umb Umbrian v. verb
xxi
Part I Getae-Dacians History and Religion in Ancient Sources
1 Introduction
By the end of the first millennium BCE, the Carpathian Mountains and the surrounding hills were populated with people living out of the deep vast forests and fertile valleys, along fast-moving springs, raising their families and tending their livestock, all the while transmitting their cultural heritage from one generation to another. The little that we know about this large group comes mostly from Greek sources, beginning with Herodotus (Histories 4. 93–96), who groups these people named Getae into the large tribes of Thracians, sharing the same language, customs, and religious beliefs (ibid.), all part of the Indo-European group, In his short depiction, the Greek historian describes them as “the most manly and law-abiding of the Thracian tribes”, recounting one of the rituals that gives us a glimpse into the beliefs these people held: According to his account, every five years one man was selected from their community to serve as a messenger to god. The messenger would be held by his arms and legs and swung in the air before being launched over three spikes. If he was impaled on the spikes and succumbed to his wounds, it was interpreted that their god was showing them favor; but if he did not die that meant the man was wicked, and they had to continue until they appeased their god with a proper sacrifice, a better man. When thunder and lightning occurred, Getae would aim their arrows toward the clouds, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_1
3
4
A. R. Chelariu
threatening the sky with angry cries. Their supreme god was Zalmoxis, (also spelt Salmoxis), a Getae who, it is said, was a slave of the Greek Pythagoras, as Herodotus writes: §1.1.1: 4.94: (1) “I understand from the Greeks who live beside the Hellespont and Pontus, that this Salmoxis was a man who was once a slave in Samos, his master being Pythagoras son of Mnesarchus.” As the historian continues, after being freed and gaining great wealth, ‘Salmoxis’ returned to his country and began teaching the Pythagorean doctrine of immortality. He built a hall in which people would gather and listen to his teachings (3) “that neither he nor his guests nor any of their descendants would ever die, but that they would go to a place where they would live forever and have all good things” (ibid.). While he was teaching, he (Zalmoxis) had built an underground chamber where he vanished and lived for three years (4). The Thracians mourned him for dead, but (5) “in the fourth year he appeared to the Thracians, and thus they came to believe what ‘Salmoxis’ had told them” (ibid.). Getae believed in immortality, and after death they go to Zalmoxis, §1.1.3: 4.96. “who is called also Gebeleizis by some among them” (ibid.). The Romanian historian Vasile Pârvan in his extensive work Getica opposes the Greek historians who attributed human condition to Zalmoxis, in the end agreeing with Herodotus on the Getae’s beliefs in a god named Zalmoxis. (Pârvan 1982: p. 92) Centuries later, Strabo (Geography 7. 3. 12) called the tribes living in the western parts of the Carpathians toward German territories “daoi, dakoi”, which directs the search to the PIE [Pokorny (1959) Indogermanisches etymologiscches worterbuch (IEW) 235] *dhau- ‘to press, squeeze, strangle’ > DRom n. Daci collective ethnonym tribe name, Lat Dāci; Phryg δάος; other DRom developments dulău n.m. ‘big aggressive dog, shepherd dog’; dolcă n. f. ‘bitch’, with cognates, as per Pokorny: “Phryg δάος … ὑπο Υρυγων λύκος Hes. (therefrom the people’s name Δᾶοι, Dāci), Lyd Καν-δαύλης (κυν-άγχης ‘Indian Hemp, dogbane, plant poisonous to dogs’), compare Καν-δάων, name of Thrac god of war, Illyr PN Can-davia; dhaunos ‘wolf ’ as ‘shrike’; Lat Faunus; Grk θαῦνον θηρίον (Hesykhios); Illyr Daunus therefrom Δαύνιοι, inhabitant of Daunia; compare Thrac Δαύνιον τεῖχος); Gk. Zεὺς Θαύλιος, that is, ‘shrike’; with ablaut Grk θώς, θω(F)ός ‘jackal’, maybe Alb dac ‘cat’; Phryg. δάος; Goth af-dauiÞs ‘rended, mangled, afflicted’; OCS davljǫ, daviti ‘embroider, choke, strangle’, Russ davítь ‘pressure, press, choke, crush’, dávka ‘crush’”. The Bulgarian linguist V. Georgiev (1960, p. 48)
1 Introduction
5
finds a Dacian plant name δαχινα (Diosc. 4, 16, nv [W. 2, 183]) λυχου χαρδια ‘wolf heart’ IE *dhău-k-ino adj ‘of wolf ’ tribe’s name Δαχοι, IE. *dh ~u-ko—‘wolf ’; older name Δαοι; περσον∋σ ναμε Dăvos, Daus town Δαουσ-δαυα in Moesia Inferior, and their close relatives who occupied the eastern parts toward the Pontic region the “Getae”. Strabo states that Zamolxis (sic!) preached to his countrymen about immortality and vegetarianism, based on Pythagorean ideas (Geography 7. 3–5). Zalmoxis’ teachings on immortality are confirmed by Plato (Charmides 156d–e), who also mentions the highly respected medical teachings of the prophet. The Getae-Dacian religious beliefs in a supreme deity, the god, Zalmoxis in some documents, or Zamolxis in others, must have impressed the Greeks with his story and preaching, as it is seen in their writings. The god’s name is subject of endless controversies mainly due to the different ways in which it was transcribed in many Greek works, varying from Zalmoxis to Samolxis, to Zamolxis. The Romanian researcher, Popa- Lisseanu (2007) published translations of Roman and Greek documents pertaining to the Dacian history, including antique references to the Dacian god Zamolxis. From his work I extract the following chronological list of writers using one or the other of these forms: in Herodotus, Histories, in fifth century BCE, we find Zalmoxis and Salmoxis (see above); Porphyry (1920), Life of Pythagoras, third century BCE uses Zamolxis: “14. Pythagoras had another youthful disciple from Thrace. Zamolxis was he named because he was born wrapped in a bear’s skin, in Thracian called Zalmus.” Even though Porphyrios uses the god’s name as Zamolxis he attempts the first etymology of this name, in a Thracian form zalmus meaning ‘bear’s skin’. Other ancient writers using the form Zalmoxis are Diodorus Siculus, first century BCE, Apuleius, second century CE, and Jordanes in the sixth century CE. Among those using the form Zamolxis are: Poseidonios, first century BCE, a native to Apamea, Syria; Strabo, 64 or 63 BCE–24 CE, geographer and historian who lived in Asia Minor; Lucian of Samosata, second century CE. Starting with the third century CE, the form Zamolxis is more commonly used by authors such as Diogenes Laertios, third century AD, or the Emperor Julian the Apostate, fourth century CE. Addressing
6
A. R. Chelariu
the stem zamol M. Eliade (1972: 45) brings to attention Matthäus Prätorius who in his Orbis Gothicus (1688) relates the god’s name with the notion of zamol ‘earth’, an idea reiterated by Cless (1852) who compares it with the name of the Lithuanian chthonic god Zameluks, an etymology later discussed by the Romanian linguist Ioan I. Russu (1967: 128). Evidently, the use of the two forms of the Getae-Dacian god’s name generated two pathways of etymological explanations, one starting from the Thracian form zalmo ‘bear skin, cover, hide’, hence, Zalmoxis, etymology that could imply a relation to the German berserkers, even though there is no description of such in Herodotus or the other historians. The Zalmoxis form is considered correct by the majority of linguists accepting Herodotus as the most ancient source, and based on Thracian form zalmon and zelmis ‘hide’, a hypothesis that receives credibility on the account of a Gaete king, Zalmodegikos. But the Thracian language offers yet another form zamol ‘earth’, leading to the god’s underground sojourn, both forms having similar semantic connotations, the hiding and the underground, which may help clarify the coexistence of both forms. The appellative Gebeleizis or Beleizis, was linked to the same root, *ǵhem-el- ‘earth’, as presumed base for Zamolxis. The PIE root *ĝhem-el (*dhǵhom-) for ‘earth’ (IEW 414–416) *ĝhðem-, ĝhðom-, gen. ĝh(ð)m-és ‘earth’ has cognates in Skt kşam, Av za, zam, zme, Grk khthōn, Lat humus, Lith žēmė, Old Slvic zemlja, Alb dhe, Hit tēkan, Toch A tkam, forms that send us to the Earth Goddess, found in the Thracian mythology as Σεμελη ‘Mother of Earth’ (West, 2007, p. 175), and Semele, διωσ ξεμελω, mother of Dionysos, in Lithuanian Zemyna ‘Earth Goddess’, Greek Ξαμ-αι, Ξαμ-υνη, Demeter’s name, and possibly with the Cretan δηαί, Ionic ζηαί meaning ‘barley’. These forms lead us to relate the well- established divinity of the Earth with the Dacian male deity Zamolxis, the god who disappeared (or hid) underground for a number of years, a connection that could be structural, etymological, or conjectural, due to an association made by the Greek speakers familiar with Thracian and Phrygian: ζεμελω ‘earth’. Around the first century BCE, the Dacian-Getae tribes unified and formed a powerful state, becoming a threat to the Roman provinces in the Balkans. According to Strabo (7.3.11, 12), “Boerebistas”, king of the Getae-Dacian tribes from 82/61 BCE to 45/44 BCE, was the first ruler
1 Introduction
7
to unify their tribes under a state that stretched from the north of the Carpathians to the Haemus Mountains in the west, from the mountainside of the Hercynian Forest to the Black Sea and the Tyregetae region. After Boerebista’s death the state disintegrated into smaller states, approximately on the same geographic territory as modern-day Romania, including: Transylvania, Moldova, Maramureș, Banat, Oltenia, and Muntenia. Following his predecessor, Decebalus unified these tribes yet again under the Dacian kingdom with its capital at Sarmizegetusa Regia. Their constant raids south of the Danube River gave the Roman Empire enough reason to start military campaigns against them. After two bloody wars, by the year 106 CE the emperor Trajan conquered Dacia, and the colonization and Romanization process began (Pârvan 1972). Despite thousands of years of turbulent history in this region, the people of Romania withstood the test of times, continuing their traditions and creating a rich oral culture. Currently, the folkloric materials collected mostly in the nineteenth century are regarded by more and more researchers as hiding archaic patterns, making a case for the Romanian songs and stories to be considered as potential sources of information in searching mythical motifs. Romanian folklore, as well as the folklore of other European countries, with careful investigation, can be a valuable resource for shedding new light on the study of European and ultimately Indo-European mythology. This work is an attempt to identify mythic traits that endured from the archaic times emerging into the Romanian oral tradition.
References Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. University of Chicago Press. Georgiev, V. I. (1960). Raporturile dintre limbile dacă, tracă şi frigiană. Studii clasice II, Societatea de studii clasice RPR, p. 39. Herodotus. (1920). Histories (A. D. Godley, Trans.). Harvard University Press. http://www.perseus.tufts.edu/cgibin/ptext?doc=Perseus%3Atext%3A1999.0 1.0126;query=chapter%3D%23103;chunk=chapter;layout=;loc=1.102.1
8
A. R. Chelariu
Pârvan, V. ([1937] 1972). Dacia. Civilizaţiile străvechi din regiunile carpato- danubiene (5th ed.). Editura Stiințifică. Platon. (1975). Opere. Ediția a II-a; ediție îngrijită de Petre Creția și Constantic Noica. Editura științifică și enciclopedică. Pokorny, J. (1959). Indogermanisches etymologiscches worterbuch. A. Francke Verlag. Popa-Lisseanu, G. (2007). Dacia în autorii clasici; I Autorii latini clasici și postclasici; II Autorii greci și bizantini. Vestala. Porphyry. (1920). Life of Pythagoras (English translation, K. S. Guthrie, Trans.). https://www.tertullian.org/fathers/porphyry_life_of_pythagoras_02_ text.htm Russu, I. I. (1967). Limba traco-dacilor. Bucharest Ştiinti̧ficǎ și enciclopedică. Strabo. The Geography of Strabo. Literally translated, with notes, in three volumes, ed. H. C. Hamilton, Esq., W. Falconer, M.A (London 1903). http:// www.perseus.tufts.edu/hopper/text?doc=Perseus%3atext%3a1999.01.0239 West, M. L. (2007). Indo-European Poetry and Myth. Oxford University Press.
2 Brief Overview of Recent Theories on the Indo-Europeans’ Homeland
The Indo-Europeans’ ‘homeland’ and the moment of their dispersal throughout Europe is a much-debated subject. In recent decades, the most prominent researchers on this topic modified their view on the arrival of the Indo-Europeans to Europe from a violent invasion to a peaceful intrusion. According to the invasion theory, postulated by Marija Gimbutas (1982, p. 238), the Indo-Europeans disrupted the lower Danubian long-standing cultures, Hamangia, Gumelnița Karanovo, Vinča, Tisa, and on the northeast the Cucuteni-Tripolye cultures, for which she established the term ‘Old Europe’. The populations flourishing for millennia in these regions prior to the arrival of the Indo-Europeans, practiced on a limited scale agriculture and herding, advanced ceramic techniques, and worshipped the Great Goddess. The Indo-European invaders appeared as a warrior-like nomadic population, essentially patriarchal, structured in social classes, who worshipped the sky and the sun. The central position occupied by the feminine deity in the ‘so-called’ Old Europe culture led to mingling between the worshippers of the goddess and the newcomers’ religious beliefs, that could explain the confusion often associated with the feminine deities’ functions in the Greek pantheon. For example, the Greek Demeter, as her name stands, is the © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_2
9
10
A. R. Chelariu
‘Mother Earth’, a representation of the Old Europe goddesses displaying a distinct relation with the Olympians. The violent invasion theory was refuted by Colin Renfrew (1987) who advanced the hypothesis that the Indo-Europeans came around 6500 BCE from Anatolia through Greece, as the first farmers that colonized the middle and lower Danube valleys. According to the author, their arrival was not a warlike intrusion, but took place rather by demic diffusion, a gradual penetration by which the ‘conquering’ process was realized through assimilation. James Mallory argues that in the millennia following 4500–4000 BCE, there was a sizeable influx of people from the Pontic region into Europe, laying “the foundation for the Indo-European languages throughout Southeastern Europe” (Mallory, 1989/1996, p. 234). Archaeological evidence of the newcomers in the Romanian area includes the presence of objects typical for the Pontic Caspian region such as scepters shaped as horseheads found in Casimcea, Sălcuța, Fedeleseni, and other sites (Mallory, ibid.). The place of the Indo-European dispersal is the subject of yet another important study offered by David Anthony (2007). His extensive archaeological research concludes that the homeland of the Proto-Indo- Europeans was north of the Black Sea in the Pontic-Caspian steppes, reinforcing the previous theories of Gimbutas and Mallory. Analyzing the archaeological data in conjunction with the linguistic reconstructions, Anthony reaches the conclusion that the penetration of the Proto-Indo- European culture and the fall of the Southeast European settlements, resulted most likely from a combination of several factors: cultural and economic exchanges, warlike incursions on horseback, and climate conditions resulting in agricultural failure (Anthony, 2007, p. 230). Between 6500 and 5500 BCE, farmers and herders of a non-Indo-European- speaking group arrived in Europe from Anatolia through Greece and Macedonia, migrating up north by the Danube Valley. Several new Old European Neolithic languages may have emerged from a Thessalian parent, a non-Indo-European language that curiously might have preserved in the Proto-Indo-European the term for bull, *tawro-s ‘bull’, considered by many linguists borrowed from an Afro-Asiatic super-family that generated both Egyptian and Semitic in the Near East (Anthony, 2007,
2 Brief Overview of Recent Theories on the Indo-Europeans’…
11
p. 147). According to him, the farmers from Anatolia did not spread into Europe the Proto-Indo-European language: PIE must be dated not to the Mesolithic but after the beginning of the Neolithic era, so after 6500 BCE in Europe, because many Indo-European (IE) cognate word roots securely assigned to PIE (Mallory & Adams, 2006: 139–172) had meanings related to Neolithic economies (cow, bull, calf, ewe, ram, lamb, wool, milk products, ard/plow, domesticated grain). The speakers of the most archaic recoverable form of PIE, preserved in the Anatolian IE languages, were already familiar with agriculture and domesticated animals—and the pre-Neolithic WHG were not. (Anthony & Brown, 2017)
At about 5800–5700 BCE the farmers of the Criș culture, developed along the Criș river on the Carpathian Mountains by farmers from Anatolia, migrated to the east of the Carpathian Mountains into the Bug- Dnister areas, coming in contact with its neighbors, the Pontic-Caspian population of foragers and hunters. They brought with them domesticated sheep and cattle, probably the source of the first domesticated cattle in the North Pontic region,. They cultivated barley, millet, peas, a variety of wheat (emmer, einkorn, spelt, barley, millet), peas that were not native to southeastern Europe, and plum orchards, etc., and yet the European cultures lacked the vocabulary for ‘seed, sowing, fields, crops’ (Anthony, 2007, p. 138–143). A note should be made here on the difficulty linguists have to reconstruct a PIE root for ‘pea’ (Mallory & Adams, 1997, 2006, p. 416) in spite of its confirmed presence on the continent for thousands of years. Relevant here is the unexplained Albanian modhullë and Romanian mazăre, both meaning ‘pea’, to which we have to add the Dacian form found in Dioscorides’ list of plants mozula, mizela ‘the plant thyme’, showing phonetic similarity, but semantic difference. Around 5800–5500 BCE, the archaeological data show that the hunters and gatherers on the Bug-Dniester valley selectively and on a limited scale adopted the Criş farmers-herders influence, consisting of small plots of grain, cattle, and pigs. From 5200 BCE, the lower Danube valley settlements and the Criş culture began to enjoy centuries of peace and prosperity, the farming hamlets turned into large settlements and larger
12
A. R. Chelariu
towns, with bigger, sturdier houses, some two-storied, painted in specific patterns. Remarkably, at some sites, villages continued on the same spot for many generations, showing signs of a very sedentary type of population. The ceramic decorations follow similar patterns, and the numerous female figurines confirm that the main religious beliefs centered on the Great Goddess. Beginning with 5200–5000 BCE, the Pontic-Caspian steppe societies became attracted to the Old European copper trade and the beautifully decorated ceramics of the Cucuteni-Tripolye cultures from the northeast Carpathian Mountains. After a thousand years of prosperous existence, by 4200 BCE the Old Europe cultures reached their peak. Unfortunately, the lower Danube valley civilization was hit by a period of terrible winters (Anthony, 2007, p. 227), which resulted in burned and abandoned settlements. From 4200 to 3800 BCE the shift in climate, with a period of 140–150 years of terrible winter colds, led to the end of farming activities, and of Old Europe cultures. This hardship was exacerbated further by the incursions of the steppe immigrants, a mobile force on horseback looking for Balkan copper and perhaps forceful accumulation of herds. Some of the inhabitants from the lower Danube valley retreated northwest into the Transylvanian territory, where they settled and developed other cultural complexes. Following centuries of very harsh winters and warlike interventions, by 3700 BCE the milder climate returned to the lower Danube valley, encouraging the further penetration of steppe settlers. The agricultural type of economy was replaced by the adoption of an Indo-European pastoral type of economy. Yet, the traditions of Old Europe continued for a longer period of time in the western part of Romania, in Transylvania and western Bulgaria, while the northeast Cucuteni-Tripolye cultures maintained economic and presumably social relations with their steppe neighbors. Beginning sometime before 4000 BCE, the steppe population adopted more of the farmers and herders way of life and economy, creating the conditions for the start of what is accepted as Proto-Indo- European proper: “The heart of the Proto-Indo-European period probably fell between 4000 and 3000 BCE, with an early phase that might go back to 4500 BCE and a late phase that ended by 2500 BCE.” (Anthony, 2007, p. 99) In accord with Gimbutas and Mallory, Anthony (2007, p. 58) concludes that linguistic reconstruction and archaeological
2 Brief Overview of Recent Theories on the Indo-Europeans’…
13
discoveries lead to the probability that Proto-Indo-European (PIE) was the language spoken in the Pontic-Caspian steppes between 4500 and 2500 BCE: By 2500 BCE the language that has been reconstructed as Proto-Indo- European had evolved into some thing else or, more accurately, into a variety of things, late dialects such as Pre-Greek and Pre-Indo-Iranian that continued to diverge in different ways in different places. The Indo- European languages that evolved after 2500 BCE did not develop from Proto-Indo-European but from a set of intermediate Indo-European languages that preserved and passed along aspects of the mother tongue.
According to this model, the PIE language was dead by 2500 BCE. Moreover, Anthony presents a scenario in which the diffusion of the Proto-Indo-European language that resulted into the formation of the Indo-European dialects, occurred in the following sequence (Anthony, 2007, p. 305): the dispersal of Yamnaya horizon began with Afanasievo culture spreading toward the east around 3700–3500 BCE with Pre- Tocharian, followed at about 3300 BCE by the diffusion across the Pontic-Caspian region. The large migration from Western Yamnaya to Danube valley and Carpathian basin took place from 3100 to 3000 BCE during the Early Bronze Age, presumably with the Pre-Italic and Pre- Celtic dialects. Around 2800–2600 BCE, the Yamnaya horizon penetrated the Corded Ware cultures northwest of the Pontic region, giving leeway to Germanic and Balto-Slavic dialects. The last diffusion took place around 2200–2000 BCE when the late Yamnaya-Poltavka cultures went east, separating into what are known as the Indo-Iranian speakers. The core of economic exchanges between the steppe tribes and the Anatolian settlers in the southeast and eastern part of Romania was material culture in nature, including: pottery, copperware, cattle, and textiles. Anthony argues that the Indo-European infiltration in Europe was successful, aside from violence, due to the employment of a specific system of contractual patron–client relations and guest–host agreements enforced by lavish public rituals that impressed their neighbors and led to building of social and economic bonds. The reconstructed Proto-Indo-European vocabulary hints at the importance of the oath-bound rituals for their
14
A. R. Chelariu
relations with gods, other members of the tribe, and even with people from different ethnic groups, or other outsiders. Such rituals reflect archaic European traditional cultures, and are recognized in a well- documented Romanian custom called judecata la hotare ‘the judgment at the border’ (Vulcǎnescu, 1972, p. 67), which took place under an old sacred oak, known in the Indo-European tradition as a sacred tree. F. Kortlandt in his recent work (Kortlandt, 2018) proposes several points that highlight the importance of the areas populated by the Romanians’ ancestors: When considering the way the Indo-Europeans took to the west, it is important to realize that mountains, forests and marshlands were prohibitive impediments. Moreover, people need fresh water, all the more so when traveling with horses. The natural way from the Russian steppe to the west is therefore along the northern bank of the river Danube. This leads to the hypothesis that the western Indo-Europeans represent successive waves of migration along the Danube and its tributaries. The Celts evidently followed the Danube all the way into southern Germany. The ancestors of the Italic tribes, including the Veneti, may have followed river Sava towards northern Italy. The ancestors of Germanic speakers apparently moved into Moravia and Bohemia and followed the Elbe into Saxony.
The same route was followed, as Kortlandt argues, by “the ancestors of speakers of Balkan languages,” who crossed the lower Danube toward the south. This scenario is in agreement with the generally accepted view of “the earliest relations between these branches of Indo-European (cf. Holzer 1989: 165 on Temematic)” (ibid., p. 327). These researchers’ important contribution regarding the early contacts of the lower Danube valley and east Carpathian civilizations with the Indo-European migrants, whether associated with violence or by commercial contracts, shines a bright light on the Romanian culture.
2 Brief Overview of Recent Theories on the Indo-Europeans’…
15
References Anthony, D. W. (2007). The Horse the Wheel and the Language. Oxford and Princeton: Princeton University Press. Anthony, D. W., & Brown, D. R. (2017). Molecular Archaeology and Indo- European Linguistics: Impressions from New Data. https://www.academia. edu/32927784/Molecular_archaeology_and_Indo_European_linguistics_ Impressions_from_new_data Dioscorides Pedanius, of Anazarbos. (1934). Greek Herbal of Dioscorides (J. Goodyer & R. T. Gunther, Eds.). (pdf ). Oxford University Press. Gimbutas, M. (1982). The Goddesses and Gods of Old Europe: Myths and Cult Images. University of California Press. Kortlandt, F. (2018). The Expansion of the Indo-European Languages # 322. www. academia.edu Mallory, J. P. (1989/1996). In Search of the Indo-Europeans, Language, Archaeology and Myth. London: Thame and Hudson Ltd. Mallory, J. P., & Adams, D. Q. (Eds.). (1997). Encyclopedia of Indo-European Culture. London and Chicago. Fitzroy Dearborn. Mallory, J. P., & Adams, D. Q. (2006). The Oxford Introduction to Proto-Indo- European and the Proto-Indo-European World. Oxford: Oxford University Press. Renfrew, C. (1987). Archaeology and Language: The Puzzle of Indo-European Origins. Cambridge: Cambridge University Press. Vulcǎnescu, R. (1972). Coloana Ceriului. Bucharest: Editura Academiei Române.
3 Cucuteni-Tripolye: The Indo-European Homeland?
In the article Indo-European Expansion Cycles Axel Kristinsson (2012) offers yet another hypothesis on the intensely disputed problem of the Indo-European (IE) homeland, suggesting that the most remarkable prehistoric culture, the Cucuteni-Tripolye (CT) culture, may very well be an Indo-European creation: …the evidence for material culture seems to fit best with a classic sedentary farming culture like the CT culture, rather than a semi-nomadic culture we would expect on the steppe although the evidence cannot be claimed to be conclusive. (Kristinsson, 2012, p. 389)
The CT culture appears by the middle of sixth millennium BCE at the foothills of the East Carpathian Mountains in Romania spreading eastward through Bessarabia toward the Dnieper and northward reaching the area near Kiev. This culture presupposes a large population benefiting from favorable conditions for developing primitive agriculture, which flourished in apparent peaceful conditions for two millennia. Kristinsson’s argument for locating the IE homeland in the CT region is based exactly on the fact that Cucuteni represents the ideal geographic position for IE
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_3
17
18
A. R. Chelariu
cultural development: It is situated at the junction between the forest and the steppe areas, thus satisfying the IE main linguistic requirements. In Kristinsson’s opinion, the large spread of the Indo-Europeans should be regarded not as a problem but rather as evidence of a migration phenomenon that started probably at the beginning of the fourth millennium BCE (Kristinsson, 2012, p. 388). As a macro-historian, the author had studied the cyclic expansions of many societies, reaching conclusions that he considers applicable to the IE expansion. Thus, careful examination of massive movement of populations from the history of Europe could offer a better solution for solving the IE spread. He gives examples of the Celtic, Germanic, Slavic, and Vikings migrations throughout the history of Europe, not to mention the populating of the Americas. The author argues that large movements function on a common mechanism that works inside these cycles. According to Kristinsson, there are two varieties of expansions capable of working separately or together: The simplest model of expansion is by colonizing, occupying new lands for agriculture needed to sustain the growth of families. In this kind of expansion by colonization, the population density is essential; even if the indigenous population from the occupied territories is practicing agriculture the large number of newcomers becomes dominant, thus generating a process of ‘democratization’, with the elites of the occupied minority disappearing in time (Kristinsson, 2012, p. 378). The second model of expansion called an ‘expansion system’ is based on competition: The cultures in which this type of expansion could be observed are more conformist and belong to the same linguistic group. Within this system, culture–society competition springs between divided political groups and constant in-fighting for political and military power. Gradually, the military elites need more soldiers and start recruiting from among the farmers, who request new lands as compensation. Over time, the intense use of existing lands and population growth leads to the expansion into new territories through military force. Kristinsson claims that such processes could ostensibly explain the IE expansion problem. Moreover, Kristinsson (2012, p. 417) argues (against the Kurgan hypothesis) that the elite invasion and domination does not imply language replacement as history demonstrates, with one example from recent European history being the Norman elite in medieval England.
3 Cucuteni-Tripolye: The Indo-European Homeland?
19
Another argument that Kristinsson brings against the elite groups’ infusion is that the warriors on horse carts are recorded around 2000 BCE, and the IE dispersal begins around the fourth millennium. Expansions of such large scale can generate language changes particularly in the agricultural societies lacking a powerful organized state. Thus, Kristinsson argues that the beginning of the IE expansion was the result of a large growth of populations on the west side of the CT culture region, which created the need for new land, determining the gradual spread toward the eastern steppe. Additional consequence of the growth in population density was the intensifying of the internecine fights among tribes shown in the increased number of arrowheads discovered around the CT sites, which could be the result of competition among groups within the society. Another aspect brought into the discussion by the author is that around the fourth millennium some important agricultural technology advancements were made, including the plough and the wheel, the use of oxen for traction vehicles, the presence of a new breed of sheep richer in wool, together with an increase in farmers’ production and use of milk and dairy in Europe and West Asia. All these innovations had a consequence in terms of population growth, causing pressure within the CT society that led to successive invasions eastwards beginning with 3500 BCE, culminating in the colonization of the steppe lands. As a result, the CT culture started to break up, and Corded Ware cultures emerged in northern Europe, characterized by a semi-nomadic lifestyle. The Yamna culture emerged in the East adapting to the new conditions; but both cultures retained identifiable characteristics of the CT culture from which they descended. Kristinsson’s argument is that in the process of colonization of the steppe regions the newcomers changed their way of life, adapting to the new environmental conditions. The big differences in the quality of the CT ceramic and the ceramics of Corded Ware or Yamna cultures are due, as the author believes, to the fact that the production of ceramic was in general a women’s occupation, while population movement was preponderantly male-oriented: Men supposedly formed new families with local women that were not acquainted with the style and quality of the CT ceramic culture.
20
A. R. Chelariu
Beginning circa 3100 BCE, elements of Yamna culture, including probably speakers of IE western dialects, for example, ancient Greek and Phrygian, could be observed moving along the Danube toward the Balkans, inducing the Pre-Anatolian local population movements across the Bosporus into Anatolia. Populations speaking Indo-European western dialects, pre-Greek and Phrygian from the Carpathian region may have had contact with speakers of north IE dialects, Italic or Celtic, forming a Schprachbund of linguistic influences. By 2800 BCE the CT culture separated into two branches that gradually merged or disappeared into other cultures. Interestingly, while the CT culture is known through its settlements, the following cultures are visible by their graves. The dispersal of IE language over the European continent is a unique phenomenon. Kristinsson’s hypothesis is based on the idea that for a language to become dominant a massive migration of democratized non- elitist populations of farmers using a common language for communication was necessary: “Under such circumstances, an indigenous language can be displaced even if the population didn’t change much as happened in parts of the Roman Empire, in Ireland under British rule or in large parts of Latin America under Spanish rule” (Kristinsson, 2012, p. 373). These two hypotheses, the colonization and the expansion system working in conjunction, could be applied to Dacian history: a country conquered by the Roman centralized state in expansion, where the local rural population did not suffer major changes as a consequence of the war, but received an infusion of colonists during and after the Roman military administration left. The prolonged intensive assault on the Dacian territories, first by Emperor Domitian between 84 and 89 BCE, followed by even a greater force organized by Trajan in two waves (101–102 and 105–106 BCE), that resulted in a total occupaton of the country, and the influx of a large number of colonists, attracted by the richness of the newly conquered territory, famous for its gold and salt mines. The Roman citizens and army veteran colonizers in search of riches were speaking a lingua franca, known as Latina Vulgata, spoken all over the territories of the empire already Romanized. The economic and social relations established primarily through transhumance between the population from conquered Dacia and the people speaking the same Romance language from the south and west of the
3 Cucuteni-Tripolye: The Indo-European Homeland?
21
Danube, favored the formation of a Schprachbund of a common vocabulary specialized for the economic needs. Transhumance, as it survived for centuries on Romanian lands, did not involve the movement of entire families, but only bands of shepherds caring for large flocks of sheep. The rest of the family stayed in their villages, leaving only when danger struck. Getting back to the Kristinsson’s hypotheses, the Bulgaro-Slavic invasion consisted of a nonelitist group of people, driven mainly by economic not military reasons, which resulted in language imposition in the areas with a weak organizational system, an area stretching from Poland to Macedonia. It did not prevail, however, in regions with a more stable social structure such as Dacia Romana. The author’s hypothesis seems to offer a solution for the IE problem; the possibility of Cucuteni-Tripolye culture as the homeland from where the IE diffusion began: the first division to the west side—Cucuteni to west and north of Europe, and Tripolye—toward the east, “similar to the classic but depreciated centum-satem split” (Kristinsson, 2012, p. 423) but probably predating by a millennium the actual formation of the two branches. Most importantly, this hypothesis brings a new light to the Cucuteni-Tripolye culture, hopefully initiating a more in-depth research of this relatively less studied culture. One note should be added here: in disagreement to Kristinsson’s opinion, that the comparative mythology should not be regarded as a valid source in the IE discussions, one should consider the several mythic patterns observable in most of the IE cultures, such as the dragon fighter, creation and foundation myths—primeval man, or the theft of fire. Certainly, a great deal of caution should be applied, and if linguists could agree on a common IE religious vocabulary then by default we could accept the existence of a common religious ideology. The existence of a number of mythical patterns in most of the IE cultures could very well be considered as expression of a common mythology, detectable by the researcher through the process of comparison and elimination. Some of the main characteristics of the Cucuteni ceramics, besides the beautiful graphic designs, are the presumptive symbols that are represented on the statuettes of the feminine character, the Great Goddess or ancestress of the group as the reproductive organs represented by triangles, ovals, circles; figurative animal suggesting bovine horns, deer, pig,
22
A. R. Chelariu
dog, snake, ram; or, the vegetation, trees, leaves, or the water as the wavy lines, all designs so well detailed and analyzed by Marija Gimbutas (1982, p. 169) in her monumental work on this subject. Even though the symbolic designs offer information of the CT peoples’ beliefs, all the efforts that the Romanian archaeologists made to decipher in more detail messages regarding the socioeconomic structures among the CT populations, encountered serious difficulties. The absence of cemeteries among the thousands of villages raises many questions of what they were doing with their dead. There are a few cases revealing human bones deposited under house floors, but these are inconclusive. The themes on the Cucuteni ceramics that may be considered mythical expressions do not seem explicitly related to the IE culture, the most identifiable being the two pair figures, believed to represent either a goddess with her consort, or a mother and daughter pair, which plausibly could be linked to the Demeter-Persephone Greek myth, a relation that may not be representative of IE mythology. The importance of fire in the Cucuteni-Tripolye culture, is exemplified particularly by the practice of complete burning of settlements every 75–80 years, together with the ritual breaking of the ceramics and cremation. Such practices could be considered sufficient grounds to speculate on a hypothetical relation between these practices and the cult of light and fire in the IE culture. On the linguistic level, the DRom language retained a few essential IE concepts related to fire; DRom n. scrum (with prothetic s- crum, frequent in language) ‘singe, cinder, ash’ < PIE (IEW 571 *k(e)r-em-) *kr-em‘burn’, with cognates in Latin cremō, −üre ‘burn’, Umbr. Krematra, pl *crematra ‘kind of vessel for frying meat’ (Orel, 1998: Albanian shkrumb, pl shkrumba ‘black ashes’; PAlb iš-krum). Another example is DRom n. jar ‘glowing coal, ember’, Dacian Γέρμαζα (Germaza), Γερμἰζερα (Germizera) < *gwhermós PIE MA 1. *g(e)ulo- ‘fire, glowing coal’; 2. *gwher- ‘warm’, *gh’ermós adj. ‘hot’, with cognates in OIr gūal ‘coal’; NE coal; Alb zjarr|i ‘fire’, Alb zjarm, Arm Jě rm Jě r; Latv. gařme, Gr. θερμὀς, ´fire, heat´; OCS goreti ‘burn’; or, DRom n. văpaie ‘flame, shooting flame’ < PIE MA *u̯ep- ‘throw, throw out’, IEW 1149–1150 *uēp-: uǝp- (*suekʷ-) ‘to blow; to soar, with cognates in Skt vàpati ‘throw out, scatter’, vàpra ‘earthen wall’, vaprā ‘fireplace’; Av vafra, OIran vafr,
3 Cucuteni-Tripolye: The Indo-European Homeland?
23
Iran bafr ‘snowdrift?’, OInd causative vāpayati ‘makes blow’; Lat vapor (old vapōs) ‘vapor, heat’; Alb vapë ‘summer heat’; DRom n. dogoare ‘blaze, heat, summer heat’, v. dogori ‘to emanate heat, blaze’ < IEW 240–241 *dhegʷh- ‘to burn, *day *dhōgwho- ‘summer heat’, (*dhogʷhlo-lā) *dhegwh- ‘burn, be hot,’ with cognates in OIr doghaim ‘burn’, daig ‘flame’; Lat foveō ‘heat, cherish’; Lith degù ‘burn’; dāgas ‘summer heat’; OCS žegǫ ‘burn’; Alb djeg ‘burn’, ndez ‘kindle’; Grk tephra ‘ash’; Av dažati ‘burns’; Skt dagdhá-ḥ (= Lith. dègtas), Kaus. dăhayati; dăha-ḥ ‘blaze, heat’, niduăgha-ḥ ‘heat, summer’, Pers. dăɣ ‘burn brand’; Av. daxša- m. ‘blaze” dáhati ‘burns’; Toch tsäk- ‘burn’ Yet, if the Cucuteni-Tripolye culture was the result of the IE group, it will be a challenge to prove it, and, for that matter, an investigation of the Romanian heritage could help resolve this impasse.
References Gimbutas, M. (1982). The Goddesses and Gods of Old Europe; Myths and Cult Images. Berkeley Los Angeles: University of California Press. Kristinsson, A. (2012). Indo-European Expansion Cycles. Journal of Indo-European Studies, 40(3–4), Fall/Winter. Institute for the Studies of Man. Orel, V. (1998). Albanian Etymological Dictionary. Leiden. Boston Kohl: Brill.
4 Some Recent Genetic and Archaeological Conclusions
The advancement of genetic research opened up new perspectives into the understanding of population migrations in ancient Europe. Studies on this subject performed by renowned scientists such as David Reich (2018), Nick Patterson, and Iosif Lazaridis from the Harvard laboratory reached surprising conclusions regarding the population migrations in the large Eurasian region. The people inhabiting this area were formed of a large mixture of people genetically related to the ancestors of ancient European hunter-gatherers, which is, in some measure, carried by the present-day Europeans. The European ancestral hunter-gatherers were preceded by migrations of farmers from the Near East circa 9000 years ago, supporting the earlier theory proposed by Colin Renfrew (1987). One millennium later, from 6000 bce, the Carpathians, the Balkans, and Greece were the most culturally advanced European societies. The development of these societies was associated with a new large migration wave, probably between 4200 bce and 3900 bce, formed mainly of pastoralists who probably spoke the Indo-European languages. This second wave came from an area stretching between the north of the Black Sea and the Caspian Sea, which archaeologists called the Yamnaya culture, or pit culture, of cattle herders equipped with horse-drawn wagons that crossed © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_4
25
26
A. R. Chelariu
the Dniester which apparently annihilated the cultures in Eastern Romania and Bulgaria. The DNA analysis conducted by Iosif Lazaridis from Reich laboratory showed that the Yamnaya population comprised a combination of lineages that did not exist previously in central Europe. Traversed and populated by many migrations, settlements, and resettlements, the European continent is veiled in mysteries that are yet to be uncovered, among which include the Yamnaya culture from Eurasia. Reich studies prove that Yamnaya invaders “were the missing ingredient, carrying exactly the type of ancestry that needed to be added to early European farmers and hunter-gatherers to produce populations with the mixture of ancestries observed in Europe today” (Reich, 2018, p. 108), supporting the previous theories of Marija Gimbutas (1982, p. 238). The invaders, strongly comprised of the patriarchal Indo-European elite, were almost exclusively R1b on the paternal side. However, they were absorbed in high proportion by a non-Indo-European maternal to lineage; they were using horses, oxen, and transportation on wheels, and produced rudimentary ceramics. Reich finds through his studies of ancient DNA that the Yamnaya culture was a male-dominant society. They buried their dead under mounds called kurgans, mostly displaying male skeletons placed at the center, with metal daggers and axes, many showing evidence of violent injuries. Even though the Cucuteni-Tripolye DNA results are missing, the decline of ‘Old Europe’, the relatively peaceful civilization analyzed by Marija Gimbutas, coincided with the kurgan invasion. David Reich, Nick Patterson, and Iosif Lazaridis, observed that data from the Corded Ware burials revealed that a large part of their ancestry related to the Yamnaya and concluded that “the makers of the Corded Ware culture were, at least in a genetic sense, a westward extension of the Yamnaya” (ibid., p. 110). The relations between Yamnaya and Corded Ware complexes, a long-standing debate among researchers, can be considered settled now with the genetic data results. The two complexes had many striking parallels, such as the construction of large burial mounds, the intensive exploitation of horses and herding, combined with a strikingly male-centered culture celebrating combat and violence, as reflected by the large maces (or hammer-axes) buried in some graves. Even though other researchers have stated that the Yamnaya culture did not expand to the north of the Tisza River, Reich agrees that “only later, during the
4 Some Recent Genetic and Archaeological Conclusions
27
contemporaneous Corded Ware and Yamna migration waves were direct contacts possibly between Yamna and Corded Ware herders on the upper Dniester region (Anthony, 2007; Gimbutas, 1977)” (ibid., p. 110). Furthermore, Reich reveals that aDNA analysis has shown traces of a plague disease linked to the Eurasian steppe that could have determined the Corded Ware culture expansion over vast areas in east-central Europe. Such unprecedentedly empirically grounded revelations demonstrate the importance of the advanced genetic technology for understanding human history. The thousands of years of the so-called Old Europe’s peaceful civilization was disrupted, as Gimbutas (1982, p. 238) eloquently demonstrates, by a male-centered society of elite dominance and warfare, as evidenced now in Reich’s book with the spread of Y chromosomes of the Yamnaya or their close relatives into Europe and India. The male-dominated Indo- European culture, revealed in the Greek, Norse, and Hindu mythologies, indicative of the Indo-European class structure society, centered on a male figure, god or hero, naturally arising as a reflex of the social behavior of conquering men with social power overwhelming women from the conquered group. The genetic admixture that resulted from these migrations still characterizes modern European populations, just as it is very likely that predecessors of one or several Indo-European languages spoken in Europe today, were carried by these migrants. (Kristiansen et al., 2020) From these new studies the conclusions point to the populating by IE speakers of a large area spreading from north-central Europe through the Pontic steppe area to the Caspian Sea. The invasion of the Indo-Europeans into southeastern Europe passing through the territories of present-day Romania to the Balkans, Greece, and Anatolia, and into the western part of Europe, changed the local civilizations, their social order and language. These facts must be taken into account, in spite of all the difficulties in proving a dialectal continuum, akin to that of the Daco-Romanian language. Even though it may be hard to accept that a Proto-Indo-European reconstructed root can be detected in certain lexeme without written ancient documentation, the fact that neighboring languages, including Albanian, Greek, Latin, Celtic, and Balto-Slavic, exhibit the same base correspondences, the same correspondences must have been sensed to an
28
A. R. Chelariu
even greater extent by the speakers during economic and social contacts, transhumance, ceramic and metal commerce, especially if we consider that these contacts were at a time when the Indo-European languages were less differentiated. As Kortlandt (2018, p. 228) notes, “the linguistic evidence takes precedence over archaeological and genetic data, which give no information about the languages spoken and can only support the linguistic evidence”. Language studies together with comparative analysis of the mythological traits may help us form a better picture of the archaic societies of Southeast Europe.
References Anthony, D. W. (2007). The Horse the Wheel and the Language. Princeton: Princeton University Press. Gimbutas, M. (1977). The First Wave of Eurasian Steppe Pastoralists into Copper Age Europe. Washington DC: JIES, 5: 277–338. Institute for the Study of Man. Gimbutas, M. (1982). The Goddesses and Gods of Old Europe; Myths and Cult Images. Berkeley Los Angeles: University of California Press. Kortlandt, F. (2018). The Expansion of the Indo-European Languages. # 322. www.academia.edu Kristiansen, K., et al. (2020). Kinship and Social Organization in Copper Age Europe. Across-Disciplinary Analysis of Archaeology, DNA, Isotopes, and Anthropology from Two Bell Beaker Cemeteries. Sjo¨gren K.-G., Olalde, I., Carver, S., Allentoft, M. E., Knowles, T., Kroonen, G., Kristiansen, K., et al. PLoS ONE 15(11): e0241278. https://doi.org/10.1371/journal. pone.0241278 Reich, D. (2018). Who We Are and How We Got Here: Ancient DNA and the New Science of the Human Past. New York: Knopf Doubleday Publishing Group. Renfrew, C. (1987). Archaeology and Language: The Puzzle of Indo-European Origins. Cambridge: Cambridge University Press.
5 Daco-Romanian Language: An Indo-European Branch
At this juncture, I should add a few lines addressing the history of the language spoken by the population of today’s Romania, known in antiquity as Dacian, which was classified by linguists as an Indo-European language. Some researchers consider Dacian an independent language, some a close relative of the Thracian language, some closer to Illyrian. There are also those who claim that the Pre-Indo-European language/ languages, spoken in this area before the Indo-European arrival, should be considered when analyzing the modern language, even though such attempt would address a society dissipated into the deep darkness of times. The first information on the Getae-Dacian and Thracian relation is found in the Histories of the Greek writer Herodotus (4. 94) who states that “the Getae are the bravest and the most law-abiding of all Thracians”. The geographer Strabo places the Getae along the Danube on the mountain side of Hercynian Forest: As for the southern part of Germany beyond the Albis, the portion which is just contiguous to that river is occupied by the Suevi; then immediately adjoining this is the land of the Getae, which, though narrow at first, stretching as it does along the Ister on its southern side and on the opposite
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_5
29
30
A. R. Chelariu
side along the mountain-side of the Hercynian Forest (for the land of the Getae also embraces a part of the mountains), afterwards broadens out towards the north as far as the Tyregetae; but I cannot tell the precise boundaries. (7.3.10; 7.3.12)
He also writes that the Dacians, Getae, and Thracians spoke the same language, information reaffirmed in documents of Cassius Dio and Pliny the Elder. The Romanian linguist I. I. Russu (1967, p. 33) proceeded in his linguistic research on the assumption that the Getae-Dacian language was closely related to Thracian. This close relation between Dacian and Thracian languages came under scrutiny from the Bulgarian linguist Vladimir I. Georgiev (1960, p. 45) whose objections were made on the ground of the Dacian name of 47 towns ending in δαυα of which 28 are on the Romanian territory, and the Thracian form παρα of which 37 (41) are found only in Thrace; these arguments show that Dacian and Thracian may have been, at least in some respects, two different languages, part of the large group of common Indo-European. Unfortunately, no Dacian language inscriptions have been discovered, and the limited direct testimony of written documents from these languages are mostly toponyms, hydronyms, and proper names, coming to us through Greek and Latin alphabets, which makes it quite difficult to reach definite conclusions. These languages, whether dialectal formations or independent, were certainly part of an Indo-European linguistic group, situated in close proximity from southern Pannonia region, to the Carpathian and Balkan Mountains to the Black Sea. As a result of the geographic proximity, the economic exchanges and pastoralism, the Dacian-Getae together with Mysian, Thracian, Dardanian, Illyrian, Macedonian, and Greeks were a large group in close contact, forming alliances or fighting among themselves, sharing specialized vocabulary for commercial exchanges, in the end, becoming the main substrata that underline the Balkan peoples of today. As Mallory and Adams state: “Some uniform proto-language may have been spoken over a geographically compact area at the same time when their neighbors had already differentiated into different language groups…” (Mallory & Adams, 2006, p. 445).
5 Daco-Romanian Language: An Indo-European Branch
31
Following the Greek influence in the region, the Roman conquest of the Balkan Peninsula resulted in a gradual Romanization of the populations. The economic relations between North and South of the Danube must have been influenced by this linguistic development. When the Dacian territory succumbed to the Roman army led by the Emperor Trajan between 101 AD and 106 AD, the official Romanization process began, facilitated by the geographic proximity and the permanent contact with the already Romanized Southern population. The lack of information regarding a state form of administration is compensated by intense pastoral transhumance and economic activities the formerly Dacian population practiced in the new conditions, developing a homogenized language. The economic relations between North and South of the Danube, especially the pastoral movements, continued uninterrupted for centuries, forming a key component in the process of language development. Even today we find among the Greek and Slavic speakers South of the Danube the word vlachos with the meaning ‘shepherd’ affirming the influence of the Vlach shepherds that lived in the Balkans, part of the large group of Latin speakers akin to the Romanians North of the Danube. The language spoken by the habitants of this large area formed the four Daco-Romance dialects: Aromanian, spoken in some areas of Albania and Macedonia, Megleno-Romanian spoken in areas from Macedonia and Greece, Istro-Romanian located on the peninsula of Istria, and Daco-Romanian (DRom) in today’s Romania. The Vlacho- Romanian shepherds carried the practice of transhumance along the high pastures of the entire Carpathian mountains, following archaic mountain paths, spreading their technology and terminology from Northern Yugoslavia, through Rhaetia, Hungary, to Czechia, Slovakia, Slovenia, all to the Southern Poland, as demonstrated by John G. Nandris in a presentation on pastoralism in Southern Europe (1999, p. 111). The author brings to attention a North-South line of villages in the Czech region, incorporating ‘Vlach’ in their names, such as Vlachovice, that indicates settlements of Vlach shepherds from around the fifteenth century in Moravia, and marking the probable western penetration of the Vlachs. He also notes that in these Slavic speaking villages the counting system of sheep is easily recognizable of Romanian (Latin) origin: doua ‘two’ , patru ‘four’: “Doua, patrzt, shase, opt, zeci …”, and continues with “deset-doua,
32
A. R. Chelariu
deset-patru … dvadeset-doua …” ‘twenty, forty, twenty two’; from ten upwards it is admixed with Slav numerals, using “deset” instead of ”zeci” for “ten” in compound numerals”. These examples testify for cohabitation of the Romanian shepherds in these regions during centuries of pastoral activities. These findings together with the topographic remains from antiquity on the Romanian territory contradict the theory emitted by some researchers that the bulk of Romanian population came from the South of the Danube around the twelve century as a result of sociopolitical events, with the implication that an entire region, as large as the Romania, would be populated by Romance speakers from today’s Bulgarian territory. Without a doubt, such a massive migration would create a big stir and would remain recorded in historical documents of the time. Many of the people migrating from the South of the Danube River were Slavic speakers, exercising an important role in the formation of the Romanian language, as well as the other way around. The reciprocal influences between these two groups, the Romanized population, carrying the prestige of the Latin language spoken on a large area, and the newcomers, the Slavic people, should be compared and evaluated, as both belong to the common Indo-European family. The position of Romanian among the European languages is, at face value, that of a Romance language, having a basic spoken vocabulary of approximately 2581 words, with over 30% directly from Latin, of which 25% are formed internally, 17% of multiple etymologies, and the Slavic influence counting for about 9% (Sala et al., 1988, p. 73). In discussing the progressive interaction between the base language spoken by the Dacians, and the new imposed one, the late Latin spoken by the Roman soldiers, most likely the Vulgar Latin, we should consider the socioeconomic and religious concepts inherited from Indo-European in both languages. Subjected to complex phonetic and conceptual changes, the two languages intermingled, resulting in a new language, referred to by historical linguistics as the Daco-Romanian (DRom) language. During the imposed cohabitation, the speakers of the host language and those of the guest language may have been able to recognize certain IE lexeme, leading to the linguistic process of ‘contamination’, or ‘merger’ between languages coming from the group of the Indo-Europeans.
5 Daco-Romanian Language: An Indo-European Branch
33
This process may have taken place at a historic moment when, presumably, the IE dialects may not have been completely separated. The process of merging between common forms among neighboring IE languages can be observed more convincingly in the Balkan area and its surrounding populations, the Greek, Thracian, Illyrian, and Celtic. On this topic, Brian D. Joseph (1983, p. 275) stated: “[…] from earliest times, the Balkans have been an important crossroads region, and numerous groups of people have moved into (and out of ) the area, absorbing, or being absorbed by—linguistically as well as socially and culturally—already existing groups. This constant process of absorption has left its mark on various languages currently and previously in the Balkans, so that it is possible for one to identify different layers of language contact; these layers, in turn, allow one to draw conclusions about shifts in Balkan language use from prehistoric times on.” This process became more complex during the Romanization, followed by further developments, resulting from the major migrations, the Gothic and more important, the Slavic one. The merging process or “contact-induced language change” as Joseph puts it, was based often on phonetic and semantic recognition and associations, adopted or rejected by the speakers of the host language, and coupled with the social and economic interactions. The analysis of language strata starting with the newcomer language, Latin in this case, in comparison with other IE roots, could lead to identifying possible forms from a language that did not survive history, the Dacian language. The survival of key PIE roots in Romanian could help in understanding the DRom evolution and its relationship to the other IE languages. One essential concept is the PIE *kwel ‘turn’, *kwelo-, *kwolo- ‘rotation, wheel’; *kwekwlóm, *kwokwlos ‘wheel’, which, in DRom, generated reflexes connoting the round shape: coacăză n. ‘gooseberry’, cocoaşǎ n. ‘hump’, cocoli, cocoloşi v. ‘to overprotect particularly a child, to overdress a child’ cocoloş n. ‘a ball of dough or other material, cocon n. ‘baby.’ The Dacian language position among the Indo-European languages is addressed by renowned researchers in their works on the southeastern IE languages, among whom the two well-known Indo-Europeanists Gamkrelidze and Ivanov (1995, p. 415) do not include Dacian or Thracian in their classification; Georgiev (1981) groups Dacian together
34
A. R. Chelariu
with Albanian and Indo-Iranian in the eastern IE branch, while Thracian and Pelasgian in the southern Aegean branch; Eric Hamp’s model of the IE classification (1990) groups Slavic, Baltic and Thracian together, and Dacian, Albanian together. Sergei Starostin (2004) using his own model of the ‘recalibrated’ glottochronology, groups Albanian with Greek and Armenian, while there is no mention of Dacian or Thracian. The Bulgarian linguist Ivan Duridanov (1969) in his study observed that comparing materials available on inscriptions, geographical and personal names, tribal names, names of deities, with the corresponding Baltic language forms helped in clarifying some Thracian isoglosses. Duridanov, based on his study of the phonetic differences between Dacian and Thracian versus the Indo-European, issued the conclusion that Dacian and Thracian were two different Indo-European languages: IE b, d, g > Dacian b, d, g, Thracian p, t, k; IE ē > Dacian ä, Thracian ē (later i); IE e (after consonant) > Dacian ie (ia), Thracian e; IE ai > Dacian a; Thracian ai; IE ei; Dacian e; Thracian ei; IE dt (tt) > Dacian s; Thracian st. Mircea M. Rădulescu (1984) offers a considerable analysis of the relationship between Albanian, Illyrian, Thracian, and the Baltic languages, a relation for which he lists a number of Baltic elements he identifies in the substratum of Romanian and Albanian, concluding that further investigation is needed for assisting with the advancement of Indo- European linguistics. As stated above, the geographic distribution of the topographic suffixes meaning ‘settlement’ –dava and –para, with 33 Dacian locations ending in -dava, -deva, -daua, -daba, located on the entire Dacian (today Romanian) territory, plus an area immediate south of the Danube, versus the Thracian—para documented down south of the Jireček line is in line with Duridanov’s conclusion. The same goes for the Daco-Getae names of kings and princes not to be found in Thracian region (Poghirc, 1976, p. 345). These certainties may prove that these languages were just relatives, part of the large family of Indo-European languages. Following Duridanov’s work, Harvey E. Mayer (1992, p. 1) takes it a step further when he argues: “[…] the ancestors of the historical Balts moved around a great deal and in so doing spread Baltic linguistic items over an immense area. Duridanov’s
5 Daco-Romanian Language: An Indo-European Branch
35
book allows us to expand that concept to include in a way even movement southward to the Balkans if we can accept the Thracians and Dacians as descendants of “Balts,” or really, “Pre-Balts” of a sort.”
The observable relations with the Baltic languages, a most noticeable one being the word for lyric song Lithuanian daina—DRom doina, or some folk traditions that will be mentioned later, such as the similar funerary ritual for young person, could be considered for this hypothesis, albeit, may not be enough to advance the idea of Pre-Balts travels south of their today’s region. If Dacians came down from a northern area at a different time than the Thracian tribes, and if the archaeological evidence would justify such hypothesis, the differences observed between Dacian and Thracian languages, especially the DRom dava and Thracian para may have some validity. Yet, more careful examination is still needed. R. Katičić and Mate (1976, pp. 151–2) reaches important conclusions: The Thracian and Dacian substratum has also left some traces in the vocabulary of the modern languages of the area. Research on Dacian substratum words in Rumanian, in spite of the many controversies it has raised, has brought forth such important relations of correspondences that it has to be taken seriously. Substratum words have also been established in Bulgarian, and some Thracian influence on Bulgarian anthroponomy has been made probable. The Dacian substratum words in Rumanian often have very interesting correspondences in Albanian… The opinions of scholars who in Rumanian etymology operate with Dacian substratum words diverge so much that it is not easy to find Rumanian words on the Dacian origin of which general consensus has been reached.
Discussing the Paleo-Balkan situation, Vaclav Blažek (2005, pp. 1–16) finds that Albanian is a descendant of Illyrian, both satem-languages, while the change *gw > b in Dacian indicates more probably a centumtype based on Lat lingua > DRom limba, to which I have to add the similar form in Logudorese limba). In addition to his argument, attention has to be brought to the DRom reflex negura, explained in Romanian Etymological Dictionary by the Lat nebula, which presupposes a reversed development b > gw. Yet, the Romanian negură ‘mist’ could be
36
A. R. Chelariu
considered an archaism, since the voiced labiovelar *gw became b in Celtic and Latin as specified in Fortson (2004, p. 276). More recently, Matasović (2013, p. 78) argues: It is a priori not improbable that there were unknown Indo-European languages spoken between the Italic, Celtic and Germanic languages in the West, and Baltic and the Slavic languages in the East. These may have included ‘Pannonian’ (Anreiter 2001), Venetic, Dacian, and several others about which we know next to nothing (Katičić & Mate, 1976). It is not unlikely, moreover, that there are loanwords from such languages in Balto- Slavic, and the only question is whether they can be recognized as such.
Among so many uncertainties regarding this region, coupled with the lack for documentation, there are yet, social certainties that should be considered, especially the perpetual economic interaction between populations of central Europe and the Cucuteni-Tripolye—Dniester— Dnieper areas forming a large regional synthesis or fusion of multiethnic cultural and linguistic diversity, over which the Indo-European speakers became dominant. The struggle to decipher the substrata of this large area continues to preoccupy linguists, anthropologists, archaeologists with relative success.
References Blažek, V. (2005). On the Internal Classification of Indo-European Languages: Survey[*] Linguistica ONLINE. Added: November 22nd 2005. http://www. phil.muni.cz/linguistica/art/blazek/bla-003.pdf Duridanov, I. (1969). Die Thrakisch-und Dakisch-Baltischen Sprachbeziehungen. Cambridge Ma: Sofia. Fortson, B. W., IV. (2004). Indo-European Language and Culture: An Introduction. Oxford: Blackwell Publishing. Gamkrelidze, T. V., & Ivanov Vjačeslav, V. (1995). Indo-European and the Indo- Europeans: Reconstruction and Historical Analysis of a Proto-language and a Protoculture (2nd ed.). Trends in Linguistics. Studies and Monographs, 80. Berlin: Mouton de Gruyter.
5 Daco-Romanian Language: An Indo-European Branch
37
Georgiev, V. I. (1960). Raporturile dintre limbile dacă, tracă şi frigiană. Studii clasice II, Societatea de studii clasice RPR, p. 39. Georgiev, V. I. (1981). Introduction to the History of the Indo-European Languages. Pub. Sofia: House of the Bulgarian Academy of Sciences. Hamp, Eric P. (1990). Celtic Language – Celtic Culture: A Festschrift for Eric P. Hamp by A. T. Matonis (Editor), Daniel F. Melia (Editor). Ford & Bailie Publishers. Joseph, B. D. (1983). Language Use in the Balkans: The Contributions of Historical Linguistics. Anthropological Linguistics, Fall, 25, 275–287. Katičić, R., & Mate, K. (1976). Ancient Languages of the Balkans. Berlin: Mouton & Co. N.V., Publishers. Mallory, J. P., & Adams, D. Q. (2006). The Oxford Introduction to Proto-Indo- European and the Proto-Indo-European World. Oxford: Oxford University Press. Matasovič, R. (2013). Substratum Words in Balto-Slavic. FILOLOGIJA, 60, 75–102. Mayer, H. E. (1992). Dacian and Thracian as Southern Baltoidic. Lithuanian Quaterly Journal of Arts and Sciences, 38(2) Summer Chicago. Lituanus Foundation, Inc. Nandris, J. G. (1999). Ethnoarchaeology and Lalinity in the Mountains of the Southern Velebit. In L. Bartosiewicz & H. J. Greenfield (Eds.), Transhumant Pastoralism in Southern Europe; Recent Perspectives from Archaeology, History and Ethnology (p. 111). Archaeolingua Foundation. Poghirc, C. (1976). Thrace et Daco-mesien: Langues ou Dialectes? Thraco-Dacica: Recueil d’’tudes. l’occasion du 2i me Congres International de Thracologie, Bucarest 4–10 Sept 1976. Institutul Român de Tracologie. Bucharest. Editura Academiei Române. Rădulescu, M. M. (1984). Illyrian, Thracian, Daco-Mysian, the Substratum of Romanian and Albanian. JIES, 12 (1–2, Spring). Institute for the Studies of Man. Russu, I. I. (1967). Limba traco-dacilor. Bucharest. Editura stiinti̧ficǎ și enciclopedică. Sala, M. (coord.), Mihaela Bârlădeanu, Maria Iliescu, Liliana Macarie, Ioana Nichita, et al. (1988). Vocabularul reprezentativ al limbilor romanice. Bucharest. Editura stiințifică și Enciclopedică. Starostin, S. (2004). Workshop on the Chronology in Linguistics, Santa Fe.
Part II Comparative Method: Myth, Fairy Tale, Folk Tale
6 The Development of the Comparative Method
In a succint and pertinent article Michael Weiss (2014) makes a brief historic of the comparative method naming Marcus Zuerius van Boxhorn (1612–1653) “the first to suggest that the lexical similarities observable among the languages of Europe resulted from a common origin in a language he called Scythian”, together with another pioneer in comparative studies, Andreas Jäger or Georg Caspar Kirchmaier, the author of De lingua vetustissima Europae (1686). In the late eighteenth century, the pioneering efforts of Rasmus Kristian Rask (1787–1832), a Danish linguist and philologist, first used the comparative-historical method. Rask traveled extensively to study languages, first to Iceland, where he studied the Origin of the Old Norse or Icelandic language; later he also visited Russia, Persia, India, and Ceylon (Sri Lanka); in his studies he demonstrated the relationship between the Germanic languages and the Balto- Slavic, Greek, and Latin languages, by establishing the phonetic correspondences between them. Jakob Grimm (1785–1863), a contemporary of Rask, devoted his entire life to studying the German language in relationship with Thracian Scythian Getae, known to us through Greek and Latin. As a pioneering linguist in the field of comparative philology, Jacob Grimm’s contribution was framing the principle © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_6
41
42
A. R. Chelariu
according to which philology must strictly comply with the laws of sound change. Jacob Grimm is recognized for articulating the Grimm’s law. Part of the same group, Franz Bopp (1791–1867) was interested less in specific languages but in language itself and its universal function of mankind; his studies in the formative processes and the connections among the Indo-European (IE) family of languages resulted in the development of Indo-European comparative grammar. The eighteenth-century development of Indo-European linguistic studies followed mainly as a result of the contact with the Hindu culture and the Sanskrit language was best expressed in the address of Sir William Jones from 1786: The Sanskrit language, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger affinity, both in the roots of verbs and the forms of grammar. (As cited in Beekes 1995, p. 13)
Based on these important linguistic studies, linguists were able to reconstruct what is known as the Proto-Indo-European (PIE) roots of the language these ancient populations used, the vehicle of religious, social, and moral concepts, opening a door to what can be construed as the Indo-European culture. What followed was an intense focus on the relations between these languages, which led to the hypothesis that the European and Indian languages derive from the same older one, a supposition that formed the foundation moving forward of efforts to reconstruct a prehistoric common language. Researchers began by observing the similarities between deities’ names from various myths, which led to further analysis of the functions of divinities within the mythological structures. These similarities were stratified in mythical motifs which created the base for the development of comparative mythology as an independent field. The comparanda of ancient documents of the Hindus, the Persians, the Greeks, the Italics, the Celts, and more recently, the Baltic- Slavic data, are important sources especially, in conjunction with the large pool of common beliefs and customs hidden in the European and Indo-Iranian folkloric traditions. Unearthing and evaluating these social
6 The Development of the Comparative Method
43
testimonies formed the basis for research in the IE culture. On the same foundation, however, such data analysis has led to unfortunate exaggerations, particularly when used beyond known cultural structures, resulting in reconstructing imaginary historical events or conclusions. This work steers clear of such circumspection by applying the comparative method to selected “known” cultural and linguistic systems, and using tried-and-true data from European and Indo-Iranian languages, which have already been individually studied and thoroughly documented. To this I add newly gathered and accepted folkloric material, as an adjuvant in the process of comparing and analyzing sets of simple corresponding structures, which I prefer to call mythical motifs.
References Beekes, R. S. P. (1995). Comparative Indo-European Linguistics. Amsterdam: John Benjamis Publishing Co. Weiss, M. (2014). The Comparative Method. Routledge Handbook of Historical Linguistics. (C. Bowern & B. Evans, Eds.). https://www.academia. edu/9535884/The_Comparative_Method
7 The Concept of Myth
In conjunction with comparing the mythical material, scholars toil with the understanding of how myth was created and what its functions in society are. As the study of myths and the beliefs of archaic societies, in essence, the science of mythology began to develop, it tended to diverge based on researchers’ understanding of myth’s formation and functions. Moreover, debates frequently centered on the differences between what is considered myth about cosmogony or theogony, versus folk stories, fairy tales, etc. Given the nature of the subject, human beliefs, the investigations often surpassed the mythic domain, extending to other fields, religion, anthropology, psychology, sociology, and, to a lesser degree, folklore, the stepsister of mythology. Gaining familiarity with the most notable researchers’ ideas on myth and its development could help understand this ambiguous and elusive concept. Considered the founder of comparative mythology and of the so-called solar mythology school, Max Müller viewed myth as the result of a ‘disease of language’ (Müller, 1862, p. 11), the misunderstanding and confusion of names, of an initial ‘nomina’ that later became ‘numina’, the name of a divinity. For Müller, myth was a symbolic expression of cosmic phenomena, a theory that may be acceptable, except that many other stories © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_7
45
46
A. R. Chelariu
not related to celestial events are commonly considered to be of mythic content. Edward Burnett Tylor in his Primitive Culture published in 1871, introduced what would come to be known as the animistic theory, based on the idea that transfiguration of daily experiences into myths is a result of the belief that all nature is animated and susceptible to personifications. He observed the universality of mythical creations, which led him to regard myth as “an organic product of mankind, expressing universal qualities of the human mind” (vol. I 285), an idea that finds a home in Jungian archetypes. Tylor argued that myths contain cultural heritages molded on the adventures of gods and heroes of the group they belong to, a thought shared and developed by other researchers, George Dumezil being one of them. By the end of the nineteenth century, Sir James Frazer’s monumental work The Golden Bough (1911–15) generated a great impact among anthropologists, classicists, and the like, among them Jane E. Harrison and Francis Cornford, who were at the center of the ‘Cambridge Ritualists’ group. Their conclusions were that mythos was the story that accompanied the ritual; in essence, that myth was an explanation of the ritual (Bruce Lincoln, 1999, p. 71). Even though not all researchers may have agreed that myth derives from ritual, or the other way around, they agreed that myth and ritual are essentially connected. Following the same ideas, Vladimir Propp (1973, pp. 12–28) argues for fairy tales as the story accompanying rituals. Other scientists regard ritual as a parallel structure between the real and the ideal, where myth has the function of translating the real into terms of the ideal, the functional into durative and transcendental (Gaster, 1961, p. 24). By stressing the function of the enactment of the story—the ritual—over the story of the divine itself, these researchers place the essential function of religion and its metaphoric and symbolic connotations in a secondary position. Even though the psychological angle represented by Sigmund Freud, Carl Jung, and their followers, led to controversial interpretations, they brought forward the idea that myth was a symbolic expression of the deepest needs of the human psyche and its visions on the meaning of existence. C. Kerényi, a close collaborator of Jung, notes that myth
7 The Concept of Myth
47
exercises power to fill the unconscious with images of the primordial times (Jung & Kerényi, 1989). The anthropological studies owe much to Boris Malinowski, who was of the opinion that myth’s primary function was to record and validate social institutions: “Myth fulfills in primitive culture an indispensable function: it expresses, enhances and codifies belief, it safeguards and enforces morality; it vouches for the efficiency of ritual and contains practical rules for the guidance of man” (Malinowski, 1955, p. 101). On a similar note, the prominent Indo-Europeanist George Dumezil based his studies on the idea that myths reveal the social structures of a society, and that classic mythology reflects the ideology and social institutions of the Indo-European society, essentially organized in three social classes, (1) kings-priests, (2) warriors, and (3) laborers, representing the famous Dumezilian tripartite school (Dumézil, 1958). Although it has often been simplified as a story dealing with gods, or as the story behind the ritual, myth embodies a complex cultural phenomenon documenting the way archaic societies functioned and interacted. Bruce Lincoln offers a discussion on the mythos concept in contextual use by the ancient writers, such as: Hesiod and Homer, for whom “mythos is an assertive discourse of power and authority that represents itself as something to be believed and obeyed…”, an oral poetic performance to celebrate gods. By the end of fifth century BCE, Plato, the voice for Socrates, regarded mythos as an inferior discourse addressing women, children, and the lower classes, a prime instrument of indoctrination (Lincoln, 1999, pp. 17–41). Even though the philosophers of the late antiquity were degrading the concept of myth as conveyed by Greek data, people in archaic societies believed myths were “true” and “sacred” stories, recounting the beginning of things, and offering exemplary models of human behavior, stories that gave meaning to life (Pettazzoni, 1954, pp. 11–36). For Mircea Eliade, myth is the narration of a sacred story, of an event that happened in primordial times, when supernatural beings took part in the creation of the cosmos, the immediate surroundings, and the social institutions. Thus, myth is relating a ‘true story’ (Eliade, 1963, p. 8), in which supernatural beings were involved in primordial actions in a primordial time, describing the combustion of divine force. Eliade argues
48
A. R. Chelariu
that by knowing the origin of his surroundings Man thought he could control and influence them, and by recreating the primordial actions through rituals, Man believed that such reenactments will immerse him in the powers of the divine. Myths describe exemplary models of all significant human activities that were revealed by divine beings, or by the mythical forefathers during the course of an acting ritual. Through myth and symbols the world is no longer an opaque mass of objects thrown arbitrarily together, but a living and meaningful cosmos (Eliade, 1963, p. 141). As a researcher of religions, Eliade analyzes myth from the perspective of ‘homo religiosus’, preoccupied with the structure of the religious conscience. Together with Cassirer, he attributes myth a religious dimension. The philosopher Ernst Cassirer brings into the discussion a different perspective on myth: Thus the attempt is made again and again to make soul mythology or nature mythology, sun or moon or thunder mythology the basis of mythology as such. But even if one of these attempts should prove successful, this would not solve the real problem which mythology presents to philosophy [….] For mythical formulation [….] remains the same miracle of the spirit and the same mystery [….]
And he continues: But if pure philosophy is necessarily restricted to a general, theoretical picture of such an evolution, it may be that philology and comparative mythology can fill in the outline draw with firm, clear strokes what philosophical speculation could only suggestively sketch. (Cassirer, 1979, p. 11)
In his attempt to define myth and its function, Cassirer brings in the idea that art, esthetics, law, or science are all related in their genesis to the mythical conscience. He compares the mythic discourse with the philosophic and religious one, arguing that we could not establish, within the development of human culture, the moment when myth ceased to exist and religion began, and thus myth could have been, from its initial form, potentially religion.
7 The Concept of Myth
49
In agreement with these researchers, I consider approaching myth as the religious experience of humans in the archaic societies that, despite the difficulties created by time and alterations, could help us to better understand the essence and vitality of these forms of expression. G. S. Kirk , in his study of myth and its functions, concludes that: [the] speculative and operative functions of myths may often develop gradually out of their narrative ones; the needs of the community to impress themselves on the story-telling, among other aspects of social life, as a basic mode of communication; and the most fantastic elements of myths, reflecting the demand for the remarkable in stories as well as the importance of the irrational and supernatural, both in waking and in sleeping experience, are gradually and erratically accreted. (Kirk, 1970, p. 285)
This seems to be an all-inclusive definition of myth, in which Kirk clasps most of the major aspects from the previous theories. On par with a psychological approach to myth, Gerald Larson (1974, p. 1) states that: “The general acceptance is that myth is different than other stories because it articulates the basis for the self-understanding of humans, and thus operates as a vehicle for the entire cultural life” of a group. To this it could be added the definition offered by Walter Burkert (1979, p. 23): “myth is a traditional tale with secondary, partial reference to something of collective importance”, which justifies the family, the clan, or the rituals and the gods’ involvement. Concluding this short sketch of prior attempts to define myth, we can surmise that each opinion has validity. For instance, it may be true that human beings were impressed by the surrounding natural phenomena, since every aspect of their life depended on it. However, the idea that nature is the sole source of mythical thinking ignores the social or psychological aspects of this process. The same can be said of social groups, and their ‘collective importance’, which cannot be regarded solely as the basis for the creation of myth without the natural phenomena and the psyche. Socioeconomic conditions may determine certain functions of myths, but the other way around is equally valid, that socioeconomic conditions may be determined by myths. Rituals definitely played a major part in myth development, but the long debate whether myth is the story
50
A. R. Chelariu
of a ritual, or that ritual is based on myth is still ongoing, even though the times of exclusivity on the matter have passed. The little certainty that may help this dilemma is that the many stories that may describe certain rituals were always tied with the human effort to reach the subliminal, the sacred realm of the divine. In addition, one should not forget that in oral traditions the spoken word carried deep respect and fear, as in the Biblical sense, ‘In the beginning was the Word’ (John 1:1 KJV). At the core of the prehistoric man’s cultural expression lay the spoken word, the oral story that was revered the moment it was voiced by the priest, shaman, or storyteller, a story loaded in sacred forces, gaining powerful, independent existence, that influenced man’s destiny and gods’ decisions—it became myth. The stories created by the power of imagination and the use of metaphoric language are Man’s myths, serving him as guides in the world in which he might feel at home. On the basis of the imagined myths he toils to achieve the society, the world, and the self for which he hopes (Miller, 1970, p. 97). Synthesizing the above definitions it could be said that myth, in the context of religious consciousness (Eliade), or of mythical thought (Cassirer), creates by symbolic language stories about the beginnings of clans, events involving heroes and families, expressing the societal and spiritual needs of the collectivity and the self, enforcing the social and cultural specifics to which it belongs. The concept of ‘religious consciousness’ (Eliade) or that of ‘mythical thought’ (Cassirer) in relation to myth may ignore some rigidities shared by current studies of mythology; yet, it is simply enough to observe that present-day social studies often involve religious thought or mythical consciousness of masses in their discourse. The above-mentioned researchers attempt to define myth and its creation, omitted to address the concept of metaphor as an essential component in the formation of mythical language. The lack of research on the cultural and linguistic context formation of myth was sensed by Boris Oguibenine: “the fundamental aspect of the formation and survival of a linguistic and cultural community has apparently never been the object of serious research” (1998, p. 5). The author recognizes that linguists attempting to reconstruct an old culture ideology encounter obstacles
7 The Concept of Myth
51
such as the fact that a culture “[…] is not presenting itself as a world of objects and concepts […] but as signs of objects and concepts in a discourse or in a set of discourses”. The complex issue of archaic mythical language could benefit from the recent studies on the subject of metaphor and its role in human thought formation.
References Burkert, W. (1979). Structure and History in Greek Mythology and Ritual. Berkeley and Los Angeles: University of California Press. Cassirer, E. (1979). In D. P. Verene (Ed.), Symbol, Myth, and Culture; Essays and Lectures of Ernst Cassirer 1935–1945. New Haven: Yale University Press. Dumezil, G. (1958). L’idéologie tripartie des Indo-Européens. Bruxelles. Collection Latomus, vol. XXXI. Eliade, M. (1963). Myth and Reality. New York: Harper Torchbooks. Gaster, T. H. (1961). Thespis; Ritual, Myth, and Drama in the Ancient Near East. Harper & Row. Jung, C. G., & Kerenyi, K. (1989). Essays on a Science of Mythology: The Myth of the Divine Child and the Mysteries of Eleusis (8th ed.). Princeton: Princeton University Press. First Edition 1949. Kirk, G. S. (1970). Myth, Its Meaning and Functions in Ancient and Other Cultures. Cambridge University Press; Berkeley Los Angeles: University of California Press. Larson editor of Myth in Indo-European antiquity (1974). Edited by Gerald James Larson, co-edited by C. Scott Littleton and Jaan Puhvel. Berkeley, Los Angele, London. University of California Press. Lincoln, B. (1999). Theorizing Myth; Narrative, Idealogy, and Scholarship. Chicago: University of Chicago Press. Malinowski, B. (1955). Magic, Science and Religion. New York. Berkeley Los Angeles: Doubleday. Miller, D. L. (1970). Gods and Games: Toward a Theology of Play. Colophon Books; Harper and Row, Publishers. Müller, M. F. (1862). Lectures of the Science of Language Delivered at the Royal Institution of Great Britain in … 1861 [and 1863]. Charles Scribner. Oguibénine, B. (1998). Essays on Vedic and Indo-European Culture. Motilal Banar-sidass. Pettazzoni, R. (1954). Essays on the History of Religions. Leiden: Brill. Propp, V. I. (1973). Rădăcinile istorice ale basmului fantastic. Editura Univers.
8 Myth Between Symbol and Metaphor
As a mental activity, myth manifests itself through language, and through its many figurative forms of expression, the most important being symbols and metaphors, figures of speech deeply intertwined with the formation and content of the stories. As Müller states, language is an instrument of the mind, and “Mythology, in the highest sense, is the power exercised by language on the thought in every possible sphere of mental activity” (1881, p. 591). The philosopher’s angle on metaphor goes even deeper, when Cassirer (1953, p. 5) characterizes it as the intellectual link between myth and language. More explicitly, Burkert (1979, p. 28) states: “…metaphor is the basic trick of language to cover the unfamiliar with familiar words on account of partial similarity; in this sense, myth can be defined as a metaphor at tale level.” Yet, in spite of all these statements, most attempts to define myth refer primarily to the concept of symbol as the vehicle by which mythic images are expressed, leaving metaphor behind. Symbol, (from Greek symbolon ‘a token’, or ‘password’), is, in mythical context, a verbal or physical representation expressing the spiritual relation between men and the unknown. In time, the meaning of symbol evolved from defining something specific to signify any ‘something’ that, by analogy, represents ‘something’ else. More so, we use the word symbol © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_8
53
54
A. R. Chelariu
to express indirectly other concepts such as sign, emblem, or code, showing conventional objects linked in a combination of tangible object and intangible emotion, as seen in the two halves of the same coin from the original definition of symbol. To further clarify and understand the complexity of myth we must add to the discussion the concept of metaphor. Wheelwright (1962, p. 68) in his study on the relation between metaphor and reality states that “…for many persons the word ‘symbol’ suggests meanings of a more permanent kind than those transient wisps of suggestiveness that are never entirely lacking in a poem and that sometimes mark its chief intent”. Whereas metaphor is “described as semantic motion, the idea of which is implicit in the very word ‘metaphor,’ since the motion (phora) that the word connotes is a semantic motion—the double imaginative act of outreaching and combining that essentially marks the metaphoric process”. Thus, a difference between the two could be that while the symbol may utilize static images, metaphors employ a movement upfront of a logical set of facts that actually belong to another set of facts. The distinction between symbol and metaphor becomes clearer when bearing in mind that metaphor, being governed by rhetorical rules of the ‘con-text’, may turn out to be illogical if it is interpreted ad literam. Starting from Aristotle’s (2005) classical definition of metaphor as “giving the thing a name that belongs to something else” (Poetics, 1457 b 6-9), the traditional grammar dealt primarily with the poetical metaphor. Recent studies addressed the concept of metaphor from a different angle, by stressing the importance of metaphor in creating new meanings. The American philosopher Max Black considers that the way metaphor works is by “bringing two separate domains into cognitive and emotional relation by using language directly appropriate to the one as a lens for seeing the other” (Black, 1962, p. 37). Umberto Eco (1984, p. 88), after admitting the difficulties of defining metaphor, offers two ways to approach the subject; first, he suggests that language itself is metaphorical, with the metaphorical mechanisms establishing the language activity, and any effort in defining it becomes a failure unless we speak metaphorically; second, considering that language is governed by rules according to which metaphor is a malfunction, an uncontrollable deviation that leads to a linguistic renewal, in which case
8 Myth Between Symbol and Metaphor
55
the speaker uses the language incorrectly, but implies something that cannot be explained. In the end, Eco accepts that metaphor is in fact an instrument of comprehension, and that any simplistic attempt to find a quick definition is a failure, exactly because the metaphoric creation is a long and difficult process. Once the premise was established that metaphor is not simply a figure of speech, but it manifests itself in speech as part of the process of thinking, the process of conceptualizing a mental realm by another, new horizons opened in the study of metaphor. G. Lakoff & M. Johnson argue many of the most basic ideas in our conceptual system of thought, such as time, quantity, stage, change, action, cause, purpose, means, modality, or even the concept of category, are comprehended through ‘metaphor- concepts’. Metaphor is the foremost mechanism through which we understand an abstract concept, and realize the abstract thought. Taking Lakoff & Johnson ideas and applying them to the study of mythology or religion we can argue that, since myths deal in the world of the divine, and the divine realms are abstract concepts, they are, therefore, embedded in the mind through conceptual metaphors. The conclusion could be that the prehistoric religious thought expressed in myth employed a metaphorical language, which recalls Berggren’s idea (1963, p. 246) that myth is the result of the ‘abuse’ of metaphors, when “the mask, lens filter, or construing subject, is mistaken for or equated with the subject construed”, in other words when the distance between the tenor and the vehicle of the metaphor is lost. The French philosopher Paul Ricoeur (2008) finds that the description of reality through metaphor is the result of the interplay between differences and similarities that generate the tension at the level of utterance, a reality that in our discussion is the world of myth. Joseph Campbell goes beyond this ‘tension’ when he argues that metaphor’s role is to transcend separation and duality (Campbell 2001). Approaching the subject from yet another angle, Robert N. St. Clair (2000) marks the distinction between the verbal metaphor and the visual one. He argues that we cannot understand an oral culture with the instruments of the so-called literate culture, a very important observation considering that myth was essentially transmitted orally. Further, the author reveals that our formal school systems is focusing on analysis, whereas
56
A. R. Chelariu
oral culture focused attention on understanding how things are related to one another. This way of understanding can be illustrated on the mythic level, by the creation of man myth from elements of nature, as found in the southeastern European tradition, in which conceptual pairings reveal the profound human relation to the environment: flesh / earth; bones / stones; hair / plants; blood / water; eyes / sun; mind / moon; though / clouds; head / sky; breath / wind. In the print or literate culture, the information is analyzed by using the verbal metaphor based on language, while the oral culture uses the visual metaphor based on reorganizing visual space. A similar distinction between the two types of metaphors is offered by the Romanian philosopher and poet, Lucian Blaga, when he refers to ‘image metaphor’ and ‘revelation metaphor’: “The genesis of image metaphor is a non-historic moment which is linked to the genesis of human consciousness as such, […] of the structural organization of human spirit” in the process of correcting the fatal conflict between the figurative and the abstract, essential to the human spirit; and he continues, whereas: “the revelation metaphor is a result of the specific human way of existence, of being in the horizon of mystery and revelation” (Blaga, 1944, p. 358). In Blaga’s opinion, only when man is positioning himself in the dimension of mystery does he really become ‘man’. A few years later, the philosopher Cassirer (1953, p. 83) made a similar distinction regarding metaphor, between the general metaphor as a conscious denotation of a thought which includes another, using a known vocabulary, and the radical metaphor as a condition of mythical and verbal expression, which operates not as a mutation into a different category, but the creation of the category itself. The perception of two sides in discussing metaphor is postulated by Wheelwright as well, when he uses two “distinguishing names —‘epiphor’ and ‘diphor’, one standing for the outreach and extension of meaning through comparison, the other for the creation of new meaning.by juxtaposition and synthesis…” (1962, p. 72). The quality of metaphor to create new meanings discussed by Wheelwright was further examined by Bernard Debatin (1995), who argues that metaphor opens up a perspective on an object and at the same time describes it:
8 Myth Between Symbol and Metaphor
57
[…] metaphor represents an ‘as if ’ prediction with an anticipatory reference to the world […] Based on its particular power of synthesis, metaphor can bridge the gaps between experience and thought, between imagination and concept, and between new and the known […] Through this selective process, metaphor makes possible not only the conceptualization of experience but also the linkage of new to prior experience.
Thus, metaphor’s function of orientation and openness toward the world comes from the cultural heritage of images forming the metaphoric systems of orientation deeply rooted within the culture, visible and accessible to reasonable thinking that can project the absolute metaphor, recalling Cassirer’s radical metaphor (1953, p. 83) as a condition of mythical verbal formulas. In an article on the conceptual metaphor theory as methodology in comparative religion, Edward Slingerland (2004) states that if we need to know what people truly think about a concept we need to look at the metaphors used in relation to that concept. More so, conceptual metaphors as a primary tool for reasoning about self and the world need to be observed in the shared conceptual structure in which they are formed. In the end, any discussion regarding myth must consider the intrinsic role of metaphor. The already accepted statement that ‘the religious discourse is metaphoric’, asserts that metaphor is a sine qua non component of myth. Thus, a tentative definition can read as: Myth is a cultural representation using symbols and metaphors, where metaphors are an essential component of myth. In addition, metaphor includes and operates with symbols; as a creative-cognitive form of expression, metaphor generates new concepts, whereas symbol, as a societal established system of codes, a representation of the general in particular (Goethe 1833, 1998 “That is true Symbolism, where the more particular represents the more general”), seems to be more static, contributory in the process of creating new metaphors. Symbol, as a verbal or material form of expression, and part of the myth–ritual discourse, can be controlled, whereas metaphor, as an intense effort to penetrate the transcendent, escapes control and becomes unconfined, leading the human mind toward new abstract concepts. Together with symbol, metaphor acts within the human mind, forming ‘the mythic or religious consciousness’.
58
A. R. Chelariu
Once we agree on the role of metaphor as an essential component in the process of forming abstract concepts, the next step would be to find out how such conclusion may influence us in revisiting the understanding of myth. In this attempt we should take into consideration our contemporary cultural group of concepts and images, with which we are trying to see through the web of symbols by which mythical metaphors transmit archaic meanings. Following St. Clair’s idea that visual metaphors controlled the mental processes in the prehistoric oral tradition societies, the physical elements such as the earth, the sky, or the trees, have transcended into symbols representing mythic divinities. Simultaneously, these symbols were part of metaphors, part of logical sets of facts that were applied to other set of facts within the context of a myth. In this complex process of myth creation, the divinity was a metaphorical manifestation, anthropomorphic or otherwise, of the divine in action. To illustrate this thought, we could consider for example, the Great Goddess Demeter as a symbolical embodiment of the Earth, that in a logical metaphorical act articulated through the expression ‘Demeter feeds us all’ becomes the metaphorical manifestation of the divine in action. Hence, myths are stories recording the metaphorical manifestations of the divine in action. This view has validity as long as the social group recognizes the divinity in its metaphorical context in relation to the divine actions. The potential of the metaphoric process to create new categories is so versatile that time and changes in the human understanding of the divine could get altered, and new meanings could develop. Stories once perceived as sacred, recounting divine actions in establishing certain social values for the community, could be transformed by new events or conditions, new revelations, or leadership. The metaphors that were once connoting a particular divinity may become obsolete, losing their spiritual sacred value, as, to use the same example, the expression ‘Demeter feeds us all’ became irrelevant for today’s understanding of the divine. Yet, there are cases in which a rather significant phenomenon takes place: A divinity may resurface as a new divinity under a different name, but, in essence, retaining the characteristics and powers of the old one. A good example is the case of the pre-Christian Slavic gods Perun and Veles/ Volos that appear to have resurfaced in early Christian iconography in the
8 Myth Between Symbol and Metaphor
59
Slavic lands as St. Ilya and St.Vlas-Vlasiy-St. Blasius. Veiled in metaphoric language the new revelation arises from previous religious imagery. By borrowing a concept from physics, these metaphors can be called collapsed metaphors. To clarify, we shall refer to the famous Schrödinger’s (1935) Gedankenexperiment, in which a hypothetical cat hidden in a box can be simultaneously considered dead and alive until we open the box and ‘observe’ it; then our quantum supposition ‘collapses’ and the cat is either dead or alive. Applying this theory to mythical metaphors that were operating within their original mythical environment, we could assume that they become collapsed metaphors the moment they were ‘observed’ through a new cultural milieu. Thus, the metaphors belonging to a previous mythical background can collapse into a new reality. What was once a metaphoric manifestation of a divinity may emerge either as simply a character in a story or as a new divinity. For example, the Hindu dragon Vrtra, obstructor of the fertile rains, slayed by the hero Indra, or the Old Norse dragon Fafnir protecting a gold treasure, slayed by Sigurdr, the Slavic Perun battling Volos, or the Romanian balaur, ready to eat the beautiful fairy, are all metaphorical manifestations of the mythical character symbolizing the trial facing the hero–dragon slayer. Hence, we may entertain the possibility that the German dragon may be a collapsed metaphor of a dragon, having a similar function with Vrtra, the obstructor of the fertilizing waters, that resurfaced with a new meaning, obstructing the hero’s access to gold, changing the myth’s heroic purpose from that of fertility to that of acquiring richness and power. Another example of a collapsed metaphor could be the story of the Romanian Trickster Păcală, who, among other inexplicable actions, proceeds to sell his cow to a tree in the forest. At first glance, his action seems laughable and foolish, but only if a metaphor is interpreted ad literam, and the denotation remains visible. However, if we look, for example, at the Hindu (and not only) meaning of a cow as a symbol of the Mother Earth, the food provider, and then we consider the well-known symbol of a tree, the axis mundi connecting the three realms, sacred to the prehistoric people, Păcală’s action may not seem so foolish. And so, the story opens itself to a range of interpretations, such as, perhaps, his action is related to an ancient ritual performed in forests, associated with an animal sacrifice to a tree-god. It may suffice to note here how an apparently
60
A. R. Chelariu
insignificant event from a folk story could hide a wide variety of imagery and ‘collapsed’ metaphorical actions, that to us seem emptied of a distinctive mythical significance. Attempting to unveil the mythological meanings of collapsed metaphors, albeit hazardous, could broaden the understanding of the role of myth in oral societies. For some researchers, the fantastic elements present in myths were believed to come from dreams, hence their chaotic and imaginative nature. Yet, the chaotic nature we perceive today in myth is most likely the result of our inability to grasp the symbols and visual metaphors at work in the religious mysteries of the past. These symbols and metaphors, ab initio, open primarily to the initiates, were intentionally veiled in the sacred and tabooistic mysteries, if not for other reasons, for protecting the group from the furies of the supernatural forces. It should be stressed that the priest or the shaman as guardians of these cultural forms, and in later times, the storyteller of the oral tradition to which they belonged, were under strict obligations to respect and follow the well-established course of action in their narration, even following the exact wordings, especially in the cases of spells and chants. Thus, knowing the Iliad by heart was a sacred duty in the Greek schools of antiquity, and even later on, in our own academic schools, guarding and transmitting the sacred stories from generation to generation. These strict requirements were determined just because of the oral means through which the tradition was transmitted, as it is known from the Vedic or Homeric legacy. The community, especially the elderly group of listeners, had the authority and the duty to reject or correct a narrator who would not respect closely their tradition, their cultural heritage, precisely because of fear of their ancestors, fear of committing a sacrilege against the gods and heroes, and the forefathers involved in the story. The popular interpreter, as a Shakespearean actor, may have had the liberty to improvise on the declamation style, but the mythical content had to be respected in its entirety. The respect for tradition and the fear of punishment coming from the divinities was the base for consolidating the sense of belonging to the social group.
8 Myth Between Symbol and Metaphor
61
References Aristotle. (2005). Poetics, Longinus: On the Sublime; Demetrius: On Style (Loeb Classical Library No. 199). Translated by Stephen Halliwell. Cambridge Ma: Harvard University Press. Revised Edition 2005. Berggren, D. (1963). The Use and Abuse of Metaphor. Review of Metaphysics, 16: 237–58. Philosophy Education Society. Black, M. (1962). Models and Metaphor: Studies in Language and Philosophy. Ithaca: Cornell University Press. Blaga, L. (1944). Trilogia culturii; geneza metaforei şi sensul culturii. Bucharest. Fundația Regală pentru Literatură și Artă. Burkert, W. (1979). Structure and History in Greek Mythology and Ritual. Berkeley, Los Angeles and London. University of California Press. Campbell, J. (2001). Thou Art That; Transforming Religious Metaphor. Ed. Eugene Kennedy, Joseph Campbell Foundation. New World Library. Cassirer, E. (1946 [1953]). Language and Myth. New York: Harper and Brothers, Dover. Debatin, B. (1995). Die Rationalität der Metapher; Eine sprachphilosophische und kommunikationstheoretische Untersuchung. Berlin: Walter de Gruyter. Eco, U. (1984). Semiotics and the Philosophy of Language. Bloomington: Indiana University Press. Goethe, Johann Wolfgang, von. The Maxims and Reflections (1833, 1998) Peter Hutchinson (Editor), Elisabeth Stopp (Translator) Penguin Classics. Lakoff, G., & Johnson, M. (1980). Metaphors We Live By. Chicago: Chicago University Press. Müller, M. F. (1881). Selected Essays on Language, Mythology and Religion. Longmans, Green. Ricoeur, P. (2008). The Rule of Metaphor; the Creation of Meaning in Language. London: Routledge Classics. Schrödinger, E. (1935, November). Die gegenwärtige Situation in der Quantenmechanik (The Present Situation in Quantum Mechanics). Die Naturwissenschaften 23(48), 807–812. Slingerland, E. (2004). Conceptual Metaphor Theory as Methodology for Comparative Religion. Journal of the American Academy of Religion, 72(10), 1–31. St. Clair, Robert N. (2000). Visual Metaphor, Cultural Knowledge, and the New Rhetoric. Flagstaff: Northern Arizona University. Wheelwright, P. (1962). Metaphor and Reality. Bloomington: Indiana University Press.
9 Myth and Fairy Tales
Over the years, the study of myth was often confronted with the problem of its relationship with fairy tales, a question that divided the researchers mainly into those of the opinion that fairy tales are desacralized myths, and those who believe there are no essential differences between the two. To add to the difficulty, most researchers agreed that certain mythical motifs are found in fairy tales, asserting that the main difference between these two forms of expression is that fairy tales are optimistic in nature. Mircea Eliade was of the opinion that the progressive desacralization of myth, resulted in the hero from fairy tales being detached of gods’ influence, while the hero of myth was dominated by gods and destiny. For Eliade (1963, pp. 199–200), fairy tales do not seem to belong to mythical world, being rather stories that carry camouflaged mythical motifs and characters. The observation that gods are absent from fairy tales is valid, although this absence may not be due to desacralization, but it may reflect a more archaic understanding of the divine. In Eliade’s understanding, the manifestation of the sacred, which the author calls hierophany, is a mysterious act, “the manifestation of something of a wholly different order, a reality that does not belong to our world, in objects that are an integral part of our natural ‘profane’ world” (1987, p. 11). These © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_9
63
64
A. R. Chelariu
statements make it more difficult to regard certain characters from fairy tales desacralized mythic motifs. For example, the magic helpers from the Romanian fairy tales assisting the hero and heroine in their trials, as the Sun and its domain, or the sacred Moon with its domain, or the Neverland, are revered entities dwelling in a sacred space, possibly collapsed metaphors from a more archaic time. These astral domains from the fairy tales do not seem to be far away from Mount Olympus, the place known in Greek myth as gods’ dwelling. The magic helpers in these fairy tales do not show the characteristics of mythic motifs that lost their sacred connotations, as the sun and the moon are essential components of many folk stories, see for example the sun and the moon marriage poem, or the many winter ritual customs meant to help the sun revive. The reverence with which these folk characters are regarded is certainly different than the Christian reverence toward God and saints, as the new religion surpassed the old beliefs, making the attempt of discerning the sacred of the profane more difficult. It could be argued that the relationship between myths and fairy tales indicates a coexistence of two different forms of cultural expression within the same community. The subject of distinction between myth and folktale was largely addressed by G. Kirk, who, together with many other researchers, does not see differences between the two cultural expressions, grouping both under the concept of folktales, and wondering if it is feasible to separate them. For methodological purposes the author does not see folktales as a distinctive kind of myth, arguing that the difference stands partially on the supernatural elements found in them. To support his argument, Kirk (1970, p. 37) offers the following definition of folktales: “traditional tales, of no firmly established form, in which supernatural elements are subsidiary; they are not primarily concerned with ‘serious’ subjects or the reflection of deep problems and preoccupations; and their first appeal lies in their narrative interest”. The argument that fairy tales have no ‘firmly established form’ is contradicted by the specific formulas used in the beginning and the end of the stories, and by their familiar structures, as classified in Aarne-Thompson (1987, 4th edition), more recently the Aarne–Thompson–Uther online catalog. As to their lack of seriousness argued by Kirk (1970), his view expresses the current disregard toward fairy tales, in modern times viewed as children’s literature, even though
9 Myth and Fairy Tales
65
these stories describe heroic acts conveying moral concepts, subjects with an important role in the archaic community that endured for centuries. For example, the many trials confronting the hero or heroine that have as the final result a matrimonial relationship do not qualify as superficial in prehistoric societies as well as nowadays; rather, they reveal basic concepts regarding human relations articulated into exemplary models of social behavior. Kirk observes, rightfully so, that in fairy tales the characters’ names are generic, like ‘the Little Red Riding Hood’, ‘the Giant’, ‘Little Claus’, ‘the Ogre’, etc., a fact that gives them a universal quality, as they do not belong to any specific location and specific time (Kirk, 1970, p. 39). It should be added here that in fairy tales the time and space are relative, time seems to expand or reduce as the needs of the action may be: the Prince grows up in one day as others in one year; the Sleeping Beauty wakes up after 100 years, the hero’s horse flies through space at incredible speed, and so on. Thus, in fairy tales time and space are not reflecting historical time or real space, and could expand or contract as the needs of the story may require. Whereas in myth, the characters are gods, semi-gods, and heroes, acting in the beginning of time, the time of world creation, and the founding and confirmation of social events and institutions. The difference between the two cultural forms of expression rests in the specific social needs they address. To clarify these differences, some examples given by Mircea Eliade from a community in which myth is still ‘alive’, might be helpful: in such a community the people make distinction between ‘false stories’, fairy tales, and ‘true stories’, myths. In the false stories, the hero performs heroic and noble acts, frees the people from famine or natural calamities. These stories could be narrated or listened to any time by anyone in the community. The true stories though, narrate events that took place at the beginning of the world, and divine characters were involved. The true stories are not to be recited in the presence of women and children, who are not initiated in sacred matters. These stories could be recited only in a certain kind of weather, and in certain seasons (Eliade, 1963, p. 8). These are perfect examples of two forms of expression coexisting in the same cultural group, having different functions, under different restrictions. We could assume that myths were addressing specific groups, under certain interdictions, whereas fairy
66
A. R. Chelariu
tales, ‘false stories’, were told to a different group involving different cultural needs. The exclusion of women from the recitals of sacred stories and myths may reflect only certain social group, since women were viewed closer to the mysteries of life, as life givers, familiar with the mysteries of the divine in most of the groups, even having power to influence gods. One could think only of the fear and interdictions concerning menstruating women, or young mothers in the first weeks after giving birth, to understand such boundaries. Together with children’s exclusion from the ‘true’ stories as they were not yet initiated into the sacred, these were acts generated by the deep need of protecting the community against the potential wrath of gods. The presence of camouflaged motifs recognizable in fairy tales may have had the intent to convey in a ‘desacralized’ mode the trials enacted during the adolescent initiation rituals, hence to prepare the children for what was to follow, the actual rite of initiation. The same can be said about the absence of the divine and sacred, characteristic to this pre- initiation stage in the life of the child, the lack of death or copulation in fairy tales, leading researchers to believe that fairy tales are desacralized myths. Yet, such omissions could have been intentional, as these stories addressed children that were not ready to recognize and understand the sacra, but offer a foretaste of supernatural powers and the Other World. The simplest way of conceptualizing the world is through binary oppositions, such as dark and non-dark, light and non-light, and this forms the basic structure operating in the fairy tale, and understanding the relationship between this world and the Other World becomes the primary mythical awareness for children. However, over time the world of fairy tales lost its connotations to the initiation rites and became pure fantasy devoid of mythical denotation.
References Aarne, A., & Thompson, S. (1987, 4th Printing). The Types of the Folk-tale (ATh). Helsinki: Soumalinen Tiedeakatemia, Academia Scientiarum Fiennica. Aarne-Thompson-Uther. Classification of Folk Tales (ATU). http://www.mftd. org/index.php?action=atu
9 Myth and Fairy Tales
67
Eliade, M. (1957, 1987). The Sacred and the Profane. New York: Harcourt, Inc. Eliade, M. (1963). Myth and Reality. New York: Harper Torchbooks. Kirk, G. S. (1970). Myth, Its Meaning and Functions in Ancient and Other Cultures. Berkeley Loas Angeles: Cambridge University Press; University of California Press.
10 Mythic Time Versus Fairy Tale Time
The way time is represented in myth versus folktale brings to mind Heraclitus’ words: αἰὼν παῖς ἐστι παίζων, πεττεύων· παιδὸς ἡ βασιληίη “Aion is a child at play, playing draughts; the kingship is a child’s.” More explicitly, another great Greek mind, Parmenides, hinted that the difference between Kronos and Aion is that time as Kronos contains the past, the future, and the present, whereas time as Aion ignores the periodical separations between the past and the future, validating all within the present (Peters, 1967, p. 7). Intertwined in a symbiotic relation, the two concepts of time, Khronos and Aion, are meandering between the story of the beginnings as myth, and the cyclical enactment of the story as rite, which, in Russian folklorist V. I. Propp’s view (1973) is preserved in fairy tale, in fact, the story associated with rituals. The Greek representation of ‘everlasting time’, Aion, was in the shape of a circle, the hidden harmony behind all change, linked to the concept of Logos. The subtle symbiosis between Kronos and Aion, the time when myths of the beginnings happened, and the circular time seen through periodical ritual enactment, conveyed in cultural expressions the two ways of perceiving time in antiquity. The child’s perception of time is devoid of clear understanding of past, present, or future, living “in his/her own kingdom” in Heraclitus’ © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_10
69
70
A. R. Chelariu
words, in Aionic time, or the ‘everlasting time’ of fairy tales with their specific set of concepts. It is by the ritual reenactment and retelling of the origin myths that the adolescent transcends into the adulthood comprehension of the past and present, attaining the perception of a symbiotic coexistence of Kronos and Aion. Writing extensively on the subject of mythic time, Mircea Eliade (1957, p. 68) also perceives time as a dual concept, when he relays the idea that time could be construed as sacred or profane: “The intervals of sacred time, the time of festivals, and the profane time, ordinary temporal duration in which acts without religious meaning have their setting.” And he continues: “The sacred time is a primordial mythical time made present. Every religious festival, any liturgical time, represents the re- actualization of a sacred event that took place in mythical past, ‘in the beginning’” (ibid.). The sacred time is indefinitely recoverable through rites and festivals, yearly fests and ritual events, unveiling, in measurable celestial moves, the intrinsic relation between the sacred mythic time in junction with ritual ceremonies, and the profane time. The in-between time, regarded as the profane time, may in fact have been very limited, because in reality, the everyday activity of archaic societies was deeply implanted in sacra, even though, for our contemporary way of thinking, that may pass as superstitions. When mythic time is diminished, the religious influence on a social group is low, and mythologies disappear. Examples of the daily activity in the sacred and profane understanding of time is found in old popular calendars, as in the Romanian religious calendar, in which we see that almost every day has set a religious significance in the group’s memory, over which the Christian church superimposed the Christian saints. Without detailing the numerous daily reminders and interdictions, it suffices to mention that each day of the week is addressed to as a saint, all regarded as feminine entities, Sfânta Duminică ‘Holy Sunday’, Sfânta Lune ‘Holy Monday’, Sfânta Miercuri ‘Holy Wednesday’, Sfânta Joie ‘Holy Thursday’ and the ever feared Sfânta Vineri ‘Holy Friday’, or in common daily talk, Sfânta zi de astăzi ‘This Holy Day’. Such religious understanding of the flow of time is deeply embedded in the cultural heritage. Separating themselves from mythic time of the beginnings, the time of creation, and the circular time of the ritual reenactments, fairy tales
10 Mythic Time Versus Fairy Tale Time
71
continue, for thousands of years, to recount stories that transcend a specific moment in time, closer to everlasting time, as in the formulaic expressions used at the start and at the end of the story, for example: ‘once upon a time’, in the Romanian formula a fost odată ca niciodată ‘it happened sometimes as it never happened’, that inserts in the mind of the listener the ambiguity of the discourse. Likewise for the end formula: ‘and they lived happily ever after’, or the Romanian counterpart și am încălecat pe-o șea și am spus povestea așa ‘and I mounted my horse saddle and I finished this story’, again projecting the relativity of space and time. The different representation of time in myth and in fairy tales extends to the Other World as well: In myth, Persephone spends 6 months with Hades and 6 months above, in a cyclical time, as she is the spring and flower goddess, while Orpheus goes to Hades to retrieve Euridice and comes back in linear time. In fairy tales the hero or heroine travels in the other world, believing that the visit lasted for three days, but in the world of their home, hundreds of years passed. We could say that in fairy tales, the Aion’s time in Parmenides’ terms was a continuous present, whereas in mythic events the time is linear or cyclic. From fairy tales the child got familiar with elementary mythic concepts before entering into the adolescent stage, when the community would starts the process of initiation, of transmitting ethical and religious information. The initiation rites and their metaphoric substance involving difficult trials with various degrees of cruelty guided teens through the process of entering into the adult group. They had to abandon childhood and motherly care, and experience through death and resurrection, the world of ancestors and gods. They were introduced to myths of creation of humans and the world, the myths of social foundations, heroic actions of the ancestors, events that happened in illo tempore. Through such myths adolescents understood the human condition in relation to the divine. Either as part of initiation rites or recited at special feasts and celebrations, myths had the role of showing the function and power of gods for the social group, aiming to reassure protection of life and procreation. Impressing the young with the standards and expectations acceptable to the group, and reinforcing the god–human relationship was the main function of myths. From such a perspective we can assume the coexistence between myth and fairy tales, each addressing specific needs
72
A. R. Chelariu
of the social group. Mythical motifs could be present in both, myth and fairy tale, such as, for example, the metamorphoses or the magic helpers, yet the essential structure could be camouflaged in fairy tales, or involving divine participation in myth. The specific elements that separated the two, as stated by M. Eliade, de Vries, or Kirk, the absence of gods and the sacred, the a-temporality, the generic names of the characters in fairy tales, can be less the result of a desacralized myth, but more essential attributes of fairy tales structure, having their own role in preparing the youth for initiation.
References de Vries, J. (1977). Perspectives in the History of Religions. Berkeley Los Angeles: University of California Press. Eliade, M. (1957). The Sacred and the Profane; the Nature of Religion. Orlando Fl. Harcourt, Inc. (1987 copyright renewed) Kirk, G. S. (1970). Myth, Its Meaning and Functions in Ancient and Other Cultures. Berkeley Los Angeles: Cambridge University Press; University of California Press. Peters, F. E. (1967). Greek Philosophical Terms: A Historical Lexicon. New York: New York University Press. Propp, V. I. (1973) Rădăcinile istorice ale basmului fantastic. Bucharest. Editura Univers.
Part III Traits of Indo-European Mythic Motifs in Romanian Folk Stories
11 The Mythic Motif of Man’s Creation
To establish the place of Romanian mythology amid the European cultural heritage is a challenging enterprise. Together with the cultural heritages of the Baltic countries, and the Albanians, Romanian folklore and mythology can be traced sporadically in the historic chronicles of sixteenth century, but mainly in popular almanacs and calendars beginning with the eighteenth century. Following the European interest in folklore represented by Brothers Grimm, among many others, in the nineteenth century Romanian academics began expressing interest in collecting and publishing oral traditions in a more sustained effort. Yet, many scholars showed reticence in comparing and analyzing these folk data, as they considered the lack of written documentation on the Romanian ancient cultural heritage a serious impediment. The Romanian researchers’ distrust toward the comparative method, coupled with lack of access to documentation on the account of political conditions, held the advancement of Romanian mythology and folklore studies in a stagnant state. Recently, the application on a large scale of comparative method expanded the search to include the comparanda of eighteenth-century folkloric data, particularly the motifs hidden in cultures less exposed to © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_11
75
76
A. R. Chelariu
modern influences, such as Albanian, Romanian, and Baltic, increasing the possibility of extracting typological aspects found in collections of folktales, folk songs, and customs. For this challenge, the comparative method is the most attractive way to identify possible traits of the ancient cultural and social expression. A relevant example of the applied comparative method is the Indo- European motif of creation surveyed by Bruce Lincoln (1986, p. 11) in a series of documents from ancient times, beginning with the “Song of Purușa” (Rg Veda 1-.90): “When they divided Purușa (‘Man’), how many pieces did they prepare? “What was his mouth? What are his arms, thighs, and feet called? The priest was his mouth, the warrior was made from his arms; His thighs were the commoner, and the servant was made from his feet. (Purușasukta) The moon was born from his mind; of his eyes, the sun was born; From his mouth, Indra and fire; from his breath, wind was born. From his navel there was the atmosphere; from his head, heaven was rolled together; From his feet, the earth, from his ear, the cardinal directions.”
His investigation continues with the ancient German poem Grimnismal (40–41) that recounts the origin of the cosmos: “The earth was formed from Ymir’s flesh, and the sea from his blood, the rocks from his bones, the trees from his hair and the sky from his skull. The happy gods formed Midgard for humans from Ymir’s eyelashes They formed all the grim clouds From his brains.” (Poetic Edda, 2015, p. 68)
11 The Mythic Motif of Man’s Creation
77
Perhaps of a lesser complex account, yet still important to bring into this discussion, is the story of the giant Atlas and his meeting with Perseus, found in Ovid’s Metamorphoses (1955, p. 102): “Well, anyway, since you will give me nothing, I have something for you!” He turned his back, Held up, with his left hand behind his body, Medusa’s terrible head, and, big as he was, Atlas was all at once a mountain: beard And hair were forests, and his arms and shoulders Were mountain-ridges; what had been his head Was the peak of the mountain, and his bones were boulders.”
Almost the exact pattern of the pairing between the body parts and nature’s basic elements from this creation motif is quite prevalent in the hagiographic literature and the folk data collections of Southeastern Europe and Russia. The remarkable vigor of this motif allowed it to penetrate in folklore in an inverted format under the Christian influence: from the motif in which gods dismember a primordial man creates the world, to a supreme god who creates man from elements of nature in accord with the Biblical story. Patrice Lajoye (2013) in his in-depth analysis of this motif gives many Slavic sources, among which he cites the following: “It is written in the Book of the Dove: ‘Our world is born Of the Holy Spirit of Sabaoth; The red sun, from the face of God, From Christ himself, Tsar Celeste; The new moon, of the chest of God; The wiry stars, of the chasuble of God; Dawn and twilight The eyes of God, of Christ, heavenly tsar; The winds are springing up on our land From the Holy Spirit of Sabaoth, From the breath of God: The thunders are born on our earth -
78
A. R. Chelariu
Words of God Tsars are born on our earth From the holy head of Adam; Princes and boyars are born Holy relics of Adam; Orthodox peasants Of a holy knee of Adam’ [my translation]
Another Slavic source cited by Patrice Lajoye is the apocryphal text of Enoch, or the Book of Secrets of Enoch, where he finds the same man’s creation pattern: “And on the sixth day I commanded my wisdom to create man out of the seven components: /1/ his flesh from earth; /second/ his blood from dew and from the sun; /third/ his eyes from bottomless sea (the sun in 2 manuscripts); /fourth/ his bones from stones; /fifth/ his reason from the mobility of angels and from clouds; /sixth/ his veins and hair from grass of the earth; /seventh/ his spirit from my spirit and from wind.”
Patrice Lajoye (2013, p. 29) calls attention to the important contribution of the French philologist of Romanian descent Emil Turdeanu (1981), who “made an inventory of songs of man’s creation ranging from an English version of the seventh century to Romanian nuptial songs collected in the nineteenth century: “And he [God] took dust, and from the earth he took clay, and he blessed this clay, and he made a man, his body of the earth, his bones, of stone, his blood, dew, his strength, the wind, his soul, of the holy spirit,
11 The Mythic Motif of Man’s Creation
79
his eyes, from the sea, its beauty, the sun, his thought, the speed of the angels.” [my translation]
Another Romanian version describing how man was created from elements of nature is found in an almanac from 1809 as a Questions and Answers article, in a technique very popular at the time: “Question: ‘From how many parts did God made man?’ Answer: ‘From eight parts: the body from soil; bones from stones; blood from dew; eyes from sun; thoughts from clouds; breath from wind; intellect from moon; the gift of prophecy from the Holy Spirit’”. (Bruce Lincoln, 1986: 11, 15, 17)
About one hundred years later, the same pattern of man’s creation, this time the Biblical Adam, was published in a collection of Romanian religious folk stories by Ion Pop-Reteganul (1901, p. 3): “God made Adam from eight parts: the body from from soil, bones from stone, blood from dew, eyes from sn, thought from clouds, the soul from wind, wisdom from moon, and the gift of prophecy from the Holy Spirit.” The environment in archaic human perception is revealed in these folk stories: from the Indo-European cosmogonic myth, the creation of the world from the primordial man’s body, transcended into the man’s own creation, following the same outline: flesh - earth; bones - stones; hair plants; blood - water; eyes - sun; mind - moon; thought - clouds; head - sky; breath - wind. This structure associating the origin of the entire nature with each part of the primeval man reveals the symbiotic relation of humans with the environment, the universe, in what the Greek philosopher, Posidonius, postulated as the cosmic “sympathy” (συμπάθεια), idea embraced by his followers, such as the Roman poet Ovid. The names of anatomical parts used in the Romanian version of this creation myth are mostly of Latin etymology, corpul din pământ, oasele din pietre, părul din plante, sângele din apă, ochii din soare, mintea din lună, gândul din nori, capul din cer, suflarea din vânt. The few exceptions with roots in the Indo-European are of special interest for the development, albeit quite controversial, of the Daco-Romanian: DRom n. gând,
80
A. R. Chelariu
gândire ‘thought’; v. gândi ‘think’; possibly v. găsi ‘to find’—cogn. with Alb gjënj, gjëj ‘find’ < PIE (Mallory-Adams) 1. *ghe(n)dh- ‘seize, take in’, with the following cognnates: OIr ro-geinn ‘finds a place in’; Lat pre-(he) ndō ‘grasp’; Lith godóti ‘guess, suppose’; OCS gadati ‘imagine, guess’; Alb gjëndem ‘be found’, gjënj, gjëj ‘find’; Grk khandánō ‘take in, comprise’; Skt gan, ganati ‘count, consider, think’. A second possibility is IEW 437 *ghend, *ghou- ‘to grab, grip, perceive, pay heed to’ with cogn.: Gk. χανδάνω (*ghend-) ‘take in, hold, contain, take; to be capable, able; catch’; Goth. bi-gitan ‘find’, OIce geta ‘reach; bring forth, assume’, OE be-gietan ‘receive, produce’, for-gietan ‘forgotten’; (Eng. get, beget, forget) Or, IEW 1125-1127 u̯(e)id-2 ‘to see; to know’: MWelsh 1. sg. gwnn, Corn gon, MBret goun ‘I know,’ Welsh gwys ‘knowledge’ from *uid-tu-s, M.Welsh gwyss, MBret gous ‘became known’; the DEX.(Dicționarul explicativ al limbi române-online) gives as etymological solution the Hungarian gond ‘care, worry, concern, anxiety, trouble’; v. gondol ‘consider, believe, think’, of unknown etymology, Finno-Ugric, Iranian? (A Magyar Nyelv Törteneti-Etimologiai Szotara, 1970) The DRom creier ‘brain’ < PIE MA *k̑ ṛrēh2 ‘head’, IEW 574-77 1: k̑ er-, k̑ erə- : k̑ rā-, k̑ erei-, k̑ ereu- ‘head, horn, upper part of body’, with cognates in ON hjarsi ‘crown of the head’; Lat cerebrum ‘brain’; Alb krye ‘head’, Gheg krëja ‘head’, Tosc krye- ‘head’; Grk kárē, karárā ‘head’, krānion ‘crown of the head’; Av sāra, sarah ‘head’; Skt śiras ‘head’, Toch B kraiñye ‘neck’; Hit kitkar ‘headlong’; Medieval Latin cranium ‘cranium, skull’. An interesting situation presents DRom ceafă ‘nape, back of the neck’; IstrRom čâfa ‘nape’ for which there is no clear etymology, but for which the PIE MA *ghebhōl could offer a solution, with the following cognates: ON gafl ‘gable, gable-side’; NE gable; Grk kephálē ‘head’; Toch śpāl ‘head’; Welsh gwdff; Alb. qafë (neck), Gheg kafë; [Alb *qafë, kjafa < Balkan-Trk kafa ‘back of the neck’, Holm: JIES vol. 39, no. 1–2; 2011, p. 45]. Another example could be the DRom creț ‘curly hair, anything curly’, but in this case there could an example of contamination, or of users’ misunderstanding of two PIE roots expressing related concepts, IEW (584-585) *kert-, kerǝt-, krāt- ‘to turn, roll, wind’ and IEW (583) *k̑ er(es)- ± ‘rough hair, bristle’. Here it may be worth noting that the PIE *haén(h1)mos ‘breath’, Lat animus ‘spirit, wind’ developed in DRom inimă ‘heart’, while all the other Romance languages preferred the Lat cōr, cordis; the
11 The Mythic Motif of Man’s Creation
81
meanings inimă ‘heart’; also, the DRom suflet (< suflu ‘breath, blow’) ‘soul,’ both offering a different perspective on the DRom spiritual associations. Aside from the nineteenth-century questionaire, many Romanian collections of folklore narrate the following myth of the creation of man: The divine chatacter Fârtat/God made the earth, the plants, the rivers, the mountains, and everything else, with the ‘help’ and under the constant criticism of Nefârtat/Devil. One day Nefârtat was complaining that the Earth has no beings to enjoy it, to work it. God agreed with his companion, but did not know what should be done. Ready to trick God, the Devil decided to make a being similar to himself; God guessed his partner’s mischievous thought and waited to see what he will make; the result was a form out of clay, with eyes, nose, mouth, but the mud creature would not move. The devil blew air through a straw into his creature mouth, but without success. After watching his struggle, God invited him to see his own creature: paraphrasing the text, God made the bones from the stone, the flesh from some clay, the blood from dew, the hearing from wind, beauty from the sun, the soul from the Holy Spirit, the mind from the angels, and so He created ‘Man’ in His image; then God said ‘move’ and His creature moved. The Devil was very upset that he could not create a helper for himself to fight against God, and since then, he has tried to get ‘men’ to be his allies (Pamfile, 1916, p. 22). As shown above, the same pattern flesh/clay/earth, bones/stone, blood/ dew/water, eyes/beauty/sun, with a few differences, the IE myth of the creation continued for centuries, resurfacing in the Southeast Europe and Romanian folklores, only, under the Christian doctrine, reversed as God’s creation of man. Careful to comply with the Bible’s teaching of God creating Adam, the first man from clay, the South European people associated the primordial being from the Indo-European heritage, with the Christian primordial man, using the pattern already existing in the arrhaic tradition. The extraordinary vitality of this mythical motif validates the importance of the folk collections in comparative studies of mythology. Although the comparative method could be quite instrumental in this effort, it could also be misused, particularly if there is the tendency to emphasize only the similarities without paying enough attention to
82
A. R. Chelariu
differences. With these observations, it is unanimously accepted that, if applied with prudence, the comparative method could provide relevant results for the study of mythology, particularly in the struggle to understand the social and philosophical systems by which the ancient populations were conducting their life. The Indo-European mythological heritage was the subject of numerous studies from which the Romanian mythology contribution is mostly absent. By comparing the mythical motifs found in Romanian mythology and folklore with motifs from other IE cultural complexes, and observing the particular aspects of Romanian culture can help to establish its position within the European cultural complex.
References Lajoye, P. (2013). Puruṣa. Nouvelle Mythologie Comparée/New Comparative Mythology, 1, 19–54. Lincoln, B. (1986). Myth, Cosmos, and Society: Indo-European Themes of Creation and Destruction. Cambridge MA: Harvard University Press. Ovid. (1955). Metamorphoses (R. Humphries, Trans.). Bloomington: Indiana University Press. Pamfile, T. (1916). Mitologie Românească; Dușmani și Prieteni ai Omului. Bucharest: Socec și C. Sfetea. Pop-Reteganul, I. (1901). Zidirea lumei; Adam și Eva; Originea Sfintei Cruci; Cele 12 vineri. (După tradițiipoporale și manuscrise vechi). Gherla. Editura Tipografiei “Aurora” A. Todoran. The Poetic Edda: Stories of the Norse, Gods and Heroes. Translated by Jackson Crawford. (2015). Indianapolis: Hackett Publishing Company. Turdeanu, E. (1981). Apocryphes slaves et roumains de l’ancien testament. Brill: Leiden.
12 Cosmogony: Fârtat and Nefârtat, the Romanian Twins among the Indo-European Divine Twins
The Romanian cultural complex, as other populations entering into the printed era at a later date, is formed of folktales, fairy tales, and folksongs collected primarily by the end of the nineteenth century and published at the beginning of the twentieth century. These collections offer information about the moral. religious beliefs and social structures of the communities, transmitted through generations from long ago. In recent studies, more and more researchers conclude that certain patterns observed in the classical literature can be recognized in the folk literature as well. Most of the Romanian folklore collections published by the end of the nineteenth century and the beginning of the twentieth century, describe the creation myth in the following sequence of events: In the beginning there was nothing but water; out of nowhere, the water started to wave, and formed a huge foam in shape of a flower; in the middle of this flower appeared a butterfly and a worm; in some versions, the two appeared directly from the foam. The butterfly turned into a handsome young man, who lit up everything around him, and that was God—Fârtat, while the worm turned into the Devil, Nefârtat (non-Fârtat), a creature
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_12
83
84
A. R. Chelariu
in human form but without light; at this time the two were living in harmony. In other versions, the two were floating above the waters or standing on a raft. For a while God/Fârtat was flying above the chaotic waters; not long after, the Devil/Nefârtat began to complain that there was nothing to rest on; to solve the problem Fârtat asked Nefârtat, to dive to the bottom of the sea and fetch some mud (sometimes called ‘earth seed’) but with one condition, to get the mud in God’s name, in Fârtat’s name. Unhappy with the idea of getting mud for his fellow, and wanting to keep it for himself, Nefârtat dives to bring mud only in his own name; for two times he dives to get mud but each time all the mud slips through his fingers and he ends up with nothing in his hands. Only after his third attempt, giving up his desire for his own piece of earth, does Nefârtat concede, and, by using Fârtat’s name, he finally returns to the surface with a handful of mud. God presses the mud flat, large enough for them to lie down upon it, then, because he is tired, he falls asleep, conveying the mythical image of Deus otiosus. Nefârtat quickly takes advantage of the sleeping God, and starts pushing him off the land into the waters, hoping to have the earth only for himself; but the more he pushes Fârtat toward the edge, the earth begins to grow beneath him; the pushing of God in all four directions, toward North, South, East, and West, ends up in the creation of the earth. In a more recent version, it is said that when God woke up, he told the Devil that now he no longer had claim to the earth because, when he pushed God in four directions, he made the cross sign, an interesting Christian insertion (Niculiţǎ-Voronca, [1903] 1998, p. 33). In the Romanian heritage, the cooperation between the Creator- God and his partner is not limited only to the creation of the earth. It includes, among other things, the creation of the sun: when Nefârtat/ Devil complains that people are walking in the dark, God creates the sun. In most of the stories the material world is created at the suggestion or instigation of Nefârtat, whereas Fârtat/God proceeds with the actual act of creation. The jealousy of Nefârtat recalls the Gnostic beliefs that matter was formed by evil angels, although the Romanian thought does not regard God’s partner with any actual creative powers. The extent to which the first Christian theological disagreements would influence the Romanian creation myth appears to be quite limited.
12 Cosmogony: Fârtat and Nefârtat, the Romanian…
85
The creation myth, as described here, is also found in the surrounding regions of Romania. A detailed acoount of this myth in Slavic mythology, particularly on the duality of the creation myth, can be found in the work of M. P. Dragomanov, translated and published by E. W. Count in 1961. In his collection, Dragomanov recorded some important versions of this myth from Bulgarian and Ukrainian folklore; the Bulgarian version of the cosmogonic legend was recorded in 1868 among the Bulgarian Diaspora in Bessarabia, and it narrates the following sequence of events: In the beginning there was only water and God and Devil lived together; ‘Once the Lord said to the devil: Let us make the earth and people.’ Since the devil didn’t know where to get dirt, God sends him diving into the water, with the specific request to get it ‘With God’s power and mine!’; the devil disregarding the instruction from God, says ‘With my power and God’s power’, thus he doesn’t reach the bottom; the same happens the second time, and only the third time he obeys and brings some dirt under his fingernails to the surface; from it God made a little dry land. ‘Then the devil got up, took the Lord in his hands, and started for the water, to throw him in.’ As he walked towards the water the earth grew underneath, and so on in three directions; after that he lay down and slept a bit when ’it occurred to him that there was still a fourth direction,’ but still he could not reach the water. (Dragomanov, 1961, p. 2)
The Ukrainian version in Dragomanov’s text does not record the primordial waters, and the creation starts as God thought of it. He tells the oldest angel, Satanail: “What say you to our going and creating the world?” “And so they went to the sea”, to the bottom of which God sent Satanail to fetch sand, instructing him to complete the task in his (God’s) name. Provoked by jealously, Satanail wanted to add his own name, to have earth in both names, but when he came out of the water, the sand was washed through his fingers. Only the third time did he succeed by conceding to God’s request to retrieve the sand in his (God’s) name. The devil then licked his hand to make some earth for himself. God blessed the earth in four directions and the ground began to grow, and the sand in devil’s mouth grew also, then God said, “Spit, Satanail!” and where he spat, there grew mountains; and
86
A. R. Chelariu
where he hawked, there grew rocks. God decided then that they should both rest as the ground beneath them grew; but the devil still wanting the earth for himself, picked God up and ran towards the water in all four quarters of the earth, day and night, without ever reaching the end. He wakes up God and tells him to bestow the earth but God says he already did it in all four directions. (Dragomanov, 1961, p. 19)
From these versions we can see that the main events are very similar, particularly the duality of creation, the diving motif, the three trials to succeed in getting mud, and the blessing/dedication in God’s name. The few differences observed may help separate what is particular to the Romanian myth: the motif of the chaotic waters is common to the Romanian and the Bulgarian versions, but is absent in the Dragomanov record. In these Slavic versions the two divine characters God/Lord and Devil, addressed with the current Christian names, are in existence from the beginning, whereas in the Romanian version they appear above the chaotic waters, on a foam in the shape of a flower, as a worm and a butterfly. An exception from this sequence of events is found in a variant from Moldova, an area closer to Ukraine, where the two divinities do not appear as butterfly and worm. Such differences may or may not have relevance, depending on the accuracy with which the collecting of data was executed. In both Bulgarian and Ukrainian stories, God decides to create the earth, but in the Romanian story the urge to create the earth is expressed by Nefârtat/Devil, who complains that there is nothing to rest on. Another interesting motif that is absent in the Romanian data, but is found in Ukraine, also in Latvia, and Little Russia, is the Devil hiding some mud in his mouth; the mud starts to grow, forcing him to spit it out, thus creating the mountains (Dragomanov, 1961). The cosmic dive motif, in which an entity submerges to the bottom of the primordial waters to get mud for creating the dry land, is to be found in numerous myths from Siberia to Central Asia, India, and North America. In versions from North America, the diver or divers are often birds, and in Eurasian versions usually there is a single diver, often a bird, suggesting a shamanic context, based on the belief that the soul can take shape of a bird. Further references to the diving motif from around the world are offered by Mircea Eliade (1972, p. 76) in his opinion, the
12 Cosmogony: Fârtat and Nefârtat, the Romanian…
87
Northern Asia is the most likely place of origin of the diving motif, where the creator, in shape of a bird, dives into the waters, which points to a single creator, not a dualistic creation myth. A similar story of the non- dual cosmic dive is found in an archaic Indian myth, where only one character is diving: a God dives into the primordial waters and brings up clay; in the Brāhmanas, the god Prajapati dives in the shape of a boar, and in the Rāmayana this act is performed by Brahma. A form of duality is found in a myth from Altai, where the pair is a god and a man: God sends the man to dive for mud, who he hides some in his mouth, but when the earth starts to grow in his mouth the man must spit it out, creating the swamps; this episode recalls the Latvian, Ukrainian, and Little Russia stories. In Mongolia, the pair Očirvani and Tsagan Sukurty climb down from the sky on top of the Primordial Waters. Tsagan Sukurty dives for mud and since he wants to get mud in his own name, it is all washed away through his fingers, as in the Romanian version. When he dives the second time, he takes it in Očirvani’s name and the earth is created. There is a second version of the story in which the two gods spread the earth on a turtle’s back and then fall asleep; while they sleep the Devil comes and starts pushing them off the earth into the water, but the earth grows beneath the gods, as in the Romanian myth. In both Mongolian versions the Devil appears after the earth’s creation is done, with no explanation from where (Eliade, 1994, vol. 3, pp. 14–16). The antagonistic duality transpired gradually into the second partner as an auxiliary entity, the devil, under the Christian influence. Many researchers thought that the adversarial duality of the divinities involved in this myth had its origin in the Bogomilic tradition. In challenging this hypothesis, M. Eliade brings up a few important points: This myth is not to be found in any Bogomilic text; more so, although an important center for Bogomilism up to the fifteenth century, Bosnian folklore does not present traces of Satanail extracting dirt from water under God’s instruction, nor can it be found in Serbia, or Herzegovina. However, the story is found in various forms in Ukraine, Western Russia, and the Baltic region, areas where the Bogomil sect never reached. Focusing his discussion on the diving theme, M. Eliade refers to the main characters in the myth as God and Satan, not as Fârtat and Nefârtat,
88
A. R. Chelariu
the names noted in the Romanian folk collections, the author insisting mostly on the antagonistic duality of the divine pair (Eliade, 1972, p. 92): “A priori, it is not impossible that certain ‘dualistic’ beliefs disseminated in the Balkans and the Carpatho-Danubian regions represent vestiges of religious beliefs from the Thracian-Scythian substratum”, therefore, it is probable that “in the folk strata in which the myth was current” there were images and symbols employed by the storyteller to impress the audience of the “mysterious structure of the divinity.”
The author concludes, together with other researchers, that this myth presents an extremely archaic narrative plot. The religious dualism, and especially the antagonistic dualism, considered of Iranian origin, will disperse through Gnostic and Manicheistic movements, reaching, via Paulicianism, the Bogomilism doctrine. But, as Eliade concludes, the Romanian cosmogonic myth is not of a Bogomilic influence. To clarify this aspect, the Romanian scholar Ion P. Culianu’s in-depth study on Gnosticism could be of assistance. Discussing the subject of cosmogonic duality, Culianu (2002, pp. 21–22) asserts that in dualistic religions the two principiae, coherent or not, create the existence, real or apparent, of all that is manifest in the world. The Gnostic texts recognize the first ‘primitive’ stage of creation, in which the quarrel between the creators does not imply their evil nature, nor the evil nature of cosmos (Culianu, 2002, p. 73), a fact suggesting that the Gnostics were initially influences by myths from the folklore of their time, free of the dual antagonisms of a later date. This ‘primitive’ stage with the absence of the good and evil antagonism of the creators is reflected in the Romanian myth, which indicates that the Romanian creation myth is part of the archaic folk data and thus could not have been influenced by Gnosticism. The Western dualism, as discussed by Culianu, the anticosmism—the world is bad—the body is corrupt, the encratism or ascetism, all the way to the refusal of procreation, are not present in the Romanian cosmogonic myth. Furthermore, Culianu (2002, p. 171) states that the only correct description of Gnosticism is represented by a group of dualist systems,
12 Cosmogony: Fârtat and Nefârtat, the Romanian…
89
involving simultaneously the myth of the female trickster and that of the male, a relation absent in the Romanian myth. The two creators from the Romanian myth are presented as cooperating in the creation of the world, proving once more that Gnosticism or Bogomilism are not reflected in this archaic myth. For the purpose of this work, I will limit the inquiry to the pair Fârtat and Nefârtat in relationship with the Indo-European Divine Twins, with particular attention to the Iranian pair Avirdada ‘lord of the waters’, and Amirdada, ‘lord of the trees’ (Darmesteter, 1875). The Indo-European creation or foundation myth centers on the primeval twins. Ancient mythologies record the pair divinities as the Indic Yama, the Avestan Yima, Latvian jùmis ‘pair’, Old Irish emon ‘twin’ (Emain Machae ‘the twins of Macha’), Latin Remus and his twin brother Romulus, to which we can add the Romanian pair Fârtat and Nefârtat. The Proto-Indo-European primeval creators, the progenitors of mankind, were *Man (Indic Manu, German Mannus) and *Twin, the latter of which was sacrificed and carved up by his brother to produce mankind, a dualism reflected by the twins myths expressing the binary opposition as an “…underlying structure of the Indo-European ideology” (Mallory, 1996, p. 140). The dual structure encountered in the Indo- European myth transcends on social level into “the twins at the beginning of time—which is endowed with a typical I-E content—one brother is the first priest, the other the first king” (Lincoln, 1975, p. 135). The subject of twins in the Indo-European mythology and their significance was extensively discussed, among many others, by Dumezil (1988), beginning with the most famous pair: Remus and Romulus, gemino Quirinos, or Rhomos and Rhomulos, as the Greek writers called them. Their proto-Roman names were ‘twin’ and ‘man’ *Yemos and *Wiros, a clear parallel with the Vedic couple Yama ‘twin’ and his brother Manu ‘man’, cognate with the Norse primordial ‘twin’ Ymir. Romulus is associated with the cults of Jupiter Feretrius and Jupiter Stator, divinities situated on the same opposition as the one between Romulus and his successor Numa, one is the terrible, the warlike, the violent, and the other is the ruler, the religious figure, the jurist. The same opposition is expressed by another pair in the Roman pantheon, between Dius Fidius and Jupiter Summanus, one diurnal, the other nocturnal. One may wonder what
90
A. R. Chelariu
could these pairs have in common with Janus, the two-faced god prefigured above doors, the one who gave his name to the month of the beginning of the year, January, the border between the old and the New Year. Similar opposition is found between the Roman Flamines and the Hindu Brahmans, who are “the guardians of the sacred order,” and the “Luperci and Gandharva”, who are “the agents of a no less sacred disorder”. Similar opposition can be found between Uranus, as symbol of the night sky, monstrous and chaotic, and Zeus, symbol of the luminous sky, the magician with his thunderbolt, although a less valid pair (Dumezil, 1988, p. 34). Mitra and Varuna follow the same pattern: Varuna is the warlike figure, the terrible one, who “creates and modifies forms, and makes the laws of nature” (Dumezil, 1988, p. 67), whose function is to protect the oath, and may be “related to an old Indic custom, that of swearing by water” (Puhvel, 1987, p. 49), if his name relates to the PIE *wod-ar meaning ‘water’, as in the Skt. var, vari ‘water’, although it is mainly considered that it comes from the root PIE *uer-, *var- meaning ‘to cover, protect, bind’. His counterpart, Mitra, is known as ‘the friend’, but a friendship which implies a contract; his name comes from the root *mei- ‘to exchange’, as in Skt. mayate ‘he exchanges’, or Lat. munus ‘gift, obligation’, or Old Slavic mena ‘change, contract’, and miru ‘peace, cosmos’. For Dumezil “Mitra and Varuna form a unit”, and “are, to some extent, synonymous. This collaboration is made possible, however, only by a congenital opposition: Varuna is also to be defined as the contrary of Mitra” (Dumezil, 1988, p. 67). Varuna represents the magical powers, the darkness, whereas Mitra is the light, the order and the justice. But these oppositions are complementary, and do not imply any incompatibility. In Tacitus’s (2009 Germania, Chap. 2) we find the following reference to the ancient Germans, who in their songs Celebrant carminibus antiquis, quod unum apud illos memoriae et annalium genus est, Tuistonem deum terra editum. ei filium Mannum originem gentis conditoresque Manno tres filios adsignant,.. (“they celebrate an earth-born god, Tuisco (‘twin’- for that is the proper etymology of Tuisco) and his son Mannus, as the origin of their race, as their founders”). In the Scandinavian world, there is the pair of Odhinn and Ullr or Tyr, alongside Wodhanaz and Tiwaz or Tiuz,
12 Cosmogony: Fârtat and Nefârtat, the Romanian…
91
and the Germans Wotan and Ziu. Ullr, Gothic wulthus has the meaning of ‘majestic glory’, “a divine person, whose activity consists in a cosmic brilliance” (Dumezil, 1988, p. 123). By contrast, Odhin, deriving from *odhr- the Scandinavian form of German Wut ‘fury’, Gothic woths, the frenetic agitation, and as an adjective it means‘rapid movement, agitation, storm, terrible, furious’ (ibid. 124), characteristics similar to those of Romulus and his warriors, the wild riders, luperci, and also of Varuna. Another Germanic pair that should be added to our discussion is that of Fjorgynn and his female counterpart, Fjorgyn, possibly reminiscent of the archaic sky-earth couple, whose names seem related to Old Icelandic fjor ‘life’, Old English feorh ‘life, living person’, Old Fris frech, Old Swedish, OHG fer(a)h ‘soul, spirit, life’, CrimGo fers ‘man’, found also in Old Irish fer ‘man’. Next to the Vedic pair, Mitra and Varuna, there is the Avestan counterparts, Mithra-Ahura, that we could trace to the Iranian couple Ohrmazd (< Av. Ahura Mazda, the ‘Knowing Lord’) and Ahriman (< Av. Aŋra Mainyu, the ‘Evil Spirit’), example of the Zoroastrian obsession for duality, the good Vayu counterbalanced by the evil one (Dumezil, 1988, p. 67). In his article on the Indo-European creation myth Bruce Lincoln (1975, p. 121) refers to a fragment from Yaçna 30.4: “And when these two spirits first met [the good and evil spirits], they instituted Life (gaem) and non-life…,” where the Avestan gaya- stands for ‘life’, that in the Younger Avesta is often associated with maratan-, “mortal”, thus the name given to the first mortal man, Gaya maratan, found in the Pahlavi (Middle Persian) of our Bundahisn text as Gayōmart. The good and evil, life–death dualities found in the Iranian mythology brings us to a more archaic pair of interest in this discussion, that of Haurvatat, lord of the waters, and Ameretat, lord of the forests. Haurvatat comes from the Zend word haurva, Skt. sarva, which means ‘guardian, the keeper, whole and wholeness, universality’, in other words, the god who produces everything. Ameretat connects to the word amereta, a-mereta meaning ‘non-mortus’ better yet ‘immortality’, plus the suffix -tat used to form an abstract noun. (Darmesteter, 1875, p. 16) Considering the meanings of these names, as they are used in various Avestan text of prayers, Haurvatat and Ameretat appear as the gods who give abundance and immortality, lords of waters and plants; Haurvatat relates to the meaning
92
A. R. Chelariu
of ‘waters, wholeness, health’, corresponding in Vedic to sarvatat and Greek holotes, ‘wholeness’, while Ameretat relates to ‘plants, long life, immortality’, in Vedic to amrtatvam and Greek ambrosia (Puhvel, 1987, p. 98). Water and plants are sources of health and immortality as it is found in myths related to Soma or the Fountain of Youth: “This belief is Indo-European; it existed already in the period of unity, in other words, the time when the ancestors of Aryans from Asia and those from Europe were living together, having the same religion, the same language” (Darmesteter, 1875, p. 71). In a Sanskrit translation of “Yaçna” the name Haurvatat appears as Avirdada, apam patim ‘lord of the waters’, and Amirdada, vanaspatinam patim ‘lord of the trees’. In the Romanian cosmogonic myth, the divine twins represent two different but interrelated features: Fârtat, is associated to the light and primordial waters, guardian of the order, the peace keeper, the luminous sky figure, while the other, Nefârtat, is associated to the earth and plants, he is the wild and terrible one, swift and ready to use tricks, in the end, the evil one. Hence, the Romanian myth might reflect the old Indo-European principle, the bi-unity of the divine, coincidentia oppositorum (Eliade, 1972), as West states “One may say that bipolarity (not trifunctionality) is the fundamental structuring principle of Indo-European thought” (West, 2007, p. 100). The name connotations of these two divine characters from the Romanian creation myth could help in the effort to understand their origin and function. In Romanian language, the word ‘Fârtat’ (vocative Fârtate!) had lost its divine associations, reduced to the use of frate de cruce, ‘blood brother,’ ‘comrade’, a meaning that kept the initial sense, that of one of the Divine Twins. More so, in the Romanian language and folklore the word ‘fârtat’ connoting ‘blood brother’, is linked to a specific ritual performed between two young men who wanted to turn their friendship into brotherhood: the two men would cut a small incision in their forearms and juxtapose the wounds in a cross, after which they would call each other fârtate [voc.], a ritual called ‘becoming fraţi de cruce’ ‘cross brothers’, young men that belong to the same clan, group, etc. Similar fraternity rituals continued until the Middle Ages, for example, in Greece these rites were known as Adelphopoiesis ‘fraternity union’, in
12 Cosmogony: Fârtat and Nefârtat, the Romanian…
93
Slavic languages as pobratimstvo ‘fraternity union’, words not related to the Latin frater, or the name of the divinity Fârtat The phonetic similarity between the Romanian word frate ‘brother’ and Fârtate led to the general consensus among linguists that these two words, Fârtate and frate ‘brother’ originate in the Latin form ‘frater’. Such conclusion presents some difficulties, most importantly, because the Roman mythology does not record a creation myth similar to the Romanian and Slavic one, or a god with a name close to Fârtat. As mentioned above, the myth of the Roman twins, Romulus and Remus, is not about the creation of the earth, but the foundation of a town/community. Assuming that after the Roman conquering of Dacia, the local population borrowed the Latin form ‘frater’ to designate two separate concepts, that of ‘brother’, DRom frate, and, phonetically modified that of fârtat, the god of light, earth-creator, a name absent in Roman mythology. If we take a look at the word ‘brother’ in all the Indo-European languages, such as: Skt. bhratar, Av. bratar, Phrygyan braterais, Grk-Lat frater, Old Ir brathair, Got. brothar, Lith. broterelis, Old Slv. brati, Toch. pracar, pratri, we can easily observe that the group of phonemes bra/fra/ pra remains constant in all the Indo-European languages, which makes the form ‘Fârtat’ as coming from ‘frate’ difficult to accept. To this argument it should be added the oldest document of Romanian language “torna, torna fratre” in Theophylactus Simocatta, (1986 Histories, c. 630 CE), that proves the existence in language of the form ‘frate’. In the Romanian language there are no other examples of the development of the sound group fra > fâr, for example, Lat frenum > DRom frâu ‘rein’. A more appropriate conclusion seems to be that Fârtat is a divinity in his own right, together with the Indian Dyauspitar, Thraco-Phrigian Zeus Pappos, the Roman Dius pater, the Greek Zeus Pater, the Celtic Dispater, etc., all compound words, expressing the concepts of sacred sky and paternity, that of father/master/lord of the sky/day light. Likewise, the Romanian Fârtat could be analyzed as a compound word, formed from the PIE *wihxrós > DRom fâr-, OIr fer ‘man’, Irish fir ‘man’, Lat vir ‘man, husband’; OE wer ‘man, husband’, in the same family with Avirdada, Tyr, Virinius, Fjorgynn, expressing the same meaning: life, water, fertility, man. The concept of ‘life’ as seen in the Germanic Fjorgynn, is evident in the Old Icelandic fjor ‘life’, Old Engilsh feorh ‘life,
94
A. R. Chelariu
living person,’ Old Frisian frech, Old Swedish, OHG fer(a)h ‘soul, spirit, life’, CrimGo fers ‘man’, and also in the Romanian noun fior, with the sense of ‘light chill, tremor’, in expressions like fiorul vieții ‘the thrill of life’. It is possible to assume that the word Fârtat is formed in the same way as the other Indo-European divinities, from the PIE *wihxrós ‘life, fertility, man’ and the PIE *t-at-, meaning lord/protector/father, the great creator, or with the suffix -tat used to create abstract nouns, as mentioned above. It may also be possible to relate the root Fâr- to the PIE *wel ‘see’, or, *wel ‘die’, as in the mythic characters: Lithuanian Velnias Russian Velesu, Volosu, Sanskrit Varuna, Gaulish Vellaunos, Hitit Walis, and especially the Armenian geł ‘visible, beauty’ and An-geł ‘invisible, god of underworld’ (Jakobson, 1969 in Petrosyan, 2016, p. 80), an internal development similar to the Romanian Ne-Fârtat (non-light). Further, the Indo-European “ability to crate negative compounds with the prefix *ṇ- made it easy to form polar expressions of the type ‘X and non-X: amrtam mártiyam ca ‘immortal and mortal’ (RV 1.35.2)…” (West, 2007, p. 101), hence the formation of the opposed pair Fârtat Nefârtat certifies their relation to the Divine Twins as an expression of the Indo-European duality structure. The images and actions of the butterfly Fârtat and the worm Nefârtat in this myth are metaphoric expressions of the divine, manifested as water/light and land/darkness intertwined symbolically into the cycle of life and transformation. The twin divinities portrayed in the Romanian myth, one signifying the waters, the other, the earth, and implicitly, the plants, one luminous, the other, dark, one kind the other terrible, one representing universal totality, the other immortality of the cycle of life, united in an inseparable relation, who enter into ‘conflict’ when the earth is being created. Only when Nefârtat accepts his role in the duality, gives up his selfishness, and admits to act in his companion’s name, to bring mud into the Light, the harmony is established, and the earth can be created. Fârtat cannot create anything alone, and Nefârtat loses the mud through his fingers, since he does not embody the Life principle of primordial waters; he is Ne-Fârtat (non-light), lightless matter and primordial action—Satan in Christian era, as the Almighty God has to prevail.
12 Cosmogony: Fârtat and Nefârtat, the Romanian…
95
In another Romanian story (mentioned above) Satan wants to create his own man: he makes a mud figure, but he could not make it move, the creature did not breathe, as the vital force was granted only by Fârtat. The Romanian creation myth testifies to an archaic Romanian belief, that only through harmonious unity could the earth’s creation be achieved, and through a balanced duality can the act of world creation take place. The essential unity of a worm and a butterfly, one emerging from the other, expresses the Romanian understanding of the Life phenomenon, as the cycle of transformation from the ground, the domain of plants and worms, into the light, conveyed by the beauty of a butterfly. Once the Christian beliefs settled in, the Romanian divine twins Fârtat/ Nefârtat, metaphoric representations of complementary forces, collapsed into the antagonistic dualities of God and Devil, as in the Bulgarian and Ukrainian versions, demonstrating the extraordinary vitality of this divine pair. The complexity of this myth and its continuity, adapted into the Christian beliefs, endures through times, revealing an archaic religious system, possibly part of the Indo-European religious substrata, or even older.
References Culianu, I. P. (2002). Gnosele dualiste ale Occidentului, Istorie și Mituri. Bucharest: Polirom. Darmesteter, J. (1875). Haurvatat et Ameretat; essai sur la mythologie de l’Avesta. Paris: Librairie A. Franck. Dragomanov, M. P. (1961). Notes on the Slavic Religio-Ethical Legends: The Dualistic Creation of the World (E. W. Count, Trans.). Bloomington: Indiana University Publications Russian and East European Series, vol. 23. Indiana University. Dumezil, G. (1988). Mitra-Varun: An Essay on Two Indo-European Representations of Sovereignt. New York: Zone Books. Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. Chicago: University of Chicago Press. Eliade, M. (1994). Istoria Credintelor si Ideilor Religioase (3 vols.). Universitas.
96
A. R. Chelariu
Lincoln, B. (1975). The Indo-European Myth of Creation. History of Religions, 15(2, Nov.), 121–145. Chicago: University of Chicago Press. Mallory, J. P. ([1989] 1996). In Search of the Indo-Europeans, Language, Archaeology and Myth. London: Thame and Hudson Ltd. Niculiţǎ-Voronca, E. ([1903] 1998). Datinele şi Credinţele Poporului Român. Bucharest: Editura Saeculum. Petrosyan, A. Y. (2016, Spring/Summer). Indo-European *wel- in Armenian Mythology. Journal of Indo-European Studies, 44(1&2), 129. Washington DC: Institute for the Studies of Man. Puhvel, J. (1987). Comparative Mythology. Baltimore: John Hopkins University Press. Tacitus. ([1948] 2009). Agricola and Germania (H. Mattingly, Trans.; revised with an Introduction and Notes by J. B. Rives). New York: Penguin. Theophylact Simocatta. (1986). The History of Theophylact Simocatta (M. Whitby & M. Whitby, English Trans.). Claredon Press. West, M. L. (2007). Indo-European Poetry and Myth. Oxford: Oxford University Press.
13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains in Romanian Beliefs
In her collection of folklore published in 1903, Elena Niculiţǎ-Voronca records well-preserved Romanian traditions and other beliefs about the world creation. Besides attributing the earth creation to the divine twins, Romanians believed that the earth was a feminine entity, mother of mankind, who gave birth to all and feeds all. Continuing and reafirming the importance of feminine deities in prehistory, Romanians believed that ‘Mother Earth’ was a living being, with a head, a heart, and waters as her blood. She must be respected for her sacrifices for people, who have to pray to her well-being. To satisfy her certain needs, special interdictions were imposed when working the fields, such as, not to work it in the afternoon or at night because Mother Earth is resting. When angry at people she trembles, or sends diseases, wars, plagues, and may swallow many inhabitants. It was believed that in the beginning, the Earth was as white as the light, but after the first man was killed (the Cain and Abel Bible story), she started to turn black. One legend says that in the beginning the flat Earth floated above the primordial waters, resting on four pillars supported by four large fish floating in the water, each with its own name: the one from the Sunrise— the Beginner—because it starts what the others have to do; the one from © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_13
97
98
A. R. Chelariu
the Sunset—the Obedient—because it gets the advise from the Beginner; the one from the Middle-day—the Guider of the Sun, and the one from the Midnight—the Sleepy—because it sleeps all day, but at night guides the Moon and the Stars. When the Fish move, the Earth shakes, thus earthquakes occur. Beliefs about the number of these pillars vary from one to sixteen, and are made of silver, iron, wax, or stone, materials that could corrode in time, but are restored by Christmas songs and prayers (Olinescu, 1944, p. 49). In another story we are told that God was not so happy with a flat Earth, therefore He started to ask around what to do. Because the hedgehog was famous for his wisdom, He sent the bee to ask for help; the hedgehog did not wish to give a solution, arguing that God was supposed to know everything; when the bee pretended to leave, the hedgehog mumbled to himself how simple would be for God to tie the Earth’s waist with a belt and make the mountains. Since then the bee is blessed (Olinescu, 1944, p. 50). In the ancient world, the bees and the wax were believed to have purifying powers, and bees were attending the cults of Demeter, Persephone, and Artemis. Another story about the hedgehog contribution to creation relates that the earth, for an unknown reason, was made larger than the sky that could not cover the whole earth; following the hedgehog idea, God wrinkles the earth, and so the mountains and the valleys came about (Pamfile, 1915). In the North part of the country, Pamfile records a story about a time before the sky was created, and the sun and the moon were driving their carts on earth, and the heat was distressing all the people, who prayed to God to make a sky on which the celestial stars could drive their carts; seven angels went under the earth, and got precious stones and crystals to make the sky, in Romanian bolta de cleștar (< Latin crystallus) ‘the crystal volt’ or ‘heavenly vault’, that was resting on seven pillars; after all was done, the sun, the moon, and God, all went beyond the volt, into the deep sky. Similarly, in The Prose Edda (Snorri, 2005) there were four dwarfs holding up the four corners of the sky. The quartz sky is known in other Indo-European cultures, as, for example, Akmon, the father of Ouranos, that personified Heaven; his name is cognate to the Indo- European *h2ek’mon meaning both ‘stone’ and “heaven”, found in Vedic ás’man-, Avestan asman-, Lithuanian akmuõ, leading to “the notion of
13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains…
99
stone sky as part of the Indo-European world view” (West, 2007, p. 342). The PIE root *haek̑ ‘sharp, pointed’ relates to a few DRom reflexes: v. ascuţi ‘to sharpen’, acăța, agăța ‘to hang’. God’s innocence and weariness, his deep sleep after creating the Earth (Deus otiosus), and his need to create the mountains and the valleys out of necessity, with the help from some of the most insignificant animals, are characteristics of the Romanian and Southeast European cosmogony myths. The divinity in these folkloric data expresses the transition from a prehistoric ‘dual’ entities, the twins divinities, to a monotheistic one. The Romanian Dumnezeu is a God walking on Earth, never alone, always accompanied by St. Peter, who, because of men’s depravity, retired to the heavens. Many of the Daco-Romanian forms naming God’s creations, the Earth, the mountain, the sky, the rivers, are concepts with roots in the reconstructed by Pokorny (1959) and Mallory-Adams (MA): (IEW 140–141) *bherg̑h, *bhereĝh- ‘high, hill, mountain’ > DRom reflexes. Bârzava, (Dacian Bersobis, Bersovia), and hydronym and toponyms Bârsa; also mountain region Țara Bârsei (the Country of Bârsa); anthroponyms Bârsan, etc., with the following cognates: MIr brī ‘hill’; NE barrow; NHG Berg ‘mountain’; Rus béreg ‘river-bank’; Av bərəz- hill’; Lith place-name Beržorai; personal names (PN): Thrac. Βεργούλη, Maced. Βέργα, Illyr. Berginium (Bruttium: Bergae), Lig. Bergomum, Celto-Lig. Bergusia, Hisp. Bergantia. Another example is the PIE IEW (938–947) (s)ker-4, (s)kerǝ-, (s)krē‘to cut’ reflected in the DRom Munții Carpați ‘Carpathian Mountains’, Thrac Καρπάτης ὄρος ‘Carpathian Mountains’, the tribe Carpi. , with cognates in Alb. karpë, karmë (*korp-n-) ‘rock, cliff’, Grk κρώπιον ‘sickle,’ Lat carpō, −ere ‘pluck’, carpinus ‘hornbeam’, Hit karpina- ‘a tree hornbeam hornbeam’ The etymology for the toponym Moldova < PIE (MA) *mḷdho/eha ‘clay’, although problematic, seems attractive. The DRom glod could be a good example of PIE root retention, since the Latin form gluten ‘glue’ presents the voiceless dental: PIE (MA) *gloiwos ‘clay’> DRom n. glod, ‘clay, mud, clod of earth’; v. glodi ‘to. rub, irritate’; adj. gloduros ‘muddy, bumpy’; also, toponyms Glodu—village in Muntenia, with the following c. ognates.: Grk gloiós ‘clay’; Lat gluten ‘glue’; NE clod.
100
A. R. Chelariu
The word for ‘small cliff (by water), shore’, DRom mal, and. Maldăr pl, maldăre ‘pile, heap, moumd’ has its roots in PIE (MA) *melh3 -,, melǝ-, mlö- ‘to rise up, rising land’, found mainly in DRom and Albanian, is verified in the archaic Thrac-Gaetae toponyms Malva, Malvensis’, with cognates in Alb mal’ ‘mountain’ (Holm: JIES vol 39, no. 1–2, 2011, p. 72), Celt malo- ‘rising, prominent’ (Falileyev et al., 2010), Ltv. mala f. ‘edge, bank, border, shore, region’; Lith. lūg-mala ‘height of edge’ (MacBain, 1982: Gaelic mol, mal ‘shingly beach’; from Norse möl, g. malar ‘pebbles, bed of pebbles on the beach’; root mel, grind.). The DRom noun in neuter pământ ‘earth’ is explained in DEX.RO (n.d.) through the Latin pavimentum (pavio, −ire ‘beat’ and the suffix mentum) ‘pavement,’ a solution phonetically and even semantically plausible; yet, the concept of Mother Earth as a powerful feminine character in Romanian folklore, carrying many interdictions and sacred beliefs, gives reason to pause and consider the transfer of meaning and connotations to a Latin concept in neuter, related to pavement. The contamination is definitely recent, perhaps revealing a merger between the Latin noun and an archaic DRom form, perhaps related to the PIE (IEW 146–150) *bheu-, bheuǝ- (bhuā-, bhuē-): bhōu-: bhū- ‘to be; to grow’, with which the following cognates the DRom seems related: OInd bhū- f. ‘earth, world’, bhūmī, bhūmih-, Av būmī-, Pers būm ‘earth’, OInd bhūman- n. ‘earth, world, being’ (= Gk. φῦμα), bhūmanm. ‘fullness, wealth, bulk, mass, wealth’; maybe Alb (*bēam) bimë ‘plant’. The general use of the DRom pământ in association with the notion of Mother Earth ‘Mama Pământ’ a feminine noun Mama, next to a neuter Pământ, pleads to a later merger with the Latin form. It should be mentioned here the few similarities between the Friulian palment, Logudorese pamentu ‘pavement, floor’, Welsh palmant, paement ‘floor’ and the DRom pământ, ARom pimintu, MeglRom pimint, IstroRom pemint, all ‘earth, ground’, while the Romance languages prefered the Latin ‘terra’ and ‘terra mater’, and the Slavic population use ‘zemlia’. This situation presents a linguistic problem that still awaits a solution. The similarity between the DRom pământ and the Logudorese
13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains…
101
dialect pamentu is not unique, as the form limba, and the DRom limba ‘language, tangue’, both from the Latin lingua ‘language’ opens up for more explorations. The concept of ‘hearth’ DRom vatră, an essential part of household, is also of Indo-European heritage: *h2ehx-tr-eha ‘hearth’, IEW 69 *āt(e)r- ‘fire, *blow the fire,’ Alb vatër / votër [f ] ‘hearth, fireplace’ ProtoAlb. *a̅t(e)ra̅ {1} (AE 410) Albanian adds an initial ‘v’ to words of ancient origin that begin with a vowel (common Alb. prothetic v- before initial bare vowels—proof of ancient laryngeal ḫ) as in vesh from Indo-European, verbër from Latin orbus; Byzantine Grk βᾰ́θρον ‘base, pedestal, scaffold’. Pokorny IEW 69—Note: “Alb Tosk vatra, Gheg votër ‘hearth’ proves that Slavs borrowed prothetic v- before bare initial vowels from Illyrian This phonetic mutation in Albanian took place before the invasion of Slavs into the Balkans because Alb and Rom share the same cognate.” The DRom form apă ‘water’ traditionally explained by Lat aqua < PIE *haekweha- ‘water’ may have a better solution in MA *h2eP- ‘living water’, particularly if considering the Dacian river name Apos; the cognates are: Skt āp- ‘water’; OIr ab ‘river’; OPrus ape; Hit hāpa- ‘river’; Toch AB āp ‘water, river’, Lat amnis ‘river’; OHG river names in -affa-. Note Sard abba ‘water’.
The PIE concept of ‘god’ is reconstructed in the root *haénsus ‘god’, that continued in the DRom noun îns, ins ‘a person, any person, an individual’, a form instrumental in the development of the DRom pronouns m. însu-mi, fem însă-mi ‘myself ’, m. însu-ți f. însă-ţi ‘yourself ’, m. însu-și f. însă-şi ‘him/herself ’, etc.; the PIE root has cognates in Latin ens, entis ‘being’, a loan from the Greek ειμι ‘I am’; Matasovič (2018) reconstructed the form *h2nsu- with cognates Hittite haššu- ‘king’, OIc. áss ‘a kind of god’, Skr. ásura- ‘a kind of god’, Avestan ahu- ‘lord’, Greek ἑνός θεοῦ (henos theou, ‘of one god’, and it can be added here the Hurrian eni ‘god’. The continuation of this form in the Daco-Romanian language, with the pronoun developments, could indicate certain religious understanding of the relationship between man as an ins ‘individual’ and the divine. The Romanian word for God—Dumnezeu, as a religious concept, carries connotations associated with adoration, together with interdictions and fear. The concept of God in any language could tell us important
102
A. R. Chelariu
facts about the people retaining it. For example, the etymology for the Romanian form Dumnezeu is established as based on the Latin Christian formula Dominus Deus. From the historical perspective, this explanation starts from the assumption that sometime in the beginning of the second century CE, and thereafter, the conquering Roman soldiers and colonists invading Dacia, a heterogeneous group of Roman citizens and veterans coming from all over the empire, did use this formula, Dominus Deus, in their prayers. We should note in this context that Latin gods’ names, such as Jove or Jupiter, Mars, Ares, Minerva, Junona, are not to be found in Romanian cultic language, or in other folk data. The only exception is the names for the days of the week, luni, marți, miercuri, joi, vinery, sâmbătă, duminică, similar to all the Romance languages. At this juncture, one may wonder what religious beliefs the ‘Roman’ soldiers encountered in the new conquered land. As Herodotus tells us (see above), the Getae, “the bravest of the Thracians and the most just”, worshipped Zalmoxis, or Zamolxis, the prophet who went underground for three years and returned to teach his people about the ascetic way of life, becoming famous for his wisdom. If so, we could assume that these tribes were more inclined to monotheistic beliefs. The very limited list of Dacian-Getae words, and even lesser references to deities, leads us to wonder, what word would have been used by these people for the concept of God before Roman conquest? What we know, though, is that in the Romanian language the fundamental concepts of worship, especially the Christian beliefs, are of Latin descent, while the organizational terminology of the Church presents a strong Greek and Slavic influence, which, again, shows the difference in time periods of those influences, with the Slavic one taking place after the eleventh century. V. Pârvan, the renowned Romanian historian from the turn of the twentieth century, discussed this subject in his work “Contributii epigrafice la istoria Creştinismului daco-roman” (Epigraphic Contributions to the History of Daco-Romanian Christianity), Bucharest, 1911, which may help shed some light on the subject. He used as his base of argument the number of sarcophagi inscriptions found on the territories from middle to lower Danube, dated from the second to fifth centuries. Unfortunately, many of the so-called pagan, or pre-Christian monument inscriptions were destroyed from the fourth century onward, after
13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains…
103
Christianity was declared a state religion in Constantinople, concomitant with the prohibition of pagan worship. As the author explains, the first Christian centers are located primarily in the southern and northern part of Greece, therefore the language used for preaching in these areas was Greek, and for a period of time the sacred Christian language was Greek. Up until 180 CE, there is no mention of Christianity in the Latin provinces of Illyricum, although St. Paul and his follower Titus did have in their intentions to travel to these territories. In fact, Christianity develops in the Latin provinces from around the Danube River by the end of the third century, when the Greek language influence is weakening in favor of the Latin language. The epigraphic Christian documents from the Danube provinces are mostly dated from the time of Constantine to Justinian, fourth–sixth centuries. Consulting these documents Pârvan makes an important observation, that the missionaries traveling to Christianize the local population in Dacia were mostly from Roman provinces located in the Middle East, priests that were fluent in Greek and Latin, but who would preach in the Latin language, when the local population would speak it, so they could be understood. Supporting this argument is the fact that in the Romanian language, the ecclesiastic terms are based on the internal developments of Latin spoken in the region, such as: Romanian credinţă ‘faith’, Treime ‘the Trinity’, Dumnezeu ‘God’, Tatăl ‘the Father’, înviere ‘Resurrection’, înălţare ‘the Rising’, biserică ‘church’, rugăciune ‘prayer’, împărăţia lui Dumnezeu ‘the God’s kingdom’, Făcătorul ‘Creator’, Fecioara ‘the Vergin’, etc., all developments that differ from the Latin ecclesiastical terminology, such as fides, Trinitas, Deus, Ecclesia, Oratio, regnum Dei, pater, incarnatio, resurectio, assumptio, carnatio, virgo, etc. This could suggest, as Pârvan stated, that either the Christian preachers were not coming from Rome, or that perhaps the religious terminology used by the Roman priests has settled in at a later date. The word for church could also indicate a local, perhaps pre-Christian, development of religious terminology in the region: Romanian biserica < Lat basilica of Greek origin, preserved also in Vegliot (bašalka), Rhaeto-Romansh (baselgia), while the rest of Romance languages retained Lat ecclesia. Regarding the religious concept of God, Romanian Dumnezeu, Pârvan observes that the post-Alexandrine Greek-Roman syncretism could have
104
A. R. Chelariu
resulted in the formation of a single supreme deity to be addressed with one general formula expressing his omnipotent character as in Deus, without a specific name. Thus, in the inscriptions from the East, particularly from Dacia Traiana (p. 103), he finds generic appellatives naming a ‘lord, absolute monarch’ divinity: Hero et domnus, Hero domnus, Domnus et Domna in Potaissa, Apulum, Ampelum, Sarmizegetusa, as unique divinities, or dom(n)a regina at domnus et bonus eventus in Naissus, Moesia Superior. The popular divine appellations domnus et domna, as found in inscriptions, were in use at an earlier date, as it appears in the name of the empress Iulia Domna, and also documented in names of pagan origin Domna, and Domnica, Domnus and Domnio. This appellative, domnus, was later given to Christian martyrs. There was also, as found by Pârvan in documents, an episcope with the name Domnus, who was sent to the Nicaean Concilium. This respectful form of address was retained in the contemporary Romanian language as domn ‘sir, lord, domna>doamna ‘lady’, domnitor ‘king’, domnie ‘kingdom’, as well as in the cultic imperative formula Doamne Dumnezeule ‘Lord God’, or in a single form in expressions like O, Doamne,‘Oh, Lord’, Doamne ajută, ‘Lord help’, Doamne iartă ‘Lord forgive’. On the right side of the Danube, the form found in Istro-Romanian dialect is domnu, Megleno-Romanian dialect domnu, dom, Aromanian dialect domnu, donu, and in Dalmatian persisted only as the fem. domna. Pârvan’s conclusion is that the word Dumnezeu was an old pagan invocation in Dacia, while in the West the Christian priests who Christianized those regions reintroduced and imposed the literary Latin form, Dominus Deus, found in Italian Domine Dio, French Mon Dieu, etc. Taking into account this conclusion according to which the form Dumnezeu and the religious concept of a monotheistic God are older than Christian terminology in Romanian language, an investigation with current linguistic tools into the origin of this word could be helpful. The reconstructed Indo-European root *dom(ha)u-no-s ‘master’ (Mallory & Adams, 2006) and the base for the Latin dominus, Sanskrit damuna ‘master of the house’, is a development from the IE *dom(ha)os ‘house’ with the suffix *-no- used to create words as ‘leader of ’. Other Indo-European languages show reflexes of this root, as OE dōm ‘fate’; Goth dōms ‘sentence, glory’; Av dāman ‘abode’, to that the form from
13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains…
105
Burushaski Daman meaning Lord or God could be added; Russian (and pan-Slavic) dom ‘house/home’; Skt dhā́man ‘law’; Greek θεμα ‘assertion’, anaϑema ‘votive offering’. In Old Irish we find domnán ‘world’, domun ‘world’, domhan ‘universe’, dumno ‘house’, and also as part of many proper names: Gaelic Domhnall, and a very interesting form Dumnorix ‘world king’; more so, the region Dumnonia in the western part of England inhabited by the Celtic tribe Dumnonii or Dumnones with their goddess Domnu; to these it could be added the location Iska (‘water’) Dumnoniorum, and an inscription dated around third century from Salonae on the Adriatic coast in which “rivis Dumnonia” is mentioned. These forms, especially the Irish ones, demonstrate the widespread development of the PIE *dom(ha)os-no to *dumno, the formation in Dumnorix, and more importantly, the religious connection brought forward by the goddess Domnu. Even though it is admitted that the consonant group m+n has changed into nn in Latin, perhaps around first century BCE (Istoria Limbii Române 1969), the proper names Domnus, Domna, Grk Δομνος are found by the end of second century CE. On the other hand, the consonantal group mn existed in Thracian, as found in anthroponyms like Lamneis (Vinereanu, 2008, p. 326), as well as in Gaelic proper names Domnotaurus, Conconnetodumnus (Caesar: De Bello Gallico). It could be justified to argue that the PIE root *dom(ha)os-no ‘master, lord’ was still used among the Thracian-Dacian speakers, as it is proven by the examples found in the other IE languages, and the influence of the Latin language reinforced its use. Contact with Roman soldiers, or later on, with priests from the Middle East, dispersed the concept of a main and all-powerful divinity, a concept well established in the regions from Egypt to Palestine, Syria and Anatolia, but also recognizable in the archaic Zamolxis. Therefore, we can conclude that the Dacian-Romanian Dumne-zeu ‘God’, together with the developments domn ‘gentleman, lord, sir’, doamnă ‘lady, madam, royalty’, vocative masculine form doamne! in expressions like Doamne Dumnezeule! ‘Oh, Lord God’, well documented in the region, could be part of an old common religious vocabulary, one to name God, the other to express a deference in forms of address.
106
A. R. Chelariu
Regarding the Latin ‘deus’(< PIE (Mallory & Adams, 2006, p. 409 *-deiwós ‘god”, OIr dīa, Lith dievas, Hit sius, Thrac Saba-zios, Grk Zeus) that in the other Romance languages developed as Italian ‘dio’, French ‘dieu’, is found in Romanian as zeu/zău, showing the phonetic changed > dz. > z, typical in this region, as per the Thracian Zio, Grk vocative Ζεῡ (Zeu), and Skt D(í)yaus *D(í)yeu. The Romanian word Dumnezeu in its present form is attested in first handwritten translations of religious texts from the fifteenth century onward. In the Scheian Psalter (Psaltirea Scheiană) probably dated from 1482 and written with Cyrillic letters, we find both forms: in Psalm III “Doamne, măntoiaște-mă/Dumned̡ăul ‘mieu ”, ‘Lord save me, my God’; in the Catechism from 1559 to 1560 we find the same formula containing both forms: “…că eu săntu domnu(l) dumnizeu(l) vostru…” (Hașdeu, 1876, p. 79); in the Psalter published by the Diacon Coresi in 1577, there is the same verse with both forms: Doamne! spăsește-mâ/Dumnezeulǔ mieu ‘Lord save me, my God’. In the same texts both forms, Dumnezeu and domn, with derivatives doamne, coexisted, even though the texts were written with Cyrillic letters and the sources for the translations were Greek and Slavic. The PIE root *dom(ha)os-no ‘master, lord’ did not limit its development in Romanian language to God, becoming instrumental in creating the polite addressing pronoun ‘you’ dumnea-voastră, dumnea-ta ‘your lordship’, possible contractions from an original formula domnia voastră, domnia ta, proving that the sound shift o>u occurred mostly in compound words, Dumne+zeu, dumnea-voastră, etc. The Dacians, as speakers of an Indo-European language, may have had the concept *deiwós, an omnipotent God. The ethnographic data testify that the population from the Romanian territory was under interdiction to use the God’s name; instead, they used special religious formulaic expressions such as Cel Ne-Numit “The Not- to-be-Named One”. This interdiction could explain the absence of a reminiscent form of the famous god Zamolxis, or the goddess Bendis, or any gods’ names from the Roman times. Instead, it persisted in the preference for a generic denomination Dumnezeu, Dumnidzău, a compound word locally formed. The concept *domno deiwós ‘lord God’ became specialized only to name a monotheistic entity God, Domne-dzău > Dumnezău, sacred and fixed, the Lord God, a main divinity with its root in the PIE *dom(ha)u-no-s ‘lord, master’.
13 God ‘Dumnezeu’ and the Creation of Earth, Sky, Mountains…
107
Although the Latin church formula Domine Deus, superimposed in the liturgical Latin by the Roman priests, may seem at first sight a valid etymological solution, the inscriptions from pre-Christian era found by Pârvan, and published one hundred years ago, coupled with the Proto- Indo-European comparative linguistic analysis, offers sufficient argument to reconsider the current explanation of Romanian Dumnezeu.
References Dicționarul Explicativ al Limbii Române DEX.RO. (n.d.). (On line) https:// dexonline.ro/ Falileyev, A., Gohil, A. E., et al. (2010). Dictionary of Continental Celtic Place- Names: A Celtic Companion to the Barrington Atlas of the Greek and Roman World. CMCS Publications. Hașdeu, B. P. (1876). Cucul și turturica la Români și la Persani (The Cuckoo and the Turtledove) Columna Traiana, pp. 40–44 MacBain, A. (1982). An Etymological Dictionary of the Gaelic Language. Gairm Publications. http://www.ceantar.org/Dicts/MB2/index.html Mallory, J. P., & Adams, D. Q. (2006). The Oxford Introduction to Proto-Indo- Europeanand the Proto-Indo-European World. Oxford: Oxford University Press. Matasovič, R. (2018). A Reader in Comparative Indo-European Religion. University of Zagreb. Niculiţǎ-Voronca, E. (1903 [1998]). Datinele şi Credinţele Poporului Român. Bucharest: Editura Saeculum. Olinescu, M. (1944). Mitologie Românească. Bucharest: Casa Şcoalelor. Pamfile, T. (1915). Cerul și podoabele lui după credințele poporului român; din Vieața poporului român; culegeri și studii, XXVI. Bucharest: Academia Română. Pârvan, V. (1911). Contributii epigrafice la istoria Creştinismului daco-roman, (Epigraphic Contributions to the History of Daco-Romanian Christianity). Bucharest: Atelierele grafice Socec. Pokorny, J. (1959). Indogermanisches etymologiscches worterbuch. Francke Verlag. Snorri, S. (2005). The Prose Edda; North Mythology (J. Byock, Trans.). New York: Penguin. Vinereanu, M. (2008). Dicționar etimologic al limbii româme pe baza cercetărilor de indo-europenistică. Bucharest: Alcor Edimpex S.R.L. West, M. L. (2007). Indo-European Poetry and Myth. Oxford: Oxford University Press.
14 The Romanian Goddess Ileana Simziana: The Sun and the Moon Marriage
The main character of the large number of statuettes from late Neolithic times, as the result of archaeological findings in the Romanian territory, part of Cucuteni culture and its surroundings, is a female figure that was considered by many as representing the Great Goddess of Neolithic, as described in detail by the renowned researcher Marjia Gimbutas. A majestic figure, she is the subject of many cultic statuettes, portrayed in various shapes, as a snake, bird, or fish, covering all forms of existence. She is the source of life, feeder of men and animals, protector of all creatures living in the air, water, or on the earth. She seems to be the dominant character of the prehistoric mythologies, creator and protector of animal and vegetal life, goddess of fertility and rejuvenation, of the mystery of birth and demise, holding powers over life and death. This mythic presence could generate life without a male, give birth to all gods, holding power over them, the ruler over the earth and the sky. This Neolithic goddess with her powerful mythic role transcended into the Indo- European pantheon, whose qualities resurfaced in almost all the feminine divinities from the classic mythologies, was feared and revered among the Romanian Christians as Sfânta Maria ‘Saint Mery’, or Sfânta Vineri ‘Saint Venera’. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_14
109
110
A. R. Chelariu
According to Duridanov (1976), the Thracians worshipped the goddess Bendis (Herodian, Choer., Hesych. and in inscriptions)—a Thracian goddess, identified by the Greeks with Artemis or Hekata. Her name is interpreted as ‘binder’, ‘patroness of the marriage, the family life’, explained by the IE *bhendh- ‘to bind, to join’, with cognates in the Avest. bandayaiti ‘to connect’, Old-Ind. bándhana- ‘connection’, Goth., Anlo-Saxon bindan, German binden ‘to bind’. The cult of Bendida was also spread in Bythinia, Asia Minor”. The Romanian folklore does not offer us any remaining from the name of the Dacian-Thracian goddess, but her character can be recognized in the Romanian Ileana Simzian, also known as Iana Sânziana or Ileana Cosânzeana, the most adorned fairy of the land, called in folk songs: “Ileana, Ileana/Ileana Simzeana/You, Lady of flowers/Of carnation bowers/. The Sun’s sister bright/Milk-and lily-white” (Balade Populare Românești, 1980, p. 25), the queen fairy, a brilliant goddess with golden hair waving as the fields of wheat. Her epithets are ‘the beautiful, the moon fairy, ‘lady of the flowers’, as white as the milk’s foam, protectress of the wild animals and the forests. Her powers are invoked in many songs and rites by young girls and women in search of a husband or to receive the gift of fertility. In Romanian Christmas carols she is featured being carried in a silk hammock hooked between a stag’s horns, or between a bull’s horns. She is the heroine of numerous carols and folktales, the most beautiful of all fairies, their queen, so beautiful that ‘one could look at the sun but not at her’, the fairy for whom Făt Frumos, ‘The Prince Charming’ from fairy tales, has to fight her abductor, the dragon, or the zmeu. ‘the ogre’ (Stăncescu, 1970). One of the most interesting representations of this Romanian fairy queen is found in the ubiquitous song The Sun and the Moon, performed regularly by popular bards at community fests. The song tells the story of the sun voicing his desire to get married; even though he looked everywhere for nine years he did not find a girl as beautiful as his own sister, Ileana; therefore, he decides to ask her hand in marriage. But the beautiful fairy does not want to commit such a sin by marrying her own brother. Wishing to stop his unwanted desires she asks him to fulfill impossible tasks, requests similar to those from many fairy tales: first she asks him to build a copper bridge over the sky, then a silver one, and then a golden
14 The Romanian Goddess Ileana Simziana: The Sun…
111
one, all tasks easily fulfilled by the astral divinity. In the end, to escape the unwanted marriage request, she throws herself into the sea from the golden bridge, turning into a silvery fish. God takes pity on her and raises her into the sky as the moon, to shine at night, always away from her brother (Alecsandri, 1852–1853). The ‘sun wishing to marry his sister’ mythic motif is widely spread in the Balkans: In the Bulgarian folklore there is the story narating the sun’s wish to marry his sister the moon; in some versions the moon is replaced by the morning star, or a mortal girl, Grozdanka. The hedgehog is concerned that this union would result in future children, new suns that would burn the earth, and so the sun gives up his marriage plans, either on his own, or under pressure from the animals. In a Macedonian version, the sun is upset that he had to give up his marriage desires, so he plunges into the sea, but he comes back up when hearing the cock’s cry; since then the Sun rises at the very time the cock cries (Berezkin, 2012, p. 625). As seen here, the Romanian story separates from the Balkan traditions, as the incestuous union between a brother and his sister represents the main concern. The motif of a fairy/nymph throwing herself into the sea to avoid an unwanted marriage has its ancient correspondence in the Minoan legend of the Britomartis, also known as Dictynna, the goddess of Mount Dikte, Zeus’ birthplace, a Greek goddess of mountains and hunting, sometimes believed to be a nymph of Gortyn, beloved by Artemis; for nine months she runs away from the attentions of Minos; in the end she jumps off a cliff into the sea, and is caught by the nets of fishermen, who save her. (Callimachus, 1921, Hymn III to Artemis). The motif of the sun marriage is also popular from a fable by Phaedrus, in which the frogs are worried that the sun wants to marry and have children, thus, more suns and more heat; they start a big noise that rises to Jupiter’s ears. (Phaedrus Ranae ad solem I-6). This version of animals’ involvement in the sun’s desires is popular south of the Danube in the Balkan region, as seen above. Similarly, the story in which the sun falls in love with a mortal girl, is common in Serbia, Bulgaria, and also in the Romanian song “Chicory”, discussed in Chap. 15 (Vrabie, 1966, p. 173). The mythical motif of the wedding or erotic relationship between the sun and the moon can be found in many classic myths, where the many
112
A. R. Chelariu
different characteristics attributed to these divinities generate sometimes confusions between the astral pair, sun-moon, and gods and goddesses associated with the sun and the moon, that are also brother and sister. Hesiod’s (2018) Theogony tells us that the Titans Hyperion and Theia, brother and sister, are the parents of Helion, the Sun God, and Selene, the Moon Goddess. The same is the case with the most famous divine pair in Greek myths, the Olympians Zeus and Hera, that are also brother/ husband and sister/wife; Hera is associated with the moon, and often represented in association with the bovine horns, thus, her Homeric epithet, the cow-eyed goddess. The younger divine brother and sister pair Apollo, the sun god, and his sister Artemis, do not seem to be involved in an erotic relationship. Selene, the known moon goddess, appears sometimes as Helios’ sister, and sometimes as his wife. In her myth, every night Selene goes down to Mount Latmos in Karia, where, in the darkness of the cave she kisses Endymion, her sleepy lover. Although in some stories Endymion is a mortal, his name is one of the many names of the sun. According to Max Müller, in some Greek dialect, enduo was used in the sense of the sun sinking by night in his mother’s purifying water, and from this was formed endumion, to express the sunset. Over time, people would come to say “Selene loves and watches Endymion”, meaning ‘it is getting late’, or, “Selene embraces Endymion” for ‘the sun is setting and the moon is rising’, and the original meaning of his name would be forgotten (Müller, 1884, pp. 52–141). In the Vedic hymns, Yama, the god of death and also the sun god, together with his sister Yami, are considered the divine twins, creators of humanity. As the Romanian fairy Ileana Simziana, the goddess Yami refuses to marry her brother, arguing the immorality of such an act, while her Persian counterpart will submit to it (Puhvel, 1987, p. 109). In another story from Hinduism the first being Prajapati feels lonely, and decides to split his body in two, giving rise to husband and wife; but when his wife realizes that he begot her from his own body, and their union would be morally wrong, she decides to hide from him (Max Müller trans. Brhadaranyaka Upanisad 1:4:2–6). The Baltic dainas tell the story of the marriage between the sun and the moon, with the difference that in these regions the moon is a male deity married to the sun as a female, and they have many children—the stars
14 The Romanian Goddess Ileana Simziana: The Sun…
113
(Dundzila, 1991, p. 144). In the Nordic sagas the sun is also female, as in the Baltic songs, and her companion, the moon is a male; being children of Mundifaeri, they are brother and sister (Snorri The Prose of Edda Snorri, 2005, p. 19). In Scandinavian mythology, Njördhr, the oldest among Vani, married his sister and had as children the divine pair Freyr, associated with the sun, the sky, and the rain, and his sister Freyja, the most beautiful and fair among the goddesses (Puhvel, 1987, p. 208). As the goddess from Chatai-Huyul holding the lunar horn, the new moon, this archaic divinity, as stated above, has the function of Terra Mater ‘Mother of the Earth’, a divinity considered to predate the arrival of the Indo-Europeans in Europe. Her paradoxical character persisting in many European myth and legends is illustrated by her duplicity manifest in her virginity-maternity, contraries that indicate the archaic mystery of the divine. Perhaps the glory of this archaic divinity is best illustrated by the Greek goddesses, Demeter, the distressed mother of the Earth, and by Hera, bo-opis ‘cow-eyed’, the Queen of Evening Sky and of full moon, while the younger ones, the virgins Artemis, Persephone, and Helen, goddesses of the new moon, the growing moon, of death and resurrection. In this group could be included the Romanian deity, Ileana, whose name sends us to the Greek goddess Helen, both related to an archaic goddess; her name is discussed by Mallory and Adams (1997, p. 232): *il(i)eha –‘goddess name’, Lat. Ilia ‘Numitor’s daughter’, Skt. Ilā-Idā ‘Manu’s daughter’. Noteworthy, Ilia was the daughter of the progenitor of Romulus and Remus, the Divine Twins, and the OInd Ilā was the granddaughter of Vivasvat, father of the twins Yama and Yami, as Helen was the twin sister of the Dioscuroi (Mallory & Adams, 1997, p. 232). Puhvel (1987, p. 143) considers Helen a pre-Doric version of the ‘tree- goddess’ having roots in the substratum: “…her name does suggest the Vedic Sūryā (Helénē perhaps from Swelenā, cognate with Helios < *Sāweliyos ‘Sun’ and with Avestan xvarэnah < *swelnos ‘solarity’)”. It seems therefore, possible that this archaic deity, this goddess Ilia, the Helen of Greek tradition, and the Romanian Ileana, the sun’s sister, belong to the same substratum. Perhaps a closer look at the myth of Helen could help understand other hidden connotations in the name of this goddess. Helen was born from the union between the goddess Nemesis as a goose and Zeus as a
114
A. R. Chelariu
swan (Stasinus of Cyprus or Hegesias of Aegina). The divine egg ends up onto Leda’s lap, or, as in some Greek representations, Hermes is the one who puts the egg on the altar. Helen is often portrayed on monuments holding the lunar horn, the growing moon, standing between her two brothers, Kastor and Polydeukes, shown with stars on their hats and tunics. The relationship between the lunar horn and Helen is also reflected in an old Greek tongue twist: ελενη σεληνη, meaning that the egg holding Helen fell from the moon (Chapouthier, 1935, p. 128). Similar to the Romanian Ileana, she was believed to be the most beautiful woman in the world, showing the same affection to flowers, as it is found in a tradition from Sparta, where young girls would hang flower wreaths from a certain tree consecrated to her. This custom may be based on the story, according to which on orders of a jealous queen, she was hanged from a tree, the source of her other name Helen Dendritis. (Chapouthier, 1935, p. 144) Likewise, there was a very important Romanian fest held on the day of the summer solstice called the day of Sânziene (fem. pl), with regional forms: Sâmzene, Sânzenii, the name of the flower fairies and forest spirits known as Ileana’s companions; during this fest girls would make flower wreaths and throw them on the house roof, while chanting a special invocation, in the hope to get married that year. These fairies’ name comes from the flower sânziana (or drăgaica), which is of two kinds, the yellow flowers (Galium verun) or white flowers (Galim mollugo) that blooms during the summer solstice. The plant’s fragrant flowers are used for making ritual wreaths, and the leaves are used for medical purpose (Ghinoiu, 2003, p. 307). Ileana’s other appellative, Simziana/Sânziana/Cosenzeana/Cosânzeana, was the subject of various controversial explanations. In the Explicative Romanian Dictionary (Dictionarul explicativ al limbii romane, DEX online), the official etymology offered for this appellative is the Latin formation sanctus dies Johannis, based on the fact that Sânziene Fest Day was held on June 24th—the Saint John’s Day. This Christian day was designated by the Council of Agde, France, in 506 CE, as the Nativity of St John the Baptist, in an attempt to associate a Christian feast to the summer solstice ceremonies celebrated at that time all over Europe, even though there was no religious connection between John the Baptist and Sânziene. The responses to N. Densușianu’s questionnaire from 1896
14 The Romanian Goddess Ileana Simziana: The Sun…
115
show that Sânziene are fairies feared by all, imagined as ‘3 beautiful holy women’ (Fochi, 1976). On the night of Sânziene summer fest, women would go into the fields to collect powerful medicinal plants under the auspices of the fairies; all the nocturnal rites, the plants collected, and the symbols related to these rituals, have little if any relation to the Christian Saint who baptized Jesus. The phonologically contracted form sanctus dies Johannis > sim(pt)-dzi-iuane, supposedly the root of Romanian sânziane (Candrea, 1927, p. 24), is a development associating this major Christian Saint and his birthday with the goddess of the land Ileana, and the folk summer solstice customs, while DRom Sântioan < Sanctus Ioannis < Greek Ιωάννης, exists as a revered Saint in the Romanian Christian calendar (Bolocan, 2012, p. 32). Thus, the etymological explanation of the appelative Sânziana through Saint John’s name is difficult to accept, especially since Saint John’s birthday was established by a Christian council in 500s, with no connection to the archaic traditions of the day of summer solstice, common to the entire Europe. It was a celebration of the power of the sun, when teens rolled burning wreaths down the hills during the longest night of the year, the night when women gathered sacred plants, singing and praying to the most beloved fairy of the land Ileana Simziana and her entourage of fairies. Mircea Eliade (1972, p. 68) used the concept of Diana Sancta, Sancta Diana > Sânziana, explaining the Romanian ziana as a development from Diana. Based on the adjective dius < *diwios, the form Diana < *Diviānā is a feminine of the sky god Dius, as in Dea Dia, related to Jupiter, and meaning just ‘divine’ (Puhvel, 1987, p. 151). The relation between the Roman deity and the Romanian Ileana, zâna florilor, sora soarelui, ‘the flowers fairy, the sun’s sister’, presents some difficulties as Diana is mainly a huntress, protecting animals, characteristics that are not obvious in Ileana Simziana. Perhaps a better approach to understand this archaic divinity would be to consider her appellative Simziana in connection with her attributes of the Queen of the flowers and the forests, the Earth Goddess, found in the Thracian mythology as Σεμελη ‘Mother of Earth’ (West, 2007, p. 175), and Semele, διωσ ξεμελω, mother of Dionysos, subject of a cult among the Phrigians as well as the Scythes. Believed to be Cadmus’ daughter, Semele’s name relates to ‘earth’: in Thracian and Phirgian: ζεμελω, Old Slvavic: zemlja ‘earth’, Lithuanian Zemyna ‘Earth
116
A. R. Chelariu
Goddess,’ cognate with Lat. Humus, Grk. Ξαμ-αι, Ξαμ-υνη, Demeter’s name, and possibly with the Cretan δηαί, Ionic ζηαί meaning “barley”, which makes her the Corn-Mother, and the giver of food. The Lithuanian goddess Zemyna, Zemyne, Zemynele, and the Latvian Zemes, “the Baltic Earth Goddess is both grain and mother-nourisher, out of whom humans arise and to whom they return” (Dundzila, 1991, p. 133). The Proto- Indo-European root for ‘earth’ as listed by Mallory and Adams (2006) is *dhéĝhōm-, Skt kşam, Av za, zam, zme, Grk khthōn, Lat humus, Lith žēmė, OCS zemlja, Alb dhe, Hit tēkan, Toch A tkam, continued as ‘human being, earthly’, and, as I mentioned above, ‘goddess’ in Phirgian ξεμελω, Thracian Σεμελη ‘Mother of Earth’. The essential characteristics of the Earth Goddess remained the same for thousands of years, as in Demeter, Semele, Zemyne, and also as a male deity in the Dacian Zamolxes, the god who disappeared underground for a number of years. From the little we know, among the few known Dacian words listed by Dioscorides, we find διεσεμα (Verbascum), with the Romanian name lumânărică ‘little candle’, a small plant with bright yellow flowers, with roots in the IE *dei-, di-, dia-, ‘bright, sun ray’ and σεμα ‘earth’, thus ‘light coming from the ground’, attesting the presence of the word naming this plant on the Dacian territory. Against this solution, Georgiev (1960, p. 43) offers different opinion, treating it as compound of Lat dies ‘day, Zeus’ and the PIE *eus- ‘to burn’, Lat ūrō, −ere, ustus (thereafter ussī) ‘to burn; to dry up’, ustus, uṣṇá- ‘hot, warm’, Gk. εὕω (*εὔhω, *eusō) ‘singe’, also a possible hypothesis, considering the Romanian plant name lumânărică ‘candle’. As shown above, the PIE *dhéĝhōm- ‘earth’ can help explain the appellative Ileana Sim-ziana, meaning the divine fairy of flowers, of the forests, and the Earth, and her companions, the simziene, forming the Romanian word for ‘fairies of the land’. Another possible explanation of the appellative Sim-ziana could be the PIE *sem, som, sm expressing the concept of ‘oneness in conjunction with others’, a very productive root in IE languages. Some examples include Lat. semper, singulus, simplex, simul, Toch. sam-, ‘equal, the same’, Skt. samtarâm ‘together’, sambhárana- ‘bring together’, sam-, ‘complete, perfect’, samvasa ‘living together’, Av. hangam, ‘to get together’, etc. (Caruba, 2000). If we apply this root to the Romanian compound form Sim-ziana it could reflect on her function as
14 The Romanian Goddess Ileana Simziana: The Sun…
117
a deity belonging to the group of fairies, DRom zîne ‘fairies’, the simziene, ‘one-together with, but unique’ among the other fairies, their mistress. The same need to express togetherness in her role as mistress of the fairies can be reflected by her other appellative Co-sân-zeana, formed with the Latin associative particle co-, com- ‘together, with’, similar to the Galatian name “*Com-onto-ro-, avec Com-, préfixe qui marque la réunion, la participation, le groupe” (Delamarre, 2016). The parallel form Sânziana may be a result of the contamination with the Latin noun sanctus, −a, as reflected in the Romanian saint’s names, Sân Petru, Sân Nicoaară, or in toponyms, Sân Giorzan, etc. The diphthongization of the form—zeana, ziana as in Sim-zeana, Cosân-zeana, seems to be the result of phonetic needs for assimilation with Ileana, zana > zeana. One may have to take into consideration the tabooistic aspect of these developments, since even pronouncing these words was regarded potentially dangerous, as is the case with the euphemistic appellative n. f. pl. iele, DRom fem. pl. pronoun iele ‘they’, implying the ‘fearful fairies’, those that ‘should not be named’ fairies that could turn very aggressive toward mankind (see below), contrasting with zâne. Beautiful and kind fairies, helping people, they are spirits of the earth, fairies of flowers, of fields and birds, fairies of springs and lakes, of unknown number. In the spring time, the tradition was that children and unmarried young people would start a fire, made only of specific kind of wood, and sitting by it with a jug of water, a loaf of bread and three chairs, they will be waiting around, in the belief that fairies will come to wash, eat, and warm up by the fire, and bring good luck to the entire community (Olinescu, 1944, p. 418). Getting back to the incestuous marriage between the two astral divine entities, brother and sister, and the moral aspect of it, it should be noted that in certain geographical areas the incest was rejected as sinful, while in others it was accepted. In the Baltic and South Slavic regions, the marriage between the two stars was considered acceptable, whereas in the Romanian folklore it was regarded as immoral, the same as in the Indian tradition. The two different approaches to the morality or immorality of this myth could signal the different time periods in the life of the myth; in some regions the hyerigamy reveals an archaic myth, that, in time, had lost the sacred connotations and was rejected as incestuous behavior.
118
A. R. Chelariu
Also, in some cases the nonhuman, astral aspect of the characters from the myth prevailed, therefore, the incest was not regarded as a moral issue, whereas in others, the human aspect prevailed. Both facets bring forth the complexity of this archaic myth, offering clues on the social development in various regions.
References Alecsandri, V. (1852–1853). Poesii populare—Balade adunte și îndreptate. Tipografia lucrătorilor asociați. Balade Populare Românești - Romanian Popular Ballads; Romanian-English bilingual edition. (1980). Translated by Leon Levițchi, Andrei Bantaș, et al. Bucharest, Minerva. Berezkin, Y. (2012). Some Motifs of Bulgarian Folk Beliefs in EurAsian Context. In Stratigraphy of Cultural Interaction in Eurasia Based on Computing of Folklore Motifs. Trames (Vol. 20, Issue 3). Estonian Academy Publishers. Bolocan, C. M. (2012). Christian Feasts Dedicated to Saints; An Interdisciplinary Perspective. In Annals of ‘Stefan cel Mare’ Univ., Suceava, Philosophy, Social and Human Disciplines Series, Vol. 1. Callimachus. (1921). Hymns and Epigrams, Lycophron and Aratus. Translated by W. Mair & G. R. Loeb (Loeb Classical Library No. 129). Cambridge MA. Hardvard Universiy Press. Candrea, I-Aurel. (1927). Românescul sânziene. Grai și suflet III fasc., 2, 428. Caruba, O. (2000). Indo-European *sem/sm- in the Pronouns. Washington DC: JIES, 28(3–4), Fall/Winter. Institute for the Studies of Man. Chapouthier, F. (1935). Les dioscures au service d’une déesse; etude d’iconograhie religieuse. Paris: de Boccard. Delamarre, X. (2016). Note galates: Kομοντόριος, Λεοννόριος, Λουτούριος, Tολιστοβόγιοι in Wekwos 2: Revue d’études indo-européennes (Errance archéologie) (French Edition) (French) Paperback—April 27, p. 49. Dundzila, A. V. (1991). Maiden, Mother, Crone: Goddesses from Prehistory to European Mythology and their Reemergence in German, Lithuanian, and Latvian. Dissertation, Madison: University of Wisconsin. Duridanov, I. (1976). The Language of the Thracians (An Abridged Translation of Ezikyt na trakite, Nauka i izkustvo, Sofia. http://www.kroraina.com/ thrac_lang/
14 The Romanian Goddess Ileana Simziana: The Sun…
119
Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. Chicago. University: Univ of Chicago Press. Fochi, A. (1976). Datini și eresuri populare de la sfârșitul secolului al XIX-lea. Bucharest: Editura Minerva. Georgiev, V. I. (1960). Raporturile dintre limbile dacă, tracă şi frigiană. Studii clasice II, Societatea de studii clasice RPR, p. 39. Ghinoiu, I. (2003). Sărbători şi obiceiuri Româneşti. Bucharest: Elion. Hesiod. (2018). Theogony, Works and Days, Testimonia. Edited and translated by Glenn W. Most. Loeb Classical Library 57. Cambridge MA: Harvard University Press. Mallory, J. P., & Adams, D. Q. (1997) Encyclopedia of Indo-European Culture. London and Chiicago: Fitzroy Dearborn. Mallory, J. P., & Adams, D. Q. (2006). The Oxford Introduction to Proto-Indo- European and the Proto-Indo-European World. Oxford: Oxford University Press. Müller, F. M. (1884). The Upanishads, Part II (SBE15) (Sacred Books of the East). http://www.sacred-texts.com/hin/sbe15/index.htm Olinescu, M. (1944). Mitologie Românească. Bucharest: Casa Şcoalelor. Puhvel, J. (1987). Comparative Mythology. Baltimore: John Hopkins University Press, pp. 208, 143, 151. Snorri, S. (2005). The Prose of Edda; North Mythology. Translated by Jesse Byock. New York: Penguin. Stăncescu Dumitru. (1970). Sora soarelui; basme culese din popor. Bucharest: Editura Minerva. Vrabie, G. (1966). Balada populară română. Bucharest: Editura Academiei Române, p. 173. West, M. L. (2007). Indo-European Poetry and Myth. Oxford: Oxford University Press.
15 The Sun and a Mortal Girl Marriage: The Romanian Song of Cicoarea ‘The Chicory’
Similar to “The Sun and the Moon” song, the motif of the love between the sun and a girl is the subject of another Romanian song Cicoarea “The Chicory”. In this song the subject of the sun’s love is no longer his sister, but a mortal girl, or, in some versions, a fairy. The motif of the sun’s love toward a mortal girl is found also in Bulgarian and Serbian folklores, with the difference that the girl will marry the sun, and will become the weaker spring sun (Vrabie, 1966, p. 188). In the Romanian song, the beautiful fairy of the flowers refuses the sun’s marriage proposal, saying that he is “always traveling / by day over villages / by night over waters”. For her refusal, the sun punishes her, turning her into a chicory flower, destined to follow him from dawn to dusk, when her petals would close. Traditionally, the chicory flower is used in rituals performed by unmarried girls, in the belief that the sacred powers of this plant will bring their pretender sooner (Ghinoiu, 2003, p. 316). The Romanian song Cicoarea shares similarities with the Greek myth of Daphne, who, as in the Romanian counterpart, refuses Apollo’s love: To escape his pressure she calls for help her father, the river Peneus, and is turned into a laurel tree (Ovid Metamorphoses 1: pp. 452–567). The motif of the young girl turning into a flower is also found in the Greek © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_15
121
122
A. R. Chelariu
myth of the nymph Clytie. However, in this myth the roles are reversed, the girl is the one in love with Apollo, who ignores her. Overwhelmed by her sorrow, Clytie keeps watching the sun until she becomes attached to the ground and turns into a plant (Ovid, 1955, Metamorphoses 4: 256). The refusal of the sun’s love from Cicoarea, as well as from Ileana Simziana, the song discussed above, has no other correspondences in Southeast European folklore, separating the Romanian songs from the surrounding regions’ folk data, setting it closer to the Greek version. The transformation of a young woman into a plant, a medicinal plant, fits well with the tradition in which these fairies and nymphs were regarded as powerful feminine flower spirits, helping women in their desires for procreation, or against diseases. The Greek Clytie is transformed into a heliotrope (Valeriana officinalis), and the Romanian girl turns into a chicory flower (Cichorium intybus), both feared by women for their magical powers, and used in ritual practices. These plants, the heliotrope and the chicory, belong to a wide range of sacred plants found in the Indo-European mythology, together with the mandrake (Atropa belladonna), the hemlock (Conium maculatum), the laurel (Laurus nobilis), to name a few. Among the enormous material written on this topic, Mircea Eliade discussed in detail the use of mandrake by the Romanian girls in a ‘wish-to-marry’ magic chant: “The mandrake is preeminently the erotic plant. It brings love, marriage and fertility …. The mandrake is personified as ‘Great Lady’, ‘Empress’, Good Mother’” (Eliade, 1972, p. 224). This relation brings to mind the agrarian festivals celebrated in Europe from antiquity to recent times. During these fests, as described by Sir James Frazer, young girls covered with flowers, played the role of a bride in a ritual marriage ceremony riding in a cart next to an image of an agrarian god. In the same context, the author mentions a ritual from Bengal, where the goddess Earth was celebrated annually in a ceremony of her marriage with the Sun-god, a ritual similar to the one from the Eleusis mysteries: “In the great mysteries solemnized at Eleusis in the month of September the union of the sky-god Zeus with the corn-goddess Demeter appears to have been represented by the union of the hierophant with the priestess of Demeter…. But their intercourse was only symbolical or theatrical, for the hierophant had temporarily deprived
15 The Sun and a Mortal Girl Marriage: The Romanian Song…
123
himself of his virility by an application of hemlock” (Frazer, 1971, p. 165). Similarly, in the Romanian folklore there was the old festival named Drăgaica, or Sânziene, as mentioned above, celebrated at the summer solstice, the beginning of harvesting: to prepare for this fest, girls from a large region would select the one who was the most beautiful among them, named her ‘Drăgaica,’ and crowned her as a bride for that day; her crown was made of wheat, braided with flowers called drăgaica, and her dress was decorated with colorful pieces of cloths. Thus dressed up, she would go dancing all around the village followed by her entourage of young girls (Pamfile, 1997, p. 66). During the same day, young men would perform the ritual called boul instruţat, boul de sânziene ‘decorated bull, the sânziene bull’, recalling the bull consort of Hera, Zeus. For this ceremony, the group usually selected a white bull with large horns, decorate it with beautiful mats, bells, and colorful fringes, and adorned it with a similar crown, symbolizing its role as a groom. The large entourage roamed through the village with music and noise makers imitating the bull’s roar, while one of the young men, dressed as a woman, would engage in sexual gestures (Ghinoiu, 2003, p. 319). The plant drăgaica or sânziana (Galium verum), also called the Lady of the Flowers, has medicinal properties, as the chicory flowers, valued for their magical powers; women would wear them around their waist during harvest to insure protection against the pains of working in the field during harvest. All these plants, chicory, hemlock, mandrake, wormwood, sânziana, drăgaica, and many more, were used in fertility rites practiced since immemorial times. To gather them, women were required to follow special ritual preparations: On the day of sânziene, the day of the fairies, certain women would get up very early in the morning, and go in groups to the meadows, chanting and reciting sacred verses, eager for good omens in their search for gathering the powerful medicinal plants. The word ‘zena’ is listed by Dioscorides (1934) as a Dacian word meaning ‘hemlock’. Investigating this word’s etymology we find Skt. jyanay-, Av. zyanay-, zyana-, ‘to overcome, conquer, weaken’, with a reconstructed PIE form *ĝiena, *ĝeie- ‘subjugate, overpower, oppress’ (Detschew, 1957). Thus, the argument can be made that the Dacian word zena ‘hemlock’ could be a valid etymology for the Romanian zână ‘fairy’, plural zâne, the
124
A. R. Chelariu
overpowering and feared fairy, with powers to ‘subjugate’ ‘oppress’, sicken, or cripple people if not revered. Zânele ‘the fairies’, companions of the Romanian goddess Ileana Simziana, also named ‘simziane, sânziene’, flower fairies, were protectresses of medicinal herbs, feared and revered by everyone but mostly by women, praised in many rituals and songs, spirits showing traits of relations with the ancient goddess of fertility, and with her consort, the sun- god. Perhaps there is not just a coincidence that almost all the versions of the song “The Sun and the Moon” start with the words: Foaie verde de cicoare ‘Green leaf of chicory’, leading us to conclude that the Romanian song “The Chicory” Cicoarea is reminiscent of an archaic medicinal chant or a shamanic ritual.
References Detschew, D. (1957). Die thrakischen Sprachreste. Österreichische Akademie der Wissenschaften. Dioscorides, P., of Anazarbos. (1934). Greek Herbal of Dioscorides (J. Goodyer & R. T. Gunther) (pdf ). Oxford University Press. Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. Chicago. University of Chicago Press. Frazer, J. G. Sir. (1971). The Golden Bough: A Study in Magic and Religion (5th ed.). Macmillan. Paperbacks edition 1963. Ghinoiu, I. (2003). Sărbători şi obiceiuri Româneşti. Elio. Ovid. (1955). Metamorphoses (R. Humphries, Trans.). Indiana University Press. Pamfile, T. ([1910–1914] 1997). Sărbătorile la români; studiu etnografic. Editura Saeculum I. O. Vrabie, G. (1966). Balada populară română. Editura Academiei Române.
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
Romanians held their wedding ceremonies in the fall or right after Christmas, and very rarely in the summer. They were married early in life, at around 18 years of age. The custom was that in the spring time, around Easter, the Sunday of St. Thomas, or by ‘Rusalii’, 50 days after Easter, boys and girls would gather by the church yard or in the cemetery, and play all sorts of games, giving boys the chance to interact with a girl they like to get to know her better (Marian, 1995, p. 17). They would joke around, play, or walk together, closely followed by the girl’s mother. When the young man comes to a decision about the girl he likes, he discusses his intentions with his parents, who will find out what her dowry may be. If everything was acceptable, the father of the young man, or a relative or a village dignitary together with a woman who was versed in marriage matters, go to the girl’s house, to discuss the engagement; during their visit, the girl is not allowed to be present. and would stay with a neighbor; after exchanging ceremonial greetings, parents would reach an agreement, and decide on the engagement date. In the evening of the official engagement, the young man with a group of friends, relatives, go to the girl’s house, and the ceremony begins with the reciting of the oratio nuptiae, a poem specific for this occasion remembered in its entirety for generations. Enacted by the ‘speaker,’ called in Romanian ‘vătaf’, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_16
125
126
A. R. Chelariu
‘captain’ or ‘god father’, a man known for his declamation gifts, and ability to conduct the entire ceremony (Nunta la români, 1989). The oratio nuptiae poem, recited for generations, recreates the story of an ideal ceremony, aiming to ensure the good omen for marying couple. The poem consists of the dialog between the speaker of the young man’s group and the girl’s entourage, her father, brothers, cousins, and relatives; after the customary greetings in formulaic speech, the father asks them what is it they are looking for; the speaker explains that they are their king’s (Romanian împărat) escort, and came from afar with good intentions, to honor an understanding [between the two families], hoping that it is still a valid arrangement. The father assures them that the understanding is still on. Then the groom’s speaker asks to see ‘the one’ for whom they came. The girl’s entourage tries to slow things down, calling the girl’s name, although she is not supposed to answer; joking about the various reasons the girl is not answering the calls, someone from the father’s group would bring out an old lady, or a very young girl, to the protests and amusement of all (Marian, 1995, p. 87). This specific ritual is shared by many European traditions. When finally the bride is brought in, all approve, the bride and the groom are seated at their reserved place at the party table. In front of them are placed two embroidered kerchiefs with money, grain, basil, and the rings. The pair’s god father asks them if they are willingly engaging to get married, and after they agree, they exchange the rings. Dinner is served, and after many hours of eating and partying, the groom’s father asks the girl’s father to take good care of her until they will come back on the wedding day (Marian, 1995, p. 125). The DRom word for engagement v. a se logodi ‘to get engaged’, n. logodnă ‘engagement’, is considered of Slavic influence, lagodı ̃nŭ as per (DEX.RO), a form that is not listed in Derksen (2008). For this essential concept in the Romanian social relations, a better solution could be the PIE root: IEW 687 *leugh-2. lugh- ‘oath’ (Note: only Celt. and Gmc.) as found in OIr lu(i)ge n., Welsh llw m., Bret. le “oath, vow, pledge” (*lughiom); Goth. liugan, −aida “marry”, liuga “matrimony” (*oath), OHG urliugi (*uz-liugja) “war, fight” OFris logia “marry”; these cognates are semantically related to the DRom logodnă, as it indicates ‘oath, matrimony, marriage’; if the Slavic form is valid, it could be considered a contamination.
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
127
Following the engagement, the pair will ask two of their friends to be their ‘callers’ to the wedding: Three days before the wedding day, the callers will go around the village, carrying specific ceremonial canes decorated with colored ribbons, flowers, and wheat, and invite everyone to the wedding ceremony. On the wedding day, a group of relatives and friends, about twenty to thirty people, gather together with the groom, forming ‘the groom’s hunting suite’, the ‘king’s suite’. The group of young men recalls the IE youth bands, particularly since the Romanian word for ‘groom’ mire, pl miri, fem. Mireasă ‘bride,’ has its roots in the PIE MA *méryos ‘young man’ and *meriha- ‘young woman’, IEW 738–739 *merio- ‘young man, woman’, with cognates in Lat marītus ‘husband, lover, suitor’; Alb shemër (MA notation) ‘co-wife, concubine, mistress’; Grk meîraks ‘young man or woman; Av mairya ‘young man’; Skt márya- ‘young man, lover, suitor’; MPers mērak, Grk μεῖραξ m. f. ‘knave, boy, girl’; Lith martì ‘bride, maid, OPruss martin acc. sg. ‘bride’. The solution offered by DEX.RO through the Latin miles, −itis ‘soldier’ is semantically unlikely; (de Vaan, 2008, p. 379: “It is tempting to connect milia [pi.] ‘thousand(s) hence *mili-it‘who goes with/by the thousand’). After assembling the entire hunting suite, under the guidance of the speaker, vătaf the captain, they go to the bride’s house, where they come upon the bride’s father, brothers, and other relatives, waiting behind the gates secured with ropes. There, the hunting suite start enacting an altercation with the bride’s suite, demanding for the gates to be opened; pretending to know nothing of a hunting group and their search, the bride’s entourage refuses to let them enter the yard. The argument for letting them enter the bride’s yard is made by the ‘speaker’, who recites again the poem oratio nuptiae, the same from the engagement party. The content of the poem relates the dialog between the groom’s speaker and his suite arriving at the gate of the bride’s house, and her father. Paraphrasing the dialog, the bride’s father sternly asks them what it is that they want; the speaker of the groom answers that they are not wrongdoers, but an army assembled in the morning by their ‘king’ (the groom), who woke up that morning, washed, combed his hair, dressed up in his best outfit, and asked them to join him in hunting, because it was the time for him to get married (Marian, 1995, p. 291). Calling the groom ‘king’ in the
128
A. R. Chelariu
Romanian ritual oratio nuptiae echoes wedding rituals seen on early Greek bas-reliefs showing a crowned pair for their wedding ceremony (Lyons, 1997, p. 8). This is similar to the Greek or Latin ancient documents, portraying the groom and his bride crowned as king and queen for their wedding day (Oakley & Sinos, 1993, p. 16), as in the depiction of the annual wedding ceremonies from antiquity by Sir James Frazer: “In earlier times the Roman king, as representative of Jupiter, would naturally play the part of the heavenly bridegroom at the sacred marriage, while his queen would figure as the heavenly bride (…) that the Roman king and queen should act the parts of Jupiter and Juno, would seem all the more natural because these deities themselves bore the title of King and Queen.” (Frazer, 1971, p. 171) The wedding day crowning tradition is alive even today in Greek orthodox religious wedding ceremonies, during which the bride and the groom are crowned by the priest. The fragment from the Romanian wedding poem describing the ritual struggle between the suite of young men with the bride’s father, and his refusal to open gates and give the bride away, is found in many European cultures, among which are the Teutons, the Celts, the Slavs, and the Scandinavians (Oakley & Sinos, 1993), except in the Hungarian wedding traditions, where neither the ‘ritual deer hunt’ motif nor the struggle by the gate is present. The speaker continues reciting how the group hunted all day long and all around the region without any luck; tired, their ‘king’ stopped by a fountain or a spring for a drink of water, and there he saw the trails of a wild beast. He consulted his followers on the nature of this trail, debating whether it was of a deer, a fairy, or a heavenly flower. It is worthwhile noticing that in the old traditions, the belief was that the footprints, humans or animals, magically retain the qualities of their owner (Hulubaș, 2009). They decided to follow the trails that guided them to ‘this’ house, the bride’s house, thus showing the entire facet of ‘the deer hunt’ mythic motif, in association with that of ‘the guiding animal’. The oratio nuptiae reciting ends here. After pretending to protect his gates, arguing that there was no deer or fairy coming into his house, the father ends up opening the gates and inviting the groom and his best men inside the house, with more greetings and best wishes for prosperity and long life, and the entire audience enjoys eating and singing (Marian, 1995, p. 310).
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
129
In the northern parts of Romania, the ceremony continues with someone from the bride’s suite questioning the groom’s knowledge and abilities through riddles, as: ‘What is the dense shadow above us?’ The groom should know the answer, ‘the sky’; ‘what is the biggest water on earth?’, ‘the dew in the fall’; ‘what is the biggest mountain among all mountains?’, and the groom should answer, ‘the anthill on top of the mountain’; ‘what food would you feed your bride?’, ‘bread of pure wheat and wine’; and so on, continuing about his work in the fields during the summer and around the house during the winter (Olinescu, 1944, p. 259). After eating, drinking, and partying, the bride has to leave her parents’ house, carried by her father over the threshold; one of the groom’s men takes her by the hand, sets her in her cart, and the entire suite goes to the groom’s house, following the bride’s decorated cart, stuffed with her dowry. At the groom’s house a woman, the groom’s relative, greets them with a bucket full of water splashing the entire group with water and grain. The newlyweds go into the garden, where an ox harnessed with a plough awaits for them to perform the ‘covering of the bride’s head’ ceremony: staying on each side of the plough the groom un-braids his bride’s hair and covers her head with a kerchief, while women sing to her that she is now changing her girlish ware with that of a married woman. Similar custom is found in Austria: after the wedding, the bride was uncrowned, and covered her head with a bonnet, the sign of her new position among married women. In some parts of Romania, after the ‘head covering’ ceremony, the bride and her suite go to a spring or river, where the bride breaks a ceramic bowl, and throws in the river a round sweet bread, ‘colac’, which is retreived from the water by a boy from the group and brought to the house, and everybody must have a piece from it (Marian, 1995, p. 312). The motif of a ‘deer hunt’ and the tracking of the animal-fairy-flower from the oratio nuptiae are spread in every region of Romania. Variations of the motif of the ‘ritual hunt’ in which the hunted animal, the doe or stag, is replaced by a bird, or a flower guiding the hero, are found in other European traditions, as for example in Ukraine, where the king ‘knyeaz’ follows an unspecified kind of bird to the bride’s house. In the Italian tradition the bird is a dove, and in the Serbian tradition the animal guide is replaced by a red flower growing by the side of the road (Marian, 1995,
130
A. R. Chelariu
p. 96). These stories, however, show a significant difference from the Romanian wedding poem that seems to retain a more archaic structure. The Romanian wedding traditions and their distant past can be compared with the images found on Greek vases that display the ancient wedding ceremonies, which followed closely those of gods. The ceremonies figured on these ancient vases display interesting similarities with certain aspects present in the Romanian oratio nuptiae poem: the purification bath taking place on the wedding morning as in the verses “Our king/this morning woke up/his white face he washed…;” the coronation of the bride and the groom on their wedding day with a crown of flowers; the bride’s suite, Greek ‘paides propempontes’, led by a ‘proegetes’, as she is taken to a cart (Oakley & Sinos, 1993, p. 27), similar to the Romanian tradition, where the leader ‘vataf’ or ‘captain’, guides her to the cart that takes her to the groom’s house, followed by the entire suite. Accordingly, being called king and queen on the wedding day may have very old roots. Throughout antiquity, the image of the goddess was always in association with a cervide symbolizing force and renewal. The chasing of a deer motif is already documented in many Greek myths, as the myth of Taygete (Ταϋγέτη): according to Apollodorus (1921, The Library 3.10.1) the nymph Taygete, one of the Pleiades, and a companion of Artemis, was pursued by Zeus; because she refused the Olympian’s advances, she invoked her protectress Artemis, who turned her into a doe with golden horns. Another case of a deer pursued by a hero is related in the Third Herculean labor: Eurystheus demands that Heracles capture the female Ceryneian Hind, the fastest in the land; Heracles chased the hind on foot for a full year before capturing it (Apollodorus. The Library 2. 81). Although the hunt of the cervide from these myths does not end in marriage, the motif of the ‘deer’ being hunted by a deity, a king or a hero, is in existence since ancient times. But the most famous story from classic mythology involves the Greek goddess Artemis, often pictured in the company of a deer or a stag, or wearing a doe skin on her shoulder. In the Greek Hymn to Artemis, the goddess is surprised by Actaeon, a passionate hunter, while she was bathing; his intrusion stirs her anger, and she punishes him by turning him into a deer; because neither his companions nor his dogs recognize him, the dogs kill him (Callimachus, 1921, Hymn V to Artemis). In another story, an archaic Actaeon aspires to Semele’s
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
131
favors, provoking the divine wrath of his rival, Zeus, which could interfere with the birth of Dionysus (Pausanias, 1918, 9.2.3). These stories may actually describe a reenactment of a hierogamy ritual between the goddess as a doe and her lover as a stag (Baring & Cashford, 1993, p. 331). The Romanian wedding song oratio nuptiae and the hunter/groom discovery of the deer the tracks near a source of water, a fountain, a spring, reveals its ritual character, as trails near a body of water indicate a passage of transformation. It is by a body of water that the groom discovers the animal tracks, a place that in most of the rites of initiation signifies entering into the magical world of the fairy (see Chap. 19), the world of the goddess of the wild. These stories lead us to conclude that they actually describe a reenactment of a hierogamy ritual between the goddess as a doe and her lover as a stag (Baring & Cashford, 1993, p. 331), suggesting a premarital rite preserved in the episode of the deer / fairy / bride whose trails near a fountain are found by the king / hunter /groom, from the Romanian song. The motif of deer hunt from the Romanian wedding song as a possible premarital rite could gain validity if we compare it with the Irish story of Oisín of the Land of Eternal Youths. Oisin’s own mother, Sadb, had been changed into a young doe by the Druid named Fear Doirche because she refused his advances. The doe is hunted by Fionn MacCumhail, who catches the fawn but refused to kill his prey. Magically, the doe changes back into the beautiful girl. They wed and Sadb is soon with child, Oisin (Flannery, 1896). In the famous stories of King Arthur we find him as pursuing a large, white deer; arriving at Sir Pellinore’s well, a magical site, he finds himself without his hunting party or his horse (Baines, 1962), similar to the Romanian groom. An interesting development from a premarital rite to the foundation of a state boundaries is recorded in an Indian story of the Visukamma (Visvakarma) who turned himself into a golden stag that the king in that country was unable to capture; the king sends his son to fulfill this task; gathering thirty-two men, the son follows the golden stag; after many months and renewed enforcements, the prince makes love with a local woman, and stays there for a longer period of time, while the stag is waiting. In the end, the prince and his entourage restart the hunt and when
132
A. R. Chelariu
they reach the end of the kingdom, the stag disappears completely. This myth could be related to a ‘ritual of settlement’, in which the stag takes the prince through his entire country up to the borders, before it disappears; the erotic relation signifies a ritual of sovereignty, that of a king copulating with a chthonic being, akin to the Irish mythic warrior’s obligation to marry a goddess as a condition to reach sovereignty (Eliade, 1972, p. 152). The numerous stories about deer and the complex beliefs surrounding them are found in many folk stories all over Europe. In an interesting communication, J. G. McKay (1932, p. 169) analyzes the deer presence in Scottish folklore, where they were looked upon as fairy cattle, supernatural animals, and only supernatural women had the ability or privilege of being able to turn themselves into deer. The author concludes that some women deer love and marry their hunter-lovers and live happily ever after. The tracking of animals in the hunting act, present in the classic mythologies, could signify the path followed by the hunter under the guidance of a sacred animal leading to a new condition. A. Krappe in his article Guiding Animals (1942) details many ancient stories narrating the mythic motif of a hero and his suite pursuing a wild animal, a cervide, a cow, a buck, an aurochs, a bird, or even a water animal, that in the end guides the hunter to a new situation, the foundation of a new settlement, a town, or state. Stories of following a guiding animal sacred to a divinity are quite common in the antiquity of Europe: the legend of Cadmus, who builds Thebes on the place where a cow lies down, (Apollodorus: The library 3.4.1), a ‘guiding cow’ motif that also recurs in the legend about Ilus the founder of Troy (Apollodorus: The library 3.12.3). The motif of the ‘the guiding animal’, often a stag, is found in legends of a foundation, be that a dynasty, a city or a state, and not related to establishing a marital situation as found in the Romanian wedding song. These stories are tangentially related to the deer as a bride from the Romanian song, and only if establishing a new family is viewed as a settlement. The guiding animal mythic motif resulting in a foundation resurfaced in the medieval literature, fulfilling similar needs, to validate the name or the location of a community, of a monastery, or to enforce the Christian beliefs. The motif is found in various legends such as the
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
133
one recorded by Gregory bishop of Tours in his History of the Franks (1916, 2: 37) where a hind showed Clovis and his army the path to cross the river Vienne in the fight against Goths, or the story recording how Charlemagne was guided by a white stag across the Alps (Krappe, 1942, p. 236). Guiding animal motif is often found in Christian legends, as for example in the legend about the Roman general Placidus, a passionate hunter, followed a stag to the top of the mountain; there he sees a cross between the animal’s antlers, and hears the words that the stag was Christ, and the general converts as St. Eustathius. As Krappe argues (1942, p. 235), the motif was used in the late middle ages in a very large group of stories clearly etiological in character, created to explain a place name, and legendary ancestries: “Procopius of Caesarea tells the story of how the ‘Cimmerians’, i.e. the Huns, followed a stag across the Sea of Azov and, beholding the fertile fields on the other side, decided to invade that country. Jordanes (n.d.) reports a similar story on the invasion of Scythia by the Huns: Hunnish hunters followed a hind through the Sea of Azov. On the opposite shore the animal disappeared, hunters returned home full of praise of the country they found, the land to be invaded by the Huns. And further Krappe (1942, p. 237) states that: True enough, no far-reaching conclusions can be drawn from the recurrence of the story among the Hungarians [54];71 for it is well known that the Magyars down to modern times erroneously believed themselves the descendants of the Huns, and the adoption of the story by their chroniclers betrays a literary, not a popular, tradition. What is more significant is the story recorded by Ricold of Monte Croce, a thirteenth century traveller, of how the enclosed nations of Gog and Magog, identified with the Tatars, at last found their way beyond the mountains, guided by a hare and an owl. Hence, adds the traveller, wonderful is the honour they still pay to owls, whose feathers they wear in memory of their deliverance.
The guiding animal motif, well represented in the European medieval literature, is also found in the Romanian legend recording the foundation of the Moldovan royal dynasty: In the earliest historical documents, Prince Dragoș from Maramureș, located in the northeast part of Romania,
134
A. R. Chelariu
was hunting an aurochs with his suite, killing the animal by the river Moldova, where they decided to settle (Eliade, 1972, p. 132). Although the motif of ‘deer hunting’ is part of the Romanian oratio nuptiae, a well- known poem recited at the beginning of the betrothal and the wedding ceremony, and recorded over the entire Romanian territory, M. Eliade does not correlate the two in his work, concentrating instead only on the motif of the foundation of a dynasty. The motif of ‘deer hunt’ extends beyond the European area, surfacing in heroic epics from the Central Eurasia, as the famous Gesar Epic of Tibetan origin, where the hero hunts a deer that is actually the daughter of a foreign king, or of the Dragon King (Mátéffy, 2017, p. 337), perhaps a mytheme borrowed from the Indo-European neighbors. The common elements from the European, Indian, and Tibetan traditions (the young prince with his escort; the hunt of a cervide; the erotic encounter; the settlement leading to a change in his situation) are all centered on the ‘ritual hunt’. The same motif shown in this summary plays a key role in the Romanian wedding song, the oratio nuptiae, that leads not to a Christian conversion or monastery foundation, but the foundation of a family. The allegoric erotic hunt revealed by the two chasing characters, the doe as a feminine principle chased by the hunter as a masculine principle, forms the subject of a large number of poems and stories in medieval literature, as presented by M. Thiébaux in her pertinent work suggestively entitled The Stag of Love. She begins her investigation with the chapter Sacred Chase, with the stag guiding the hero into the other world through a process of initiation into knowledge / self-knowledge and passionate love to death. In the Love Chase chapter, the hunted–hunter relation is expressing an erotic game, the chase including the hunter’s sense of conquest and pleasure. Marriage among humans made the subject of many studies, of which the old work of Eduard Westermarck (1894) touches on a custom spread around the world, called ‘marriage by capture’ (p. 389). At a later date, Dumezil addresses the marriage customs of the Indo-Europeans, giving special attention to the Romulus and the rape of Sabines’ legend (Dumezil, 1979, p. 73). Customs of similar nature are also known to the Romanians: The ‘groom’ together with a group of his friends organize the kidnaping
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
135
of the ‘bride,’ sometimes with the girl’s consent, and sometimes against her will; most of the time the ‘marriage by capture’ is the result of the parents’ refusal to give their consent to the wedding plans and dowry, and the two youngsters consummate the marriage on the fields, a ritual act that could not be undone by relatives (Marian, 1995, p. 115). In case of a violent capture, the girl in forcefully captured by a young man who is not liked either by her or her family; such an act results most of the time in family feud and violence. In those instances, the girl had no choice in the matter but to stay in the man’s house and become his wife without a wedding ceremony (Olinescu, 1944, p. 286). In an interesting study, Ernst S. Dick (1966) offers the hypothesis that marriage was a rite of passage, particularly of the young woman, who had to experience separation from her parents and also acceptance in the new family. From this approach, the author distinguishes between the rites of separation and rites of acceptance, and seizing the bride becomes part of the separation rites, in which the hostilities are only ritual play. This idea corresponds to the scene from the Romanian oratio nuptiae, of the bride’s father mocking a fight with the groom and his escort. The same pretend hostilities, as related by the author, were common in Norway, where the maiden was taken by force by the groom’s friends. The ‘hunting of a deer or stag’ ritual, as a process which ‘leads to a radical change in the hunter’s situation’—establishing of a dynasty, a settlement, and so on, could very well describe an archaic wedding ritual, the transition of the young man ‘the king of the day’ and his bride ‘the fairy queen of the day’ into the married life, fundamentally changing their status as they are about to establish a family. Perhaps an even older mythic level of the ‘deer hunt’ motif transpires in a number of Romanian Christmas carols, ‘colinde’, in which the main subject is the ritual dismembering of the stag, recalling the myth of Artemis and Actaeon. In a Christmas carol addressed to a young man, the caroler invokes the sacred stag of mythic proportions, the color of gold, perhaps a solar manifestation, a towering animal that invaded and damaged the community’s fields; the place to find the stag is under a singular tree, seemingly the cosmic tree; the caroler wishes the young man to show courage and strength to shoot the stag with his deadly arrow, and restore fertility to the fields (Hulubaș, 2009, p. 122). The description of ritual
136
A. R. Chelariu
killing of the stag from the Romanian carol resembles the ritual killing of the ‘primodial being’, followed by its dismembering, the symbolic creation of a new world, the regrowth of the fields, and the building of a house for the future couple. The caroler gives details instructions for the young man on how to use each of the stag’s body parts: The horns should be used as stilts for his house, the blood should color the house in red paint, and the flesh should be used to feed the wedding guests. The events recorded in the Romanian carol, the killing of the sacred animal and the dismembering ritual, suggest initiation acts of a young man (Coman, 1986: I. 163). The same process takes place in another Christmas carol: The hero hunts a deer, hind, or doe; after a long and difficult hunt the deer is killed, and the hunter proceeds to perform a sacrificial fertility ritual: he gives the deer’s blood to the earth, thus fertilizing nature and ensuring prosperity of the community. In yet another carol, a young man wearing the mask of a stag wishes the people of the house prosperity, assuring them that the stag will bring them sacred objects: a table symbolizing food, crops and wealth, weapons for the boys of the house, and a wreath of flowers, cununa, Latin corona, signifying marriage for the young girl of the house. Also, in carols addressed to young girls soon to get married, there is a description of the sacred stag with a silk hammock hooked between its horns, on which the girl will be carried to her groom’s home; at her wedding the young girl should have the animal killed and, following the same rite, the blood will fertilize the earth, the horns will be used to build her house, and the meat to feed the guests. In another song, the stag carries on this hammock the soul of a dead young person, which seems to further confirm the connection between this animal and the afterlife realm, the sacred world of the divine (Coman, 1986: I. 163). The stories of the stag as offered in these carols suggest that the deer hunt motif is not isolated to the oratio nuptiae, but has deep roots in the Romanian cultural heritage. As mentioned before, such mythic motifs portraying the archaic powerful goddess, protectress of the wild animals, often featured with a deer or a stag by her side, goddess of life, death and rejuvenation, can be found in many European folklore as the fairy queen, surrounded by her companions, feared and revered by all. In an interesting article discussing the collision between two ideologies, one belonging to the Old European
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
137
cultural complex and the other of the New European imvaders, Marjia Gimbutas (1990, p. 175) states that: [The] functions and images of gods, of beliefs in afterlife, and the different sets of symbols prove the existence of two religions and mythologies, the Old European indigenous, inherited from Paleolithic, and the Indo- European, intrusive, related to the Near Eastern. Their collision on European territory resulted in the hybridization of two symbol structures. The Indo-European prevailed, but the Old European survived as an undercurrent. … Both sets are still existent today in the mythologies and folklore of Europe.
Even though some of Gimbutas’ theories are now contested, the life- giving and life-protecting goddess often represented by the zoomorphic image of a deer, preserved on cult vases from the Neolithic period, particularly on the Romanian pictorial representations created on the beautiful ceramics of Cucuteni, have continued her presence in the cultural tradition of Europe (Gimbutas, 1982). Traits of this Old Europe magnificent goddess could be recognized in the Romanian fairy Ileana Cosenzeana, ‘zâna zânelor, doanma florilor’ “the fairy of fairies, mistress of flowers” often portrayed as carried on the silk hammock between the stag’s horns, as described in the Christmas carol from above. It could be argued that the oratio nuptiae poem from the Romanian wedding ceremony stands as a viable proof of a very old ritual scenario reflecting a mythological motif of the union between a god, sometimes in the shape of a stag, as the ‘king’ of the forests, symbol of eternal rejuvenating youth, and the goddess of fertility and procreation, associated with a doe, This union was mirrored in the rite of marriage described in the song oratio nuptiae, invocation addressed to the divine powers to help the young couple into their new status, that of a husband and wife, and wishing them good luck on the way to establish their new family.
138
A. R. Chelariu
References Apollodorus. (1921). The Libray (J. G. Frazer, Trans.). Harvard University Press. http://www.perseus.tufts.edu/ Baines, K., Trans. (1962) Malory’s Le Morte D’Arthur: King Arthur and the Legends of the Round Table. New York: Penguin Books. Baring, A., & Cashford, J. (1993). Myth of the Goddess; Evolution of an Image. New York. Penguin Books. Arkana. Callimachus. (1921). Hymns and Epigrams, Lycophron and Aratus (W. Mair & G. R. Loeb, Trans.). Loeb Classical Library No. 129. Hardcover. Coman, M. (1986). Mitologie populara romaneasca, vol. I. Minerva. de Vaan, Michiel (2008). Etymological Dictionary of Latin and the other Italic languages. Liedern, Boston. Brill. Derksen, R. (2008). Etymological Dict of the Slavic Inherited Lexicon. Vol. 4. Leiden Indo-European Etymological Dictionary Series (A. Lubotsky, Ed.). Brill. Dick, Ernst S. (1966). The Bridesman in the Indo-European Tradition: Ritual and Myth in Marriage Ceremonies. The Journal of American Folklore, vol. 79, no. 312, University of Illinois Press, pp. 338–47. Dumezil, G. (1979). Le Mariage Indo-européens, suivi de Quinze questions romaines. Bibliothèque historique Payot. Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. Chicago: University of Chicago Press. Frazer, J. G., Sir. (1971). The Golden Bough, a Study in Magic and Religion (5th ed.). New York: Macmillan. Paperbacks edition 1963. Gimbutas, M. (1982). The Goddesses and Gods of Old Europe; 6500–3500 BC. Myths and Cult Images. Berkeley Los Angeles: University of California. Gimbutas, M. (1990). The Collision of Two Ideologies: When Worlds Collide: Indo- Europeans and Pre-Indo-Europeans (T. L. Markey & A. C. Greppin, Eds.). Kasoma, pp. 171–8. Gregory, Saint, Bishop of Tours. (1916). History of the Franks (E. Brehaut, Trans.). New York: Columbia University Press, pp. 538–594. Hulubaș, A. (2009). Trasee inițiatice în folklorul literar românesc. Iași. Editura Univesității ‘Alexandru Ioan Cuza’. Jordanes. (n.d.). The Origin and Deeds of the Goths (Charles C. Mierow, Trans.) with Introductory Note by J. Vanderspoel, Department of Greek, Latin and
16 The ‘Deer Hunt’ Motif in the Romanian Wedding Ceremony
139
Ancient History, University of Calgary. http://people.ucalgary.ca/~vandersp/ Courses/texts/jordgeti.html Krappe, A. H. (1942). Guiding Animals. The Journal of American Folklore, 55(218), 228–246. Laoi Oisín as ṫir na n-óg—The Lay of Oisín in the Land of Youth. (1896). Coimín, Micheál, 1676–1760 (Thomas: Thomas, Flannery, Ed.): M. H. Gill and Son, Ltd. Lyons, D. (1997). Gender and Immortality: Heroines in Ancient Greek Myth and Cult. Princeton: Princeton University Press. Marian, S. F. (1995 [1890]). Nunta la români; studio istorico-etnografic comparativ. Bucharest: Grai și suflet cultura națională. Mátéffy, A. (2017). The Wonderful Deer (ATU 401): A Pre-Buddhist Inner Asian Cultural Substratum Element in Tibetan Cosmology. In A. Mátéffy & G. Szabados (Eds.), Shamanhood and Mythology Archaic Techniques of Ecstasy and Current Techniques of Research (In Honour of Mihály Hoppál, Celebrating His 75th Birthday) (p. 337). Hungarian Association for the Academic Study of Religions. McKay, J. G. (1932). The Deer-Cult and the Deer-Goddess Cult of the Ancient Caledonians Folklore, Vol. 43, No. 2 (Jun. 30, 1932), pp. 144–174. Nunta la români; orații. 1989. Edited by Ion Moanță. Bucharest: Editura Minerva. Oakley, J. H., & Sinos, R. H. (1993). The Wedding in Ancient Athens. Madison: University of Wisconsin Press. Olinescu, M. (1944). Mitologie Românească. Bucharest, Casa Şcoalelor. Ovid. (1955). Metamorphoses (R. Humphries, Trans.). Indiana University Press. Pausanias. (1918). Description of Greece—With an English Translation by W. H. S. Jones & H. A. Ormerod, 4 Volumes. Harvard University Press; William Heinemann Ltd. http://www.perseus.tufts.edu/hopper/text?doc=Pe rseus%3Atext%3A1999.01.0160%3Abook%3D1%3Achapter%3D1%3 Asection%3D1 Thiébaux, M. (1974). The Stag of Love: The Chase in Medieval Literature. Ithaca and London. Cornell University Press. Westermarck, E. (1894). The History of Human Marriage. London and New York. Macmillan.
17 Romanian Feminine Spirits iele, rusalii, șoimane
The chimerical features of birds, or snakes, expressed by the prehistoric figurines of goddesses seem to have perpetuated in the malefic feminine spirits, the feared rusalki and/or vili, characters with a strong presence in the folk beliefs of southeastern regions, in Bulgarian, Serbian, Croatian, or Macedonian traditions. In these regions perhaps lingered in the collective memory the exposed dead on high peaks des-carnated by vultures, giving way to imagination of terrifying vulture-goddess, the goddess of death. To such terrifying memories surely may have contributed the ritual linching of Orpheus by Ciconian women on the Thracian hills (Ovid, 1955) that has so impressed the Greeks, or, the madness associated with the Maenads or Bacchantes, followers of Dionysus, Roman Bacchus, are stories that terrified humans. These demonic feminine figures with negative powers, serving the ancient goddesses, creatures with bird heads, woman-like body, and lower extremities of birds, known by Greeks as the Sirens, Erynies, or Furies, snaky virgin goddesses, Dirae, twin plaguesnaky coil with wings, Harpies, and more, flourished in Greek imagination. The Romanian oral tradition inherited, besides the good fairies, sânziene discussed above, the bad ones, iele, and the rusalii, feminine spirits related to rusalki from the southeastern European population. They were © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_17
141
142
A. R. Chelariu
feminine spirits imagined as beautiful girls with long light hair, usually living near a body of water. Three or nine in number, the rusalii ran through clouds armed with sharp tools, danced in a circle ‘hora’, leaving dark circle marks in the grass, danced wildly, following their own music, and if a human happened to see them he/she could go mad, get paralyzed, or suffer painful rheumatism. Their magic powers could also blind, cripple, or deafen the one who saw them. The rusalii would get more powerful during their feast day, the fiftieth day after Easter, when it was believed that the dead return to their tombs after fifty days of rummaging the earth, connecting these fairies with death and cemeteries (Pamfile, 1997, pp. 33–57). In Moldova, the day of rusalii was called also the Trinity holiday, a day during which people did not quarrel, fight, or work out of fear of grave repercussions. To protect themselves from the powerful fairies, Romanians wear around their waist flowers of pelin ‘wormwood’ (Artemisia absintium), decorate their houses with this plant, or with ropes of garlic (Ciauşanu, 2005, p. 203). Romanians call these fairies by many names, such as the Merciful ones, the Glorious, the Ladies, the Powerful, the Beautiful, the Field’s Daughters, the Forest’s Daughters, the Courageous ones, the Laborious, etc., all attributes describing them in good terms, hoping they will be impressed in a positive way, and lessen their negative force toward humans. As mentioned above, they also go by the name iele, DRom pronoun ele ‘they’, an euphemism used for fear of disturbing them by pronouncing their name, rusalii. A frequent practice in the Romanian tradition was the interdiction of saying or even thinking of the real name of a divinity, particularly the malefic ones. Instead, fearing the consequences of loud calling, people used made-up names, nicknames, as ‘they’ iele for bad fairies, Naiba for the ‘devil’, or Necuratul ‘the unclean one, the devil’ (Olinescu, 1944, pp. 397; 441). The DRom naiba, euphemism for ‘devil, negatve force’, as in la naiba! ‘to heck!’, offers an interesting relation to the PIE IEW 760 *noi-bhos ‘holy’, with cognates in OIr noīb, nìab ‘vital force’; OPers naiba- ‘holy’; Matasovič (2018) PIE *noybho-? ‘holy’ > OIr. noíb ‘holy’, OPers. naiba- ‘good, beautiful’, a semantic change in Romanian, proving the popular tendency to compliment the negative forces so they will be mollified toward humans.
17 Romanian Feminine Spirits iele, rusalii, șoimane
143
Romanians belived that these negative spirits were the nine daughters of Iraclie, [Heracles/Hercules] or of Rusalim (Ierusalim?) king, the owner of the fountain of immortality. One legend says that Alexander the Great went to get a jar of water from this fountain of immortality, and took the nine girls with him. When he went to war he told the girls not to touch the jar with water, but they did anyway, hence, they flew away as falcons, Romanian şoims, from which comes their other name, Şoimanele (Olinescu, 1944, p. 415), a feature close to the archaic bird goddess imagery. Franz Miklosich (1886, p. 283) argued that the rusalii name derived from the spring festivity called rosalia in Latin, via the Medieval Greek rusalia (ῥουσάλια), or pascha rosata, pascha rosarum, dominica de rosa, meaning Pentecost, or ‘Easter of Roses’, borrowed into the OCS rusalьja or rusalьe. In the Greek tradition, Rousalia was a feast of the dead, reminiscent of an ancient ‘festival of flowers’: “In Athens the ancient Anthestereia, the so-called ‘flower festival’ which was actually a feast of the dead, was replaced by the medieval and modern ‘feast of roses’ the Rousalia, held on Easter Tuesday. (Rodd, 1892) The author reports that the modern Rousalia was also celebrated elsewhere in Greece on the feast of All Souls, this time explicitly in memory of the dead. At the ancient Anthestereia, the casks of new wine were broached; interestingly, the Greek word anthos (flower), like the modern Spanish word flor, and the DRom floare ‘flower’, also refers to the yeast that forms on the top of the weak or old wine” (Jones & Pennick, 1995, p. 194). The author’s conclusion is that this festival was probably a celebration of the blooming of the new life from the old and from the decaying. Other sources relate this holiday to the Jewish feast of harvest called Pentecost by Jews speaking Greek, celebrated fifty days after Passover, when they used to decorate their homes with the fruits of the harvest. Adopted by Christians, it appeared first in Italy by the fourth century as Festa Rosalia; being accepted by the Byzantines with the same name, transcribed into Greek as Rousalia, and spread throughout southeast Europe. Once the Southern Slaves settled and adopted Christianity, the holiday became celebrated in the entire Slavic area. In Bulgarian, pycaлиа [rusalia] is the name of the lovage plant [Levisticum officiale], and also as ‘rusalca’, a fairy with magical powers, often confused with samodiva, or
144
A. R. Chelariu
samovila, a feminine demonic figure, believed to live around lakes or rivers. In Serbian, rusâlie means ‘Christian holiday’, and rusalije are women who dance and sing in the streets on the Rusalia holiday. In Russian tradition, the word is present in the 1008 Kiev Chronicle; it was believed that rusalki were ghostly creatures, mostly suffered an accidental or violent death (Warner, 2002, p. 42). From the Greek festival of wine and the dead souls to the Latin rosalia, it seems quite possible that in the Christian time, this holiday was adopted to justify an old spring festival of death and rebirth, festival in which the Goddess and her companions were part of the ritual enactments. In the end, the Rusalii with a Latinized name, are powerful fairies, living around waters and in forests, adopted all over the Balkans and Eastern Europe, and celebrated around fifty days after Easter (Jones & Pennick, 1995, p. 191). In spite of some confusing attributes, these fairies are of archaic extraction, reminiscent of an old festival of death and rebirth, of the blooming of new flowers, and the fertilizing role of waters. Their mixed attributes, mostly negative and sometimes positive, reveal their divine ambiguity of archaic origin.
References Ciauşanu, G. F. (2005). Superstiţiile poporului roman. Bucharest: Editura Saeculum I.O. Jones, P., & Pennick, N. (1995). A History of Pagan Europe. Oxfordshire: Routledge. Matasovič, R. (2018). A Reader in Comparative Indo-European Religion. Zagreb: University of Zagreb. Miklosich, F. (1886). Etymologisches Wörterbuch der 8lavischen Sprachen. Wien: Wilhelm Braumüller. Olinescu, M. (1944). Mitologie Românească. Bucgarest: Casa Şcoalelor. Ovid. (1955). Metamorphoses (R. Humphries, Trans.). Indiana University Press. Pamfile. T. (1910–1914, 1997). Sărbătorile la români; studiu etnografic. Bucharest, Editura Saeculum I. O.Press. Rodd, R. (1892). The Customs and Lore of Modern Greece. David Stott. Warner, E. (2002). Russian Myths (Legendary Past Series) London: British Museum Press. University of Texas Press.
18 The Romanian Păcală among the Indo-European Tricksters
The Trickster, as he is referred to in all the studies of mythology and folklore, is a fascinating character, whose function in the Indo-European social-cultural context deserves more attention. His presence is visible from classic mythology to the current oral traditions, as Hermes or Mercury in the Greek and Roman mythologies, Loki in German sagas, Bricriu in the Irish songs, Pekulis, Patullos in the Baltic heritage, or Varuna in Hinduism. In the European folklore, the Trickster is found in each cultural complex: he is well known as Thyl Ulenspiegel in German culture, or Peik in Scandinavia, Pooca, Puca in Ireland, Velnius in the Baltic region, or Păcală (pronounced Pəkálə) in the Romanian tradition. According to Herodotus (1920, Histories 2. 51), Hermes was adopted by the Athenians from Pelasgi, and was believed to have been born in Arcadia, where he was particularly honored by the Arcadian shepherds as the protector of their flocks and huts. As a symbol of veneration, a rudimentary image of him was often found by the shepherds’ huts doors. This image was a pillar featuring a bearded head, or realistically modeled erect phallus. These phallus-shaped stones or painted images, these ‘hermai’, were considered aggressive symbols of boundaries and fecundity, placed
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_18
145
146
A. R. Chelariu
in front of houses, in market places, but also at crossroads, as reminders of the god’s role as the ethereal guide of the soul into the world of the dead. Hermes the Trickster, is the son of Zeus and Maia as Apollodorus gives a short version of the Homeric Hymn to Hermes (Morford & Lenardon, 1985, p. 179): A few hours after he was born he steals Apollo’s herds, and to trick the sun god and hide his act he wears huge sandals to hide his footprints, and makes all the cows walk backwards so that no one could tell where they are hidden. After that on the banks of Alpheios (the place of the Olympic Games), Hermes performs a noteworthy ritual: He sacrifices two cows, dividing the meat into twelve portions in honor of the twelve gods. Then, by rubbing two laurel twigs, he makes a fire and cooks the meat; although the aroma is appealing to him, he, as a god, must restrain from eating the flesh (Apollodorus, 1921, The Library: 3.10.2). This episode may explain why Hermes is often regarded as the god who taught people how to make fire with twigs. On his way home, contrary to time of the beginning of his adventures, he is not seen by humans, and dogs don’t bark at him; perhaps his sacrificial act may signal his transformation into the flying god, invisible at wish, the messenger of gods, conveying the divine wisdom to mankind. Apollo’s gratification for all his troubles is a musical instrument, a lyre that makes beautiful sounds by vibrating the air, perhaps in relation to the flying attributes of the young god. In turn, Apollo entrusts him with protecting his herds, thus confirming Hermes’ position as the protector of shepherds, their flocks, and even their thieves (Homeric Hymn 4 to Hermes, trans. Evelyn-White). In one of his adventures Hermes is tending the sheep of a mortal man, and falls in love with the man’s daughter; from their love adventure, she has a son with horns on his head and hooves instead of feet (Homeric Hymn 19 to Pan, Hugh G. Evelyn-White, Ed., 1914). Although he scares his nurse, the gods are happy to see the half-human half- goat child, and called him Pan. Being a very lascivious creature, Pan follows the nymph Syrinx one day, but she runs away from him, and her sisters, to hide her, turn her into reeds; this prompts him to cut seven or nine reeds, and tie them together in what became the musical instrument, the reed pipe. The Lithuanian Pekulis, or the Prussian Patollus, are regarded as chthonic deities, agrarian helpers, who live near bodies of water. In the ancient times, Pekulis was feared as the god of death. He became known
18 The Romanian Păcală among the Indo-European…
147
in the Christian era as Velnias, the flying evil spirit, in which we recognize the element of air as a medium and an attribute of this mythical character. According to Gimbutas, Velnias is a chthonic divinity, a pastoral god, who helps the poor, displaying a perpetual resistance to Perkunas, the sovereign god (Gimbutas, 1963, p. 202). As Hermes’ symbol was a phallus, so Pekulis/Velnias often shows his enormous phallus to women. Velnias goes to weddings and frightens women who will not dance with him. He is known as the one who punishes the unfaithful. This character’s connection with animals is found in a Prussian triptych, where Patullos, whose sacred objects are “the skulls of a man, a horse, and a cow” (Puhvel, 1987, p. 225) is represented in the form of a horse’s skull. Likewise, the old Russian Veles, also named Volos, a parallel of Velnias, is a chthonic god of horned cattle in permanent conflict with Perun, the thunder god; his function is that of a shepherd of the dead (Eliade, 1994, pp. 3. 37). In Northern mythology, the Trickster, Loki, is a companion of gods, a giant able to change his sex and shape at will. In his female form he gives birth to monsters, thus causing considerable troubles for the other gods. He is the ‘mother’ of Hell, the giantess ruling the realm of death, he scares the gods with old age and dying. Loki is a thief stealing or helping others to steal gods’ treasures, then helping them to recover the goods in exchange for favors. In the poem “Lokasenna” (Poetic Edda, 2015), the gods try to keep him out of the Hall of Aegir where they are having a feast, but he manages to enter anyway. Once inside he starts insulting and betraying shameful secrets about the gods’ cowardice and the goddesses’ infidelities. The Irish god Bricriu, whose nickname was Nemthenga, Poison- Tongue, builds a splendid hall, preparing a feast to which he invites all gods of Ulster. When they refuse his invitation, he threatens to make them kill each other, and to turn daughter against mother. After arriving at the party they try to keep him out of the hall, but he, as in the German story, manages to get in, and incite the gods against one another. In the end, the hall is damaged, and he is covered with grime (Early Irish Myths and Sagas, 1981, p. 219). In the Indian mythology, the pair Varuna and Mitra’s essential function is to maintain the universal order. While Mitra watches over
148
A. R. Chelariu
friendships and ratifies contracts, Varuna is the lord of moral order watching for sinful behavior. As the guardian of the oaths he upholds ṛtá, the cosmic moral norm, a quality which links him to one aspect of the Trickster’s role. Present at every gathering, witnessing every action, he could be ferocious, chaotic in his impulses if the rules are broken, punishing those who break them. Varuna sees, or ‘shines’ at night, thus he is linked to the moon, the place of the dead, sharing with Yama the title of ‘King of the Dead’ (Dumezil, 1988, p. 76). The tricksters from the folklore share many common features, showing little differences from the mythical character, essentially, antagonizing the authority and watching over the moral order. To illustrate this point, we will analyze the Romanian folk story of Păcală, and observe the correspondences with the other Tricksters, as well as the differences. In most anthologies of Romanian folktales, the story of Păcală begins with the trickster building a big fire on top of a mountain, adding plenty of resin to it, causing a thick smoke to rise up to God, who happens to have a cold. The smoke cures God’s cold, who graciously offers Păcală anything he wants. The Trickster asks for a flute, a musical instrument very common among shepherds. He receives a flute that has the magic powers to make anyone dance incessantly until the music stops, similar to Pan’s flute making the satyrs and the fairies dance. The magic flute episode is found in the Neo-Greek, Irish, and German folklores as well (Săineanu, 1978, p. 604). Other mythical characters creating magic with their musical instrumants are Amphion, Orpheus, Oberon, and Hameln. As the story goes, Păcală inherits a cow, and goes to the market to sell it. On his way to the market he stops in the forest, and sells his cow to a tree, but he gets upset because the tree would not answer him, or give him his money. In anger he strikes the tree only to find a treasure in its hollow. In the Western folklore the tree is replaced by a statue of a god or saint. In Aesop’s fables, we find a story of a man who asked Mercury’s statue for money, and because the god does not seem to hear him, he gets upset, breaks it, and finds a treasure in the statue’s head. In Cosquin’s French collection, the character Cadet Cruchon sells a piece of linen to a saint’s statue, and because the saint does not want to pay him, he breaks the statue and finds a treasure. In the Neapolitan version Vardiello does the same thing, and Giufa from Sicily wants to dye his linen green, sells
18 The Romanian Păcală among the Indo-European…
149
it to a lizard, and the treasure is then found in the lizard’s house (Săineanu, 1978, p. 607). As it is evident from these examples, in the East European folklore, this motif includes the cow/bull and the tree elements, whereas in the West European we find a saint statue and a piece of linen. Next, Păcală gets hired as a shepherd by a greedy priest, with whom he enters into a work contract; they vouch to keep their agreement until one of them gets angry, and the one who breaks the contract ought to be punished in various ways, usually to have a piece of skin cut off his back, or to have his nose cut off. Next, the trickster starts his malicious pranks, trying to force the priest to lose his temper. Some of his pranks are: making the priest and his wife dance until they are completely exhausted; going up in the meadows with the herd, and because the priest’s wife doesn’t give him any lunch, he cuts a cow and eats it on the spot; the next day he cuts a pig, then a lamb; another nasty work he does is peeling off the child’s skin when told to clean him. Another time, the priest’s wife tells him to cook soup, and to add parsley to it, but he cooks the dog with the name Parsley. In the end, the priest loses his patience and gets punished. In stories from Corsica, Picardia, or Serbo-Croatia, instead of a priest the Trickster enters into a contract with a rich man or a king, who endures almost the same types of pranks, and gets punished just the same (Săineanu, 1978, p. 607). In the Irish folklore, the trickster Pooka is considered essentially an animal spirit, whose name ‘poc’ means ‘he-goat’, but he takes many shapes, horse, ass, bull, goat, etc. As a horse he has to be kept away from any sites of water, because he will plunge in with his rider and kill him. In an Irish story, a piper, on his way to sing at a party, meets with Pooka, who plays a few spiteful pranks on him, then gives him a pipe that makes him a famous piper (Treasury of Irish Folklore, 1985, p. 509). The Scandinavian Peik gets horses and cattle from the bottom of the sea. Both, the Irish Pooka and the Scandinavian Peik own the magic flute that makes people dance until exhaustion. Peik is a feared trickster, always challenging the king, who attempts to outsmart him, but with no success (Scandinavian Folk and Fairy Tales, 1984, p. 96). The names of all these European folk tricksters, Pekulis, Patulos, Pooka, Pooc, Peik, Puc, Păcală, point to a common Indo-European root *pek̑ - (PIE IEW 797) ‘livestock, domestic animal, cattle’, with cognates
150
A. R. Chelariu
in Latin pecū, −ūs n. ‘a head of cattle, beast, brute, animal, one of a herd’, next to which pecus, −oris n. (formal = Grk τὸ πέκος), pecus, −ŭdis f. ds.; derivatives pecūnia ‘property, riches, wealth’, pecūlium ‘property’; Umbr. pequo pl. n. ‘pecua’. Comparing Hermes’ actions to those of the Romanian Păcală, such as the use of fire, the musical instrument, the cows/animals sacrifice, and being entrusted as shepherds to take care of herds (Hermes was entrusted by Apollo with protecting his herds—Păcală is hired as a shepherd), all show many similarities between the mythic and the folk characters. The tricksters from the Indo-European myths, Hermes, Pekulis, Loki, Brikriu, etc., fulfill an essential function as guardians of moral principles, passing down the ethical teachings through tricks, a function that has endured over the millennia. The trickster fulfills his function by performing actions that always puts a devious character on trial through his numerous pranks. It may be possible that under his tricks there are hidden encoded messages waiting to be deciphered, revealing his function of messenger of the divine rules, to which we can add the role of catalyst that he plays in myths and folktales, affecting the fortunes and behavior of gods and men. As we have seen, the trickster of ancient mythology plays tricks on the other gods, while the folk character plays tricks on the authority, the god’s representative on Earth, a priest or a king. In myths his pranks address the divine authority, Apollo, or other gods, whereas in the West European folklore this divine authority is represented by the king of the land, the statues of a god, or the landlord, that in the Eastern Europe is replaced by the priest. When the Romanian trickster Păcală enters into a contract with the priest, forcing him to break it, he maintains the moral order. Păcală always punishes the greedy ‘devil’s priest’, the unfaithful wife, the thrifty fellow, and the perverts. He is in permanent conflict with the bad spirits and the devil. He is the moral compass, and watches over the ethical order within the community. He seems to be closer in his vigilance to Varuna, the god who keeps a very sharp eye on every contract, and punishes bad behavior. Besides similarities, there are some rather important differences between the mythic Trickster and the folk character: Păcală, with all the other Tricksters from the European folklore, never steal a herd from
18 The Romanian Păcală among the Indo-European…
151
authority, but Hermes does. In myth, Apollo entrusts him with his herds after the boy plays tricks on him, whereas in the folk story the trickster enters into a contract with the authority, and then starts his malicious pranks. Another difference is that Hermes, as a god himself, doesn’t eat the meat he has sacrificed, while Păcală eats it. The significance of these differences may rest with the fact that, over time, ritual practices and their significance, suffered modifications from myth to oral tradition. Following the Dumezilian social “class” structure in the Indo-European society, we may speculate that Păcală, together with Pooka, Peik, and the god Pekulis, as shepherds dwelling among the shepherds and fieldworkers, belongs to the third class, which governs wealth and fertility. Yet, his role as the guardian of the moral balance positions him within the first class, together with the keepers of the cosmic and judicial order, a divine ambiguity which defines the trickster as a character able to move freely between classes. Why is this character playing tricks on the other gods? Is it because, as suggested by Gimbutas and others, he may be part of an older agrarian pantheon, a pre-Indo-European god? This question may be difficult to answer, and Varuna, or even Pekulis, may stand to prove otherwise. The parallels between the mythic character and the folk one display a proclivity toward moral balance, regardless who the guilty is, god or man. When the Trickster exposes the faults of gods or goddesses, the greediness and the stupidity of the king or the priest, the story enhances moral values within the social group, and the ethical and moral principles are transmitted from generation to generation.
References Apollodorus. (1921). The Libray (J. G. Frazer, Trans.). Harvard University Press. http://www.perseus.tufts.edu/ Dumezil, G. (1988). Mitra-Varun; An Essay on Two Indo-European Representations of Sovereignty. New York: Zone Books. Early Irish Myths and Sagas. (1981). Translated by Jeffrey Gantz. New York: Penguin Classics. Eliade, M. (1994). Istoria Credintelor si Ideilor Religioase (3 vols.) Chişinău. Universitas.
152
A. R. Chelariu
Gimbutas, M. (1963). The Balts. London Hudson: Frederick A. Praeger Publishers. Herodotus. (1920). Histories (A. D. Godley, Trans.). Harvard University Press. http://www.perseus.tufts.edu/cgibin/ptext?doc=Perseus%3Atext%3A1999.0 1.0126;query=chapter%3D%23103;chunk=chapter;layout=;loc=1.102.1 Hymni Homerici; Homeric Hymns: Homeric Hymn 19 to Pan. (1914). Hugh G. Evelyn-White, Ed. http://www.perseus.tufts.edu/hopper/text?doc=urn:cts :greekLit:tlg0013.tlg019.perseus-eng1:1 Morford, M. P. O., & Lenardon, R. (1985). Classical Mythology. New York and London: Longman, Inc. Puhvel, J. (1987). Comparative Mythology. Baltimore: John Hopkins University Press. Săineanu, L. (1895 [1978]). Basmele Române în comparaţiune cu legendele antice clasice. Bucharest: Editura Minerva. Scandinavian Folk and Fairy Tales; Tales from Norway, Sweden, Denmark, Finland, Iceland. (1984). Edited by Claire Booss. New York: Avenel Books. The Poetic Edda: Stories of the Norse, Gods and Heroes. (2015). Translated by Jackson Crawford. Indianapolis: Hackett Publishing Company. Treasury of Irish Folklore. (1985). Padraic Colum. New York: Random House.
19 The Hero Slaying the Dragon in the Romanian Song Iovan Iorgovan
The hero slaying the dragon is believed to be the quintessential Indo- European myth that has endured for thousands of years. Numerous songs, ballads, and fairy tales retell the story of the hero who must restore cosmic order by slaying the malefic dragon that created the havoc. Many European folktales describes the dragons as monstrous and fierce beasts, in the shape of a very large snake, spitting fire through their nostrils, with one or more heads, and mouths with many tongues; sharp fangs, and sometimes, a set of bat-like wings complete the image of this fantastic beast that belongs to the pre-cosmic era, a creature of the chaotic elements of nature. Gods and heroes must reinforce their sovereign powers over the dragon’s force of destruction and chaos, and create or restore the cosmic order. As obstructer of waters, the dragon must be vanquished by the storm god, who frees the rain, and returns fertility to the land. and prosperity to the community. These monstrous creatures are, sometimes, guardians of the tree of life, or the Golden Fleece, or guard some other kind of wealth, as the gold from the Germanic sagas, for which the hero has to fight the monster. Stories in which dragons threaten the life of women, or request the sacrifice of a young girl in exchange for the
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_19
153
154
A. R. Chelariu
fertilizing water, or something else valuable to the community, are also widespread. In more recent times, dragons became agents of the devil, as depicted in the well-known Christian portrait of St. George slaying the serpent, the enemy of God, or, as in some legends, the monster threatens the life of a princess, and the Saint saves her from perish. Aside from the Christian influence, most of the folk traditions retained the archaic function of the serpent-dragons as a source of fecundity and life, or, embodying the souls of the ancestors of a household. The mythical motif the dragon slayer from the Romanian and European folk songs has its roots deep in the ancient hymns and prayers, addressed to the storm god, who must fight the obstructer of rain, and restore the cosmic order. One of the oldest versions of the dragon slaying myth is found in a Hittite ancient prayer song, asking gods for rain and abundance of crops. In the text published by W. Burkert (1979, p. 8), the Storm god Tarhunt fights the dragon Illuyankas, and loses the first encounter; the Storm god goes for help to the goddess Inaras; with the help of a mortal man, the dragon is defeated, and dies together with the human. In Gaster’s opinion, the Hittite myth of the dragon slaying is a seasonal ritual: “…at the beginning of the agricultural year,…the ancient Hittites held a festival, which they called Purulia… a seasonal celebration designed to regulate the subterranean waters and to insure due med of rainfall for the crops” (Gaster, 1961, p. 137). The Greek tradition knows many versions of this myth: the Earth monster Typhoeus or Typhon, the son of Gaia. linked to strong storms and hurricanes, is struck by Zeus’s lightening. Hesiod (2018, Theog 820–835) describes this monster as having a hundred heads with black tongues flickering and fire spreading, with terrible voices coming from each head, a frightening monster challenging the sovereignty of Zeus. Apollodorus (The Library 1.6.3) gives the following account of the fight: After being struck by the god, the dragon coils around Zeus, and, taking his sickle, cuts the god’s sinews from hands and feet, leaving him in the Corycian cave; but Hermes recovers them, and gives them back to Zeus, who defeats the monster, and, in the end, imprisons it beneath Mount Etna, where it is still causing havoc with the volcanic fires.
19 The Hero Slaying the Dragon in the Romanian Song…
155
The motif of fighting with monsters can be related to the foundation myths, for example, one such myth is that of Apollo slewing the huge serpent, Python, at Delphi where he could establish his oracle. In another story told by Apollodorus, Cadmus (Latinized form of the Greek Kadmos) has to follow a guiding cow to the place the animal would lay down in order to establish the city of Thebes, as discussed above under ‘the guiding animals’; Cadmus wants to sacrifice the guiding cow to Athena, and sends his men to bring water from the nearby spring, but the waters are guarded by a dragon, an offspring of Ares. Cadmus kills the dragon, and, as instructed by Athena, the hero sows its teeth, from which the armed men called Spartoi (Sown Men) arise (Apollodorus: The Library 3.4.1), as Chrysaor and Pegasus sprang from Medusa’s neck when Perseus slayed her. (Apollodorus 1921, The Library 2.42). The famous hero Heracles in his labors encounters monsters, as when he saves Hesiona from a seadragon, or smashes each of Hydra’s heads with his club (Apollodorus: The Library 2.4.8–2.7.7). In the Indian tradition, the god Indra fights with the monster Vṛtra, obstructer of the cows symbolizing water and dawn; upon seeing the serpent, Indra gets scared and turns away, a similar first defeat that was suffered by the Hittite Storm God or Zeus; then he recovers, and strikes the serpent with his thunderbolt, releasing the waters needed for the crops. From the monster Vṛtra’s teeth comes out the fierce fighters Maruti, as the Spartoi from the dragon’s teeth in Cadmus’ story; then, from each of his heads fly away a flock of partridge, a sparrow, and quail birds. In the Iranian version of the myth, the hero must vanquish the monster with three heads, named Azhi (serpent) Dahaka, who lived in a place named “the palace of the Stork”, perhaps in a connection with the birds coming out of the monster Vṛtra’s head. In some later version, the monster is imprisoned beneath a volcanic mountain as in the Greek myth (West, 2007, p. 262). The German thunder god, Thor, plans to vanquish Jörmungandr, also known as Midgard serpent, coiled like a belt around the world, using his hammer called Mjollnir or ‘Crusher, maul’ (wooden club), generally identified as a thunderbolt (Snorri 2005, The Prose of Edda: 48) In an Irish story, the hero Fraich (Fráech Fróech, Fraoch) swims across a boundary, suggesting the crossing into the other realm, a pool in which
156
A. R. Chelariu
lives the dragon Lord of the Otherworld, a passage describing perhaps a rite of initiation. During the encounter with the monster, Findabair, the hero’s lover, comes to his aid, and gives him the sword with which he slays the serpent, but in the fight he is badly wounded (Brenneman, 1991, p. 73). In another Irish story, Fergus mac Léti goes underwater where he sees the water monster, ‘muirdris’. He becomes disfigured by fear, and, for seven years, he is under interdiction to see his face, and a slave girl washes his face for him; one time, because she is not answering his request quick enough, Fergus kills her, and plunges underwater. After one night and one day, he comes out holding the monster’s head, then he falls down dead (Watkins, 1995, pp. 441–447). In a less known Irish story, the dragon Méiche, son of Morrigan, had three hearts in the shape of three serpent heads, and the hero Diancecht had to kill it, or those snakes would grow in his belly, and no animals would be alive in the country of Ireland (Shaw, 2006, p. 163). The Russian folklore records the dragon fight story in the well-known ballad Dobrynya Nikitich slaying Zmey Gorynych: Dobrynya Nikitich was bathing in the Puchai River when the dragon appeared; initially he thought he was going to die because he had no weapons at hand to defend himself, but he finds “a hat of the Greek land” and uses it to defeat the dragon. The dragon asked Dobrynya Nikitich not to kill him and they agreed not to attack each other again. Soon after, the dragon captures the niece of Prince Vladimir, Zabava Potyatichna. The prince asks the hero to rescue his niece; Dobrynya Nikitich goes to the Saracen Mountains and the fight starts; on the third day of fighting, when Dobrynya Nikitich is ready to give up, he hears a voice from heaven telling him to go on fighting, and after three more hours the hero kills the dragon; as the story goes, because the dragon’s blood did not sink into the ground the hero is stuck in the monster’s blood for three days. Once again, he hears a voice from heaven telling him to stick his spear into the ground while uttering some magic words, and so the blood disappears into the earth and Dobrynya Nikitich rescues Zabava (Bailey & Ivanova, 1992, p. 81) In the Romanian, as in all the European folklores, the motif of the hero killing the dragon or monster, or giant (ATU 300 The Dragon- Slayer), threatening the life of a fairy/princess, is frequent in fairy tales, in which the Prince Charmig Făt Frumos fights the dragon, DRom balaur,
19 The Hero Slaying the Dragon in the Romanian Song…
157
to free the fairy, Ileana Sânziana. Aside from fairy tales, the mythic motif of the hero fighting the dragon is featured in the Romanian song “Iovan Iorgovan” (Vrabie, 1966, p. 159). Recited at fests all over the country, more frequently in the Southern part of the Romanian territory, the song is interpreted by old bards during holidays, performed in the specific chanting rhythm. The story narrate that three sisters were going for a walk in the forest; needing the rest, the older sisters abandon the youngest while she sleeps. The ‘three sisters’ motif, and the abandonment of the youngest one, may not be found in all versions, some folklorists considering it a contamination with the ballad “Three Sisters”, or with the fairy tale of Ileana Sânziana mentioned above. Alone in the forest, the girl finds herself threatened by a dragon coming out from a nearby cave; her screams are heard by the hero Iovan, who runs to her rescue. To reach the girl, Iovan has to cross the very turbulent dark River Cerna; he must offer magical objects and gifts in exchange for letting him cross it. Among these gifts are: a magic silver spinning wheel that turns by itself, and a magic fish; these enchanted offerings may indicate old rituals revealing the river’s sacrality and divine functions. The DRom hydronym Cerna, and the Dacian toponyms Dierna, Tierna, Zerna, explained by the PIE MA *dhergh- ‘sloetree, blackthorn’, are cognate with the Dacian προ- διαρνα (Veratrum nigrum), perhaps used to die clothes dark; other developments in DRom n. zârnă, zărnă, dzărnă ´poisonous plant’ Solanum nigrum, euphemism for a black sheep, and the verb a zârni ‘to darken’ (Istoria limbii române, 1969). Reaching the girl’s place in the forest, and await the dragon’s reappearance, Iovan lays his head on her lap, and falls asleep, in accord with the known motif of Deus otiosus ‘the sleeping god,’ frequently found in myth and folklore of Southeastern Europe. He is awakened by the girl’s tears at the site of the dragon. Before fighting, the dragon warns Iovan that, if he will cut off his head, a very dangerous horsefly will come out of it, a fly that could kill horses, cattle, and even people. The hero tells the monster that he will teach people how to start fumes to kill the fly, and save everyone, his instructions confirming his divine role as a culture hero. In some variants the story ends here, without further reference to an actual fight; in others the hero slays each of the three dragon’s heads with his arrows. There are some versions in which the hero is mesmerized by the young
158
A. R. Chelariu
girl’s beauty and wants to marry her, but amazed by their resemblance, she recognizes him as her brother, thus, their marriage would be incestuous (Vrabie, 1966, p. 159). By recognizing the hero as her brother, the story alludes to the hierogamy motif as preserved in another Romanian song The Sun and the Moon (see Chap. 14) establishing her position as a divine character. According to the Dumezilian classification of the Indo- European social functions, the girl/fairy, with her beauty, and her association with a body of water, the river Cerna, belongs to the third function, that of fertility, productivity, and wealth, and the hero, through his actions, the magic gifts offered to cross the magic river into the realm of the goddess of fertility, and the fighting with the dragon for saving her, position him into the role of a storm god. In other versions of the Romanian poem, the episode of the three sisters is replaced by that of a young man whose life is threatened by a huge snake, DRom balaur (Vrabie, 1966, p. 146). In this version, we are told that the lad’s mother cursed him when he was just an infant ‘to be taken’ away by the snake living under the threshold of their house. In the Romanian tradition, the snake is perceived as a fantastic creature living under the house threshold, that, after seven years, reaches maturity, becomes a huge dragon, and rises to the sky as a storm cloud. The ‘cursed’ boy grows simultaneously with the snake; when they both reached adulthood the lad must fulfill his destiny, and fight the house snake, that grew into a huge dragon. During the fight, the dragon swallows the young man up to his waist, up to his weapons, which are ‘ferecate’, a word that could be translated as ‘locked (by a spell)’. Iovan hears the lad’s cry and runs to his help, killing the dragon. Afterward, they become, fraţi de cruce ‘blood brothers’. This motif from the Romanian song can be associated with the Thracian-Phrygian mysteries of the god Sabazios, whose priests were snake-keepers. His initiates handled snakes as in the warriors’ military initiation rites, during which a serpent was drawn across the bosom of the neophyte to give him hope for attaining immortality. According to the tenth century Byzantine Encyclopedia, known as the Suda Lexicon, during this mystery cult, a cry called ‘Sabasmos’ was recorded, supposedly by Clement of Alexandria: “On attaining manhood you abetted your mother in her initiations and the other rituals, and read aloud from the cultic writings. …You rubbed the fat-cheeked snakes and swung
19 The Hero Slaying the Dragon in the Romanian Song…
159
them above your head, crying Euoi saboi and hues attes, attes hues.” This ancient rite shows possible relationship to the Romanian belief about the snake living under the house threshold as described in the Iovan Iorgovan song, a snake that embodied the soul of the ancestors. Romanian traditions record other beliefs about house snakes, as being harmless creatures, that bring good luck to the family, and they should never be killed for fear of killing the good luck bestowed upon the house (Olteanu, 1998, p. 309). Iovan saving of the young man episode from this song may describe a military initiation rite during which the neophyte fighting the ‘dragon’ is swallowed up to his weapons that are locked, not ready for fighting; he is immersed into the magic forces of the house serpent/dragon, symbolizing his ancestor warriors. After passing through this rite under the guidance of a master warrior, the neophyte is admitted into the group of warriors, with his weapons unlocked, invested with the magical powers coming from his ancestors. This conclusion is in accord with George Dumezil’s statement accoring to which warriors motifs from the Indo-European myths “may now be interpreted literally point for point, as a memory of much older rituals and myths of initiation or military promotion” (Dumezil, 1970, p. 160). The initiation of the young man ends with Iovan killing ‘the dragon’, and admitting the young warrior into his entourage, making him his ‘blood brother’, a widespread and well- documented custom in the Romanian tradition among young men. As Dumezil continues, “This does not, of course, prevent the myths from having also been—and even congenitally—myths of the storms. It is the destiny of the warrior gods, patrons of the terrestrial warriors, to be storm gods as well, or to have a tendency to become confused with them. Thor, the thunder with his hammer, like Indra with his thunderbolt, has obvious nature god significance…” (Dumezil, 1970, p. 161). The presence in the Romanian folklore of these two songs, one involving saving the fairy from the monster, and the other saving/initiating of the young man, could bring to light the multiple connotations and complexities of the Indo-European myth of the dragon-slayer. Stories about snakes and dragons are numerous as well in the folklore of the Southern Europe. The region is also known for the many monuments, funerary or dedicated to certain gods, depicting the famous
160
A. R. Chelariu
Danubian horseman statues and bas-reliefs, portrayed riding to the right, and holding a spear in his right hand, ready to hunt, perhaps reflecting the Thracian, Dacian and Celtic forces, known for their remarkable cavalry. These graphic images resurfaced later in the images of the Christian Saint George holding a spear, ready to kill the snake. Amid the main characters involved in this myth, the hero and the dragon, special attention should be given to the motif of water, be that of Vrtra obstructing the falling of the rain, or the lake in which both, the Ireland hero and the Russian one, are sinking, to which we add the River Cerna, which the Romanian hero has to cross. From ancient times, people believed that in any body of water, spring, river, or lake, fantastic creatures are settled. Among the ancient Greeks there was the belief that springs and wells were protected by a dragon or a large snake, known as lamnia; the same belief is found in the Macedonian folklore where this monster was called lamia or lamnia, and in the Bulgarian tradition, lamiia (Șăineanu, 1978, p. 531). This monstrous character is also present in the Lithuanian folklore under the name Laume, who “lives and appears in the vicinity of water” (Dundzila, 1991, p. 135). The main belief was that this feminine monster, lamia/lamnia, living in streams, wells, or lakes, would not allow anyone to take water for daily needs without a human sacrifice, in most cases, a girl, beliefs reflecting a strong association between water, dragons, and other monstrous creatures. Other beliefs on the relation between dragon and water are found in the Romanian traditions, where the dragons are regarded as rainy clouds, imagined as the sun’s bulls, pulling his cart with water; when they got lazy they would let the water pour over the cart, and thus, it rained. Other stories describe the clouds as dragons driven by ‘solomonari’ (untranslatable, of unknown etymology), highly trained wizards with special powers, and able to ride these creatures, whipping (lightening) and driving them over the sky (thundering), forcing them to burst the water, and start the rain (Niculiţǎ-Voronca, 1998, II p. 160). Other kind of dragons from the Romanian oral tradition are believed to be fierce cruel creatures, with several heads (three or seven), each with a large mouth that could swallow a human being. Before becoming large dragons, they were just snakes, that every seven years in a spring day would get together in a certain place, and collect their own drools, that
19 The Hero Slaying the Dragon in the Romanian Song…
161
will turn into a precious stone; a fierce fight would start among them to swallow that stone; the one managing to swallow the precious stone would run away into the wild forest, where if, for seven years it would not meet any human being it could become a dragon. Another belief tells as these snakes/dragons that are ruling the wells and the springs, use the rainbow as their road. Their most widespread image among the Romanian folks is that of a strong storm (Șăineanu, 1978, p. 312). This creature is also present in the Romanian cosmogony: The creator Fârtat punishes the balaur ‘dragon’ for the continuous mischiefs to coil nine times around the Earth, and protect it from floods (Vulcǎnescu, 1987, p. 425), a motif reminiscent of Jörmungandr, also known as Midgard serpent, the Old Norse dragon coiling around the Earth in the depth of the ocean. The Romanian word for dragon ‘bală, balaur’, has its roots in the PIE MA *bhel- ‘blow, swell’, *bhleu- ‘swell, overflow, roar’, and IEW 96 *bel-2 ‘strong’, also in IEW 120–122 *bhel-3, bhlē- ‘to grow, spread, swell,’ found in the Dacian anthroponyms Balas, Balius, Areibalos, Bazobalis, Decebalus, Dinibalis (I. I. Russu, 1967, p. 92), meaning ‘strong man, king’, with cognates in Skt. bala ‘physical power’, Ir bale ‘strong’, Alb bollë ‘snake’, Skt bála-m n. ‘force, strength, power’, Phrig. βαλήν• βασιλεύς. Φρυγιστί ‘βαλήν—king (Hsch); the Serbian and Bulgarian forms blavor ‘hurricane, terrible event’ are considered loans from DRom; Lith bljaúju ‘spew’; OCS bljujǫ ‘spew’; Grk phleō ‘gush, teem, overflow’, and Lat dēbilis ‘feeble, weak’. The DRom forms for ‘monster, dragon’ bală, balaur, could be related to negative divinities developed in the Russian tradition, Velesu, Volosu, Lithuanian Velnias, Skt Varuna, Gaul Vellaunos; Hit Walis (with a possible etymology in PIE *wel ‘see’; 2. *wel ‘die’ as per Jakobson, 1969 in Petrosyan, 2016), all mythic characters linked to death and the underworld, overseers of the world. The slaying of the dragon myth is generally interpreted as the symbolic victory of order over chaos, of growth over stagnation, of rebirth over death. It is a myth that must be recited during fests celebrating the annual cycles, enacted cyclically, in order to maintain its magic force. As Calvert Watkins (1995, p. 446) confirms, the dragon represents the chaotic world that the sovereignty must subdue in order to restore order. The hero has
162
A. R. Chelariu
the duty to fight the dragon causing drought by obstructing of fertilizing waters, to free them, and to ensure the abundance of crops. Being a fertility myth it has to be reenacted cyclically, and Michael Witzel (2007, p. 203) places the cyclical reenactment of the waters release at midsummer. To this West (2007, p. 261) adds the two separate dragon stories from the Ṛgveda, one relating to the release of the sun, and the other, to the release of the waters, and Emily Lyle mentions, “the full winter story is one of the disappearance and re-appearance of the sun goddess while the summer story treats her birth” (Lyle, 2012, pp. 105–107). In the Romanian tradition, the sun is a masculine entity with a sister, the moon; the dragon balaur is also a masculine form, essentially a serpent—șarpe (masculine), therefore, not a feminine dragoness. In another Romanian poem, the hero fighting monsters was revered as a god of warm waters, that sprang from the Mehadia Mountains, near the Iron Gates in the southwest area of Romania. The Romanian researcher Nicolae Densusianu regarded this god as the Dacian version of Hercules, worshipped as a god of warm waters: “in popular traditions from Banat Iovan Iorgovan was spending time in a cave near the Mehadia thermal bays, named after him, the Iorgovan Cave. Here, near the warm waters by the Cerna river is the place mentioned in the legend about the killing of the lion or lions in the Nemean forests: El din peșteră-mi pleca/From the cave he would get away Si la Cerna dobora/By the Cerna river he would crush them Pe tustrei îi omora/All three he would slaughter Si pieile le lua/And their skins he would take Cu ele se îmbrăca/Wearing them on him Si mai departe pleca/And away he would go
His weapons were those from the North of Ister: the bow, the sword (ἄρπη), and the club. Son of Joe, named Herculus Jovius or Hercules Jovius, his co-name was Iorgovan corresponding to the Greek Γεωπγος (Αgricola), either because he was the first who himself used a plough, or because he was considered the son of Ζευς Γεωργος…” (Densușianu, 1913/1986, p. 544). Following his argument, we may entertain a possible relation between Hercules Jovius, son of Ζευς Γεωργος, the mythical hero, whose function was to kill monsters, and the Romanian Iovan Iorgovan.
19 The Hero Slaying the Dragon in the Romanian Song…
163
The core of this popular version of the myth, in which the hero, be that Perseus, Hercules, or Iovan Iorgovan, saves a girl, a fairy or a princess from a monster suggests a second function of the dragon: next to keeping the fertilizing waters, he is impeding her main feminine function, that of procreation, of creating life; by threatening the life of a girl/fairy, the dragon interferes in the essential role of a woman and/or goddess of fertility, threatening the community in which she, as a goddess protects life, or, as a mortal, she marries and procreates. Hence, it is the duty of the hero to fight and restore her position as a wife and mother. The connection between the feminine entity in this myth and the dragon-monster, could indicate a forgotten initiation rite of a girl, similar to the one involving the young man’s initiation into the military group. These functions of the sovereign power under obligation to restore order, and fight for the release of waters to bring prosperity on the land, or to protect against the endangering of a girl/fairy fertility function, underline a common archaic mythical structure, in which the hero has to fight an opposing magic force, to reinstate order in the community. This basic mythic motif suffered in time various influences, in some cases the sovereignty aspect of the story prevailed, while in others the fertility was more important to the group. Synthesizing, the common motifs in these folk data in comparison with the ancient versions discussed here, lead to position the hero from the Romanian folk song toward fertility as his main function. Here are the common mythic features: • Iovan Iorgovan, the dragon fighter, uses as his weapon a wooden club, his heroic recognizable mark, Iovan’s buzdugan ‘mace’ as it appears in the leitmotif formula in the song, “Iovan Iorgovan/Braţ de buzdugan/Iovan Iorgovan/Arm for wooden club”, a defining weapon of gods, similar to that of Thor, ‘Crusher maul’, Indra’s ‘whizzing club’, or Zeus’s thunderbolt, • the dragon tries to discourage the hero to fight it, by threatening the community with the malefic consequences of his beheading—the flies that would kill horses and cattle, a motif that resonates with most of the ancient versions relating what can come from the dragon’s severed head: a bird or a flock of birds as in the Indian version, and from his
164
A. R. Chelariu
teeth, the Maruti, or the Spartoi, fierce fighters, or the Earth men, as in another Greek versions, or Medusa’s gaze turning men to stone; when Perseus cut off her head, Chrysaor and Pegasus are born from her neck (Ovid’s 1955, Metamorphoses, 9.322); instead of an army of young fighters, in the Romanian song the severed monster’s head releases an army of aggressive fierce horseflies, perhaps equally dangerous to the community, that could vanquish all the animals. By telling the dragon that he will teach people how to fight the dangerous horseflies, the hero states his sovereign role of giving the divine instruction to people, and his function as healer and protector of horses and cattle, as the Iranian Θraitauna revered as a healer, or the Indian dragon-slayer Trita Aptya associated with long life (Shaw, 2006, p. 159). As seen from this comparison, the Romanian hero does not explicitly free the fertilizing waters, yet his encounter with the river Cerna, and especially the need to offer magical gifts for crossing into the magical world of the heroine/goddess, could offer another way of connecting the powers of water with the symbol of fertility, the feminine principle. Noteworthy to mention is the relation between the two heroic figures as brother and sister, suggesting possible connections with a cosmogonic myth in which the creators were brother and sister. The Romanian song have retained the main motif of the ancient myth, that of the hero, a god-like presence, who fights the dragon, young and strong, having a mace as his distinct sign of power, whose name, Iovan Iorgovan, could offer a hint of his archaic connections. The origin of the hero’s name is presumably related to the Serbian form Jovan, believed to have its origin in Hebrew Yohānān, through the Greek Iōannēs, hence the Slavic forms Ivan, Iovan. According to this argument, the Romanian song and the name of the hero is a borrowing from the Serbian tradition, where it is found in the song “Iovan and the Leader of the Giants” (Songs of the Serbian People, 1997, p. 21). In this song the hero Iovan kills many giants and a dragon, at his mother’s request; although the hero kills a dragon, this act is not the main subject of the song, mixed among many others, primarily the killing of giants, ending in adversity toward his mother. Contrary to this hypothesis, the Romanian mythological character named Iovan Iorgovan is not to be found in popular songs southwest of the Danube, where the dragon slayer is mostly encountered as Saint George (Vrabie,
19 The Hero Slaying the Dragon in the Romanian Song…
165
1966, p. 144), showing a preference that is not found in the Romanian tradition, even though the Saint is well represented in Christian iconography. The legend of Saint George killing the dragon is first mentioned in a Church manuscript from the seventeenth century, but the Saint George’s biography is absent from Dosoftei’s work The Life and work of Saints (1682). Noteworthy, in the Romanian folklore the form Iovan Iorgovan is exclusively reserved to this mythical character, the dragon slayer. If the Indo-European myth of the dragon slayer was present on the Romanian territory prior to the Christian influence, then the Slavic name Iovan could not be justified because the Slavs arrived in the southern part of the Danube beginning around sixth century CE. If the myth entered north of the Danube after the beginning of Christian era, then it should have been known as Saint George, the same as it is in Serbian and Bulgarian traditions, not Iovan Iorgovan. More so, it seems difficult to accept an association between the Biblical Saint John the Baptist, in Greek Iōannēs, developed into Jovan in Serbian language, with the Romanian dragon slayer Iovan, while, in fact, the sea monster Leviathan is slayed by God, not by St. John, and the Christian iconography portrays Saint George as the snake slayer. It is well accepted that the development of the form ‘Ivan’ from the Greek Iōannēs took place in all Slavic languages; yet, it is less discussed the innovation ‘Iovan,’ especially since a similar form is found in the Armenian history as ‘Yovan’, the name of an army chief from the ninth century (Petrosyan, 2002, p. 138) and his successor Hovan, as well as in the common Armenian name Hovanes. This innovation brings the claim of the Slavic origin of the Romanian ‘Iovan’ into question, leaving room for other possibilities to be explored. One such possibility can be an older form found in the European Southeastt substrata, meaning ‘young’, ‘youth’ DRom n. miel ‘ young lamb’, fem. mială, diminutive mioară (with rotacism), cognate with OIr mil ‘small-animal’, Gaul mīlon, ī ‘(small) animal’; NDutch maal; NE small; Grk melon ‘sheep, goat’, a solution closer to the DRom reflex, different than the solution offered by DEX.RO: Lat agnellus < agnus ‘lamb’: de Vaan (2008): agnellus ‘young lamb’, ambiegnus ‘sacrificial animal’; PIt *agwno- [m.] ‘lamb*, dim. *agwnelo- > DRom miel ‘young lamb’. The form ortoman ‘stronger, wealthier, handsome, standing out’, has its roots in PIE MA *worhxdhus ‘upright, high’ with reflexes in Greek (w)orthós ‘upright, standing’; Skt ūrdhvá- ‘upright, high’; Toch A orto ‘from above’; Lat ortus; Umbrian
20 Youth Rivalry Fights: From Nart Sagas to the Romanian Song…
175
orto ‘rising, standing up’. The place requested by the young shepherds to be buried, în strunga oilor ‘a narrow path for milking sheep’, can be explained by PIE IEW 1036–37 *strenk-, streng- ‘narrow, tight, stiff’, with cognates in Grk στραγγοσ ‘twiddled, twisted’ < στραγγουρια ‘stricture’, Ir strengim ‘pull, drag’, Lith stringt ‘shrink up’, OIce strengr ‘cord, stripe’, strangi ‘tree trunk’, OE streng ‘strand’, OHG stranc ‘lash’, and the Alb shtrungë ‘milking enclosure,’ example of the many common lexeme between Romanian and Albanian. Another development of this PIE root in DRom includes the diminutive n. strungăreaţă ‘midline, teeth gap’. During the summer, the shepherds house their sheep in a sheepfold known in DRom as n. stână ‘summer sheepfold’ of PIE MA etymology *stéh2-no- ‘standing, place’; IEW 1004–1010 *stā- : stǝ- ‘to stand’, IE n- present *stǝ- nă- related to Skt sthãna ‘place, abode’; Grk ástēnos ‘unfortunate’; Lith stónas ‘place’; Alb stan ‘winter shelter for sheep, stall’ from medieval Latin stantia f. L stant- pres. ppl stem of stare ‘to stand’; OCS stanŭ ‘stand’, stanь ‘camp’, Rus stan ‘figure, torso, camp’, Cz, Slk stan ‘tent’, SCr stān ‘habitation, loom’, Blg stan ‘loom, camp’ (Derksen, 2008), a form considered Thracian-Dacian by many Romanian linguists, and borrowed by pastoralists from Serbia, Slovenia, etc. Brotherly jealousy and killing is found in the famous fratricide from the archaic myth of Romulus and Remus, as Livy tells us: “The commoner story is that Remus leaped over the new walls in mockery of his brother, whereupon Romulus in great anger slew him, and in menacing wise added these words withal, ‘So perish whoever else shall leap over my walls!’” (Livy, 1.6.3–7.3). Fratricide is also known in the Irish story of the two brothers, Eber Finn and Eremon, who are fighting over the reign of the kingdom, and the latter kills his brother, Eber Finn. The Indian and the Iranian mythologies tell the story of Trita, and Thraetaon, respectively, the third of the three brothers who is the subject of the jealousy of his older brothers; they abandon him in a deep well that leads to the other world, where he kills the triple or three-cephalic adversary (Dumezil, 1970, p. 14). The killing of heroes/brothers, where two of them murder or abandon the younger one who is different, is destined to overcome the enemy, the dragon, and inherit the kingdom, is a very popular motif in the Euro-Asian folklore, and also widely spread in the Romanian fairy tales. The same pattern can be followed in the Mioriţa song: the third
176
A. R. Chelariu
younger fellow, who is different from the others, has more sheep and dogs, and is wealthier, is destined to be killed. In a Transylvanian Christmas carol, instead of the Mioritic three shepherds, only two cousins, and the older one of them has to kill the younger, not because of wealth, but because the law requires it (Fochi, 1964, p. 555). We are not given any details about what law or why; such a law seems to imply a forgotten rite. The little sheep, Mioriţa, the clairvoyant oracular animal, is another mythical character, as Mircea Eliade argues: “The pastoral societies inherited from the archaic cultures the belief that animal actions and revelations have oracular functions, because the animals know the future” (Eliade, 1972, p. 252). Animal offering was an important part of Romanian traditions. Even in recent times, the Romanian funerary customs include animals as offerings: “The sheep is frequently given as offer in the funeral rites. It is handed as offer (pomană) over the grave after it was splashed with sacred water” (Ghinoiu, 1994, p. 270). The role of animals in the Romanian funerary traditions helps us better understand the oracular and funeral functions Mioriţa plays in our song. The mythic motif of animal sacrifice is found in the Indo-European myth of world creation by the sacrifice of the first human, and the proximity of a sacrificial animal, a bull in India, or a cow in Germany (Lincoln, 1975, pp. 121–145). B. Lincoln brings to attention the cultural differences between the Indo-Iranian and the European branches of the Indo- European family, noting the Europeans more display of an agricultural development, while the Indo-Iranians were more pastoral in their economic orientation; to which the author adds, the individual view of the animals they care for: the agriculturalist raises his cattle for milk, while a pastoralist gets from his animals milk, meat, and also clothing, leather used for bags, bone for tools, dung for fuel, and even urine, used as a disinfectant. The sheep in Romanian song may play such a sacrificial companion role, as foreseer of their destiny. The Romanian funerary customs were, and still are, rich in interdictions and ritual formalities. Great attention was given to each detail of the procession, evidently out of respect and fear of the dead. However, the shepherd’s ‘last’ wishes is to be buried in strunga oilor, the narrow fenced path, through which the sheep go one by one for milking, this particular location is quite out of the ordinary for an interment, as the
20 Youth Rivalry Fights: From Nart Sagas to the Romanian Song…
177
main function of this narrow place is milking the animals. Why then would the shepherd insist to be buried in such a narrow pathway, in spite of the rich and strictly followed funeral traditions, according to which burials should be placed only in designated areas, such as cemeteries? This inconsistency leads us to interpret the so-called shepherd’s testament not as a ‘testament’ at all, but as a set of ritual instructions related to a symbolic death. His instruction to be buried by the sheep milking path may, in fact, describe an actual rite of fertility. In favor of the fertility rite argument are the findings in a version from Maramureş (Fochi, 1964, p. 555) where the shepherd asks to be buried by ‘the barren sheep’, in other words, the place that most needed such a rite of fertility. Moreover, this interpretation is supported by another Romanian custom, according to which shepherds will erect a sacred tree, an oak or an evergreen, in the middle of the barren sheep’s place (Vulcǎnescu, 1972, p. 65). This custom brings to mind all the connotations related to the sacred tree as axis mundi, connecting the three realms of the world, the location where the young shepherd’s wish to be buried, by the barren sheep, implying, a fertility ritual. It could also be postulated that the motif of burial in a specific area under specific requirements suggest relations to the well- known myth of the underground disappearance of gods or goddesses of fertility, such as Zamolxis, Kore/Persephone, or Balder. Perhaps, the mythic motifs of the rivalry fights, or killing of a brother, describes two types of rituals, first, a rite of fertility, and second, a rite of transformation in the social position of the young man. Comparable to the German myth of Balder, the Romanian young shepherd must be lamented by the entire environment, summoned by the sound of his flute, in fact, a cosmic lament. A similar universal lament is found in the Baltic tradition, in the belief that the human spirit continues to stay around the grave, particularly in the tree flowers, in the grass, as in the following Lithuanian folk song, that reminds us of verses from Mioriţa: “The green oak will be his daddy/The white sands will be his mommy/ The green maple trees will be his brothers/The white linden trees will be his sisters” (Gimbutas, 1958, p. 98). These funerary songs describe the same symbiosis with the environment from the myth of man’s creation discussed above, a symbiosis with the surroundings embedded in human spirit from creation out of nature to interment into nature.
178
A. R. Chelariu
The Romanian folklore data are rich in funerary songs and customs for an unmarried girl or boy, based on the belief that the life of a young person is not fulfilled without knowledge of the mystery of marriage, a belief largely spread in Europe. Very similar to the Romanian wedding ceremonies for a dead youth, are those described by Marija Gimbutas in the Baltic folklore: “...at the funeral of a betrothed girl or boy, the burial ceremony is more like a wedding, wedding songs are sung, dances are danced, and both the living and the deceased partner are dressed in wedding costume. The wedding of dead is not simply connected with the belief in the continuity of the earthly life after death, but also with the belief that people who die unmarried, and all those who die an unnatural death, are a danger to the living since they have not lived through the whole span of life” (Gimbutas, 1963, p. 189). The similarities between the Baltic and the Romanian oral tradition are remarkable, suggesting that both traditions are part of a common archaic cultural heritage. The last part of the song, the cosmic wedding of the young man with the world bride, brings in, as previously stated, the Indo-European trans- functional goddess in her role of a goddess bride and mother of heroes. The destiny of the Romanian shepherd indicates his special position among his peers, his passing through a ritual death and rebirth into his new position among the warriors’ group. His marriage to the ‘World Bride’ shares similarities with the hero in the Irish tradition, who can be validated as the ruler of the group only if he marries the fairy queen, symbol of sovereignty (Waddell, 2018, p. 37). Thus, the apotheosis wedding, beautifully described in the poem, may not necessarily be interpreted as a marriage in death, but a symbolic marriage in ‘kingship’, a change in the shepherd’s social status. …I have gone to marry/A princess—my bride/Is the whole world’s pride/ At my wedding tell/How a bright star fell/Sun and moon came down/ To hold my bridal crown/Firs and maple trees/Were my guests; my priests/ Were the mountains high/Fiddlers, birds that fly/All birds of the sky/ Torchlights, stars on high…
The cosmic wedding with a goddess/fairy queen relates yet with another well-known ritual, the Eleusis’s mysteries and the union between
20 Youth Rivalry Fights: From Nart Sagas to the Romanian Song…
179
the hierophant and the Great Goddess, the world’s queen bride; better yet, the role of the beautiful Daena in the Iranian mythology, where the soul of the dead arrives at the gods’ tribunal for judgment, and if he passes, the goddess awaits for him to be taken into the divine presence of light (Puhvel, 1987, p. 115). Perhaps it should be mentioned that in Rome, Pales, the goddess of herds and flocks, had her own celebration called Parilia or Palilia, whose date coincided with that of the Foundation of Rome, maybe the date when Romulus traced his furrow and killed his brother (Dumezil, 1996, p. 383). The Great Goddess, majestic presence in the archaic mythologies is associated with life and death, with heroes and their achievements, be that the instatement as a new king, and/or a wedding, as the beginning of a new life, or death and resurrection. Her presence in the Romanian song, if interpreted as a psychopomp entity, proves her role in resurrection and regeneration, attesting to the considerable archaism of the story. The social structures and mythical motifs revealed in the song Mioriţa help integrate the Romanian heritage within the European, implicitly Indo-European cultures. With the passing of time, the meanings of the early rituals dissipated in favor of the emotional inclinations toward the tragic destiny of a young man. Fortunately, for the popular interpreters, as well as for the large audience, the song carries with it the obligation to be recited exactly in its entirety and with great respect: “Mioriţa is an ‘old song’ for which the peasant has a cult; it is forbidden to change its text of its melody, because an elderly will cut you off and shame you” (Maior, 1922, p. 495). One can only admire the popular interpreter’s ability to pair artistically beautiful forms and archaic mythical motifs well into the twentieth century.
References Alecsandri, V. (1852–1853). Poesii populare—Balade adunte și îndreptate. Tipografia lucrătorilor asociați. Colarusso, J. (2002). Nart Sagas from the Caucasus: Myths and Legends from the Circassians, Abazas, Abkhaz, and Ubykhs. Princeton: Princeton University Press.
180
A. R. Chelariu
Colarusso, J. (2006). The Functions Revisited, a Nart God of War and Three Nart Heroes. Journal of Indo-European Studies, 34, 1–2: 27–54. Institute for Study of Man. de la Saussaye, C. (1902). Religion of the Teutons. Ginn & Company. de Vaan, Michiel (2008). Etymological Dictionary of Latin and the other Italic languages. Liedern, Boston. Brill. Derksen, R. (2008). Etymological Dictionary of the Slavic Inherited Lexicon (A. Lubotsky, Ed.). Leiden Indo-European Etymological Dictionary Series, vol. 4. Leiden: Brill. Dumezil, G. (1970). The Destiny of the Warrior. Chicago: University of Chicago Press. Dumezil, G. (1996). Archaic Roman Religion: With an Appendix on the Religion of the Etruscan (P. Krapp, Trans.). Baltimore: John Hopkins University Press. Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. Chicago: University of Chicago Press. Eretescu, C. (2002). Miorița—Pe marginea motivului Testamentul ciobanului. Limbă și literatură, An. XLVII, vol. III–IV, 66–73. Fiske, J. (1900). Myth and Myth-Makers, Old Tales and Superstitions Interpreted by Comparative Mythology. Boston, New York. Houghton, Mifflin and Co. Fochi, A. (1964). Mioriţa; tipologie, circulaţie, genezǎ, texte. Bucharest: Editura Academiei Române. Ghinoiu, I. (1994). Vîrstele timpului, Chişinău, Știința. Ghinoiu, I. (2001). Panteonul Românesc, Dicționar. Bucharest: Editura Enciclopedică. Gimbutas, M. (1958). Ancient Symbolism in Lithuanian Folk Art. American Folk Society. Gimbutas, M. (1963). The Balts. New York, Washington, Frederick A. Praeger Publishers. Lincoln, B. (1975, November). The Indo-European Myth of Creation. History of Religions, 15(2), 121–145. Chicago. University of Chicago Press. Titus Livius (Livy). The History of Rome, Rev. (C. Roberts, Ed.). www.perseus.tufts Maior, G. (1922). Ceva despre Mioriţa. Lamura, 3(6–7), 495–496. Puhvel, J. (1987). Comparative Mythology. Baltimore: John Hopkins University Press. Vulcǎnescu, R. (1972). Coloana Ceriului. Bucharest: Editrua Academiei Române. Waddell, J. (2018). Myth and Materiality. Oxford: Oxbow Books.
21 Metamorphoses in Myth and Fairy Tales
The dispute on the relationship between rite and myth was the subject of many studies from nineteenth century onward. Sir James Frazer and his followers, the Cambridge Ritualists School, reached the conclusion that rituals are the source of myths. While many researchers agree that “myth … is the counterpart of ritual; myth implies ritual, ritual implies myth, they are one and the same” (E. R. Leach, 1954, pp. 13–14), others may not, even though they are still considering them essentially connected. Against expediently associating myth and ritual, G. Kirk cautions us that their “relations are complex and varied…”, and only if mythical and folkloric material cooperate in the story, the rite-myth is validated (Kirk, 1970, pp. 16–17). In the same direction, V. I. Propp (1973, pp. 12–28) extended the field of research to include fairy tales as the story accompanying rituals. In agreement with the importance of folkloric data in myth-rite corroboration, we should proceed to analyze the motif of shape changings, particularly in relation to rites of youth initiation, that could better illustrate their relationship. Metamorphosis was viewed by some antiquity thinkers as the outcome of a misunderstanding of daily events, the result of an exaggeration; others, together with Greek philosophers, considered shape changings as expressing the belief in the substance present in any form. As mentioned © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_21
181
182
A. R. Chelariu
before, the Greek philosopher Posidonius (135–51 bc) regarded the human symbiosis with the environment an organic interrelation with all appearances in the world, based on the “cosmic sympathy” (συμπάθεια, sumpatheia), theories embraced by the Roman poet, Ovid. Such principles preceded the contemporary concern on the environment, and the role humans have in relation to nature. The ‘cosmic sympathy’ sensed by the Greek thinkers, the human perception of the medium and spiritual need to be integrated within the surroundings, can be considered the base for the development of the shape changings in rituals of initiation, and also in transformation flight of gods, goddesses, people, and animals. Entangled in a complex and sacred net covering the relation to the surroundings and the community, the prehistoric man preserved the sacredness of the traditions and religious beliefs through ritual acts. This goal was achieved by two forms of expression, enacting scenarios of established practices, and verbal communication. The ritual enactments operated with specific sets of controlled visual forms and verbal expressions, created by the group to convey the ethics and the experiences that were upheld into complex social and religious manifestations. The verbal expressions used in storytelling, tropes, symbols, metaphors, allegories, metonyms, and the like, were all elements of communication chosen to amplify the abstract and religious concepts embedded within the cultural heritage. The environmental ‘sympathy’ includes not only the ground, or the plants, but also the animal kingdom. From times immemorial, humans had a special relation with the hunted animals, a relation that had a deep impact over the social and spiritual development of the community. Animals gained totemic values, they were seen to be holding the souls of ancestors, and were feared and revered. The need to appease the slain totemic animal in this relation was done through ritual enactments, in which the animal’s skin, horns, hooves were used to impersonate it, in the belief that by wearing part of the animal the ritual participant would obtain its powers and qualities, its life spirit. Thus, through metaphoric conceptualization, the virtual shape changings entered the world of myth. The intrinsic animal–human relations expressed in rites, a fundamental part of the community life, in historical times, became supported by the artistic forms, sculpture, murals, etc., and in literary forms, by the abundance of shape changings from myths and folktales. The large number of stories in which metamorphoses play an essential part, are separated in
21 Metamorphoses in Myth and Fairy Tales
183
those in which the hero or heroine change shapes at will usually to escape from an adversarial follower, or to avoid some danger, and those resulting from a punishment by a divine force. Transformations as the result of a divine punishment are quite common in myth, for example, the famous myth in which Artemis turns Actaeon into a stag, or Callisto into a bear; and one could not omit the metamorphosis of Odysseus’ sailors into pigs on Circe’s island. This type of metamorphosis will be discussed in the second part of this chapter. The folktales in which characters change shapes at will in an effort to escape from an eminent danger are called in the Aarne-Thompson catalog of folk data transformations flight (Aarne & Thompson, 1987, ATh 325), a formula used here as well. In this narrative, a female or a male character, or both as a couple, are able to change their shape, and go through a series of transformations at will to avoid a follower, who, consequently, changes his/ her shape into a more powerful adversary. In the end, the follower is overpowered. The flight of transformations generally leads to a change in the character’s position within the community, change that could indicate reminiscences of a journey of initiation. The sequence in which the metamorphoses in myths and folktales take place could offer a better understanding of the meaning hidden behind this popular mythical motif. Amid classic myths, the most common transformation flight takes place between a reluctant goddess, or nymph, and a god-seducer. One could only think of the numerous stories in which Zeus, taking different shapes, such as a cow, a swan, an eagle, or golden rain, pursues and overpowers goddesses and nymphs. A more complete example of this motif of the transformations flight is the famous myth of Nemesis followed by Zeus, perhaps the oldest one. From very early times, the concept of nemesis ‘vengeance’ overwhelmed her actual story, turning the goddess into a symbol of divine indignation. Her sanctuary was said to be under a tree in a meadow (Greek nemos), in a forest where no mortal was allowed. The sacred place was protected day and night by a priest, who had to be killed before another one could take his place (Frazer, 1971, p. 14). In most of the sources of antiquity, the transformations sequence is as follows: Nemesis did not want to receive Zeus’s love, and ran away from him, first by throwing herself into waters and turning into a fish; Zeus followed her as an otter; then she got out on the shore, and
184
A. R. Chelariu
turned into a wild beast, but he turned into an even more ferocious animal; finally, she rose in the air as a wild goose, but Zeus overpowered her as a swan (Stasinus of Cyprus or Hegesias of Aegina; Apollodorus, 1921, The Library 3.127). To complicate the matter somewhat, there is another version of this myth according to which, because Nemesis ignored Zeus’s advances, he staged his own transformations flight: he turns himself into a swan, and then asks Aphrodite to follow him as an eagle. Pretending to be terrified, swan-Zeus seeks salvation under Nemesis’s power. Impressed by the beauty of the animal the goddess protects him, but, while asleep, she receives his love. From this union the goddess gives birth to an egg that Hermes will drop onto Leda’s lap, from which Helen comes into being (Hyginus, 1960, Astronomica 2.8). This version is confirmed by the existence of a fourth century BCE statue representing a goddess that seems to protect a swan, while her eyes are watching the sky (Graves, 1960, p. 15). There are quite a few other stories about goddesses fleeing Zeus or other gods’ advances, as in a story told in Arcadia about Demeter who was pursued by Poseidon; to escape from her seducer, she turned into a mare, but was overpowered by the god in the shape of a horse (Pausanias, 1918, 8. 25. 5). There is also the story of Metis who, like Nemesis, had the art of changing herself into many different shapes, yet Zeus overpowered her, and thus the birth of Pallas Athena (Apollodorus, 1921, The Library 1. 20). Zeus also sought to seduce Leto’s sister, Asteria, who turned herself into a quail, fell into the sea, and remained hidden beneath the waves, thus became the island called Ortygia, the quail island, or Delos ‘visible’, as it became visible by rising from the depths (Apollodorus, 1921, The Library 1.4). Thetis, the Nereid mother of Achilles, was courted by both, Zeus and Poseidon, but because they were told she will have a son more powerful than his father, they gave her up, and made her marry the mortal Peleus. Not agreeing with this arrangement Thetis put up a transformations flight in order to avoid the marriage to a mortal; she took various shapes, as fire and water, lion, serpent, and was finally caught by Peleus, in the shape of a cuttlefish (Apollodorus, 1921, The Library 3. 168). The Roman world knows the, presumably, Etruscan god, Vertumnus, who was in love with Pomona, the goddess of fruits of trees; to convince her to marry him, he takes various shapes; in the end, she consents to his love (Ovid Metamorphoses, 1955, 14, pp. 623–771). This story is somewhat
21 Metamorphoses in Myth and Fairy Tales
185
tangential to our transformations flight motif, as the male is the only one changing his shapes. In Hinduism, the story of Prajapati may shed some light on the meaning of the transformations flight: In the beginning, the first being, Prajapati, was all alone and in need of a companion; he decides to split his body in two, giving rise to husband and wife; from their union the human beings were born. When his wife realized that he begot her from his own body, and their union was morally wrong, she decided to hide from him; first, she took the shape of a cow, but he turned into a bull, and from their union, the cattle were born; next, she took the form of a mare, and he followed her as a stallion; then, she turned into a female donkey and he in the shape of a male donkey, thus creating all the one- hoofed animals; the shape changings continued with goats, sheep, and so on, creating all the female-male pairs in existence (Max Müller trans. Brhadaranyaka Upanișad 1:4:2-6). This Hindu myth suggests that the motif of the reluctant goddess putting up a transformations flight from the god seducer has retained an archaic creation myth, perhaps of the Indo-European origin, while the European traditions preserved only the motif of transformations flight, without that of the creation of all species. The mythical motif of shape changings flight between a reluctant young girl or fairy and a seducer is documented in many European folk songs and stories, leading researchers such as Forbes Irving (1990, p. 172) to argue that the roots of these stories take us back to the Greek mythology. Many of the European folk songs relate a playful dialog between a girl and her seducer: the girl says she will run away from her pretender as a bird, a fish, a rose, a flower, or turning into a chapel; to which the lad responds that he will catch her as a hunter, a fisherman, a gardener, or a monk; these are essentially the most common transformations used in these folksongs. An excellent example of this kind of magical transformations is found in the collection The English and Scottish Popular Ballads, # 44, the Two Magicians, (Child & Kittredge, 1882) of Francis James Child, in which a blacksmith threatens to take a lady, who vows to keep herself a maiden; she tells him she will run away as a hare, but he will follow as a greyhound, next she will turn into a duck, and the follower into a water dog or a drake. In the end, she does not escape his chase. Variations of this song are known all over Europe, in Italy, Greece, Poland,
186
A. R. Chelariu
Romania, etc. In the Romanian tradition we have the folksong The Ugly Man, ‘Bărbatul urât’ and Tudoriţa and the Old Man, ‘Tudorița și moșneagul’ (Hasdeu, 1879, 1984); in these songs the girl runs away from an unwanted pretender in this sequence of transformations: first as a fish followed by him as fisherman; she turns into an elm branch, and he turns into a young lad; then she turns into a flower, and he into a young mower; the song clearly describes the girl’s dislike of her follower. A particular version of this song is The Cuckoo and the Turtledove ‘Cucul și turturica’ (Hasdeu, 1879, 1984) in which the two ‘lovers’ are birds, not humans, and the shape changes are mixed, from a ‘girl’ turtledove changing to loaf of bread, and her seducer as a fire shovel to take the bread out of the stove; the ‘girl’ turns into a reed, and he into a shepherd, then she turns into an icon, and he into a deacon, displayng a loss of the archaic sequence of transformations. Besides songs and poems, fairy tales preserved well the motif of transformations flight: the daughter of a monster/devil/witch runs away with a servant of her parent, who follows them, and the pair goes through a series of metamorphoses to escape the follower; the motif of transformations from these type of fairy tales is well preserved in the European and Indian folklores. A similar set of transformations flight is found in folk stories in which the pursued is a male character, a boy-apprentice or a servant, as, for example, the famous Song of Taliesin: Gwion, the servant of an Old witch, is accidentally splashed with a few drops of her boiling potion, which consequently, makes him gain some of her powers; as a result, he has to run away, and the transformations flight begins: first, he runs away as a hare followed by the witch as a greyhound; he turns into an otter, followed by her as a hound, then he is a bird and she is a hawk, then he turns into a grain of wheat in a heap of chaff, and she turns into a hen and swallows him. Inside her womb he begins to dream; after nine months she will give birth to a child, who will be the famous poet Taliesin (Matthews, 2001, p. 25). In another type of folktale—the promised child, a father unknowingly promises his child to a supernatural being; while in the monster’s house, the lad falls in love with the daughter of the supernatural being and they run away together; to escape their follower they turn into various objects,
21 Metamorphoses in Myth and Fairy Tales
187
first, the girl turns into a monastery or church and the young man turns into a monk, then he turns into a little boy and she into a corn field, then she turns into a lake and he becomes a duck. The transformations in this type of folktale are meant only to trick the follower, that does not change his/her shape. The transformations flight motif is present in many Romanian folk stories, from which the story entitled “Oh” or “The Boy at the Devil’s school” (Bârlea 1966, p. 493) deserves special attention, as the followed character in the transformation flight is a boy, not a girl. The story begins with a poor man who decides to take his boy to a school to learn a craft; on the way they stop by a well, where the tired father groans, ‘Oh’. That brings out from the well a little red bearded man, whose name is Oh, or, in other versions, he is the Devil; the little man offers to take the child into his school for one or two years, to teach him his craft. During the training, the wizard teaches the boy how to turn into various animals. At the end of the schooling time, the father comes to get back his son. To do that, first, he must recognize him from a group of seven or twelve kids absolutely identical to his own. However, before the trial the son reveals to his father signs by which he can be recognized: He will bring his hand to his ear, to his lips, or to his forehead. Following his son’s instructions, the father recognizes his son, and they go home. The boy tells his father how they could make some money, by selling him as an ox at the market for a good price, after which he will come back home as himself. After spending all the money, the boy tells his father to sell him once again at the market, this time as a horse. However, he warns the father to make sure not to give away the horse with the rein on, so he could come back, and not to sell the boy-horse to the wizard. Yet, the wizard comes to the market, and offers the father a very large sum of money. Forgetting his son’s warnings, the father sells the boy-horse to the wizard with the rein on. Soon after, as the wizard and his apprentice-horse cross a body of water, or the horse’s bridle is released to let the animal drink from a river, the horse turns into a fish and gets away. At this point the transformations flight begins: The wizard follows him as a bigger fish; the boy turns into a dove and flies away, followed by the wizard as a hawk; next, the boy falls as a ring into a princess’s lap, and the wizard goes to the king and claims the ring. Meanwhile, the ring-boy tells the princess not to give the
188
A. R. Chelariu
ring to the wizard, and instead, to throw it to the floor; the ring turns into grains, the wizard turns into a rooster, and starts eating the grains, but one seed that hides under the leg of the princess’s chair, turns into a fox and eats the rooster-wizard. Then the fox turns into the young man, and marries the princess. The antiquity of the motif of the father selling his child in various animal forms from the Romanian version is confirmed by the story of Erysichthon (Ovid Metamorphoses, 1955, 8, pp. 738–878) who sells his daughter in many shapes, as a mare, bird, cow, or deer. The particular sequence of shape changings taking place in the transformations flight, always in the same order, from ground into water, through the sky, and ending up on earth again, suggests a metaphysical journey through the entire spectrum of life. This symbolic cosmic journey is featured in a very archaic document, dating from 1450 to 1400 bce, a gold ring, known as the “Ring of Minos” found at Knossos, Crete. On it the engraving shows a goddess in a sequence of images, first, descending from heavens, then seated in a shrine on earth, and lastly floating on water in a boat that transports a shrine: “the passage of the goddess through the three elements of nature—air, land and sea—symbolically unites the visible world…” (Diomopoulou and Rethemiotakis, 2004: 15–19, in Shaw 2012). As revealed in most of these stories, the process of metamorphoses into various animals and/or objects brings the best result for the young people: a new situation, a wife, a kingdom. Transformations, an essential part of many myths and fairy tales, could be connected to the belief in the immortal soul, as Empedocles thought “Once I was born a youth, a maid,/A shrub, a bird, and a briny fish of the sea” (Fragment B117), and later articulated by Ovid in his famous “Metamorphosis”, Book XV: “..….All things are always changing, But nothing dies. The spirit comes and goes, Is housed wherever it wills, shifts residence From beast to men, from men to beasts, but always It keeps on living. …”
21 Metamorphoses in Myth and Fairy Tales
189
The belief in eternal life, expressed by metempsychosis, was most likely part of the cultural heritage imbedded in human thought since prehistoric times, as mentioned above, when humans, animals, and plants were intertwined in a sacred net, animated by the same divine soul. The symbolic transformations onto beings from land, water, and air, the entire spectrum of life, was an intrinsic part of initiation ritual enactments, in synchronicity with the rituals involving pairs of gods and goddesses, creators of all existence on earth. The idea that metamorphosis was a way for the human imagination to understand and conquer the fear of death, based on the belief that life would continue eternally in a different form of existence, as animal, stone, bird, etc. (Forbes, 1990, p. 11), seems limited to the basic human instincts, whereas, the series of transformations from myths and folktales point to a journey into understanding the cosmic unity and structure, always ending with a change in the character’s social status, that of a groom, or a bride, and the entering into a new way of life. The hero’s journey of adventure and discovery, and the obstacles to overcome as told in these stories, are rather relating journeys of initiation into the miracles of life, into the mystery of the divine, in which metamorphosis in various animals, or objects, played an important role. In the Romanian story discussed above, the hero first goes underground, into the Other World, where he acquires ‘the animal powers’, then he is confronted by an outside force, that of the wizard, who threatens to keep him in an animal shape, and deny him the normal course of life; only after he goes through the entire set of transformations through water, air, earth, the marriage can take place.
References Aarne, A., & Thompson, S. (1987, 4th Printing). The Types of the Folk-tale (ATh). Helsinki: Soumalinen Tiedeakatemia, Academia Scientiarum Fiennica. Apollodorus. The Library; Translator J. G. Frazer. (1921). Harvard University Press. http://www.perseus.tufts.edu/ Bârlea, O. (1966). Antologie de proză popular epică. Editura pentru literatură: Bucharest.
190
A. R. Chelariu
Child, F. J., & Kittredge, G. L. (1882). The English and Scottish Popular Ballads. Boston: Houghton Mifflin. Empedocles (c. 492—432 B.C.E.) Fragments B117 https://iep.utm. edu/empedocl/ The English and Scottish Popular Ballads. (1882) Child, F. J., & Kittredge, G. L. Editors. Houghton Mifflin. Forbes Irving, P. M. C. (1990). Metamorphosis in Greek Myths. Oxford: Clarendon Press. Frazer, J. G., & Sir. (1971). The Golden Bough, a Study in Magic and Religion. Macmillan. Graves, R. (1960). Greek Myths. New York: Penguin. Hașdeu, B. P. (1879, 1984). Cuvente den bătrîni:Cărțile poporane ale românilor în seculul XVI, vol. II. Bucharest: Editura didactică și pedagogică. Kirk, G. S. (1970). Myth, Its Meaning and Functions in ancient and Other Cultures. Cambridge University Press; University of California Press. Leach Edmund Ronald. (1954). Political Systems of Highland Burma: A Study of Kachin Social Structure. Cambridge MA. Harvard University Press. Matthews, J. (2001). The Song of Taliesin: Tales from King Arthur’s Bard. ILL. Quest Books: Wheaton. Müller, F. Max The Upanishads Volume Pt. II (SBE15) (Sacred Books of the East). http://www.sacred-texts.com/hin/sbe15/index.htm The Myths of Hyginus, Translated and edited by Mary Grant. (1960) University of Kansas Publications in Humanistic Studies, No. 34. University of Kansas Press. http://www.theoi.com Ovid. (1955). Metamorphoses (R. Humphries, Trans.). Bloomington: Indiana University Press. Pausanias, Description of Greece – With an English Translation by W.H.S. Jones, Litt, D., and H. A. Ormerod, M. A., in 4 Volumes. (1918). Harvard University Press; William Heinemann Ltd. http://www.perseus.tufts.edu/ hopper/text?doc=Perseus%3Atext%3A1999.01.0160%3Abook%3D1%3Ac hapter%3D1%3Asection%3D1 Propp, V. I. (1973). Rădăcinile istorice ale basmului fantastic, Bucharest. Univers: Editura. Shaw, John. (2012). On Indo-European Cosmic Structure: Models, Comparisons, Contexts. Cosmos 28: 57–76.
22 Metamorphoses in Youth Initiation Rites
The process of initiating young people started at the end of childhood and the beginning of puberty, the most important moment in the life of the adolescents. Organized during major feasts, the initiation practices were intended to induct the novice into the sacred myths and heroic stories of the social group. Generally, at this moment of the transitional stage the novice was considered asexual, in a sphere of ambiguity. Socially, novices did not belong to any category, and metaphysically they were crossing a period similar to a cosmic chaos (van Gennep, 1960, p. 65) from which they needed to be reborn into a new beginning. This ambiguous stage positioned the novices outside the social and the sacral order, on a limb associated with the primordial chaos, a state potentially dangerous and threatening for the entire group’s well-being, which determined the need for their isolation. To restore cosmic and social order in the community it was imperative for the novices to successfully get through, and complete the initiation rite. Part of this process was the ritual reenactment of divine creation, by which the novices received the mythical concepts needed to understand the existential human condition, and the expectations the group had from them, and to help the youth to step into a new ontological experience. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_22
191
192
A. R. Chelariu
Through enacted rites of passage, the group introduced their children into the world of adulthood and its sacred past, helping the neophyte on the pass from childhood to the adult community, assuming the new role as a wife, a husband, a warrior, or a king. At the center of these enactments were symbolic transformations, associated with changes in the neophyte appearance, for example, the hair cutting, dressing in special garments, covering in animal skins, and so on. Greek literature records various customs involving young men changing their apparel during the festival of Apatouria at Athens: after giving offerings of a sacrificial animal, adolescents, that had their hair grown especially for this ritual, went through the rite of koureia, the hair “cutting ceremony”, koureion, followed by their induction into phratries ‘brotherhoods’, being recognized as citizens of the polis (Gernet, 1981, p. 23). Prior to this ceremony, it seems that boys not only had long hair, but they also wore girls’ outfits, as Achilles, who, for the first part of his life, was hidden among the daughters of Lycomedes, dressed as a girl. The same was the story of Theseus who was ridiculed as a girl until he hurled the sacrificial animal up in the air (Burkert, 1985, p. 261). The young men, or kouroi, offered their hair locks to the ancient heroes, not to gods, whereas the girls performed their offerings to their protective goddess, once again revealing the special place goddesses and their protected female followers had in the archaic society. The Greek girls’ change of apparel is described at the Greek Brauronian festival, attested in the late fifth century BCE., when girls partook in the ritual named arkteia “bear (artktos) ritual”, during which they raced semi- nude in honor of Artemis, the goddess of purity and childbirth. Although the ‘bear ritual’ may not qualify as an initiation rite, the ceremonies of young girls, dressed in saffron garments, ‘playing bear’ just before their marriage, for appeasing the goddess, or ‘serving as a bear’ in her temple, (Initiation in Ancient Greek Rituals: Faraone, 2003, p. 43) show a form of inducing the young girls into the sacred world of a goddess. The ‘bear’ ritual is reminiscent of the myth of the nymph Callisto, the virginal companion of Artemis, turned by jealous Hera into a bear because she was seduced by Zeus. There was also the custom for girls soon to get married, to go by the river Scamander, bathe, and chant: “Scamander, take my virginity”, perhaps a reminder of the myth in which Hera Parthenos, after
22 Metamorphoses in Youth Initiation Rites
193
losing her virginity, periodically recovered it after bathing in a river (Gernet, 1981, p. 24). The institution of initiation male war bands from the Indo-European myths, was formed of adolescent young men, expelled from their community, who migrated to another cultural frontier, where they engaged in raids. These Indo-European war bands, called the luperci or suodales in Latin, kouros or ephebes in Greek, fian in Celtic, männerbünde or jungmannschaft in Germanic, and vrātyas or Maruts in Indic, were described as dressed in wolf or dog skins, or were named after dogs, or they became like wolves and behaved that way. Recent archaeological research revealed “canid sacrifices ritual, discovered at Krasnosamarskoe, Russia, during the Samara Valley Project in the Volga steppes. These shared tropes are evidence of a shared Proto-Indo-European institution of male initiation connected with dog and wolf symbols” (Anthony, 2021). To the dogs and wolves presence in the male initiation rites discussed here, we have to add the Dacian tribes whose name, according to Strabo (1903, p. 304: 7. 3. 12) was daoi, considered a Phrigian word that meant ‘wolf ’ (see Chap. 1). Unfortunately, we have no documents relating the daoi name of Dacian warriors to youth initiations, or young bands dressed as wolves, but we do have documents showing their flag, which was made in the form of the head of a wolf, with a dragon body (Latin draco, −onis), which, during the rides, made a very loud noise, meant to scare the enemy. It is safe to assume that the Dacians called ‘wolves’, as the luperci (Lat lupus ‘wolf ’) were brotherhoods behaving as real wolves (Eliade, 1972, p. 5), during the military initiation, wars, or during raids. In Romanian tradition, ‘wolf ’ is believed to be a fantastic creature of cosmic proportions, able to eat the sun and the moon, blamed for eclipses (Olteanu, 1998, p. 445). One of the most feared animals, the wolf is frequently invoked in magic spells, and is subject of numerous feasts, ritual practices, and interdictions, recorded in old calendars as the Wolf Days. Romanians believed that the feared wolves helped women gain fertility, reminding of the Roman luperci, who, during Lupercalia celebrated on February 15, hit people with pieces of animal fur, while running wild “about Rome in a mitigated form of flagellant ritual, striking especially women as a magical aid to procreation” (Puhvel, 1987, p. 160).
194
A. R. Chelariu
The ritual change in the young persons’ appearance, either of hairdo or clothes, as described in youth ritual ceremonies of the ancient Greek world, is the subject of many analysis and interpretations. In a recent article, Reyes Bertolin Cebrián discusses the ‘shape shifting warriors’ in Greek traditions, mainly the youth contingents covered in skins, found in Iliad as the ‘long-haired’ Acheans, Myrmidones, Lykians, the Wolf- People. The author lists the most important characteristic of these shape- shifters: “use of animal skins, which allowed the warriors to assume the nature of the animal”. The author lists among the most common shape- shifters the wolf warriors as luperci, the bear-skin warriors, the Norse berserkers, also called úlfheðnar ‘wolf-skinned’, frenzy warriors covered in the skins of wolves, úlfheðinn ‘wolf-coat’, youth wearing “long hair or strange hairdos, which was supposed to give them a horrifying aspect”, “they also impersonated ghosts or dead ancestors”; the physical change “made them fight in a state of fury that increased their performance” (Cebrián, 2010, p. 343). Practices of this kind are relevant to our understanding of metamorphoses, showing that, by wearing animal skins, the young men ‘become’ that animal, it changed their personality, it brought them closer to the totemic animal spirit, and the realm of their dead ancestors. These customs and stories may be reminiscences of very old rites of initiation, rites that involved the enactment of neophytes being ingurgitated by a totemic primordial animal, and regurgitated initiated into the sacra: “In other words, the mythical animal returns to life together with the initiate” (Eliade, 1972, p. 16). Such ritual enactment of symbolical ingurgitation and regurgitation by a primordial totemic animal, or by imaginary monsters, dragons and so on, relates directly to the mythical motif of metamorphoses, so popular in myths and folktales. A very suggestive description of the role of the sacrificial animal in initiation enactment was given by Prudentius “Peristephanon 10” on the “taurobolium”, and the cult of Kybele (Cybele), the great Phrygian Mother of gods, also known as Magna Mater, goddess of motherhood and fertility, whose jealousy drove Attis to castrate himself. During her spring celebration rites, her devotees sat in a pit, and the blood of a sacrificed bull was poured over them, in the belief that they would come out of the pit, as if coming from the mother’s womb, covered with blood, reborn (McLynn, 1996).
22 Metamorphoses in Youth Initiation Rites
195
Descriptions of rites of passage practiced in recent times in Australia, retold by Arnold van Gennep, can clarify further this process: As the novice is considered dead “he is resurrected and taught how to live, but differently than in childhood”. The rite would allow the neophyte to get in touch with the other world, the world of the dead and the mysterious, to acquire “the ‘spirit’ who is the protector of the entire clan…” (van Gennep, 1960, pp. 75–77). Through his contact with the sacra, be it a ‘bull-roarer’, a bull’s blood or other means, the novice enters the world of the divine, where he has to capture the secret qualities and powers of the animal or the dead ancestors. The emotionally charged actions from such rites merged through metaphorical imagery in stories retelling the initiation process. In the effort to penetrate the mysteries of the divine the story enters the metaphysical realm. The neophyte initiation enactment was associated with initiation into the spiritual world of myths, in which gods, as metaphorical manifestations, were construed as the source of everything, able to take any form, be everywhere, surrounding the human beings in any shape. At the core of myths and sacred rituals were also the beliefs that the spirits of the ancestors were all around, and can influence men, animals, and crops. It was an important task to impress the young minds through the enacted stories of transformations, of ingurgitation-death and regurgitation- rebirth anew onto the sacred traditions of the community, in the belief that human spirit is an eternal part of the larger entity of nature. As Frazer states, the scenario of killing the lad and bringing him back to life, was a collective act in the initiation practices: “Such rites became intelligible if we suppose that their substance consists in extracting the youth’s soul in order to transfer it to its totem” (Frazer, 1971, p. 802). Traditional stories captured from these ritual enactments the dramatic encounter between the sacred world that would not let the neophyte leave without resistance, as in story of the boy at the wizard’s school, and the hero’s avatar to deceive the opposing forces from the Other World through a flight of metaphoric transformations, in a symbolic contest of skills with the enchanted follower, and to return home enriched with magic powers. Other shape changes from the archaic rites involving symbolic copulation between humans and gods in shape of animals emerged in many
196
A. R. Chelariu
myths and folktales. For example, the ceremony performed in Athens during the Choes early spring festival, when the king’s wife underwent a ritual marriage to a representation of Dionysos, the god of growth and fertility. The success of the rite, meaning, the success of the king’s role in assuring prosperity for the community, depended on the queen’s mating with the god’s substitute, perhaps, the king in the god’s apparel, or, in a symbolic scenario, a statue, an animal, a goat or a bull, representing the god (West, 2007, p. 417). This rite has mythic correspondences in the story of Minotaur, the offspring of Minos’s wife, Pasiphae of Crete, who fell in love with a bull, and coupled with it while hidden in a wooden cow wrapped in bovine-skin. This myth parallels the Indian rite, Ašvamedha, during which a white stallion was sacrificed, and the queen symbolically lay down under covers with the animal, while people were chanting and jeering (Puhvel, 1987, p. 271). These rites, involving the queen-goddess ritual copulation with an animal symbolizing a divinity, performed to help the king’s induction, and bestow his duty to bring growth and fertility to the community, validate the role of heroes’ metamorphoses leading to the change in their social status.
References Anthony, D. W. (2021). Migration, ancient DNA, and Bronze Age pastoralists from the Eurasian steppes. In Daniels, Megan (ed.), Homo Migrans: Modeling Mobility and Migration in Human History. Albany: SUNY-Press, IEMA Distinguished Monograph Series. https://www.academia.edu/44892216/ Anthony_2021_Migration_nomads_from_the_east_IEMA_SUNY_Buffalo Burkert, W. (1985). Greek Religion (J. Raffan, Trans.). Blackwell Publishing Ltd. & Cambridge MAL: Harvard University Press. Cebrián, R. B. (2010). Some Greek Evidence for Indo-European Youth Contingents of Shape Shifters. JIES, 38(3&4). Fall/Winter, 343. Institute for Study of Man. Eliade, M. (1972). Zalmoxis, the Vanishing God: Comparative Studies in the Religions and Folklore of Dacia and Eastern Europe. Chicago: University of Chicago Press. Frazer, J. G., Sir. (1971). The Golden Bough, a Study in Magic and Religion. New York: Macmillan. Paperback edition 1963.
22 Metamorphoses in Youth Initiation Rites
197
Gernet, L. (1981). The Anthropology of the Ancient Greece. Baltimore: John Hopkins University. Initiation in Ancient Greek Rituals and Narratives: New Critical Perspectives. (2003). D. D. Christopher & A. Faraone (Eds.). Oxfordshire: Routledge. McLynn, N. (1996). The Fourth-Century ‘Taurobolium’. Phoenix, 50.(3/4) (Autumn–Winter), 312–330 (19 pages). Olteanu, A. (1998). Metamorfozele sacrului. Bucharest: Paideia. Puhvel, J. (1987). Comparative Mythology. Baltimore: John Hopkins University Press. Strabo. (1903). The Geography of Strabo. Literally Translated, with Notes, in Three Volumes, ed. H.C. Hamilton, Esq., W. Falconer, M.A. London. http:// www.perseus.tufts.edu/hopper/text?doc=Perseus%3atext%3a1999.01.0239 van Gennep, Arnold. (1909 [1960]). The Rites of Passage. Oxfordshire: Routledge & Kegan Paul. 1960 Reprint. West, M. L. (2007). Indo-European Poetry and Myth. Oxford: Oxford University Press.
23 The Romanian Folk Story “The Enchanted Pig”: The Motif ‘Beauty and the Beast’
Besides transformations at will to escape the magic force of the follower, there is a second type of metamorphosis resulting from a punishment by a divine being, or a wicked character. This kind of transformation, the result of a divine punishment, is very well represented in myths, as that of Artemis punishing Actaeon, or the punishment of the nymph Callisto, or the story of Circe turning Odysseus’ sailors into pigs. In folklore, the transformation as punishment is the well-known folktale Beauty and the Beast (Aarne & Thompson, 1987, ATh 402) type Monster (Animal) as a Bridegroom, the story of a beautiful girl’s marriage to a young man in an animal shape. Noteworthy, the two types of stories relating the metamorphoses of young people, show an interesting difference: in the transformations flight story discussed earlier, the reason for the shape changings is clear—running away from an adversarial power—while in the second type of folktale no clear motive is given for the young character’s punishment, except the mythic environment, where the divine power gets upset for a given reason. The most representative version of this type of fairy tale in the Romanian folklore is the story of The Enchanted Pig (Ispirescu, 1969, p. 46): an unsuspecting father promises to marry his daughter to ‘someone’ waiting home. In the Romanian story, she has to marry a pig, in fact, © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_23
199
200
A. R. Chelariu
a young man cursed by a witch into the shape of a pig (or snake in some areas of the country, as in Serbia or Italy). During the night the pig sheds his skin, turning into a handsome man, thus consummating the marriage, as she becomes with child; the husband interdicts his wife to tamper with his skin, under the threat of losing him forever. This story displays connections with the Greek or the Indian fertility ritual discussed above, as the princess parallels the queen from the ancient ritual, laying in bed with a character in animal shape, or covered in an animal skin. Further, the story continues with details that seem to describe a young woman’s initiation ritual: During the night, the young woman oversteps her husband’s interdiction, and while he sleeps she throws his animal skin into fire, hoping to prevent him from turning into a pig again. Her action happens only three days before the end of the spell, triggering his departure and the beginning of her journey in search of him, her journey of initiation. Her first action, throwing his pig skin into fire is reminiscent of Psyche burning Cupid’s face with a candle. The power of fire, and its sacred functions, is found in many classic myths, a well-known myth is that of Demeter, who holds the child Demophon on fire, to make him immortal (Morford & Lenardon, 1985: 228 Homeric Hymn to Demeter), to his mother’s horror and misunderstanding. In conjunction with the cursed ‘pig’ from the Romanian story, it is worth mentioning that Demeter’s sacrificial animals at the Eleusine’s ceremonies, were young piglets or a pregnant swine, cast into pits. According to the myth, Persephone’s abduction by Hades while she was picking flowers in the meadow was witnessed by, among others, Eubuleus, a swineherd (after some authors, another name of Hades), whose pigs are swallowed up by the earth along with the goddess. To honor this moment, a three-day harvest festival, Thesmophoria, was held only by women, dedicating to Demeter little pigs and phallic symbols cast into a hole in the ground (Frazer, 1971, p. 410). The pig sacrifices from this festival may indicate a common contextual source between a ritual of fertility, the young woman getting pregnant, and the young man in pig’s skin from our story. The journey in which the heroine must go in search of her departed husband, the motif of ‘the quest for the vanished husband’, has its correspondences in the myth of Cupid and Psyche (Aarne & Thompson, 1987, ATh 425C), and even in the Egyptian story of Isis and Osiris. The young
23 The Romanian Folk Story “The Enchanted Pig”: The Motif…
201
woman’s journey from the Romanian folktale, follows the same pattern: After she spends some time as a maid, she begins her journey to the magic world, reaching the House of the Moon, then the House of the Sun, ending in the House of the Wind; in each of these houses she is received with kindness, each giving her some advice, and the chicken bones left from her dinner. After wasting three pairs of iron shoes, she reaches her husband’s house, a house without a door or windows, suggesting that it may be a tomb, interpreted as a womb, perhaps the Earth womb. As Eliade explained, the archaic image of boys’ initiation was that of entering into a womb, an introduction into the invisible world from where they would come out anew; however, girls’ initiation consisted of revelations of the secret meaning of the natural phenomena, the visible (M. Eliade, 2005, pp. 47–55), which in our story is represented by the houses of the Sun, the Moon, and the Wind. Facing his womb/tomb, the girl makes a ladder of chicken bones received from the miraculous helpers, the cosmic houses, as the only way to enter it would be from atop, describing here unique elements of a possible archaic funerary ritual. Reaching him, she acts like a goddess, taking him out of the womb, bringing him to life, reborn into his new role, that of a husband and a father, marking the end of their initiation journey. As mentioned before, in their initiation rites the Greek girls always gave sacrificial offerings to a goddess, indicating that they were in the goddess’s service, or even impersonating the goddess herself, whereas the boys offered sacrifices to past heroes. The famous animal-groom folktales grouped under the ‘Beauty and the Beast’ type of story, become significantly flawless if interpreted as a ritual of initiation of a young woman, and her essential role in her husband’s initiation into his new position. The Enchanted Pig folktale may actually record a double initiation ritual, of a young man-husband and a girl-wife, perhaps a marriage ritual. Although at the core of the story is the wife’s journey in search of the departed husband, it is obvious that his journey of initiation could not begin without her actions in the process of his initiation, the burning of the animal skin. Their ritual actions are interdependent of each other: She must burn his animal skin in order to start her journey as a pregnant woman, while he goes into isolation into the Other World, similar to the seed she is carrying in her womb. Akin to the sacred fire from Demeter’s
202
A. R. Chelariu
myth (Homeric Hymn to Demeter) by burning her husband’s animal skin, the young woman performs an act of purification, ending a stage of his life, while she must begin her journey looking for him, a journey through her pregnancy. As Demeter, she acts as a fertility goddess: She treats her husband as harvested grain, and death by fire is the fate of grain into the baked bread. During the Eleusinian ceremonies, people were retold the mystery of the immortal grain, and the fire as the divine gift. These understandings were common in the minds of people from antiquity, symbols and metaphors mirrored in folktales, reminding of the powers of fire as the way to a journey of purification and transformation. Similarly, the hero needs the girl’s journey of initiation fulfilled, so he could reach the end of his own. Remarkably, while the young man’s initiation involves his ‘ingurgitation’ into the shape of an animal, the girl’s initiation and function is to take him out of that condition, burning the skin, helping him to become a husband and a father. Thus, the story of the Enchanted Pig or the Beauty and the Beast could retain an ancient tradition, a marriage ritual in which ‘the prince and the princess’ as portrayed in the early Greek bas-reliefs, showing the hero and heroine crowned on their wedding daiy ceremony (Lyons, 1997, p. 8) as bride and groom, are going together through a recounting ritual marriage initiation. Almost mirroring the Beauty and the Beast story are folktales recording the hero’s journey of initiation: After he marries a girl in an animal shape, a fish, bird, a turtle, or a deer; she turns into a beautiful girl/fairy at night; following the same pattern, he disposes her animal skin, resulting in her leaving him, and the beginning of his journey. The heroes and heroines’ ritual metamorphoses into animals living in the entire spectrum of nature, land, water, air, served to convey the sacra of life to adolescents, ready to enter into their new role in the community, that of a man among the men’s group, a shaman, a warrior, a groom, or a wife and a mother. Concluding, the mythical motif of shape changings from folktales could be seen as a symbolic projection of some premarital ceremonies, perhaps a reflection of a myth involving the copulation between divinities in animal shape as an act of creation of the world, a cosmogonic myth. The transformations flight motif carried from myth to fairy tales, with its symbolic connotations, were most likely, over time, no longer
23 The Romanian Folk Story “The Enchanted Pig”: The Motif…
203
understood by the folk storytellers and their audience, yet, nevertheless, mysterious and appealing. The reimagining of the traditional fairy tales and myths keeps resurfacing due to the relevance of the perennial themes from antiquity to present-day society. Significant themes often reproduced in contemporary media, film, TV shows, animated stories, such as the reimagined fairy tale Beauty and the Beast, are adapting an old story to the needs and understandings of our contemporary society. A careful analysis of the archaic stories reveals the woman position of powerful influence in the community. In the Romanian version of this type of fairy tale, The Enchanted Pig, the princess marries a man cursed in an animal skin, and acts on her own, performing the rite of burning animal his skin. She is not a victim, and, akin to the Demeter’s use of the sacred fire, she performs an act of purification as a goddess does, and in the end, she is the one taking him out of the tomb, bringing him to life, molding his destiny, reborn into his new role, that of a husband and a father. Without her, he would not have risen to life, remaining in his tomb forever. The outcome in this story and the other traditional ones discussed in the book can guide readers to discover, among the perennial motifs from the oldest stories, new aspects relevant to our current world.
References Aarne, A., & Thompson, S. (1987, 4th printing). The Types of the Folk-tale (ATh). Soumalinen Tiedeakatemia, Helsinki: Academia Scientiarum Fiennica. Eliade, M. (2005). Rites and Symbols of Initiation: The Mysteries of Birth and Rebirth, translated from French by W. Task. 3rd printing. New York: Spring Publications, Inc. Frazer, J. G. Sir. (1971). The Golden Bough: A Study in Magic and Religion. New York: Macmillan. Paperback edition 1963. Ispirescu, P. (1969). Opere. Bucharest: Editura pentru literatură. Lyons, D. (1997). Gender and Immortality: Heroines in Ancient Greek Myth and Cult. Priinceton: Princeton University Press. Morford, M. P. O., & Lenardon, R. (1985). Classical Mythology. New York and London: Longman, Inc. van Gennep, A. (1909 [1960]). The Rites of Passage.Oxfordshire. Routledge & Kegan Paul. 1960 Reprint.
24 Indo-European Social Structures and Youth Initiation Rites in Romanian Folk Customs
The examples of youth initiation rites given in Chap. 23 raise the question: Could we find a relationship between the patterns in the rites of initiations and the Indo-European social classes as postulated by Dumezil and other researchers of the last decades? Many researchers observed that the hero’s rites of passage are structured in three major phases: the separation from an existing condition, generally, the parents’ house, the initiation process proper, and lastly, the return to the community. From M. Eliade’s perspective (1994, p. 47), the fundamental pattern of all initiation rituals includes: first, the torture at the hands of masters of initiation, demons, or spirits; second, the ritual death, in which the novice descends into hell or ascends to heaven; and third, the resurrection into a new mode of being. Arnold van Gennep (1960, p. 11) discussed the same pattern, separation from a previous status, then ‘transition’, the liminal rites of transformative actions within a symbolic boundary, and lastly, reentering the community into the new social status. The persistence of the threefold pattern brings investigation to the G. Dumezil’s (1958, p. 18) tripartite ideology of the Indo-European social classes, the king, the warrior, and the providers. Recently, Emily Lyle and Kim McCone extended the discussion to reflect the ‘age-grades © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_24
205
206
A. R. Chelariu
division’ from human life, namely, youth with power and vigor, maturity with fertility and production, and old age associated with the class of shamans, priests, or kings (McCone, 1986, pp. 1–22; Lyle, 1997, p. 63). To the threefold social classes Nick J. Allen (1999, p. 241) added a fourth one, to reflect ‘the other’, ‘the outside’, or ‘the beyond.’ Taking a step further in this discussion, could the tripartite social structure be applied to the initiation rites, in an effort to bring new perspectives to the Indo- European comparanda? The tripartite social structure of the archaic society, the king/priest class, followed by the warrior and the producer, seems to endure as leitmotif in the Indo-European culture, and the effort to align the three-step structure visible in the youth initiation’s rites, with the Indo-European tripartite social organization, could be challenging as the archaic data on this topic is limited. The foremost scope of the initiation was to introduce the novices into the adult group, the sacral heritage, bestowing upon them the pathway of their heroic and divine ancestors, through a process experienced by their parents as well, for generations. According to the social structure, we could assume that the various initiation rites introduced and guided the youth into the new position within the group, that of a warrior, a leader, or a producer. Analyzing the data from this perspective could help differentiate among initiation patterns associated with preparing the youth for specific roles in the community. For this purpose, the examples selected, mainly from the Romanian folklore, in association, when possible, with classical myths. The material used for this discussion belongs to oral literature, the folk stories transmitted for centuries through a process that had to rely primarily on established patterns necessary for memorization: “The action pattern establishes a principle of synthesis which is a priori with respect to any specific tale. It explains why it is possible that the listener becomes speaker in turn—which is the principle of the traditional tale—and why good tales can be memorized so easily, by hearing them only once; there are not terribly many items to memorize, since the structure has largely been known in advance” (Burkert, 1979, p. 17). It should be added here the esteemed implications of any ideological heritage reflecting the archaic values, to which the use of current expressions such as ‘a good
24 Indo-European Social Structures and Youth Initiation Rites…
207
tale’ may not apply. In the same context of oral memorization, rituals were based on repetition of patterns, in conjunction with seasonal feasts and celebrations, through rhythms and chants, accompanied by music and dance, all creating the complex process of remembering the cultural heritage of the community. In the effort to untangle ritual descriptions in myth, Graff takes as examples the three-step pattern of initiation in Theseus and Jason myths: departure from home, experience far from home, introduction to sexuality and identity, and concludes that the tripartite structure is a very basic narrative structure (Initiation in Ancient Greek Rituals: Graff, 2003, p. 3). The author argues that Theseus’ career served as a role model for the young, who, after years of separation and military training, received a shield and a spear, items signaling his entering into the new social position, from the status of a boy to that of an adult, the status of a warrior. In folklore, the Romanian tradition offers a descriptive example of puberty initiation through the transformations flight, as found in the narrative Oh or The Boy at the Devil’s school (Bârlea, 1966, p. 493), where the young man’s journey follows the traditional hero’s initiatory pattern: first, separation (from his father), followed by the underground sojourn in the wizard’s school, where he acquires a cosmic understanding of the world through ‘the animal powers’, a world that will not release him without a confrontation with the force of the wizard, ending with his rebirth into a new position, as the prince/king/husband, the first class position from the IE tripartite social structure. The sequence of the metamorphoses into animals from the three realms, the sky, the earth, and the water, widespread in the European culture, reveals again, a tripartite structure, at cosmic level. The initiation of the boy at the devil’s school in ‘becoming’ and getting powers from animals from the three realms, points to a metaphysical journey through the entire spectrum of life. As John Shaw demonstrated in his study, the tripartite structure is visible from Herodotus and the Scythians’ encounter with Darius, through funerary rites from Rg Veda, the Mahabharata episode of god Indra born of the sky, the earth, and the waters, or the god Tvastr, and through Odysseus’s threefold oath, on earth, heavens, and water, all raises substantial evidence that the tripartite cosmological structure existed from common Indo-European times (Shaw, 2007, p. 262).
208
A. R. Chelariu
In his work, Dumezil discussed extensively the warriors and their animal shape changings, and the many shapes a mythical character can take, in particular the series of ten forms the Avestan god Vərəθragna takes, concluding that the transformation motif attested in many mythologies, derives from ancient beliefs in “a gift of metamorphosis or to a monstrous heredity, the eminent warrior possesses a veritable animal nature” (Dumezil, 1970, p. 139). This “gift of metamorphosis” of the warriors may have been based on the beliefs in the existence of ‘souls, spirits’ attached to individuals and their appearance, the exterior costume being a distinct feature of the personality, from which the development of ritual shape changes led to a metaphysical journey into the mysteries of life. The tripartite cosmic ritual pattern, sky, earth, water, appears to be part of a very archaic system, as it is seen on the gold ring (see above) dating from 1450 to 1400 BCE, the “Ring of Minos” (Shaw, 2007). On the action level, the data pertaining to initiation rites from the Greek documents present a diversity of denominations used locally throughout Greece for the puberty rites of initiation of young men or women, showing no pan-Hellenic initiation rites but a large level of variation, coming from rites tied to individual communities (Initiation in Ancient Greek Rituals, Graff, 2003, p. 9). Most of these rites are associated with entering the brotherhoods of arms, preparing the young for becoming the protector of the community. Following the rite of koureia, the koureion “the cutting ceremony” of the young long hair and changing girls’ outfits, as we know from Achilles’ story, they were admitted into the military band and recognized as citizens of the polis (Gernet, 1981, p. 23). The military education and initiation of Spartan boys, the agoge attributed to Lycurgus, show the same tri-fold pattern in three stages, the paídes, ages 7–17, the paidískoi, ages 17–19, and the hēbōntes ages 20–29. Around the age of 18, the students became reserve members of the Spartan army; at the stage of hēbōntes, roughly age 20, the students became fully part of the Spartan army, and at the age of 30, men were permitted to marry and to become full citizens of Sparta, who could vote and hold office (Burkert, 1985, p. 260). These Spartan military stages fit well with Emily Lyle tripartite structure based on age groups. Ceremonies of inducing adolescents into a band of young men are very common in the European traditions. Until recently, during the Romanian
24 Indo-European Social Structures and Youth Initiation Rites…
209
winter festivals, the young people in a community were organized in a group named ceata de flăcăi ‘the youth band’, that shows traces of possibly old customs related to male initiation rites. December 6th, the day of Saint Nicholas, was, in Romanian tradition, the day of entering into the sacred primordial time, signaling the end of the year before the coming of the new year. On this day, a group of six to eighteen young unmarried men between 17 and 20 years of age were released by their parents into the custody of the leader of the group, housed by a designated host house, to prepare for the winter fest. To start, the group elected by vote a leader from among themselves, a young man with more experience and ability, named vătaf; then they elected a little leader, vătaful mic, a judge, an accountant, and a cook. For the festivities, they hired a singer to accompany them in the caroling procession (Herseni, 1997). These functions are reminiscent of the initiation activities at Kato Syme in Crete, were the votives show males that can be classified in the following social categories: “the warrior/hunter, the symposiast (figurine holding a cup), the musician (flute and lyre player)” (Initiation in Ancient Greek Rituals, Marinatos, 2003, p. 132), similarities that indicate a basis for ceremonial rituals. The leader of the Romanian ‘ceată’ had as main responsibilities to prepare the group for Christmas Caroling, supervising their behavior, organizing and coordinating dance lessons, and planning dance parties during each community festival for the entire year. Beginning with December 6th, the group prepared for the day of Caroling: learning carols, making different garment ribbons by which to differentiate among the position of each participant, decorating the ceremonial flag, in DRom n. steag, steajer, steajăr ‘pole, flag’, with roots in PIE MA *(s)teg ‘pole, post’; IEW 1014 (s)teg-2 ‘pole, stick, beam’, with cognates in Lat tignum ‘wooden beam’; NE stake; Alb shtëngezë ‘pole, support’; other DRom developments are n. stejar ‘oak’; n. stinghie ‘wooden beam’; n. stînjen ‘measurement—1.366 meters’. To mark festively the host house, a decorated with colorful ribbons evergreen tree was set in front by the fence gate, in DRom brad, ‘fir tree’ with PIE root IEW 109–110: *bhrozdh- < *bhares-: bhores- ‘point, stubble’, formant bhrezdh-, bhrozdh-; Dacian PN Bersovia; Vasmer I:109 *bhers- ‘point’; Alb. bredh (< earlier *berdh) breth, bredhi ‘spruce, fir-tree’ Palb. *braδ- (AE 107) (common Alb. –ĝ- > −dh- phonetic mut ation); other cognates: OHG burst ‘bristle’, Oir brot
210
A. R. Chelariu
‘sting, prick’, Ocorn bros, Bret broud, OHG brart ‘edge, border, stem, stem bar, stem post’, Oice barr- ‘needle, conifer’, Swe dial. Brad, Lith. béržas, Latv bērzs ‘birch’, Rus berëza ‘birch’, Oss bärz ‘birch’; Skt bŭrjá ‘birch’; also Lithuanian place-names Beržorai. The group’s most exciting task was creating the mask to be worn by the main character at the ceremony, the Turca (unknown etymology) or Capra ‘goat’, featuring a fantastic animal with the beak of a stork and antlers of a goat. During the time spent in the host house, the leader vătaf would invite girls, pair each boy with a girl, and teach them how to dance, and shout short rhymes, strigături ‘callings’, a development from the DRom v. striga ‘call out, shout, cry out’, reflex from PIE MA.*(s)trep‘cry out, dispute’ with cognates in Lat strepō ‘cry loudly, make noise’; OE þrafian ‘restrain, reprove, urge, demand’; or, from the root *(s)preg‘speak’, IEW 1036 (s)treig-, streid(h)- ‘to hiss’, in Lat strı ̑deo ‘creak, rattle, hiss’; other cognates NE speak, Alb shpreg ‘express, voice, utter’, a basic Romanian verb considered by DEX.RO as related to the Lat *strigare < strix, strigis ‘owl’, Grk ςτριγξ -γγος ‘night bird’, relation semantically very difficult to accept; the Latin form is most probably the origin for the DRom noun strigoi ‘spirit of the dead’ < Lat striga ‘evil spirit, witch’. The strigături ‘callings’ shouted during the dance, are short rhymes, usually funny positive sayings, but also some ironic comments on the account of lads or girls with bed habits, laziness, or making stupid mistakes, all meant to correct unacceptable behavior. Being the subject of a negative ‘calling’ was very embarrassing, and young people did their best not to be called out like that. Special emphasis was put on the proper behavior of the youngsters during the entire time: they were doing everything together, were not allowed to leave the host house by themselves, and any bad behavior resulted in shameful exclusion from the group. On Christmas and New Year’s Eve, the group will go caroling each house in the village, and received from each, money and treats, nuts, sweet breads, gifts that were divided equally among them. Turca, the zoomorphic mask worn by a designated man, would follow them everywhere in total silence, scaring women, girls, and children (Herseni, 1997). On Christmas day, the masked Turca performed in the center of the village, in front of the entire community, a long and spectacular dance; although the dance suggests a ritual scope, the mythical significance is lost.
24 Indo-European Social Structures and Youth Initiation Rites…
211
Similar customs are found in ancient Greek tradition: During the period of seclusion, adolescents received instruction in adult activities, such as hunting for boys, and spinning for girls, and equally important, they received introductions into tribal tradition through learning of songs and dances (Burkert, 1985, p. 260). Another youth band is found in the Romanian wedding entourage, the leader ‘vătaf’ or ‘captain’, and the groom’s best men, guiding the bride’s cart taking her to the groom’s house, followed by the entire suite, analogous to the Greek wedding ceremony and the bride’s suite, ‘paides propempontes’ (Oakley & Sinos, 1993) (see above). The data analyzed here seem to offer a mixture of archaic elements altered with more recent folk customs. Despite these difficulties, certain patterns could be observed and be considered relics of the ancient social structure. Van Gennep argues that in discussing initiation rites, we should consider classes and professions, next to the ritual use of occupational tools that could lead to ritual inclusion in certain restrictive group. Consequently, separation and taboos are important, and separation of youth in distinct groups is signaling social organization, as warriors do not keep company with blacksmith (van Gennep, 1960, pp. 101; 192). Likewise, David Leitao (Initiation in Ancient Greek Rituals, Leitao, 2003, p. 111) argues that rites should be discussed in the context of “the social practices out of which such rites originate;” traditional small-scale societies in ancient Greece were mostly organized according to kinship and gender, but in polis the society was more complex, differentiated by social classes (liturgical, free citizens, slaves, and so on) or, by membership in voluntary associations, religious cults, political clubs, etc. The young man, first, had to claim his identity as a man and a citizen, but equally important, he had to claim his position as a member in one of the social groups, “member of the elite (kaloi kai agathoi)”, “member of the Alcmenoid genos”, or, “initiate of the Eleusinian Mysteries”, or, “practitioner of the metallurgy trade”, and so on (Initiation in Ancient Greek Rituals, Leitao, 2003, p. 111). Such complexities in the social organization presuppose selective initiation rituals, in accord with the social position of the novice, enabling us to observe possible traces of tripartite social structures. It can be quite difficult to detect from all the data, or lack of it, specific aspects involved in specific rituals.
212
A. R. Chelariu
The information describing recent Romanian youth training for Christmas traditions, reveals an agricultural society with a simple social structure, the leader and the young people. The masked man, personifying the mythical animal Turca, or Capra, can be interpreted as reminiscent of the shaman, respected and feared, whose significance is mysterious, as it is his name, but when he scares women and young girls, he could very well be associated with the luperci or berserkers, the first Dumezilian class. The social structure discerned in the fairy tales, for example, in the Romanian The Enchanted Pig story, the characters addressed as prince and princes, qualify them as part of the first class, in conjunction with the folk traditions of all Europe, reflecting the archaic customs, by which a marrying couple was addressed as emperor and empress, or prince and princes, on their wedding day. This relation gives room to ambiguity, as, for one day, the first class characters in marriage customs are, in fact, part of the third class in the tripartite social classification, that of producers. Considering the many fertility connotations in myths and fairy tales, as expressed for example, in The Enchanting Pig story, the heroine actions in relation to a goddess of fertility, position in the third class, that of producers from the Indo-European the social organization perspective. But, the girls’ protectress, the goddess of fertility and procreation, a feared divinity, with strong ties to the Other World, with power of life and death over kings, warriors, and producers, transcends all Indo-European social limits, revealing her Neolithic roots, situated in a class of her own, in a fourth function, as Nick Allen suggested.
References Allen, N. J. (1999). Hinduism, Structuralism and Dumézil. Published in E. C. Polomé (Ed.), Miscellanea Indo-Europea. [Journal of Indo-European Studies Monograph No. 33: pp. 241–260. Institute for the Study of Man.] Bârlea, Ovidiu. (1966). Antologie de proză popular epică. Bucharest: Editura pentru Literatură. Burkert, W. (1979). Structure and History in Greek Mythology and Ritual. Berkeley Los Angeles: University of California.
24 Indo-European Social Structures and Youth Initiation Rites…
213
Burkert, W. (1985). Greek Religion (J. Raffan, Trans.). Blackwell Publishing Ltd. & Cambridge MA: Harvard University Press. Dumezil, G. (1958). L’idéologie tripartie des Indo-Européens. Bruxelles: Collection Latomus, Vol. XXXI. Dumezil, G. (1970). The Destiny of the Warrior. Chicago: University of Chicago Press. Eliade, M. (1994). Istoria Credintelor si Ideilor Religioase (3 vols. ed.). Chişinău. Universitas. Gernet, L. (1981). The Anthropology of the Ancient Greece. Baltimore: John Hopkins University. Herseni, T. (1997). Colinde și obiceiuri de Crăciun. Bucharest: Grai și suflet. Initiation in Ancient Greek Rituals and Narratives: New Critical Perspectives. (2003). D. D. Christopher & A. Faraone (Eds.) London: Routledge. Lyle, E. (1997). Age Grades, Age Classes and Alternate Succession: A Restatement of the Basis at the Societal Level of Indo-European Symbolic Partition. Emania, 16, 63–71. McCone, K. R. (1986). Werewolves, Cyclopes, Dibergs and Fianna: Juvenile Delinquency in Early Ireland. Cambridge Medieval Celtic Studies, Winter, 12: 1–22. CELT Corpus Electronic Texts. https://celt.ucc.ie//cmcs.html Oakley, J. H., & Sinos, R. H. (1993). The Wedding in Ancient Athens. Madison: University of Wisconsin Press. Shaw, J. (2007). A Gaelic Eschatological Folktale, Celtic Cosmology and Dumezil’s “Three Realms”. JIES, 35(3&4), Fall/Winter, p. 249. Institute for the Studies of Man. van Gennep, Arnold. (1909 [1960] Reprint). The Rites of Passage. Oxfordshire: Routledge & Kegan Paul.
25 Father Christmas: Romanian Moş Crăciun—A Solar Myth
Winter solstice was the time of many festivals in Europe, from antiquity to present day. The most widely spread customs were centered on the belief that the sun dies in the shortest day of the year, a perilous period of time for the entire community. It was a time when chanting rituals and fire works processions were performed, in the hope that their prayers would help to revive the astral divinity. In turn, the sun would assure abundant crops, prosperity, and wealth for men and animals. Folklore from all over Europe, as well as Romania, is rich in customs related to the winter solstice. On December 20th in every Romanians household began the winter ceremonies, starting with the ‘day of Ignat’, Latin ignis, the day of fire, recorded in the Christian calendar as the day of Saint Ignaţie (Muşu, 1982). On this day, the most important Romanian winter rituals celebrated the power of fire; it was the most joyous time, when young people gathered on top of the hills, rolling burning branch wreaths down to the valleys, recreating the rolling of the sun in the sky. Christmas Eve was marked by the popular tradition of colind, ‘caroling’; after weeks of preparations, groups of young people, colindători, went caroling from house to house, singing the colinde Christmas carols, wishing everyone good health and prosperity. For their singing and good © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_25
215
216
A. R. Chelariu
wishes, each household rewarded the caroling group with dry fruits, nuts, offering drinks to the older ones. It was the day when people mixed in the animals’ food certain breads and seeds; on the house floor was spread grain to encourage prosperity, and gifts of food and clothes were offered to the neighbors and the less fortunate. The sacred function of fire was linked to many interdictions, and it was forbidden on Christmas Eve to give away fire from the hearth. During the caroling, young boys singing in houses would rummage the fireplace to increase the power of fire (Herseni, 1997). The Christmas Eve dinner was a special event, for which the woman of the house baked specific round braided breads, colaci, symbolizing the sun. In some parts of the country, the bread was made in the shape of hands, most likely representing the hands of Crăciun’s wife (Christmas’s wife), Mrs Santa Claus, from a very popular folktale, a very intersting example of pagan and Christian syncretism: Virgin Mary, looking for a place to give birth, knocks at the house where the Old shepherd Crăciun, Old Christmas—Saint Nicholas, and his wife (Crăciuneasa) lived; because young Mary was having a child out of the wedlock Crăciun (Christmas) was not letting her give birth in his house. In spite of her husband’s refusal, Crăciuneasa houses Virgin Mary in the stable, and helps her to give birth to Jesus. That angers Crăciun, who cuts off his wife’s hands; Virgin Mary performs a miracle, and Crăciuneasa gets new golden hands (Marian, 1994, p. 54). Some researchers associate the traditional Christmas Eve bread in the form of hands with this story, others consider the golden hands part of an ancient cult of the sun (Muşu, 1982, p. 21). Impressed by the miracle of golden hands, Crăciun becomes the first Christian, an indication of his condition as a pre-Christian deity. The story ends with Crăciun starting a bonfire of evergreen branches, and inviting everyone to dance around it; the Old shepherd Crăciun offers all cheese, milk, and sour cream, a feast prefiguring his function of a giver of goods. In the Romanian language, the name for the Christian holiday, Christmas, is Crăciun, also the name of the character that brings presents to kids, Moș ‘old’ Crăciun Santa Claus, Father Christmas. The DRom moș ‘old man’, and moașă ‘old woman, midwife’ are archaic forms, with roots in the PIE IEW 699 *māk̑ - ‘long,’ with developments in Alb moš ‘old(er)’, Lith móša ‘sister of man’, OPr moazo ‘aunt’, Latv mãsa ‘sister’, Toch A mok, B moko ‘old’,
25 Father Christmas: Romanian Moş Crăciun—A Solar…
217
mokone ‘age’; Pasto məšər ‘elder, patriarchs chief of village’ (Rădulescu, 1984). The motif of mutilated hands is well represented in the IndoEuropean tradition, for example, in the German myth of Tyr, who loses his hand to Fenris-wolf. Some Romanian Christmas carols describe in verses the birth of Jesus in the Old Crăciun’s yard: Bună seara lui Ajun/Good evening to the Eve Night Mâni îi ziua lui Crăciun/Tomorrow is Christmas Day Moș Crăciun îi om bătrân/Father Christmas is an old man Moașă mare-i Crăciunoaie/Middwife is Mrs Christmas Că-i născut un domn prea mare/Born is a greater Lord În curtea lui Moș Crăciun/In the Father Christmas’ yard S-a născut un Domn prea bun/Born is a kind Lord S-a născut un Domn frumos/Born is a handsome Lord Cu numele lui Hristos/His name is Christ Mititel, îi mititel/Little, little is He Râde Dumnezeu la el/God smiles upon Him.… (Pamfile, 1997, p. 302) (my translation)
In another Christmas carol, the divine hierarchy is explained as follows: St. John, St. Mary, and Jesus Christ visit Moș Crăciun—Father Christma; after eating and drinking, St. John says to Moș Crăciun: ‘We eat and drink but we don’t ask who among us is higher in status; in his answer, Moș Crăciun addresses Jesus directly: ‘higher is you, Jesus, since you were born, I was there and I received you in my mantle, and I didn’t let you touch the ground’ (Pamfile, 1997, p. 272). This protective act by which Father Christmas receives the new God, and makes sure He would not touch the Earth, validates the divine nature of Moș Crăciun as an archaic celestial divinity. On the Eve of Christmas there is a traditional custom in most European households, to bring inside the house a log or stump named the Christmas log, an oak log, known in English as the Yule log, and in Romanian as butucul crăciunului ‘the Christmas log’. Symbol of the sun and its powers, the log was kept burning until spring, when the remaining ashes were spread on the fields before plowing. In some areas the ashes were mixed
218
A. R. Chelariu
with the food prepared for cattle and poultry, in the belief that they will become healthier and more fertile. The oak log burnt until spring in the belief that the weakened sun would be helped to rise and shine again,: “…it is no very far-fetched conjecture to suppose that the Yule log, which figures so prominently in the popular celebration of Christmas, was originally designed to help the laboring sun of midwinter to rekindle his seemingly expiring light” (Frazer, 1971, p. 746). Of the same opinion is the Romanian folklorist Gheorghe Muşu (1982, p. 21) who concludes that this yule log brought inside the home on the eve of winter solstice was burned in the hearth to help the sun revive. The custom was the same in all the regions: On Christmas Eve, a log was carried inside the house to burn continually until the January holiday of ‘baptism’ bobotează, when the ashes were spread at the roots of the fruit trees in the orchards to reensure their fertility (Pamfile, 1997, p. 284). The burning the oak stump, called ‘bușteanul Crăciunului’, recorded in many areas of the country (Fochi, 1976, p. 36), is found also south of the Danube, among the Aromanians, where it was named ‘boadnic’. In the French tradition this log is called ‘chalendal, chalendaou, calignaou, calnos’ (Ionescu, 1978, p. 120), to which we can add the Romanian călindău, another name used to designate the Christmas log. The custom is also well known among the surrounding Slavic populations, Bulgarians, Serbian-Slovaks, Ukrainians, Russians. The same tradition, to bring ceremoniously the oak log on Christmas Eve into the house and burn it in the fireplace for as long as possible, was well known in the northwestern part of Europe. This custom marks the sacred role of the oak in the European tradition: “The worship of the oak tree or of the oak god appears to have been shared by all the branches of the Aryan stock in Europe. Both Greeks and Italians associated the tree with their highest god, Zeus or Jupiter, the divinity of the sky, the rain, and the thunder. Perhaps the oldest and certainly one of the most famous sanctuaries in Greece was that of Dodona, where Zeus was revered in the oracular oak” (Frazer, 1971, p. 184) Similarly, it is well known that ancient Celts performed all their rituals with oak leaves. For the Germans the oak was the holy tree, symbol of the god Donar or Thunar, the same with the Scandinavian Thor, all meaning ‘thunder’. The Slavic god of thunder Perun or the Lithuanian god of
25 Father Christmas: Romanian Moş Crăciun—A Solar…
219
thunder and lightning, Perkunas or Perkun, were honored by sacred fires of a certain kind of oak, burning continuously. In Lithuania, men sacrificed to the oak-tree, while women sacrificed to the lime-tree, a custom suggesting that the oak was considered a male spirit. The tree, regarded as the center of the world, a source of good luck and protection, continued to be celebrated in Germany and Scandinavia as late as the nineteenth century. It was believed that gods assemble beneath the World Tree or the World Pillar, and people gathered to sacrifice at its roots (Frazer, 1971, p. 184). In the Germanic world, the name used for the World Tree was Yggdrasil, and Yggr was one of the many names of Odin (Wodan), thus the usual interpretation of its name was ‘the horse of Yggr’ (Davidson, 1988, p. 170). The oak function in the Indo-European traditions can be related to the incineration of the dead, as, most likely, the funerary pyres used oak wood, associating the sacred oak to the light and fires of funerary ceremonies, in the belief that the soul will go into the light of the sun. In the Romanian and the Balkans folklores, the trees cult survived well into the twentieth century. A well-known Romanian custom, attested from the seventeenth century, was ‘the judgment by the borders under the oak trees’, a ceremonial trial intended to settle disputes, particularly over neighboring properties, in which the oak had the role of a witness next to a border location (Vulcǎnescu, 1972, p. 67). The sacred role of the oak in the Romanian tradition is proved again by the custom that the shepherds settled barren sheep under an oak or an evergreen tree, as mentioned in Chap. 20. The Romanian word for the Yule log, butucul Crăciunului, ‘the Christmas log’, and the name of Christmas holiday, Crăciun, are subject of long disputes and controversial etymologies. The most accepted etymology for the noun ‘Crăciun’ was the Latin form creatio, −onis ‘creation’ presented by Ovid Densuşianu in his work (1929–1932), and reiterated by Al. Rosetti (1971, p. 423). It should be stated here that the Latin form creatio, −onis, is developed from the verb creō, − āre ‘to procreate; be bom; cause,’ cognate with Lat crēscō, −ere ‘to be born; increase’ with the original meaning ‘to make grow’, (de Vaan, 2008) that developed in DRom basic vocabulary a crește, creștere ‘to grow’; the PIE *k(w)rehr [pr./ aor.] ‘to become bigger, stronger’. IE cognates: SeCS okrijati ‘to recover’, OCz. krati, pr. krejU Ru.dial. Krejat’ ‘to heal’, Ukr. krijaty ‘to become
220
A. R. Chelariu
healthy’, Bulg. kreja ‘to become weak, be ill’ < PS1. *krejp ‘to heal’; Pokorny IEW 577 ḱer-2, ḱerǝ-, ḱrē- ‘to grow, make grow, nurture;’ the argument against the etymology from the Latin noun creatio, onis, from the verb creō, − āre ‘to procreate; be bom; cause,’ > crēscō, −ere ‘to be born; increase’ is the DRom development crește ‘grow’ without any connotations to the notion of birth, as in ‘to be born’. In the Christian dogma regarding Jesus birth, natus et non creatus,.the ‘natus’ concept is reflected in all the Christmas formulaic wishes: Italian ‘Buon Natale’, Spanish ‘Felis Navidad’, Russian rozhdestvo ‘Christmas’ (v. roditi ‘birth’˝) Ukrainian rizhdvo, with the same meaning. The Romanian form Crăciun, explained by the Latin creatio ‘creation’ in the context of Christianity in Europe, would imply the existence on the Romanian territory of a highly formed clergy, who would prefer the Latin concept ‘creatio’ to name ‘the birth of Christ’ instead of ‘nătus, -ālis ‘birth’ existing in DRom naște, naștere, as it is found in the rest of the Romance languages. It is also important to note that the date of December 25 was adopted as the birth of Christ in fourth century, a time when the Latin religious influence on the Dacian territory was weakening. More so, as shown above, the oral tradition retained the word Crăciun as the appellative for the character from the story about a mean old man, Moș Crăciun, who would not let the Virgin Mary into his house to give birth to baby Jesus. Could an argument be made that this Old Man, refusing the let Mary in his yard, but present at Jesus’ birth with midwife Mrs. Crăciuneasa, was witnessing the ‘Creation’ of Jesus? Other Romanian researchers, among whom Vasile Pârvan, Sextil Puscariu, Nicolae Drăganu, offered another controvertial hypothesis for the word Crăciun, relating it to the Latin form calatio ‘calling’, refering to the practice of the Roman priests, calling every first day of the month to announce the holidays to follow, developed in Lat calendae > DRom colindă ‘(Christmas) carol’ and DRom calendar (de Vaan calo, −are ‘to announce, summon’, calatio ‘convoking’, PIE *kelh1/klh1 ‘to call’). This solution, without correspondences in other Romance languages, suggests that only the Romanian language retained the Latin denomination for a practice that took place every month in Roman times, but specialized to name only the winter solstice celebration.
25 Father Christmas: Romanian Moş Crăciun—A Solar…
221
Another suggestion was that the word has its root in the Slavic languages, hypothesis contradicted by the fact that “the geographic spread of the word Crăciun outside the Romanian territory includes only the areas adjacent to the Romanians” (Caraman, 1931, p. 79). A. I. Ionescu approaches the subject of borrowing of ‘pagan’ cultic words from Balkan Latin into Slavic languages, such as calende, or Crăciun, “at a very early date” (Ionescu, 1978, p. 120). The unspecified dating, ‘at a very early date’, could not be before the seventh century CE, the recognized date of the Slavic invasion into southeast Europe. To which we may add that other researchers do not accept the Slavic influence on the Romanian language any earlier than the tenth century. The timing of the Slavic influence is very important in any discussion pertaining to the Romanian language etymologies, and particularly in that of Crăciun. Assuming that the form Crăciun, has its roots in the Latin ‘creatio- onis’, with the meaning ‘birth of Christ’, it was already established among the Christian speakers of Latin from the Balkans, it becomes difficult to explain that the Slavic forms Kračun, or Kračunec, or Korocjun, do no relate to the ‘birth of Christ’, but only to a period of time when traditional festivities took place. In the Novgorod Chronicle from 1143, the form Kracun was used in reference to the time starting from August 15th. In some cases, the time frame began by the day of the Saint Spiridon, which falls on December 12th, or, as in the Bulgarian tradition, on June 8th. These different dates relating to various Saints names, spread throughout the entire year, linked to many other fests, explains the hesitation in understanding the meaning of this winter solstice holiday on the part of these populations. The Romanian Crăciun, Aromanian Crăciun, Caraciun, Meglenoromanian Cărciun, naming a single fix winter celebration, had influenced the Slavic world, and not the other way around. The Albanian linguist, E. Çabej, (1965, pp. 101–115) believes that the Romanian form is borrowed from the Albanian reconstructed form *kerçun, “log, stump” (Çabej, 1961, pp. 313–317), even though the Albanian word never meant anything else but “log”, and it did not extend its meaning to such an important winter holiday. It seems somewhat difficult to believe that the population from the entire Romanian territory borrowed this word from Albanian language, and used it to name a very old and important winter fest, with its many rituals, and the mythic character
222
A. R. Chelariu
Moș Crăciun, whereas the Albanian language limited its sense to name only a “log, stump”, while for the Christmas Eve nata e buzmit, “the log night”, was preferred. A solution for this difficult etymology could be to approach the problem from a different angle, starting from the hypothesis that the initial meaning of the word Crăciun was not related to Latin forms creatio, or calatio, but rather related to the concept of ‘oak, sacred oak’, the log carried ceremoniously inside the house on the eve of the winter fest. For this solution, the argument could start from a Proto-Indo-European root for ‘oak’ and its reflex in Latin quercus ‘oak’, found in the Mediterranean region as quercus coccifera, ‘kermes’ or ‘holly oak’, a tree that was revered among people from this region as it stays green all year round, with leaves resembling that of holly, a sacred plant in pre-Christian Europe (Williamson, 1986, p. 65). Ancient Europe was covered in wild and frightful forests, such as a large and dense forest known as Hercynia Silva, described by Julius Caesar in De Bello Gallico as stretching along the Danube from the region inhabited by Helvetii to Dacia. The same description of Hercynian Forest and its gigantic oaks is found in Pliny the Elder, in his Natural History. In Dacia proper the southwestern part was covered by a deep forest, known in more recent times as Codrii Vlăsiei ‘Vlăsiei/Walachia Woods. The DRom codru ‘forest’ has its roots in the PIE MA *dóru, gen. dréu̯ -s ‘tree’; (Pokorny, 1959, IEW 214–217) *deru-, dōru-, dr(e)u-, drou-; dreuǝ-: drū- ‘tree’, an internal development with the associative particle co- + −dru; for this reflex we have cognates in OIr daur ‘oak’; Grk doru ‘tree trunk, wood, spear; Hit taru ‘tree, wood; Av dauru ‘tree, tree trunk’; Skt daru ‘wood’, gen. Dróș ‘wood, timber’; NE ‘tree’ is a derived form, as are Grk drus ‘tree, oak, OCS drŭva ‘wood’, drěvo ‘tree’; Alb dru ‘wood, tree’, kodrë ‘hills’; PAlb. *dru(n)- {1} Alb. drỹ [m] (g) ‘lock, door-bolt (of wood)’ (AE 146); in Irish and Greek this form tends to mean oak; Ir druid ‘priest’ that would always use oak leaves in service. Falileyev (2010 *dru ‘oak’) offers an interesting solution reconstructing the form *kaito- ‘forest’ + dru ‘forest + (oak) trees’? with cognates in OWelsh coit, Welsh coed ‘wood, forest’, OCorn cuit, MCorn coys, cos ds, Bret coet, coat ‘wood, forest, spinney’. The wildness and density of these forests, populated with large powerful aurochs, wolves, bears, boars, elks, reindeer, and such, must have
25 Father Christmas: Romanian Moş Crăciun—A Solar…
223
deeply impressed the people living in the nearby areas. Coupled with the proven religious functions of the ‘oak’ tree in the Indo-European mythology, the investigation in the etymology of the ‘oak’ can help clarify its mythic role. The reconstructed Indo-European root for ‘oak’ is *perku-s, < earlier *kʷerkʷu-s, ‘oak, force, vigor, the world tree’ (Pokorny, 1959, IEW 822–823). Related forms are the Irish ceirt (quiert) “apple tree” (Proto-Celtic perkunia, later ercunia), Welsh perth, “bush, hedge” (assuming a Proto-Celtic form kwerkwti), Latin quercus, ‘oak’, the Italian form quercia ‘oak’, and the Romanian cer ‘oak’. It is worth noting that in the Italian tradition there are also other names for Saint Mary, such as Madona della Quercia, Quercia della Vergine, perhaps echoing a Roman fest in which a virgin dressed in flowers personifying Junona, was carried in a cart next to an oak log decorated as her consort, allegorically representing Jupiter (Williamson, 1986, p. 65). The Romanian linguist Ariton Vraciu, without referring to the winter fest Crăciun, explained the Romanian scrum ‘ash’ through the Indo- European root *ker- (Pokorny, 1959, IEW 571) ‘to burn, ash’, maybe ‘oak ash’. This root, perhaps reflected in the Meglenoromanian Cărciun, can offer another possible solution, based on the burning of the log, and the ash used in the spring rituals of fertility. In the Balto-Slavic area, the IE root *kʷerkʷu-s > *perku-s, ‘oak, force, vigor, the world tree’, can be recognized in the Old Lithuanian religious character Perkunas, the thunder god, in the Lettonian Perkuons, thunder and the thunder god, as in the Old Prussian percunis, thunder, and the Old Russian Perun, thunder god. Roman Jakobson, in discussing some aspects of the Slavic divinities and the IE *perku- s, observes: “This root appears, for example, in Latin and Germanic languages, as substitute for the noun ‘oak’, a preferred tree for storms and devoted to the thunder god; and in the Indo-European tradition the same root with a nasal suffix means ‘a hill covered with oak, an oak forest’, in Celto-Latin Hercynia (silva), Gothic fairguni, Slavic *pergynja, Old Slavic prgynja, Old Russian peregynja, Polish przeginia” (Jakobson, 1985). The root is also responsible for another group of gods from the Germanic area, as Old Icelandic Fjorgynn, father of Frigg and Freya, and Fjorgyn, mother of Thor, not excluding the possibility of a relationship with fjor meaning ‘life’
224
A. R. Chelariu
(Lehmann, 1986) probably related to the Romanian form fior ‘shiver’, as in DRom fiorul vieţii ‘vital vibration’ (explained in DEXRO as Lat febris). In Romanian folklore, the name of the winter fest bears the name of the pagan deity Crăciun, the same name of the god’s log, while in mostly all the other cultures the name of Saint Nicholas prevailed. The sacred Yule log, butucul Crăciunului, brought inside the house on the Eve of Christmas to burn all winter, could have its roots in a solar cult honoring the highest god in the Indo-European mythological pantheon, the god of thunder, lightning, of fire, the sun god, idea expressed by Petru Caraman (1931) “Crăciun seems to include elements of the solar cult, related to Mithra, later replaced by dies natales Solis invicti”. Perhaps no other celebration fest could exemplify better the pagan and Christian syncretism, as the winter fest of Christmas celebrated on December 25th. It coincided with the Roman fest of Sol Invictus ‘the undefeated sun god’, declared by Aurelianus in 274 CE, the official patron of soldiers. The earliest literary reference to the pagan fest celebrating the birth of Sol Invictus and the “natus Christus in Betleem Iudeae” on December 25th is found in Furius Dionysius Filocalus’s Chronography of 354. Venerated by the Roman soldiers, Sol Invictus shared its place among them with the Mithraic cults, proven by inscriptions, as Deo Sol Invicto Mithre. The many Mithraic sanctuaries excavated in Roman Dacia signal an important presence, and the festival Dies Natalis Solis Invicti ‘Birthday of the Unconquerable Sun’ celebrated on December 25th, coincided with that of Mithra, also celebrated on December 25th at the winter solstice. They were probably worshipped by the Roman soldiers of the Danubian army around the time of the Dacian wars, and after the abandonment of Dacian provinces, as testified by the many religious inscriptions dedicated by military leaders of the Dacian legions, which most likely established a precedent for the winter solstice celebrations. During the Christmas fest, the Romanian folklore records a special event detailed by Ghinoiu (2003, p. 165), named The Burial of Old Christmas/Moș Crăciun: On December 28th, young men from the region of Someș (Transylvania) and the surrounding villages got together and organized a parody burial of Moș Crăciun; under a big secrecy they elected one among them to play the role of the dead Christmas/Crăciun, known to no one outside the group; in that morning, they gather together
25 Father Christmas: Romanian Moş Crăciun—A Solar…
225
singing a specific repertoire; the young man portraying Old Christmas is laid on a ladder covered in blankets, while the others get dressed as a priest, a deacon, a cantor, and one of them is in charge with keeping the fire on in the priest’s censer burning at all times; six lads carry the ladder with the ‘dead’ Old Crăciun on their shoulders, and the entire cortege goes around the village, followed by a funerary entourage accompanied by funerary music and loud dirges, reciting the following song (Ghinoiu, 2003, p. 169): Hey you Crăciun, you old man Today we will bury you Let us all young and old Take Crăciun down the hill And put him in a barrel (or in an ice hole) Over him we will put a stump Oh, Crăciun, oh, old man Go from here in peace Go on the Saturday water (down the river) And don’t ever come back here Cause another Crăciun came And it would be better than you. [my translation]
The ceremony is meant to make everyone laugh; after the ‘priest’ forgave the Old Crăciun for his sins, the lad jumps of the ladder as a New Crăciun as a young man, and everyone returns to the guest house dancing, drinking, and singing all night. The custom was known in other areas of the country as the burial of the old year. During the end of year, Romanians believed that the time of crossing from the old year to the birth of the new year, it meant ‘the time’ broke into chaos, when the dead came out among the living. People were not afraid of their dead ancestors, instead, received them with food offerings; everyone, relatives, neighbors, and especially the poor, were invited to accept the food, and in return, they requested the dead relatives to help in with current wants, marriage, good crops, conflict solutions, sickness, and so on, believing that all would be fulfilled (Ghinoiu, 1994, p. 320). It was the time when the old saints were celebrated in the Romanian
226
A. R. Chelariu
tradition, suggesting respect and tribute toward the ancestors, the moși ‘the elders’, believed to be forty moși, were celebrated on special dates in the winter, fall, and summer. The most revered ‘elders’ were Moș Crăciun ‘Old Christmas’ envisioned as an old white bearded shepherd, with his younger brother Moș Ajun ‘Old Christmas Eve’, Moș Nicoară ‘Old Nicoară/Nicholas’, to name just a few. A very important part of the winter ceremonies was the masked caroling accompanied by songs and jokes, sometimes very lascivious, believed to be related to ancient fertility rites. The most common masks were Turca, a buck head, or Capra, a goat head, and/or a deer head. As described in the previous chapter, the ceremonial dance of Turca performed in front of the entire community was a long and spectacular dance, whose mythical significance has been lost.
References Çabej, E. (1961). Crăciun, Studii și cercetări lingvistice, XII, 4, Bucharest. Çabej, E. (1965) Betrachtubgen uber die rumanisch-albanischen Sprachbeziehungen. RRL, X. 1-2, 101-115, Bucharest. Caraman, P. (1931) Substratul mitologic al sărbătorilor de iarnă la români şi slavi. Bucharest: Institutul de Arte Grafice Presa Bună. Davidson, E. H. R. (1988). Myth and Symbols in Pagan Europe. Syracuse: Syracuse University Press. Falileyev, A., Gohil, A. E., et al. (2010). Dictionary of Continental Celtic Place- Names: A Celtic Companion to the Barrington Atlas of the Greek and Roman World. CMCS Publications. Fochi, A. (1976). Datini și eresuri populare de la sfârșitul secolului al XIX-lea. Bucharest: Editura Minerva. Frazer, James G. Sir. (1971) The Golden Bough, a Study in Magic and Religion. New York: Macmillan. Paperback edition 1963. Ghinoiu, I. (1994). Vîrstele timpului, Chişinău, Știința. Ghinoiu, I. (2003). Sărbători şi obiceiuri Româneşti. Bucharest: Elion. Herseni, T. (1997). Colinde și obiceiuri de Crăciun. Bucharest: Grai și suflet. Ionescu, A. I. (1978). Lingvistică și mitologie: contribuții la studierea terminologiei credințelor populare ale slavilor. Bucharest: Litera.
25 Father Christmas: Romanian Moş Crăciun—A Solar…
227
Jakobson, R. (1985). Selected Writings VII: Contributions to Comparative Mythology S tudies in Linguisics and Philology 1972–1982 (S. Rudy, Ed.). Berlin. De Gruyter: Mouton Publishers. Lehmann, W. P. (1986). A Gothic Etymological Dictionary, Leiden. Brill. Pamfile, T. (1910-1914, 1997). Sărbătorile la români; studiu etnografic. Bucharest, Editura Saeculum I. O. de Vaan, Michiel. (2008). Etymological Dictionary of Latin and the other Italic languages. Liedern, Boston. Brill. Marian, S. F. (1994) Sărbătorile la români; studio etnografic. Vols. 1 & II. Bucharest: Editura Fundației Culturale Române. Muşu, G. (1982). Din Mitologia Tracilor. Buchares. Cartea Românească. Pokorny, J. (1959). Indogermanisches etymologiscches worterbuch. Bern Munich Francke Verlag. Rădulescu, M. M. (1984). Illyrian, Thracian, Daco-Mysian, the Substratum of Romanian and Albanian. JIES, 12(1–2), Spring. Institute for the Studies of Man. Rosetti, A. (1971). Crăciun. Studii şi cercetări lingvistice, XXII (1), 3–5. Vulcǎnescu, R. (1972). Coloana Ceriului. Bucharest: Editrua Academiei Române. Williamson, J. (1986). The Oak King, The Holly King and the Unicorn. New York: Harper & Row.
26 Conclusions
In closing, this study attempts to persuade the inquisitive reader that the folk data from the Romanian region represent a valid source of information, in spite of the lack of written documentation, and orality was, for centuries, an essential form of perpetuating and recording the cultural manifestations. This hypothesis falls in line with the general consensus among many researchers, who consider the folkloric materials collected mostly in the nineteenth century to hide a rich pool of mythical motifs, reminiscent of the archaic religious beliefs. Such views on the folkloric data and its validity as a source of documentation was held by George Dumezil: “Even from the most ancient times variants have existed, each as legitimate as the next. By the same token, narratives about similar but distinct subjects—for example, Indra’s various battles—have many times, even before the first documents, resulted in mixed forms, some more stable than others, and not unlike those encountered today by the student of folklore and living oral tradition in general” (Dumezil, 1970, p. 160, my underlining). This statement summarizes the current convictions, and that can helps validate the Romanian songs and tales as potential sources in finding mythical motifs among the collapsed metaphors recorded in folk data. In spite of the speculative nature of these © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_26
229
230
A. R. Chelariu
associations, the aim to reveal traits of the lost cultures from ancient times in the cultural heritage from countries as Romania can be of use. The search and the unveiling of mythical motifs concealed in the Romanian folklore and their relation to the cotroversial Indo-European cultural complex could shed more light on both. The structural similarities observed, for example, between the archaic world creation by the dismemberment of the primordial being, and the Romanian folk data mirroring the same sequence of elements of nature, flesh from the earth, bones from stones, hair from plants, blood from water, eyes getting light from the sun, breath as the wind, the mind linked to the moon, and thoughts from clouds, the head imagined as the sky volt, a structure readopted for the creation of man, proves the endurance of this archaic belief. The creation of man mythic motif, common throughout Eastern Europe, shows how an old view of the ‘world creation’ was transferred and adopted into the new Christian dogma, a pattern deeply embedded within the structural ancient understanding of the world. The archaic vew of the world creation is also reflected in the motif of cosmic dive in the primordial waters, a universal cultural pattern extending to many groups, from Southeastern Europe, through North America. The archaic vew of the world creation is also reflected in the motif of cosmic dive in the primordial waters, a universal cultural pattern scattered from Southeastern Europe, through North America, a motif expressing the intrinsic mankind view of creation from waters, as it is recorded in the Romanian story of the divine twins occurrence. By comparing the mythic data we can distinguish the specific Romanian understanding of the world creation: the result of the interdependence between the divine pair, Fârtat as a butterfly and Nefârtat as a worm. This view conveys the archaic perception of the duality of nature, light-darkness, water-earth, plants-dirt, and most importantly, the eternal metaphor of the life cycle, expressed in the core existence within a larva. This metaphor articulates the harmony between the light, earth, and water, the archaic understanding of nature and the symbiosis between humans and their environment. The divine twin characters from the Romanian creation myth display the ambiguity of the divinities, associated to fertility and the producers, position in which the Romanian twins, Fârtat and Nefârtat, are well fitted, and simultaneous, one, as the god of light, the other, as the god of the earth, are both united within the powers of fertilizing water, creation gods adorned as divinities of fertility and prosperity.
26 Conclusions
231
The powerful goddess of the Neolithic prevailed for millenia in the Romanian region, always feared, always revered. Protectress of life and birth, the deity of the moon, and sun’s sister, Ileana Simziana with entourage of zâne ‘fairies’, made her majestic presence from myth to fairy tales, to ballads and songs. Next to the other European goddesses, revered under different names, but having the same functions, Ileana continued to be respected, overpowering the patriarchal opposing millieu. The archaic mythic motif of hierogamy between the majestic goddess of the earth and fertility, the Romanian Ileana Simziana, the Fairy Queen, the Moon-earth divinity, and the Sun, her brother, preserved in Romanian folklore, shows the changing in the views on a ‘brother and sister’ marriage, rejected as an incest by social norms in the group, and the loss of the old divine unity in creation. The song also is recorded as a legendary story of the sun’s sister transformed into the moon, as she is taken from the sea by God and set on the sky to light the night, showing the popular creators’ versatility in combining old motifs into new ones, acceptable to the group. The Romanian folk data records the quintessential Indo-European myth of the dragon slayer, Iovan Iorgovan, the hero embodying another case of divine ambiguity, by the hero’s double role, as the dragon fighter, coupled with protector of the divine feminine procreative principle threatened by the dragon. His name, Iorgovan, stands for earth worker and giver of fertility. Noteworthy, the myth of the dragon slaying is represented in Romanian folklore in many versions, ranging from stories in which the hero hits the dragon with arrows, or cuts all the three heads, as found mostly in fairy tales, to stories in which the hero and the dragon argue about the consequences of the killing. His divine functions in our song are confirmed by him giving instructions on ways to protect the animals against the dragon’s fly coming from his severed heads, and becoming a culture hero. He is also a leader and a youth initiation instructor; the episode of the young lad being swallowed by the dragon reveals an important initiation motif included in versions of this song, most likely a warriors’ initiation ritual practiced among soldiers from the Balkan region, revealing the conservative tendencies of the Romanian cultural heritage. The youth initiation rituals recorded in many folktales, document shape changings, either to escape an adversarial follower, the result of magic punishment, or as a marriage ritual, a ritual process showing the
232
A. R. Chelariu
importance the ancient society gave to inducing the novices into the deep understanding of the human symbiosis with the environment. Another interesting mythic structure that survived in the Romanian folklore is that of the old mythic motif of the union between a god taking various shapes, sometimes as a stag, the ‘emperor’ of the forests, symbol of eternal rejuvenating youth, and a goddess of fertility and procreation, in the shape of a doe, as reflected in the ‘oratio nuptiae’ poem from the Romanian wedding ceremony. The ritual hunting of a cervide resulting in the start of a family, resonates in another mythic motif, the flight of metamorphoses between a reluctant goddess and a god in various animal shapes from all three realms, water, air, earth, signaling an intiation rite in the sacra of nature. The motif resurfaced as a male initiation in the Romanian folklore, with the same sequence of transformations, describing an initiation journey through the spectrum of life, ending with the change in the social status of the novice. Metamorphosis as punishment from the famous type of fairy tale, known as Beauty and the Beast, could be interpreted as a double initiation process, as analyzed in the Romanian story, The Enchanted Pig. The transformations from these folk stories, seem to record some marriage rituals reaching us from prehistoric times, showing the importance of impressing the young with societal requirements, establishing a family, connoting myths of world creation. Widely spread in many European folk traditions, the character of the Trickster exemplifies a common heritage, albeit the fact that many researchers do not recognize him as an Indo-European character, being more inclined to consider him a relic from the Old Europe pantheon. Either way, his function as a moral compass in the community sets him aside among the archaic divinities. The Romanian Trickster, Păcală, sits well among the other Indo-European ones, Pooka, Peik, Pekulis, or Patulos. The remarkable persistence of the tripartite pattern from Indo-European social classes (Dumezilian theory) to the tripartite initiation patterns, separation, initiation, and return, rises to the level of general postulate of the Indo-European ideology. The Romanian winter customs discussed here are intended to relate the youth bands preparing for Christmas caroling to the youth bands from archaic heritage and other European cultures. As an argarian society, the Romanian society preserved a less clear delimitation of the social structures, and pastoralism—the seasonal movement of livestock
26 Conclusions
233
was the basic activity of young shepherds. The rough conditions of shepherd life developed specific rules and interdictions, many relating to the initiation process. Among the young men living away from home for long periods of time, the cohabitation was dominated by many rituals, most remarkably the ritual hostility among young men from different tribes, stirring rivalry fights among them, as noted in the song Mioriţa. The winter solstice divine character Moș Crăciun/Father Christmas, in relation to the Sun god Mitra, Sol Invictus, could be part of a virtual Daco-Romanian pantheon that survived within the Christian environment, the main character of the winter fest, giving his name to the most important fest of the year. The conceptual milieu specific to the cultural and folkloric heritage of Romania reflects the Latin heritage, placing the language among the other Romance languages. Besides the basic vocabulary, the Romanian folklore did not retain much of the Roman divine pantheon, except, as in all the other Romance languages, the names of week days: Luni-Monday, Marți-Tuesday, Miercuri-Wednesday, Joi-Thursday, Vineri-Friday, (Sâmbătă-Sabat-Saturday), Duminică (Dominus day)-Sunday. The current months’ names entered the language most likely in more recent times, while the DRom folk names continued to be in use: ianuarie genar, gerar / cărindar (January), februarie făurar / faur (February), martie marț / mărțișor (March), aprilie prier (April), mai florar (May), iunie cireșar (June), iulie cuptor (July), august gustar (August), septembrie răpciune (September), octombrie brumărel (October), noiembrie brumar (November), decembrie undrea / îndrea (December). The Presence of the Vlahs (Valahs) populations South of the Danube, and the economic relations with the surounding people from the Balkan region, could explain the many common words, as names of the produce resulting from the pastoral occupation of the entire area. The cultural complexes from this region show a strong symbiosis between pastoralism and small-scale agriculture, revealing, perhaps, the old substrata, in which the fighters, warrior-like class is less noticeable. The Romanian oral heritage shows trends of mainly a rural society, with a limited scale social organization, in which the council called sfatul bătrânilor ‘the elders council’ occupies the most important place, with attributes that can be associated with the first Dumezilian function, responsible for the entire well-being of the
234
A. R. Chelariu
community. Besides the judecata la hotare ‘the judgement at the border’, settling land disputes, the elders council would perform the judgment of moral and behavior disputes in the following manner: the person to be judged would stand in the middle of a ‘circle’ drawn on the ground, while the elders and the community around it would perform the judging. Romanian folklore reflects a form of social organization in which a dominant abstract figure is împăratul ‘the emperor’, a first-class character, present mostly in fairy tales, reminiscent of the Roman influence. The preference for the Latin form could be the result of the relations between the Roman soldiers and Dacian-Roman population, with a faraway figure, the Roman Emperor, without a real role in their community. The social organization of young shepherds engaging in ritual, and on occasion, real fights, the brotherhoods and the numerous rules and regulations observed by this group, situates them in a class of their own that could be considered the warrior class. The autority class is represented in the Romanian folklore by the priest figure, the subject of many satirical anecdotes, perhaps due to recent intrusion of foreign priests from South of the Danube, whose first language was either Slavic or Greek, and were not understood ny the locals, and tried to fit into an old established social order. The mythic motifs detected in the folk literature transmitted to us from antiquity, and analyzed in this work, can offer a glimpse into the social-cultural, and religious practices among the Eurpean people, in hoping to recognize the social and spiritual continuity of the Romanian people. In this endeavor, language development plays a very important role, which deserves to be explored in light of the many new research works, starting from the reconstructed Proto-Indo-European roots with reflexes in Daco-Romanian language.
References Dumezil, G. (1970). The Destiny of the Warrior. Chicago. University of Chicago Press. Vulcǎnescu, R. (1972). Coloana Ceriului. Bucharest. Editrua Academiei Române.
Part IV Daco-Romanian Language Position Among the Indo-European Languages
27 The Daco-Romanian Cultural Vocabulary
Classified together with the other Romance languages, the Romanian language is still uses a large number of isoglosses of unknown etymology. The scope of this analysis is to observe relations between the Daco- Romanian language and the larger familiy of the Indo-European languages. The possibility that the substratum of the southeast European languages contains old Indo-European forms coexisting and influencing the newer Indo-European arrivals is a hypothesis that needs to be investigated. As mentioned earlier, the Daco-Romanian used approximately 2581 words, with over 30% directly from Latin, 25% formed internally, to which can be added 17% with multiple etymologies, the Slavic influence counting for about 9%, and other smaller contributions (Sala, 1988, p. 73), and a considerable percentage of about 20% with unsolved etymology. The cultural vocabulary preserved by the Daco-Romanian population, and analysis of other Indo-European language vocabularies, can contribute to a better understanding of a civilization lost in time. The Proto-Indo-European roots recognizable in the DRom reflexes presented in this work will help not only to reveal the preferences and particularities As a general consensus, here I used single quotes for the translation of words in discussion, and double quotes when citing a text. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1_27
237
238
A. R. Chelariu
of the Daco-Romanian language, but also help advance the Indo- European linguistic research. A preliminary examination of a list of words that qualify as the basic IE vocabulary according to The Cultural Lexicon of Indo-European in Europe: Quantifying Stability and Change (2018), next to the Romanian translation and the etymology for each, can give a glimpse of the development of the language. Domestic animals English
DRom
sheep
oi (pl., fem. sg. Lat ovis; IE *h2ówis, oaie) berbece Lat berbex, -ecis (=vervex)] miel Lat agnellus—PIE *meh1l- sg. m. miel, f. mială, mioară capră Lat capra cloșcă Slv Bg kločka găină Lat gallina cocoș Slv kokoši ‘hen’ cal Lat caballus; mânz IE *mendyos măgar IE *markobivol Slv byvolŭ? (Derksen Ru býdlo ‘cattle’; ORu bydlo ‘animal’; PIE *bhHu-tlom cireadă Slv OCS črěda
ram lamb goat hen chicken rooster horse (colt) donkey ox cattle (generic) bull cow pig duck dog cat bee
bou vacă porc rață câine, câne pisică albină
Etymology
Lat vulgar *bovus, -um Lat vacca Lar porcus ??? Lat canis ??? Onomatopoeic DRom pis-pis Lat alvus ‘beehive, cavity’
In this subject category, from the total of nineteen, the majority of nouns are inherited from Latin, with four of Slavic contribution, raising the question: What equivalent forms existed in the language previous to the Latin or Slavic contact? Of the four reflexes considered here of IE origin, two could be controversial, sheep—oi, oaie < Lat ovis, lamb—miel < Lat agnellus, and yet, the IE reconstructed form MA *meh1l- ‘small animal’; IEW 724 *mēlo-, smēlo- could be a better option.
27 The Daco-Romanian Cultural Vocabulary
239
Wild animals wolf bear owl cheetah leopard lion lynx fox jackal snake raven eagle bison wild boar hare rabbit deer
lup urs bufniță n/a n/a n/a râs vulpe n/a șarpe corb vulture zimbru mistreț iepure iepure căprioară
Lat lupus Lat ursus IE *b(e)u, *b(h)eu;
Slv rysı ̆. Lat vulpes Lat serpens Lat corvus Lat vŭltŭr, vúlturis IE *ĝ(h)ombhros Lat mixticius? Lat lepus, -oris Lat lepus, -oris ‘doe’ Lat capriole
Most of the words for wild animals in this list are of Latin origin. Material vocabulary scythe sickle spade plow (n.) saw ax (hatchet) (ax) yoke wagon wheel axle cultivated field weave spin (thread)
coasă seceră spadă plug
Slv kosa (controversial) Lat sicilis Italian neolog. DRom > Slv < Germ < Celt > DRom (archaeological replica found in Dacia in third century BCE)—wonderword ferăstrău ? (Lat fero?) secure Lat secures bardă IE *bhar- s ´pointed´ topor Slv toporŭ jug Lat jugum car Lat carrus roată Lat rota osie OCS osь a) arătură, Lat arare ‘plow land’: Lat campus câmp b) lan PIE MA *lendh- ‘open land’ OIr lann, Eng land? țese Lat texere toarce Lat torquere
(continued)
240
A. R. Chelariu
(continued) Material vocabulary sew plow (v.) sow wool fur
coase ara semăna lână blană
Lat cosere (consuere) Lat arare Lat seminare Lat lana Bg блана/blana f. membrane? Slv kozina ‘fur’
The basic material vocabulary is mostly of Latin origin, with a few Slavic forms. The IE form bardă could be controversial. Edible honey milk salt wine mead wax (bees) meat grease hops wheat barley rye oats grain(generic) turnip apple grape flax
miere lapte sare vin mied ceară carne grăsime hamei grâu orz secară ovăz grâne ridiche măr struguri in
Lat mel, -is Lat lac, -tis Lat sal, salis Lat vinum IE *médhu Lat cera Lat caro, carnis Lat gras Slv chmeli Lat granum Lat hordeum Lat secale Slv ovesŭ. Lat granum Lat radicula Lat malus IE *sre/ohags- ‘berry, fruit’ Lat linum
In the Edible subject list fourteen are of Latin origin, two Slavic, and two IE. From the total words in this list, considered the basic IE vocabulary, the majority of DRom reflexes are of Latin influence, with nine Slavic and eight from the substratum. The Latin loans are in majority forms designating animals, food, and verbs related to domestic work; some developments are the result of contamination or merger phenomenon between existing Dacian/Thracian forms, and the conquering language of the Roman colonists, some speakers of the Vulgar Latin. For example, the
27 The Daco-Romanian Cultural Vocabulary
241
DRom v. a vătăma,‘to hurt, injure, damage; 1st sg. eu va’tăm (first vowel accentuated) ‘I hurt you’ or, ‘someone’, explained in DEX.RO by Lat victimare ´to give offering, sacrifice, slaughtering,´ could have been a merger between the Latin form with an existing Dacian form related to the PIE (MA) root *wedh- ‘push, strike’; Pokorny IEW 1108 u̯ ā-, u̯ ō-, u̯ ǝ- ‘to hit, wound’ with cognates in Grk éthei ´destroys´; Hit wezz‘strike, urge’; Skt vadh- ‘strikes, pushes, slays’; Toch B wät- ‘fight’; Ltv vâts ‘wound’, Lith votis ‘open wound’, all indicating an archaic reflex, a possible example of a merger. The formative process in the history of Daco-Romanian language is shown more conclusively by the coexistence of archaic forms for horse: DRom mânz ‘colt’, murg ‘dark colt’ and iapă ‘mare’, next to DRom cal < Lat caballus < Celtic. Similarly, the subject of daily ‘activities’ has many correspondences in IE, while the domesticated animals are almost all inherited from Latin. The coexistance between the inherited vocabulary and the Latin one, is also exemplified by the many daily produce for use in house or for trade, brânză, urdă ‘cheese’, zer/zară ‘whey’, of PIE heritage, next to lapte ‘milk’ of Latin origin. The vocabulary for ‘cultivated field’ is of Latin influence, while the plow (n.)—DRom plug, has the character of a wonderword. Many of the religious terms, loan words through Christian influence, are also of Latin origin: church—DRom biserică < basilica; pray/prayer > DRom ruga/ rugăcine, priest > DRom preot, etc.; in some cases the merger hypothesis is yet another way to regard the Latin contribution to the Daco-Romanian vocabulary. The Christian Greek Orthodox preachings, coming from the South of the Danube with the Greek and Bulgarian priests, form the main corpus of the Slavic influence on the Romanian religious vocabulary. The attempt to clarify the many unexplained and regional reflexes in use in Daco-Romanian language is partial, mostly due to the author’s limitations. Besides the approximately 360 words from the substratum listed in the Appendix, for which a PIE root was found, the DRom language uses quite a few ‘everyday’ isoglosses for which there is no solution as yet, many of them in common with Albanian language: băiat n.‘boy’, copil n. ‘child’, moș m. moașă f. ‘old man, midwife, old woman’; măgură n.´small hill´, Alb achaic magulë ‘hill’; Sardinian mogoro ‘hill’, Basque
242
A. R. Chelariu
mokor ‘lump of (dry) earth’? mărar n.´deel´ Alb maraj ‘fennel’; mazăre ‘pea,’ Dacian form found in Dioscoride’s list mozula, mizela ‘the plant thyme’; țap n. ´he-goat´, Alb sqap, tsap, Dalm tsap, Abruzz tsappe; pururea adv. ‘ever’; răbda v. ‘bear, endure’; șopârlă n. ´lizard´ Alb shapi ‘id’; șoric n. ‘pig skin’; spânz, spînț n. ´plant Helleborus pupurescens’; strepede, pl. strepezi n. ´larva in cheese or vinegar´; viezure ´badger´ Alb vjedhullë ‘id’, etc. The Daco-Romanian language includes particularities that are the result of the substrate influence such as: the sound ă (schwa ə); the laringeal h; the rotacism of -n- > -r- ; the change of Lat ct > pt. (Albanian ft); the postposition of the article (similar to Bulgarian and Albanian); the neuter gender; the identity between dative and genitive cases; the particle -ne attached to personal pronouns mine, tine, sine ´me, you, self´, and the interogative cine ´who´; • the future tense with Lat volere as in eu voi merge ‘I will go’; • the sufix -esc for apartenance românesc ‘Romanian’, to list the most important ones. • • • • • • • •
The phonological developments from Proto-Indo-Euroipean (PIE) to Daco-Romanian language (DRom), as presented in the Istoria limbii române (ILR) published in 1969, and by I. I. Russu (IIR) in his study Etnogeneza românilor published in 1981, instrumental in the analysis of unexplained Romanian forms, are listed here, to which I added examples from the present study: PIE a (ə) > a in DRom (ILR, 1969, p. 316); á > a; a > ə in Thraco-Dacian (IIR: 123); here: *ap-1 (proper ǝp-) ēp- ‘to take, grab, reach’ DRom v. apuca ‘to grab’ 50–51; *areg- ‘to lock’ DRom n. argea ‘underground room’ PIE a, ā > a (ə) o (ILR: 316, IIR: 123); here: *balba-, *balbal-, *barbarDRom n. bală, pl. bale ‘slobber, drool’, v. bolborosi ‘mumble’;
27 The Daco-Romanian Cultural Vocabulary
243
PIE e > e (ILR: 316, IIR: 123); here: *se(n)k- ‘dry up’ > DRom adj. sec ‘dry’ PIE e > schwa ǝ (Georgiev, Vraciu) DRom ă > â; here *kerdheha > DRom n. cârd ‘flock, herd’, the schwa ǝ sound in DRom is generally accepted as inherited from the substratum (ILR 320); PIE é > ie (ILR: 316) Dacian toponyms Tierna, Dierna, ie/ia *dheu(hx) ´being stirred (like dust or smoke) > DRom adia v. ‘to breeze, soft wind’; n. adiere ‘breeze’; *mel-6, melǝ- ‘dark color, black, dirty’ DRom adj. mieriu ‘blue’; PIE ē > a > o (ILR: 316, IIR: 123); here: *melh3 -, *mel-, melǝ-, mlö- ‘to rise up,rising land’—DRom n. mal ‘small cliff, beach’ PIE o > a (ILR: 316); (IIR: 123); here: *ghórdhos or *ghórtos ‘fence, enclosure’—DRom n. gard ‘fence’; PIE ō > ö > e (ILR: 316); o > o; ō > a (IIR: 123); here: *h3or- ‘eagle’— DRom n. erete, harete ‘hawk’ PIE ō > u; here: *bhólĝhis ‘skin, belly’—DRom n. burtă ‘belly’ PIE ū > ü > i (ILR: 316); u > u(o) (IIR: 123); here: *bulis ‘rump’—DRom n. buric ‘belly button’; *dhúbhos- DRom n. dop ‘cork’ PIE ai̯ > a (ILR: 316); here—possibly *haek̑ —‘point, sharp’; a derivative *haek̑ sti-. ‘bristle’; DRom v. ascuți, pers. I (eu) ascut ‘I sharpen’ PIE ei̯ > ẹ (ILR: 316); here: (ha)ei > i / â: *haeis- DRom v. isca ‘appear’; *rei-, *reik- ‘scratch’—DRom v. râcâi ‘scratch’, n. râie ‘scabia’ PIE aṵ > a (ILR: 316); here: no examples; only aṵ > au *auo-s (*ḫuḫḫaš) ‘grandfather’ DRom arch. n. auş ‘grandfather, old man’; PIE eu > e (ILR: 316); here: eu > ău (ǝu): *keus- ‘hollow out’—DRom n. căuc, căuş ‘cavity as in cup, spoon, middle of hand’; *reus- ‘anger’— DRom adj. rău ‘bad’; *ĝhēu-, *ĝhō(u)-, *ĝhəu- ´yawn, gap´—DRom n. gaură ‘hole’ eu > o, u *bhreu- ‘brew’- DRom n. burtă ‘belly’, borţ ‘belly’, borhot ‘fermented fruit’. eu > i, iu *leubh- ‘love, desire’—DRom v. iubi, n. iubire ‘love’, *leuk‘shine’ DRom n. licăr, v. licări ‘glitter’. PIE ṇ > a (ILR: 316); here: *ṇbh(ro/ri) ‘rain’—DRom n. abure ‘mist’, v. aburí; *andher-, *ņdher- ‘stem, spike’ DRom n. andrea ‘knitting needle’ PIE ṛ > ri (ILR: 316); here: ṛ > re: *k̑ ṛrēh2 ‘head’—DRom n. creier ‘brain’
244
A. R. Chelariu
PIE bh > b (ILR: 316); here: *bhag- ‘divide, apportion’—DRom eu bag ‘I insert’; *bhreu- ‘brew’—DRom borhot ‘marc’ bh > p *dhúbhos- ‘wedge’—DRom n. dop ‘cork’ (later development?); PIE d h > d (ILR: 316); here: *dhúbhos- ‘wedge’—DRom n. dop ‘cork’; in some cases dh > d > ġ 1. *udero- ‘uterus/womb’, 2. *h1óuhxdhŗ ‘breast, udder’—DRom n. uger ARom udzire, IstrRom uger ‘cow udder’ d h > d > t: *dhren- ‘drone’—DRom trăncăni ‘speak endlessly’; *dergh- ‘grasp’—DRom targă ‘stretcher’ PIE gh > g (ILR: 316); here: *ghórdhos or *ghórtos ‘fence, enclosure’DRom n. gard ‘fence’; *dergh- ‘grasp’—DRom targă ‘stretcher’; PIE k̑ > s (ILR: 316); k̑ > s (IIR: 124); here: * k̑ súlom ‘post, stake’— DRom sul ‘role, tube’; *haek̑ ‘sharp’—DRom as-cut ‘I sharpen’; *k̑ ouh1ros ‘powerful’—DRom în-surat ‘married man’;
But also, PIE k̑ > c *k̑ ostrom/dhrom ‘knife’—DRom custure ‘knife’; *k̑ er- ‘grow’— DRom cârlan ‘young lamb or foal’; *k̑ ṛrēh2 ‘head’—DRom n. creier ‘brain’ PIE k̑ > h *k̑ euhx- ‘hollow out’—DRom n. hău ‘abyss, precipice’;
These examples: k̑ > s, but also k̑ > c, or k̑ > h, presents Daco-Romanian with both centum and satem reflexes, offering the possibility to analyze linguistic strata in different historic stages, taken from linguistic comparison when s > h changes in Greek, Phrygian, Armenian, Iranian, but not Illyrian. PIE k > k (c, g) (IIR: 124); here: *kerk- ‘hen’—DRom n. fem. Curcă ‘turkey’; *krob- ‘hurry’—DRom v. grăbi, n. grabă ‘rush, hurry’; PIE ĝ, ĝh > z (d)̇ (ILR: 316; IIR: 124); here: *ĝ(h)ombhros ‘bison’— DRom zimbru ‘bison’ [Common Illyr gh- > d- then Lat. d- > f- phonetic mutation (IEW)] PIE kw > k (p) (ILR: 316); kw > k (IIR: 124); here: *kwekwlóm, *kwokwlos ‘wheel’
27 The Daco-Romanian Cultural Vocabulary
245
< *kwel ‘turn’—DRom n. cocǎ ‘dough’, cocoaşǎ ‘hump’n. cocoloș ‘ball of dough or other material’; v. cocoli, cocoloşi ‘to overprotect mostly of children’; kw > p *h1ek̑ wos ‘horse’, fem. *h1ék̑ weha—DRom n. fem. Iapă ‘mare’, Dacian personal names: Βετεσπιος, and Ουτασπιος (in Katičić & Mate, 1976, p. 149). PIE gw, gwh > g (b?) (ILR: 316); gw, gwh > g (IIR: 124); here: *gworhx ‘mountain’—DRom n. gorun ‘oak’; *gwr(e)ha(−u); *gwrehx-u- ‘heavy’— DRom greu, fem. grea ‘heavy’; PIE n > n; m > m; l > l (intervocalic > r) li > I; p > p; weak n intervovalic > r (ILR: 316–322; IIR: 124) PIE s > s / ș (IIR: 124) *h1ēs- ‘sit’ > DRom v (se reflexive particle) aşeza ‘sit oneself, set; something’; *septm̥ > DRom șapte ‘seven’; *k̑ os-trom/dhrom ‘knife, dagger’ > DRom n. custure ‘knife’ PIE sr > str (ILR: 316): *sre/ohags- ‘berry, fruit’; DRom n. sg. strugure ‘grapes’. A special situation is offered by the development of consonantal Lat kt, ks > DRom pt, ps and ft, fs in Alb, considered specific to the Balkan region; the Albabian development ft, fs is not accepted in DRom.
Noteworthy, the s- mobil with an unclear function, which can be present or absent (Beekes, 1995, p. 163), could be observed in a few Romanian reflexes: *(s)kel- ‘crooked’—DRom n. cârcă ‘(uper) back’, cârjă ‘croch’, cârlig ‘hook’, cârlionţi ‘ringlets’, cârnat ‘sausage’, v. cârni ‘turn’ *(s)keng- ´limp´—DRom n. ciung, ciump ‘crippled’, ciunt ´unihorned´ v. ciunti ´cut short´, ciot ‘stmp’ *(s)kēp-2, (s)kōp- and (s)kāp-; (s)kēb(h)-, skob(h)- and skāb(h)- ‘work with a sharp instrument’—DRom n. scoabă ‘clamp’, v. scobi ‘to hollow, scoop’ *(s)kerb-~(s)kerbh- ‘shrink, shrivel’—DRom v. scoroji ‘dry out, as of leather or skin’, *sphaen ‘flat-shaped piece of wood’—DRom n. pană ‘shim’, *spleiĝh- ‘step, go’—DRom v. plec(a) ‘leave’,
246
A. R. Chelariu
*(s)teg ‘pole, post’—DRom n. stejar ‘oak’; stinghie ‘wooden beam’; steag ‘flag’; steajer, steajăr ‘pole’; stînjen ‘measurement, 1,366 meters’.
In a more comprehensive comparative analysis these examples could help determine various stages in the historical development of the Daco- Romanian language. As part of the Balkan Sprachbund, the Romanian language is characterized by the existence of the ‘schwa’ vowel noted as ă in DRom, as ӗ in Albanian, and ъ in Bulgarian. Romanian language knows a proclitic particle a (also a mark of the ARomanian dialect) that in time received various explanations. Mallory and Adams (2006) argue that “the augment particle h1e added at the beginning of a root indicated the past tense, therefore an association with the imperfect and aorist, as in Skt á-bharam, Grk é-pheren, Arm e-ber as reflexes of *h1e-bher-on ‘I carried’” (MA: 65). Hamp (1985, p. 70) explained this particle as the prefix ad- ‘to’, which he derives as a special syntactic use from *ad- ‘conformity, goal’. Other linguists consider this particle/prefix a non-Indo-European mark (Kroonen, 2012, p. 240) The past tense nuance is not visible in DRom, thus this particle may be either “the augment particle h1e” (Mallory & Adams, 2006), or related to the prefix ad- (Hamp, 1985), or it may be of a non-IE origin; here are some DRom examples: adia v. ‘to breeze, soft wind’ *[h1e]dheu(hx) ´be stirred like dust or smoke´; or, dialectal Greek ἅϝησι (áwēsi) = “it blows”, from IE h2-w-h1 agonisi v. ‘gather wealth’ *[h1e]gwhonós ‘fullness, thickness’; amăgi v. ‘allure, trick’ *[h1e] h2mey-gw- ‘change, exchange’ ameți v. ‘be dizzy’ *[h1e]médhu ‘mead’ ardica (archaic) v. ‘to lift up, raise’ see ridica *[h1e]h3er- ‘set in motion vertically’ ascuţi v.’to sharpen’, *[h1e]haek̑ ‘sharp, pointed’. (se) asemăna v. ‘to look the same’ *[h1e]somos ‘same’ abure n. ‘mist’ if we take the IEW 162 *bholo- ‘smoke steam’ > a+bure ‘mist’.
27 The Daco-Romanian Cultural Vocabulary
247
Traditionally, Romanian linguists considered the DRom particle a as a development from the Lat preposition ad, as in v. mesteca ´chew’—amesteca ‘mix´; yet, in the examples listed above the Latin preposition ad- (a-) could not be considered an etymological solution since there are no Latin words such as (ad)+dia, (ad)+gonisi, to justify the formation, nor are there DRom words dia, gonisi, etc., as in regular Latin developments: Lat adducere >DRom aduce ‘to bring’, Lat adiutare > DRom ajutor ‘help’, etc. This very productive particle marks the infinitive in DRom (omitted in this dictionary to avoid redundancy): a avea ‘to have’, a fi ‘to be’, a mânca ‘to eat’, a pleca ‘to leave’, etc.; it also forms nouns: a-casă ‘(at) home’ [casă ‘house’], adverbs: adeseori ‘often’ [des ‘often’], etc. By itself the particle a indicates affiliation ‘of ’ as in: miroase a mâncare ‘smells of food’, tată a doi copii ‘father of two kids’, etc. The DRom and ARom particle a could be the development of Lat ad- in clearly Latin loans; it could also be preexisting the Latin influence, related to the PIE *haed with the sense of ‘direction, of nearness, toward, or belonging’ > DRom a-, but Lat ad, NE at, Oscan az ‘at’. Gaulish ad ‘at’, Welsh add ‘id’, Gothic at ‘at, next to’, OHG az ‘at, next to’, even though it should have developed in az as in Oscan. PIE short o before the consonantal allophones of the PIE resonants became in open syllables in Ind-Ir ā, as in Grk. δόρυ, Skt. dā́ru, but was dropped in DRom druete, codru. The change of Lat b+u into DRom gw established by the Romanian linguists based on a few examples, such as negură ‘fog’ < Lat nebula from IEW 315–316 *(enebh-2): nebh-, embh-, mbh- ‘wet, damp, water, clouds’, should be reconsidered on account of the DRom form murg, Alb murg ‘colt’ < PIE *(ha)merhxgw ‘dark’ that developed in Greek in amorbós ‘dark’; and also on account of the Alb mjegull/njegull ‘fog, mist darkness’. This change is stated in IEW (106) “Common Lat. kʷ > p phonetic mutation corresponds to common Gk. gʷ > b phonetic mutation”, and specified in Fortson: “At an earlier date, the voiced labiovelar *gw became b in Celtic” (2004); these conclusions advance the possibility that the DRom language reflex negu-ră shows the retention of a more archaic form. The other DRom forms, fagur(e) ‘honeycomb’ < Lat favus (“etymology unknown” de Vaan, 2008, p. 207), and DRom n rug with a
248
A. R. Chelariu
double meaning: ‘bramble, shrub of blackberries’ < Lat rubus ‘blackberry perennial, blackberry’, but also n rug ‘stake’ < Lat rogus (possible neologism), should also be analyzed through the PIE reconstructed roots. The phonetic modifications are even more complicated with the reversed development Lat gw > DRom b, as in Latin lingua > DRom limba ‘tongue’, considered a general rule even though this is the only example, and without further reference to DRom murg, or negură. Other examples such as k̑ > s, but also k̑ > c, or k̑ > h, may stand proof for certain tendencies at some historical moments [what used to be called centum or satem], that could help identify chronological changes during various linguistic strata in the Daco-Romanian language. The laryngeal [−h], well attested in DRom, was considered by some linguists of Slavic influence (based on OCS examples as har ‘bliss’, hrană ‘food’) since it disappeared from Latin long before the Dacian wars. Others counteragrued signaling the existence of the laryngeal in some Dacian reflexes, which may plead for an older origin: chodela plant name, hydronyms Auha, Helivacia, Hierasus (Lat Gerasus), Hister (Histros), toponyms Histria (also Istria), Helis, Lat Carsium today Hârsova, antroponym Heptapor, granted that the graphic reproduction may be problematic. Common reflexes with Albanian, hameș, hutui, lehăi, for example, or other forms that could not be explained through Slavic, or other languages, as habă (see below), hârșie ‘black sheep fur’, horoi ‘woodpecker’, hoț ‘thief ’, hud(r)ă ‘hole’ (ILR 321) attest to a Romanian laryngeal. Here are some examples of possible PIE laryngeal h developments in DRom: PIE h1 + é > h as in: *h1édmi ‘eat’ > DRom n. m. hameș, f. -ă ‘gluttonous, big eater’, v. part. Hămesit ‘starving, hungry’, verified by the Alb form ha ´eat´; *h2/3webh ´weave´—DRom habă ‘gathering of women to spin and sew’. *hxolu-, or *alu- ‘spell’—DRom hală, (bală, balaur) ală ‘monster creator of storm’. *h2ṛg̑ (u)- ‘white’—DRom hârcă ‘skull’.
27 The Daco-Romanian Cultural Vocabulary
249
The DRom can also offer the development of a laryngeal from a velar as in *kap- ‘have, seize’—DRom hapcă/japca ‘fishing rod, catch by force’; interj. Hap ‘catch with mouth’ hăpăi ‘gulp dwon’; and, *ghrebh- ´dig´— DRom n. hrubă (?) ‘underground room, gallery, cellar’. Among the lexical derivative particles inherited from the substrate are the following suffixes: -esc forming adjectives românesc ‘of romanian’, frățesc ‘brotherly’, ciobănesc ‘of shepherds’; also adverbial particle -esc, -eşte as in românește, frățește, ciobanește; -uş sufix forming diminutives sfedeluș, jucăus, bebeluș, etc. -ză associative sufix, also forming diminutives (Rosetti, 1968) as in pupăză ‘hoopoe’, coacăză ‘currant’, căcărează ‘pejorative—little piece of caca’, spetează ‘back of a chair, from spate ‘back’, gălbează ‘flatworm’ The Slavic influence on the Daco-Romanian language is undisputable, particularly during the Christian influx of monks from the South following the Turkish conquering of the area. Many of the Bible translations from the fourteenth to the sixteenth centuries used Greek and South- Slavic sources; the monks and priests performing the translations inserted Slavic words, mostly because they were not familiar with the Romanian language. However, in addressing this influence, the history of the Balkan languages and their common substratum should be taken into consideration. The separation of Slavic languages into East and South East is first attested in manuscripts from the ninth to the tenth centuries. This separation is confirmed, among other phonological phenomena, by the metathesis of the liquids l / r in the so-called open syllable law; for example, the PIE *ghórdhos or *ghórtos ‘fence, enclosure’ > OCS gradŭ ‘city’, but DRom gard ‘fence’ and Alb gardh ‘fence’, where the metathesis phenomenon is absent, pleading to the conclusion that the DRom and Albanian forms were present in the common substrate; other examples could be: PIE *del- ‘cut’, OCS dlato, dlěto ‘chisel’ and DRom daltă ‘chisel’; PIE *tolkw- ‘speak’ > OCS tlŭ-, DRom n. tâlc ‘meaningful talk, sense’, see further. In these cases, the metathesis did not take place in Daco-Romanian, either because these loans entered the language by contact with Slavs before the open syllable law occurred, in other words,
250
A. R. Chelariu
before the ninth century, a fact historically not accepted, or, most likely, these reflexes were already present in DRom, and a later regressing metathesis to a previous form, from OCS dlato, dlěto ‘chisel’ to DRom daltă, is less plausible. The same applies to DRom baltă ‘puddle, swamp’ and Blg blato ‘swamp’. To bring this argument further, here are a few examples of DRom reflexes, with correspondences in OCS, but without metatheses: • DRom tâlc ‘meaningful, wise speech’ < OCS tlŭkŭ < Proto-Slavic tъlkъ < Proto-Indo-European (IEW 1088) *tolkʷ- ‘to speak’ the number of cognates in I-E could indicate a merger rather than an earlier loan from a Slavic source prior to the ninth century: Latin loquor ‘to speak’, Old Irish do-tluchethar ‘to ask’, ad-tluchedar ‘to thank’; in the Old East Slavic the phenomenon did not occur: тълкъ (tŭlkŭ), Belarusian толк (tolk); Russian толк (tolk) ‘interpretation, explanation’; Ukrainian толк (tolk); Danish tolk; Dutch tolk; Estonian tulk; Latvian tulks; Lithuanian tùlkas; Middle Low German; tolk; Norwegian tolk; Old Norse tulkr; Swedish tolk. • DRom n. vârf ‘peak’ < OCS vrŭhŭ ‘top’ < PIE *wers- ‘peak’, IEW 1151–1152 u̯ er- (*su̯ er-) ‘highland, high place, top, high’, with cognates OIr ferr ‘better’[ Alb fshat > DRom *fsat > sat ‘village’, even though the Romanian villages are not surrounded by fossa ‘ditches.’ In a Byzantine document from the sixth century, Strategikon (2001), we find fosaton, Greek ϕοςςατον, used regularly, meaning ‘military compound’, ϕοςςατον being used overall to designate the army itself in the Byzantine military; moreover, we should consider that by the fourth century, the Roman administration and the army were no longer present in Dacia. Pokorny (IEW 626) argues that Albanian fshat, developed from *k̑þei- ‘to settle’ through an old Alb *kṣati > fshati > DRom *fsat, following a common development in Alb-DRom kwhs > phs > fs; this rule was applicable in the case of Lat coxa > DRom coapsă, and Abl kofsha ‘thigh’. The phonetic group ps-, fs- in initial position is not accepted in Romanian language. The etymology for DRom sat < fsat < Alb fshat ‘village’ is based on the occurrence of this form in only one document, the manuscript named Psaltirea Scheiană (dated approximately 1576–1578), a translation in Romanian of a religious text using the Kirilic alphabet, perhaps, that could have been translated by an Albanian monk, or a monk with limited knowledge of Romanian language. Incidentally, the use of DRom sat in this Psalter coincided with the time when Coresi’s Tetraevanghel was printed, between 1560–1561, in which the DRom sat ‘village’ is frequently used, and the form fsat is totally absent. The same can be said about the Slavo-Romanian Psalter printed by Deacon Coresi
252
A. R. Chelariu
in 1577 in which only the form sat ‘village’ is used in the entire work, proving that fsat, Albanian fshat was used only in a single publication dated from the same period with Coresi’s documents, and should not be considered as a possible solution for the DRom sat. More so, in T. Cunia’s Dictsiunar a Limbljei Armãneascã, the author specifies that the word: “fsat (fsátŭ) sn—scriari neaprucheatã tu-aestu dictsiunar; vedz fusati— ‘Sriere neacceptată in acest dictionar, vezi fusati’ (fsat (fsátŭ) [unacceptable writing in this dictionary, see fusati.] Further: “fusati/fusate (fu-sá-ti) sf fusãts (fu-săt́ sı ̆)—hãndachi (adancitura) strimtã shi lungã … {ro: tranşee, şanţ} ‘fusati/fusate (fu-sá-ti) f. n. fusãts” [long narrow ditch, Romanian trench, ditch.] http://www.dixionline.net/index.php?inputWord=f%C3%A3nt%C3 %A3nel%C3%A3 Another argument against the foston origin of the DRom sat is the fact that the Aromanians, living in close proximity with the Albanians, use for ‘village’ the form hoară, a Greek loan. The PIE offers some possible solution in MA *sed- ‘sit’; IEW 884–889 *sed ‘to sit’, *sed-to ‘sit, settlement, chair’; sed-ter ‘seat, settlement?’; sedti ‘gathering’ > DRom n. sat, pl. sate ‘village, settlement’ (e > a in DRom); cogn.: NWels sedd ‘seat’; OIr saidid, said ‘to sit’, and OIr sid ‘peace’: Grk hédos ‘seat’; Av hadiš ‘home’; Skt sádas ‘place’, sad-; sādayati ‘to sit/place down’; satti, sadas ‘seat, place to stay’; sattá- ‘sitted’; ppp. Satta’to ‘sit down, settle down’; sat ‘existence’? Av pasuš-hasta- ‘hurdle’ (*settlement); Av had- ‘I sit’; Lat sedeo, ere, sedi ‘to sit’; OIsl set ‘raised on ground’; Etruscan sath-, śat- ‘to put, establish’ (B-B 2002:218). The Albanian fshat has cognates in OInd kṣḗti, kṣiyáti ‘stays, dwells’, Av šaēiti ds; OInd kṣití-, Av šiti- ‘residence, settlement’, OInd kṣḗtra-, Av šōiϑra- n. ‘estate, residence’, OInd kṣēma- m. ‘quiet, peaceful staying’; Maybe Alb (*kṣati) fshati ‘village’. Note: H. W. Vallis, in The Cosmology of the RIGVEDA. 1887, discusses in an interesting aspect regarding this word. “Lastly we come to the expressions asat and sat, the ‘ non-existent’ and the ‘existent’. The word asat is used in the Rigveda in two senses, as an adjective with vacas ‘speech’ and as the converse of sat. The philosophic comment of Suyana on verse X. 129. 1, is disproved by the expression ‘sato bdndhum asati nir avindan’ in verse 4. If we treat the hymn philosophically, we must assume a stage
27 The Daco-Romanian Cultural Vocabulary
253
between those states described in verses 1 and 4 in which asat was present, but there was as yet no sat. The context, however, shows that the poet merely wished to shadow forth a condition in which absolutely nothing existed; and the presence of asat is denied because it was inseparably associated with ‘sat’”. Besides the cognates with Albanian or Slavic languages, the Daco- Romanian has a few reflexes in common with the Hungarian language, resulting from the Maghiars that settled in the tenth century in the proximity of the former Dacian territory. For some of these reflexes considered loans from Hungarian language I used the A Magyar Nyelv Törteneti-Etimologiai Szotarat-Etimologiai Szotara, 1970 (The Historical-Etymological Dictionary of the Hungarian Language), with special attention to forms listed in this dictionary as of unknown etymology in Hungarian; these solutions are not without controversy, and are open for further investigation. The Romanian language uses certain diacritics for its specific sounds, which include: ă, a ‘schwa’ as already mentioned, corresponding to the sound ə as in ‘hurt’; â used only in the middle of a word, and î only at the beginning of a word, corresponding to a sound similar to French ‘en’; ș stands for the sound sh; ț for the sound tz; gh is the sound found in ‘gift’, ch indicates the sound in ‘key’, and ce/ci stands for the sound č as in ‘change’. Verbal forms listed here are given in the infinitive without the mark a as Engl ‘to’; there are four conjugations, as in the following examples: -a—ex. a pleca ‘to leave’; -ea—ex. a tăcea ‘to keep quiet’; -e—ex. a merge ‘to walk’; -i—ex. a privi ‘to look’; the conjugations, -a, -ea -e, -i, are of Latin influence, The following list of Proto-Indo-European roots and their Daco- Romanian correspondences are arranged by subject, hoping to offer a better understanding of the cultural environment of the Dacian people. The appendix includes about 360 DRom words of basic vocabulary from substratum, for which there is no satisfactory etymological solution, either in the conquering language or other influences, offering a contribution to future investigations into a cultural complex research less available in the English language.
254
A. R. Chelariu
For the present Daco-Romanian unexplained forms I used Pokorny’s Indogermanisches Etymologisches Wōrterbuch (IEW) Pokorny, 1959, (Indo-European Language Association—http://dnghu.org/), Robert S. P. Beekes Comparative Indo-European Linguistics 1995, Mallory & Adams (MA) The Oxford Introduction to Proto-Indo-European and Proto- Indo-European World, 2006, and other available sources. Among the Romanian linguists used are I. I. Russu, whose studies addressed the Daco-Romanian language relation to the Indo-European stock, Ariton Vraciu Limba daco-geților (1980), Grigore Brâncuși Vocabularul autohton al limbii române (1983), Alexandru Rosetti Istoria limbii române (1968), and a more comprehensive work, the Istoria limbii române (1969) under the coordination of I. Coteanu. The following list attempts to cover mostly words from the main vocabulary, part of a large number of Romanian words that could not be explained through the usual sources, Latin, Slavic, Germanic, Hungarian, etc., and are listed in the Romanian sources as of an unclear (unknown) etymology, such as the Dictionarul Explicativ al Limbii Române (DEX.RO, https://dexonline.ro/). In an effort to clarify these isoglosses, considered as part of the Romanian substratum, I proposed the Proto-Indo-European (PIE) reconstructed root that could offer a better solution. I did not venture into reconstructing Dacian roots, a language of which we have very few documents. Due to the limited nature of this list I decided not to include the solutions found in previous Romanian studies, most of which do not include any reference to Pokorny’s (IEW) dictionary, or other IE sources. I. I, Russu, Limba traco-dacilor, 1967, lists of Dacian lexicon was used when it was relevant to the current discussion. The arrangement of the PIE roots to DRom reflexes follows the subject order corresponding, more or less to Mallory–Adams’ dictionary, hoping to offer a glimpse on the material and spiritual preferences of the population from Dacian territories, and thus learn more about this little known culture. Each entry includes the Mallory–Adams and Pokorny reconstructed form, the Romanian reflexes, n (noun), v (verb), etc., with the English translation, followed by the cognates taken from MA, IEW, other current sources, and in Romanian documents when applicable. Each subject area is preceded by a short overview. Occasionally, the Daco- Romanian form could be explained by more than one PIE reconstructed root, leaving room for further research.
27 The Daco-Romanian Cultural Vocabulary
255
Alphabetical lists of the DRom words and the PIE roots in connection to the DRom ones help a quick reference.
References A Magyar Nyelv Törteneti-Etimologiai Szotarat-Etimologiai Szotara. (1967 [1970]). Eds. Lóránd Benkõ, László Papp, & László Kubínyi. Budapest. Publisher Akadémiai Kiadó. Beekes, R. S. P. (1995). Comparative Indo-European Linguistics. Baltimore. John Benjamis Publishing Co. Brâncuși, Grigore. (1983). Vocabularul autohton al limbii române. Bucharest. Editura Științifică și Enciclopedică. The Cultural Lexicon of Indo-European in Europe: Quantifying Stability and Change Talking Neolithic. (2013 [2018]). Proceedings of the Workshop on Indo-European Origins Held at the Max Planck Institute for Evolutionary Anthropology, Leipzig. Edited by Guus Kroonen, James P. Mallory and Bernard Comrie. Journal of Indo-European Studies Monograph No.65. Dictionarul Explicativ al Limbii Române (DEX.RO, https://dexonline.ro/). Derksen, Rick. (2008). Etymological Dict of the Slavic Inherited Lexicon, Leiden Indo-European Etymological Dictionary Series, ed. by A. Lubotsky, vol. 4, Leiden-Boston, Brill. de Vaan, Michiel. (2008). Etymological Dictionary of Latin and the Other Italic Languages. Liedern, Boston. Brill. Fortson, Benjamin W., IV. (2004). Indo-European Language and Culture: An Introduction. Oxford: Blackwell Publishing. Hamp, E. P. (1985). Notes on Indo-European Dialects: Indogermanische Forschungen (1985), vol. 90, no. 1985, pp. 70–71. https://doi.org/10.1515/ 9783110243321.70 Istoria limbii române, vol. II, redactor responsabil Ion Coteanu. (1969). Bucharest. Editura Academiei Române. Katičić, R., & Mate, K. (1976). Ancient Languages of the Balkans. Hague, Paris. Mouton & Co. N.V., Publishers. Mallory, J. P., & Adams, D. Q. (2006). The Oxford Introduction to Proto-Indo- European and the Proto-Indo-European World. Leiden: Oxford University Press. Pokorny, J. (1959). Indogermanisches etymologiscches worterbuch. Bern Munich A. Francke. Verlag. Rosetti, A. (1968). Istoria limbii române. Bucharest. Editura pentru litertură.
256
A. R. Chelariu
Russu, I. I. (1967). Limba traco-dacilor. Bucuresti. Editura Ştiinti̧ficǎ și enciclopedică. Russu, I. I. (1981). Etnogeneza românilor. Bucharest. Editura stiințifică și enciclopedică. Sala, M. (coord.), Mihaela Bârlădeanu, Maria Iliescu, Liliana Macarie, Ioana Nichita et al. (1988). Vocabularul reprezentativ al limbilor romanice. Philadelphia. Bucharest Științifică și Enciclopedică. Strategikon. (2001). Maurice’s Strategikon: Handbook of Byzantine Military Strategy by Flavius Maurice (G. T. Dennis, Trans.). University of Pennsylvania Press. Vraciu, A. (1980). Limba daco-geților. Bucharest. Facla.
Appendix
F rom Proto-Indo-European to Daco-Romanian: A Short List Mountain, Forest, Valley, Ground PIE MA *bherg̑h ‘high, hill’; IEW 140-141 *bhereĝh- high; mountain DRom mountain river name Bârzava (Dacian Bersobis, Bersovia), and hydronym and toponyms Bârsa, also mountain region Țara Bârsei (the Country of Bârsa); anthroponyms Bârsan, etc. Cogn: MIr brī ‘hill’; NE barrow; NHG Berg ‘mountain’; Rus béreg ‘riverbank’; Av bərəz- hill’; Lith place-name Beržorai; PN the Mediterranean countries: Thrac. Βεργούλη, Maced. Βέργα, Illyr.Berginium (Bruttium: Bergae), Lig. Bergomum, Celto-Lig. Bergusia, Hisp. Bergantia (The mountain name Bârgău could be a Germanic borrowing.) The Bulgarian бърз and the Serbian brz ‘rapid’ helped maintain these nominations, offering a good example of merger.
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 A. R. Chelariu, Romanian Folklore and its Archaic Heritage, https://doi.org/10.1007/978-3-031-04051-1
257
258 Appendix
PIE IEW109-110 *bhares- bhores- bhrsti-, bhorsti- ‘point, stubble’; DRom adj. bârzoi, v. bârzoia ‘of hair, spiking, sticking up on the head, or an animal tail’ Cogn.: Skt bhrstí-h ‘prong, spike, cusp, peak, edge, point’, Gmc *bursti- ; OIce burst ‘bristle, ridge of the roof ’; OE byrst ‘bristle’, OHG burst, borst bursta ‘bristle’; MHG burste ‘bristle, brush’; Slav *bъrstio- in Russ. borščь ‘acanthus’; PIE IEW 938-947 (s)ker-4, (s)kerǝ-, (s)krē- ‘to cut’; DRom Munții Carpați ‘Carpathian Mountains’; Thrac Καρπάτης ὄρος ‘Carpathian Mountains,’ the tribe Carpi Cogn.: Alb. karpë, karmë (*korp-n-) ‘rock, cliff’, Gk. κρώπιον ‘sickle’; Lat. carpō, -ere ‘pluck’, carpinus ‘hornbeam’; Hit karpina- ‘a tree hornbeam’ PIE IEW 477-78 *gworhx ‘mountain, forest’; DRom n. gorun (Quercus pedunculata) ‘oak’; n. grui, gruiu, arch. gruni ‘hillock’; toponym/ athroponym Gruiu; Cogn.: Lith. gìrė, girià ‘wood, forest’; Ltv. dziŕē ds., and O.Pruss. garian n (acc. garrin) ‘tree’; further Ltv. garš ‘wood, forest’, gãršas ‘swamp, marsh”; OCS gora “mountain”, Ser.-Cr. gòra ‘mountain, wood, forest’, Alb gur ‘rock’; Av gairi- ‘mountain’; Skt giri- ‘mountain’; uniformly ‘forest’ in Baltic languages, Lith giriá ‘forest’; Or, IEW 472-73 *gwelha- ‘acorn’; *gʷel-3, *gʷel-, *gʷlā- ‘acorn; oak’; DRom n. gorun (Quercus pedunculata) ‘oak’ (IIR, 1981) Cogn.: Lat glans, Lith gilé; ‘acorn’; OPruss gile ‘acorn, oak’; Rus želudı ̆, Grk bálanos, Arm kalin, Skt gula- ‘acorn, oak’ DEX.RO Bulgarian & Serbian горун, possibly Balkan substratum PIE MA *(s)teg ‘pole, post’; IEW 1014 (s)teg-2 ‘pole, stick, beam’ DRom n. steajer, steajăr ‘pole’, n. stejar ‘oak’; n. stinghie ‘wooden beam’; n. steag ‘flag’; n. stînjen ‘measurement—1.366 meters’ Cogn.: Lat tignum ‘wooden beam’; NE stake; Alb shtëngezë ‘pole, support’ Blazek (2019) Goth. *stingils ‘Stengel’; cf. WGmc.: OEng. stengil, OHG. stengil, stingil ‘Stengel’, dim. from NWGmc. *stangō f. ‘Stange’ >
Appendix
259
ON. stDng, OSax. stanga, OHG. stanga (EWD 793, 787) Rom.: Rum. stinghe “dünne Stange” (ML 8261 a). PIE; IEW 1015 *(s)tei- ‘sharp, spike’; DRom n. stei ‘stone, sharp mountain rock’ Cogn.: Av staēra- “mountaintop, mountain peak, summit’; ἅπαξ λεγ., taēra‘mountaintop, mountain peak, summit, acme, apex’ (probably *stoi-lo-) Lat. stilus ‘pole, stem, stalk, stylus for writing; with formant smo: Lat. sti-mu-lus ‘sting, prick; in addition Lat. stīva ‘plough handle, plough stilt’ as *stei-uü? compare OInd tīvra- ‘violent, sharp’ PIE IEW 626-627 *k̑ū- ‘sharp; pike’; DRom n. ciucă ‘mountain peak’ Cogn.: OInd śū-la- m. n. ‘spit, pike, pole; pain’, Arm. slak (from *sulak) ‘spit, pike’; Bulg čuka ‘small hill’; Latv čuk-ur-s ‘summit of roof ’; Alb çuka ‘crest’, suka ‘hillock’; Greek τσούϰα ‘tumulus’; OIr cuil ‘culex’, Welsh cylion-en, Lat. culex ‘mosquito’; Av sū-kü- f. ‘needle’, Pers sōzan ‘needle’ PIE IEW 1014-1015 *stegh-, nasal. *stengh- ‘stick; pole, stalk, etc.’; DRom n. stâncă ‘rock’ Cogn.: nasalized: Gk. στόνυξ, -υχος (formed after ὄνυχ- ‘sharp cusp, peak’) PIE IEW 683 *lēu-2 : lǝu- ‘stone’; DRom n. lespede ‘large flat stone, slab, gravestone’ Cogn.: Grk Hom. λᾶας, gen. λᾶος ‘stone’ originally *ληFας; λάFα[σ]ος, Att. λᾶας and λᾶς m., gen. λᾱου; Alb lerë, -a ‘rock, rockfall’; OIr līe, newer līa, gen. līac (disyllabic) ‘stone’ (Celt. *līuank- from IE *lēuank- or -ǝnk-) PIE MA *dhólhaos ‘valley’; IEW 245-246 *dhel-, dholo- ‘curve; hollow’; DRom n. dolie ‘slow river, bed of a slow river; junction between two roofs’; v. doli ‘go around [with raft], streaming slowly’; in toponyms Doljiu [Dol ‘valley’ + river Jiu], Dolhasca, Dolheşti, Dolheni villages.
260 Appendix
Cogn.: NWels dôl ‘valley, mountain’; NE dale; Rus dol ‘valley, under side’; Grk thólos ‘vault, sort of upside-down valley’; OCS—Proto-Slavic *dolъ ’below, valley’ (Derksen, 2008). ON døll m. ‘valley inhabitant’ (*dōlja-), dial. døl ‘small valley, long gully resembling dent (*dōljō) = OHG tuolla, MHG tũele ‘small valley, dent’, doel ‘ditch, trench’; NGer dole ‘small pit, pothole’ (OHG dola ‘gully, ditch, trench, channel, duct, tube, pipe’ probably actually NGer dölle ‘short duct, tube, pipe’; OCS dolъ ‘hole, pit, pothole, valley’, dolu ‘downwards’, dole ‘under’. This can be a good example of contamination. PIE MA *lónko/eha ‘valley’; DRom luncă ‘meadow, river meadow, everglade’ Cogn.: Lith lankà ‘valley, river-meadow’; OCS lǫka ‘bay, swamp, ruse ‘; Ukr lukà ‘flood plain’; Ru. luká ‘pommel, bend, (dial.) flood plain’; Toch B leṅke ‘valley’; late Lat (Gaulish?) lanca. DEX.RO Slv lǫnka (Derksen *lǫkà f. ‘low-lying medow, water-meadow’), if accepted, perhaps an early merger since the Ukrainian and Russian forms lack the nasal aspect. PIE MA *gloiwos ‘clay’; DRom n. glod, ‘clay, mud, clod of earth’; v. glodi ‘to rub, irritate’; adj. gloduros ‘muddy, bumpy’; also, toponyms Glodu—village in Muntenia; Cogn.: Grk gloiós ‘clay’; Lat gluten ‘glue’; NE clod. PIE MA *melh3 -, melǝ-, mlö- ‘to rise up, rising land’; IEW 721-722 *mel-8, melǝ-, mlō- ‘to appear, come up’; DRom n. mal ‘small cliff; shore’; archaic Thrac-Gaet toponyms Malva, Malvensis; deriv. n. maldăr, pl. maldăre ‘pile, heap, moumd’; Cogn.: Alb mal’ ‘mountain’ (Holm: JIES vol 39, no. 1–2, 2011, p. 72); Celt malo- ‘rising, prominent’ (Falileyev et al., 2010), cf. MJr mell ‘clump’; Ltv. mala f.‚ edge, bank, border, shore, region’; Lith. lūg-mala ‚height of edge’; (MacBain, 1982: Gaelic mol , mal ‘shingly beach’; from Norse möl, g. malar ‘pebbles, bed of pebbles on the beach’; root mel, grind.)
Appendix
261
And MA *mḷdho/eha ‘clay’; possibly DRom hydronym and toponym Moldova, old Moldava (suffix Slv –ova) archaic: Dacia Malvensis. Cogn.: OE molde, molda ‘sand, dust, soil’; Goth mulda, OIce mold, OHG molta ‘dust, powder, earth’ (*ml-tü); Grk málthē ‘modeling mixture’. PIE MA *dhembh ‘to burry’, *dh3mbhos ‘mound, grave, swelling, tombstone’ (Watkins); IEW 248-249 *dhembh-, dhm̥bh- ‘to dig’; DRom dâmb n. ‘mound, knoll’ Cogn.: Grk τυμβοσ ‘mound’ (Chantraine); *tumbo-, hillock > tom a hillock, Irish tom, Middle Irish tomm, Welsh tom (MacBain); NE tomb; Skr tunga, high, hight; Lat tumulus; Arm damban ‘grave’; Av daxma‘grave’; without clear relation: Slv dǫbъ ‘oak, tree’, Ru. dub ‘oak’, Bulg. dăb ‘oak’ (Derksen, 2008) DEX.RO Hun domb ‘hill’ (< Slv dąbŭ, dǫbъ ‘oak, tree’?); most likely < DRom PIE IEW 249 *dhen-2 ‘surface of hand/land, etc (*dry land) meaning ‘arid flat area’; Dacian dina ‘place, region, plane’ PN Amlaidina, Asbolodina; possibly DRom toponym Padina, Bulg Padina < Thracian? Cogn.: OInd dhánus- n., dhánvan- m. n. ‘dry land, mainland, beach, desert’, dhánu-, dhanū- f. ‘sandbank, seashore, island”; VLat. danea ‘area’, OHG tenni n., MHG tenne. Germ Tenne ‘barn floor, threshing floor, flattened loam ground or wooden floor as a threshing place, hallway, ground, place, surface generally’, Dutch denne ‘area, pavement of tiles, brick, stone, floored’; MLG denne ‘lowland, depression’ (and ‘valley forest’; MDu denne ‘den of wild animals’, dan ‘waste, from shrubbery surrounded place, place generally; valley forest’, OE denn ‘cave, wild den’, Eng den ‘cave, pit, pothole’, Fris dann(e) ‘bed, garden bed, garden plot’.
Fire PIE MA *dhōgwho- ‘summer heat’, (*dhogʷh-lo-lā) *dhegwh- ‘burn, be hot’; IEW 240-241 *dhegʷh- ‘to burn, *day; DRom n. dogoare ‘blaze, heat, summer heat’, v. dogori ‘to emanate heat, blaze’
262 Appendix
Cogn.: OIr doghaim ‘burn’, daig ‘flame’; Lat foveō ‘heat, cherish’; Lith degù ‘burn’; dāgas ‘summer heat’; OCS žegǫ ‘burn’; Alb djeg ‘burn’, ndez ‘kindle’; Grk tephra ‘ash’; Av dažati ‘burns’; Skt dagdhá-h (= Lith. dègtas), Kaus. dăhayati; dăha-h ‘blaze, heat’, niduăgha-h ‘heat, summer’, Pers. dăɣ ‘burn brand’; Av. daxša- m. ‘blaze” dáhati ‘burns’; Toch tsäk- ‘burn’, DEX.RO Slv goreti ‘burn’ ( DRom dogoare ‘blaze, heat, summer heat’, which presupposes a local development *do+goreti, difficult to accept, esp. with the following: Cz dehna ‘devil”, ablaut. dahněti “burn”; Russ dëgotь ‘tar’; and *degǫ to *gegǫ: OCS žego, žešti “burn”, maybe Alb zheg ‘summer heat’. PIE MA *u̯ep- ‘throw, throw out’, IEW 1149-1150 *uēp- : uǝp- (*suekʷ-) ‘to blow; to soar; DRom n. văpaie ‘flame, shooting flame’ Cogn.: Skt vàpati ‘throw out, scatter’, vàpra ‘earthen wall’, vaprā ‘fireplace’; Av vafra, OIran vafr, Iran bafr ‘snowdrift?’, OInd causative vāpayati ‘makes blow’; Lat vapor (old vapōs) ‘vapor, heat’; Alb vapë ‘summer heat’. PIE MA *peh2ur ‘fire’; DRom n. pară ‘intense heat from fire, blaze’ Cogn. : OPrus panno, Czech pýř ‘ashes’; Grk pûr, Arm hur, Hit paḫḫur, Toch B puwar; Umbrian pir—all ‘fire’; Blg < OCS para ‘steam, vapor’ < DRom para; DEX.RO Slv para ‘steam, field steam, evaporation, soul’. PIE MA 1. *g(e)ulo- ‘fire, glowing coal’; 2. *gwher- ‘warm’, *gh’ermós adj. ‘hot’; DRom n. jar ‘glowing coal, ember’, Dacian Γέρμαζα (Germaza), Γερμἰζερα (Germizera) < *gwhermós Cogn.: OIr gūal ‘coal’; NE coal; Alb zjarr|i ‘fire’, Alb zjarm, Arm ǰerm ǰer; Latv. gařme, Gr. θερμὀς, ‘fire, heat’; OCS goreti ‘burn’, DEX.RO OCS zaru ‘glow, heat’ (< *gwher- Kortlandt, Dreksen). PIE MA *kr-em- ‘burn’; IEW 571 *k(e)r-em-); DRom n. scrum (s-crum) ‘singe, cinder, ash’. Cogn.: Lat cremō ‘burn’. Lat cremō, -üre ‘burn’, Umbr. Krematra, pl *crematra ‘kind of vessel for frying meat’ (Orel, 1998: shkrumb, pl shkrumba ‘black ashes’; PAlb iš-krum)
Appendix
263
PIE MA *h2ehx-tr-eha ‘hearth’; IEW 69 *āt(e)r- ‘fire, *blow the fire’; DRom n. vatră ‘hearth’ Cogn.: Lat ātrium ‘open space for smoke, hall’; Slavic [ < from Rom and Alb] vatrë ‘hearth’; Alb vatër / votër [f ] ‘hearth, fireplace’ PAlb. *a̅t(e)ra̅ {1} Albanian adds an initial ‘v’ to words of ancient origin that begin with a vowel (common Alb. prothetic v- before initial bare vowels—proof of ancient laryngeal ḫ) as in vesh from Indo-European, verbër from Latin orbus,; Byzantine Grk βᾰ́θρον ‘base, pedestal, scaffold’ PIE IEW 69 *āt(e)r- ‘fire, blow the fire’; Cogn.: Lat. a̅ter ’black, sombre (blackened by fire)’; a̅trium ’main hall of a house, containing the fire’ Av. a̅tarš ’fire’; IEW 69Note: “Alb Tosk vatra, Gheg votër ‘hearth’ proves that Slavs borrowed prothetic v- before bare initial vowels from Illyr. This phonetic mutation in Alb took place before the invasion of Slavs into the Balkans because Alb and Rom share the same cognate.
Water and Hydronyms PIE MA *h2eP- ‘living water’; DRom n. apă ‘water’, Dacian river name Apos Cogn.: Skt āp- ‘water’; OIr ab ‘river’; OPrus ape; Hit hāpa- ‘river’; Toch AB āp ‘water, river’, Lat amnis ‘river’; OHG river names in -affa-. DEX.RO Lat aqua< *haekweha- ‘water’> DRom apă, IE *h2eP- seems a better solution. Note Sard abba ‘water’. PIE MA 1.*h1wers-, 2. *we/ohxr ‘rain, water on the move’; DRom v. vărsa ‘pour’, n. vărsare ‘flow, spill’ Cogn.: Luv war(sa) ‘rain’, Av var, Skt var(i) ON ur ‘fine rain’; Lat. verrēs ‘rain’; Skt varsá- m ‘rain’, várșati ‘rains’; Hit warsa- ‘rainfall’ Grk. eérsē ‘dew’; OPrus wurs ‘pool’; Indo-Iranian *hwáršati; Indo-Aryan: *hwársati ‘rain’; Sanskrit: वषित (vársati) DEX.RO Lat versare; (de Vaan 666: vo/ersare 'to keep turning round, move to and fro, be busy,’ from ve/orto, -ere ‘to turn')
264 Appendix
PIE MA 1. *nbh(ro/ri) ‘rain’; IEW 315-316 (enebh-2): nebh-, embh-, mbh- ‘wet, damp; water; clouds;’ DRom n. abure ‘mist’, v. aburí; adj. aburos ‘misty’; ARom abur, aburare, buredzu ‘mist’; v. bura ‘to mist’(IEW 162 bholo- ‘smoke, steam?); DRom hydronyms: Ampoi, Împoiţa < Dacian Ampee; Proto-Celtic *ambe- ‘river’. (PIE: *h2ep-h3on- ‘river’ (JEW: 1)? Matasovič, 2009) Cogn.: Lat imber ‘shower’; Skt abhrá- ‘rain-cloud’; Grk? ómbros ‘rain’; Toch B epprer ‘sky’; Alb avull, old: avullë, Geg abull ‘mist’; Also Alb borë ‘(fallen) snow’(IEW 128-132); Grk αϕρός ‘sea foam, spray, mist’; Arm amp ‘cloud’, genitive ampoy; Av aβra ‘rain cloud’. *abra- YAv. aβra-; OPers. *abra- (also in compound PNs like *Abra-dāta- and *Abrakāma-); Skt ámbhas- n. ‘rainwater’; ambu n. ‘water’, abhrá- ‘gloomy weather, cloudiness’; Gk. ὄμβρος m. ‘rain’; Ambía (Ambia, 949 CE): a tributary to the Arnoia. Pokorny page 865: rriver names Gaul *Ambrü, M.Welsh Amir, Amyr as well as Ger. Amper and Ammer (Celt. *Ambrü), Emmer (Celt. *Ambriü); in addition also Eng. Amber; Fr. Ambre, Sp Ambrole; O.N. Ambron, Ambror; Ital. Ambra, Ambria, Ambro, Ambrio etc., latter are particular Ven.-Illyrian; compare without formant r Gaul. inter ambes ‘between streams’, ambe ‘a small stream, brook’. In AE PIE *h2euh1-Vl-, root * h2uh1- ‘vapour, steam’ _+_Pok. 82;—> IEW 81-84 Cogn.: ON ūr “fine rain”, yra ‘to rain subtly’, ūrigr ‘dewcovered’, root form auer- in Thrac. FlN Αὔρας, Grk (Persson IF. 35, 199) *αὔρα ‘water, spring’ PIE IEW 162 *bholo- ‘smoke steam’; DRom n. boare ‘soft breeze’, v. bura ‘to mist’, burează ‘it sprinkles (rains lightly)’. Cogn.: OIr bolad, Ir boladh, baladh ‘smell, odor’; Ltv buls, bula ‘misty muggy air, height smoke, dryness’; Arm bal ‘fog, mist, darkness’. Also Alb borë ‘(fallen) snow’; mythological traces: Boreas Thracian wind god. PIE MA 1. *rốs ‘dew’; IEW 336-337 ere-s-2 (ers-, rs-, eres-) ‘to flow’, and rēs-, rōs-; DRom n. f. rouă ? ‘dew’< *ro-o-a with the feminine mark?; devel. v. înroura ‘to get dewy’
Appendix
265
Cogn.: Lat rốs, rōris; Lith rasà; Rus rosá ‘dew’; Alb; resh ‘it rains’; Skt rása ‘sap, juice’ 2. *h1res- or h1ers- (*h1ros per John Colarusso) ‘dew’. Cogn.: Lat rōs, rōris, Lith rasà, OCS rosa; Alb resh ‘rains’; Skt rása ‘liquid, moisture’. PIE IEW 809-810 *per-1, perǝ- : prē-, preu- ‘to drizzle, sprinkle, jet’; DRom n. pârâu, părău, pârău n. ‘brook, creek’, perhaps compound *per + Lat rivus > DRom râu ‘river;’ possibly related to hydronym Prut. Cogn.: Alb përrúa PAlb *perron, per-ren (Jokl in Russu 372) ‘id’; Grk πρηστήρ ‘lightning, whirlwind, impetuous stream’; OCS para ‘smoke, vapor’, (OPruss pore ‘vapor’ from Pol. para); Hit parüi- ‘blow’; or, (Russu 372) IEW 326-332 *er-3, or-, r- ‘to move, stir, animate, fight, struggle, rise; to spring up’, […] *re-n- (maybe grown from a present *re-neu-mi, *re-nu-o) one also seeks in Alb. përrua ‘riverbed, brook bed’ (për-rēn-, lengthened grade), prrua ‘spring’ (*prër-rua ‘effluence’) Goth rinnan, rann ‘rush, run’ (*re-nu-ō), urrinnan ‘rise, from the sun’, OIce rinna ‘flow, run’, OHG, OS rinnan ‘flow, swim, run’; OE ryne m. ‘run, flow, river’, OIce run n. ‘rivulet, brook’, etc. PIE IEW 765 ner ‘under’; DRom v. nărui ‘colapse of a wall, cliff, and such, ruine’ Dacian ? RN Nǎruja. Cogn.: Lith. nérti ‘to dive’, ‘swim underwater’; Alb. hum-nerë ‘precipice, chasm’; Lith RN Neris, Narùpis; Illyr RN Νάρον PIE
MA *drewentih2 ‘river names’; DRom [toponym near Danube] Drobeta Cognate river names: Gaul Druentia; India Dravanī; related to DRom n. drui, druete ‘thick piece of wood’ PIE MA *sreu- ‘to flow’; DRom hyronym Siret, Seret, river name in Moldova, arch. Tiarantos, Gerasos, Gerasus, Seretos; Σέρετου in Dalmatia, Σερετος in Dardania. Cogn.: Thrc strumā , strumōn ‘current, river’; OHG stroum, German Strom ‘current’ river’, Lith sraumuõ, -eñs ‘fast current’, srūti (srūvù, dial.
266 Appendix
srūnù) ‘to fill with water’ and ‘to flow, to outflow the banks (for a river)’; *strūna ‘current, river’; Lith sriti ‘to fill with water, to outflow’ (Duridanov http://www.kroraina.com/thrac_lang/) IEW 909-910 *ser-1 (*ker- < *kel-) ‘to flow’ Cogn.: Old Indic sísarti, sárati `flows, hurries’; Thrc Germi-sara ‘warm water’ PIE IEW 175 *dā- ‘to flow; river’; DRom hydronym Dunăre Danubius River Cogn.: OInd dā-na- n. ‘liquid flowing from the temples of the elephant for the rutting’, dānu- n. f. ‘every dripping liquid, drop, dew’; Av dā-nuf. ‘river, stream’; Osset. don ‘water, river’; Russ. Don, (Greek) Skyth, Illyr (*don-don) Dodona Epirus; Russ Dniepr and Dniestr, old Danapris and Danastius from Skyth. *Dānu apara ‘back River’ and *Dānu nazdya‘front river’; Ir *dānu- YAv dānu-, Alan *δānu-, Oss don ‘river, water’ (Handbook, 2017); Celt Dānuvius ‘Danube river’ PIE MA *h4ṛg̑-es- adj. ‘white’; DRom hydronym Argeș; in antiquity Ordessos ‘Ορδησσός, [Herodotus] old Arghiș, Argyas. Cogn.: Grk argė́s ‘white’; Hit ḫar-ki-iš ‘white’ ̑ ‘dark color, PIE MA *kwṛsnós ‘black’; IEW 573-574 *ker-6 and kerdirt’> IEW 583 *kers- ‘kind of color/black’; DRom hydronym Criş-ul—(in antiquity Κρισος, Crisia, Grisia, Gresia) rivulets Crişul Negru, Alb, Pietros Repede ‘Black, White, Stony, Rapid Criș’; derive. toponym Crișana; Cerna hydronym (in antique documents: Dierna, Tsierna in Latin or Greek transcription where the sound č was absent); see DRom v. cerni ‘to darken’; and n. cioară ‘crow’ in the subchapter Colors. Cogn.: OPrus krisnan ‘black’; OCS črŭnŭ ‘black’; OBg črъnъ, Russ čërenъ; Skt kṛsná ‘black’; Alb sorrë ‘crow’; Arm saṘn, gen. saṙin, saṙnum ‘ice’, Ven.-Illyr PN Carmō (Austrian province), Rätorom ON carmún ‘weasel’; see under kor-men-; Alb thjer-më ‘gray’, per-thjerm ‘lazuline’ (*ker-uo- with secondary -më); i surme ‘ashen’ (*kor-mo-); s. Jokl Mel. Pedersen 153 ff.; Lith širvas ‘gray, greyish-blue’ (*kr-uo-s), širmas ds
Appendix
267
(*kr-mo-s), Ltv. sirms ‘gray’; Proto-Slav *sernъ: ORuss serenyj ‘white’ (of horses), Russ. serën ‘frozen snow’ PIE IEW 748 *mori, mōri ‘sea’; DRom Mureș River Cogn.: : Lat mare, -is n (i-stem; gen. pl. marum secondary; with unclear a); OIr muir gen. mora, Welsh Corn Bret mor ‘sea’, Gaul Mor-inī, Aremori-cī VN “Meeranwohner”, morimarusa ‘mare mortum’; PN Moridūnum ‘Murten’; Goth marei ‘sea’, OS meri, OHG marī, merī ; Goth. mari-saiws ‘sea”, OIce marr, ‘sea’; OE mere ‘sea, sea, pond, pool”, OS mersc, MNGer mersch, marsch ‘marsh’ PIE IEW 152-153 *bheug-3, bheugh- ‘to bow’; *bhug(h)-to- ‘pliable’; DRom Buzău river that twists, as in the name of the area Întorsura Buzăului ‘The Turn of Buzău’; mentioned in a Greek document from 376 as Mουσεος, Βουσέος Cogn.: OInd bhujáti ‘bends, pushes away’, bhugná-h ‘bent, curved’, bhúja-h ‘arm’, bhujü ‘twist, arm’, bhōga-h ‘coil of a snake; ring’; Gmc. *bheugh-: Goth. biugan, OHG biogan ‘bend’, OIce participle boginn ‘bent, curved’ (OHG boug) PIE IEW 902 905 som- derives somo-s: DRom Someș River documented in Lat Samus Cogn.: OInd samá- ‘the same, even’, samám adv. and preposition ‘together’, samáyü, ‘same wise, right through it’, *sama; possible relation to Lith sąmyšis ‘stir, unpredictable, confusion’ PIE IEW1080-1085 *tēu-, tǝu-, teuǝ-, tuō-, tū- ‘to swell; crowd, folk; fat; strong’; DRom Timiș River; Dacian townTibiscum (Tibisco, Tibiscus, Tibiskon) mentioned by Ptolemy; later by Jordanes Tibisia, Timisas; this solution assumes a relation with the concept of ‘strong, powerful, swell’; IEW 1053-1054 *tā-, tǝ-; tāi-, tǝi-, tī-; [tāu-], tǝu-, tū- ‘to melt, dissipate, decay’ with b(h)-extension (compare under τῖ-φ-ος) Lat tübēs ‘gradual melting, decay, disease, etc.’, tübeō, ēre, tübēscō, ēre ‘melt’; Gk. τῖφος n.’marshy place, damp ground’; dak. FlN Tιβίσκος; about Lat.
268 Appendix
Tiberis (*Thubris, Θυβρίς < *dhubris) s. Szemerenyi Arch. Ling. 5, 3 ff.; D. Here still die urBrit. FlN *Tamü > Eng. T(h)ame, *Tamēssü > Eng. Thames, etc., *Taniü> Eng. Tain, Tean, *Tauü > Eng. Taw, Tay, in addition *Tilü > Eng. Till, Ven.Tiliaventus, Illyr. Tilurius, tirol. Ziller, etc. Other hydronyms: the Place Nname Αρίνα, (Arina) fort in Moesia Inf. > DRom Arieș River; DRom Vedea River (Dacian left side affluent of Danube) < PIE *ued- ‘water’; DRom Mara River < PIE *mar- ‘water’ (Xenopol, 1985: pp. 50–51) PIE IEW 306-307 *el-6, elǝ- : lā-; el-eu-(dh-) ‘to drive; to move, go’; DRom hydronyms Olt (Aloutas, Alutas, Alutus, Alutum) left-side affluent of Danube; Altus in Illyria; Altos (oikonym); Cogn.: Arm. eɫanim ‘I become’, elanem ‘I rise up, climb, ascend, come out, emerge’; Grk ἐλα- ; imper. koisch ἐλάτω, fut ἐλᾶντι (*ἐλαοντι), aor. ἐλάσαντες and poet ἐλάω ‘drive’; suppletive to ἄγω (see under Celt. el-), aor. ἤλασα; mostly ἐλαύνω ‘drive, travel’ (from a noun *ἐλα-υν-ος, Brugmann Grundriß II, 1, 321); with dh-extension ‘come’: aor. ἦλθον (from ἤλυθον), out of it Dor. etc. ἦνθον; perf. Hom. εἰλήλουθα, Att. ἐλήλυθα; Fut. Ion. ἐλεύσομαι; about perf. ἐλήλυμεν (*elu-), adj. προσήλυτος ‘someone who comes’, ἔπηλυς, -υδος ds., s. Schwyzer Gk. I 7042, 7697; one places still here ἰάλλω ‘send, throw, cast’ (*i-el-iō), Aor. Hom. ἴηλα, Dor. ἴηλα (Schwyzer Gk. I 648, 717) PIE IEW 798-801 pel-1, pelǝ-, plē- ‘full, to fill, to pour, swim, make flow, flutter; town (?)’; DRom hydronyms Prut Herodot Piretos (Ποράτα, Πυρετός, Βουρατ) left-side Danube affluent, with large river bed; possibly related to DRom pârâu ‘brook, creek’, Albanian përrua (creek, stream); Cogn.: Welsh llanw ‘flood’, llanw, llenwi ‘fill, flow’, MBret lano, lanv ‘flood’; Lith. trans. pilu, pìlti ‘pour, fill’, intrans. ‘flow’; Persian *parutaYAv. pauruta-, ‘large river’; Chwrsm. pwrd, Digor ford, Iron furd (Handbook, 2017, p. 571) or,
Appendix
269
PIE IEW 835-837 *pleu- ‘to run, flow; to swim’—extensin *pleud-: OIr im-lūadi ‘exagitat’, imlūad ‘agitatio’, for-lūadi ‘schwenkt’, lūaid- ‘move,erwähnen, äußern’; in addition MIr loscann ‘frog’; OIce fljōta, OE flēotan, OS fliotan, OHG fliozan ‘flow’; OE flotian ‘swim’, flota ‘ship’, floterian ‘to flutter’. PIE IEW 1053-1054 *tā-, tǝ-; tāi-, tǝi-, tī-; [tāu-], tǝu-, tū- ‘to melt, dissipate, decay’ DRom hydronyms Tisa (cf. ILR II: 359; Tisa (Tisia, Pathissus, Titza) Dacian *Tibisio? Cogn.: i- forms: OIce Þīðr (*tī-tó-s) ‘melted, thawed’, whereof Þīða ‘thawing’, Þīðenn ‘melted, thawed’; bh- forms: (compare above tübēs) Osset ćirwä, cirw “yeast” from *tibh-no-; Gk. τῖφος ‘marshy place, damp’; Σιβίσκος; Lat Tiberis (*Thubris, Θυβρίς < *dhubris) PIE MA *mesg- ‘dive’, *mozg-eye/o- ‘wash, submerge’ (Anthony, 2007); DRom n. mâzgă ‘slime, soft mud, any plant sap, sticky film deposited on any surface’, v. mâzgăli ‘smear, smudge’ Cogn.: Alb muzgë ‘swamp’; Lat mergo ‘submerge’; Av mazga ‘swamp’; OCS mezga ‘tree sap’; Bg mazga ‘sap’; Rus mjezgá ‘gruel’, Lith mazgóti ‘wash up, dip repeatedly’; Rus mezgá (dial.) ‘sap-wood, pulp, membrane, remnants of meat on the inside of a hide’; ORu. mézga ‘sap, sap-wood, resin’; mjazga ‘sap-wood, resin’; and *muzga; *muzgъ f.; m. o ‘pool, mud’ ESSJa XX 202-203 CS CS muzga f. ‘pool’; E Ru. múzga (dial.) f. ‘cavity (often filled with water), pool’; muzgá (dial.) f. ‘cavity filled with water in the summertime’; ORu. muzgъ m. ‘mud’ W Pl. muzga (dial.) f. ‘grass of superior quality, grass near water’; S SCr. mŭzga (Vuk) f. ‘stripe, trail’; Sln. múzga f. ‘tree-sap, silt, mud’; mŭzga f. ‘batter, quagmire’; Bulg. múzga f. ‘snout of a pig’ (Derksen, 2008) *mazga- ‘marrow, core, brain’ YAv. mazga-, MPers. mazg, Sogd. mɣz-, Pšt. māɣz’ə ~ māzɣ’ə PIIr. *masǰ (h )án- (Ved. majján-, with loss of aspiration regular in *žǰ h, Lubotsky, 2001, p. 39) > PIr. *mazʤan’marrow’ Khotan. mäjsā, mijsā (Handbook, 2017, p. 574)
270 Appendix
MacBain, 1982: Gaelic mèag ‘whey’, Irish meadhg, Early Irish medg, Welsh maidd (*meðjo-), Cornish maith, Old Breton meid, Gallo-Latin mesga ‘whey’, whence French mègue: *mezgâ, ‘whey’; Old Slavonic mozgu ‘succus, marrow’ (Thurneysen), to which Brugmann adds Old High German marg ‘marrow’. DEX Slv, mĕzga, Bg măzga > DRom mâzgă Contamination? PIE IEW 698 *māk-1 ‘damp, to soak’; DRom n. mocirlă ‘slush, slime’ Cogn.: Arm mōr ‘ordure, slime, mud, swamp, marsh’ (*mük-ri-); Alb makë ‘glue’; Lith makonė ‘puddle, slop’; O.Bulg. mokrъ ‘damp, humid, wet’, močo, -iti, Russ. móknutь ‘become damp’; Slov močilo ‘wetland’, máčivatь ‘damp, moisten’, moča ‘urine’, močag ‘damp, marshy place’.
Atmosphere, Light and Dark PIE MA *haég̑hṛ ‘day’; DRom azi n. ‘today’ [astă-zi ‘this-day’] Cogn.: Av azan- ‘day’; Skt ahar- ‘day’; with initial *d- in Proto-Germ *dāʒwar ‘warm time of the year’ < *dhōgwho- ‘burning’; Romansh hoz ‘today’ IEW 87 *h2eug-, *h2ug ‘day-light’ Grk αὐγή ‘day-light, splendor’, (Demiraj, 1997) Cogn.: Alb ag [m] (tg) ‘dawn, early morning; black mark round the eyes’ PAlb. *(h)aug-; Alb. agull [m] (g) ‘alba, aurora’ {1} (AE 72) ) DEX.RO Lat hac die > DRom azi PIE MA *nébhos ‘clouds, sky’, possibly an older *negwos; DRom n. negură ‘mist, fog, darkness’; devel. înegura ‘get foggy, dark’, negureală ‘fog, darkness’, neguros ‘foggy, dark’ Cogn.: Lat nebula ‘mist, fog’; OIr nem ‘heaven’; OE nifol ‘dark’; OCS nebo ‘sky’; Grk nephos ‘sky’; Skt nabhas ‘mist, cloud, sky’; Hit nepis ‘sky’; Alb mjegull/njegull ‘mist, fog, darkness’; (Fortson, 2004: gw > b, see above) -other possible solution linked to the Alb mjegull is IEW 712-713: *meigh-, also meik- ‘to glimmer, twinkle; mist’
Appendix
271
Cogn.: OInd mēgha- m. ‘cloud’, míh ‘fog, watery, precipitation’, Av maēɣa- ‘cloud’; Arm mēg ‘fog’; Grk ὀμίχλη (Att ὁμίχλη) ‘cloud, fog’); Lith. miglà ‘fog’, Dutch miggelen ‘dust rain’; OIce mistr ‘weather’ (*miχstu-); Ltv migla (= ὀμίχλη); OBulg mъgla ‘fog’, Russ. mgɫa ‘snow flurry, fog’, Cz mha ‘fog’, Russ mžit ‘dust rain, fog’. DEX.RO Lat nebula ‘mist, fog’ PIE IEW 81-84 *au(e)-10, auē(o)-, uē- ‘to blow’; MA *dheu(hx)- ‘being stirred’ (like dust or smoke); DRom v. adia (a-dia?) ‘soft blow, puff, breeze’; n. adiere ‘breeze’; ARom adil’a, adil’are ‘breeze’ Cogn.: MWelsh awyđ ‘violent gust of wind’, OCorn awit ‘air’ (*aueido-); Grk αὔρα ‘aerial breath, draft’; Grk ἄελλα, Eol. αὔελλα ‘storm’ (*ἄFελ-ιᾰ); Welsh awen ‘inspiration’, awel f. ‘wind, breath’, OCorn auhel ‘aura, heaven, breeze’, MCorn awel ‘weather’, Brit. Lw. MIr ahél (h hiatus sign), aial ‘wind, breath’; Lat suf-fiō ‘smoke’; ON dỹja ‘shake’; Goth dauns ‘dust, smoke’; Lith dujà ‘dust’; OCS dunǫ ‘blow’; Alb deh ‘intoxicate, make drunk’; Grk thúō ‘rush on’; Arm dedevim ‘shake’; Av dvžaiti ‘flutters’; Skt dhūnóti ‘shakes, moves about’, dhūli ‘dust’; Toch B tweye ‘dust’. DEX.RO Lat *ad-iliare ? De Vaan Pit. *aip- ‘fireplace’. It. cognates: Fal. efiles, efile [nom.pl. or sg.]; O. aldil [nom.sg.], aidilis [nom.pl.], Vol. aidiles [nom.pl.] borrowed from Latin.aidilis [nom.pl.] PIE IEW 81-84 *au̯(e)-10, *au̯ē(o)-, *u̯ē-; d. *u̯ē-s- ‘to blow’ DRom n. viscol ‘blizzard’ Cogn.: OInd vāsa-h, vāsaka-h ‘fragrance’, vāsayati ‘fills with fragrance’, samāsitah ‘makes stinking’; Ice. vās ‘frigid aura’, væsa ‘exhale, blow, breathe’, Dutch waas ‘white frost, ripe, smell, fragrance’, Lith vestu, vesti ‘cool off, become chill or become aerial’, vėsà ‘chill air, coolness’, vesus ‘chilly, aerial’, viesula ‘whirlwind’; perhaps related to the dialectal Greek ἅϝησι (áwēsi) ‘it blows’ PIE MA *(ha)merhxgw ‘dark’; *haemrhxgw ‘dusk’; DRom n. amurg ‘dusk’, n. murg(a) ‘evening’; n. murg ‘colt;’ ARom n. amurgu ‘dusk’
272 Appendix
Cogn.: ON myrkr ‘darkness’; NE murk; Lith márgas ‘variegated’; Alb murg ‘black’; Grk amorbós ‘dark.’ (gw > b) Possibly related to OIr marc-; OGaul marco- ‘horse’ PIE MA *leuk- ‘shine’; DRom n. licăr,( deriv. licur) sparkle; v. licări ‘glitter’, n. licărire ‘sparkle’; n. licurici ‘fire fly’ Cogn.: Lat lūceō ‘shine, kindle’; Hit lukke ‘shine, kindle’; Skt rócate ‘shines’; Toch AB luk- ‘shine’; Toch B lyuke ‘light’; OIr luchair ‘bright’; Gaul *louk- ‘bright’; Lith laũkas ‘pale’. Illyr ʌευκαρος—f (Vinereanu, 2008) PIE MA 1. *h1regw- es ‘place of darkness’; DRom n. argea ‘underground room for women to keep the loom and wool, used to store produse in the winter’; toponyms Argeaua, Argel, Argele; adj. rece ‘cold’ Cogn.: Goth riqis ‘darkness’; Toch B orkamo ‘dark’; Arm erek ‘night’; Skt rajas- ‘night’; Grk érebos ‘underworld’; Alb rragalë ‘hut’; 2. IEW 64 *areg- ‘to lock’; DRom n. argea ‘underground room’ Cogn.: OInd argala-h, argalü ‘latch, bolt’, Maced. ἄργελλα ‘bathing hut, bath hut’ > Alb ragál f. ‘cottage, hut’; Cimmer ἄργιλλα (*arg-el-iü) ‘subterranean dwelling’; OS racud, OE reced m. ‘building, house’. Strabo αργιλλας ‘underground room’ as in Macedonian αργελλα, (IIR 1981) DEX.RO Lat recēns, -entis ‘new, fresh, young, recent’; the Latin form is analized by de Vaan 2008: PIE *ken-t- ‘young, new’, *kent(i) ‘young, first’ + re- ‘again, back’, which does not send to DRom adj. rece ‘cold’.
Fauna PIE MA *mendyos ‘horse’; DRom n. mânz ‘colt’ Dacian Cogn.: Alb mëz ‘colt’. [It. manzo ‘veal’ is considered of ante-Roman era, compare with DRom mânzat ‘calf under one year old’]; Gaul mandu‘poney, small horse’; Messap Jupiter Menzana ‘name of god offered horses as sacrifice’; Thracian toponym Μηργίσκη and name of a tribe Μόργητες; “Deo Saromando (Dacia), *Saro-mandu- ? Celticité incertaine” (Delamarre, 2018)
Appendix
273
PIE MA *demha’ IEW 199 *demǝ-, domǝ-, domǝ- ‘tame, subdue’; DRom n. domol ‘calm’, v. domoli ‘subdue, calm down’; possibly, n arch daș ‘lamb’; ARom daș ‘lamb’ Cogn.: Alb dash [m] (tg) ‘ram, wether’; PAlb. *dam(e)ś- < QIE *d(o) mh2-(e)s-, cf. dem (AE 124); Lat domō, perf. domui ‘break, tame, master’, perf. domuī (from *domǝ-uai), participle domitus (reshaped after domuī and domitor from *dmütos, IE *dm-to-s), domitor ‘tamer’, NE tame; Grk dámnēmi ‘break’; Skt dāmáyati ‘subdues’; OIr damnaid ‘binds, break a horse’; Hit damaszi ‘presses, pushes’; NPers dām ‘tamed animal’; Osset domun ‘tame’; OIr. dam ‘ox’; W. dafad ‘sheep’; Bret. dañvad ‘sheep’; OIr. dam ‘ox’; W. dafad ‘sheep’; Bret. dañvad ‘sheep’ (Demiraj, 1997); NE tame; Grk dámnēmi ‘break’; Skt dāmáyati ‘subdues’; OIr damnaid ‘binds, break a horse’; Hit damaszi ‘presses, pushes’; NPers dām ‘tamed animal’; Osset domun ‘tame’. PIE MA *h1ek̑wos ‘horse’, fem. *h1ék̑weha; DRom n. fem. iapă ‘mare’; Dacian personal names: Βετεσπιος, Ουτασπιος (IE ek̑wo ‘horse’) in Katičić and Mate (1976, p. 149). Cogn.: OIr ech; Lat equus, equa; OE eoh; Grk hippos; Av aspa, aspā; Skt áśva-; Toch B yakwe, HierLuv azu(wa)- all ‘horse’; Lith ešvá ~ ašvá, ašvvienis ‘stallion’; Gaulish Epo- (in names), Epona ‘horses goddess’; [Ossetic yœfs (Iron) œfsœ (Digoron) ‘mare’ as per: J. Colarusso More Pontic…, 2003] PIE MA *h2ówis ‘sheep’; IEW 784 h3owi- (> *owi-) ‘sheep’; DRom n. pl. oi, sg. oaie ‘sheep’ Cogn.: OIr oí ‘sheep’; Lat ovis; Lith avis; OCS ovı ̆nŭ ‘sheep’; Grk óis; Skt ávi; Lyc xawa, Luv hawiš ‘sheep’ (John Colarusso personal communication). DEX.RO Lat ovis > DRom oaie, pl. oi. PIE MA *meh1l- ‘small animal’; IEW 724 *mēlo-, smēlo- ‘id’; DRom n. miel ‘lamb’, fem. mială, diminutive mioară Cogn.: OIr mil ‘small-animal’, Gaul mīlon, ī ‘(small) animal’; NDutch maal; NE small; Grk melon ‘sheep, goat’.
274 Appendix
DEX.RO Lat agnellus < agnus ‘lamb’: De Vaaan (2008): agnellus ‘young lamb’, ambiegnus ‘sacrificial animal’; PIt *agwno- [m.] ‘lamb*, dim. *agwnelo-. PIE MA *ĝ(h)ombhros ‘bison’; DRom n. zimbru ‘bison’ Cogn.: Lith stum̃bras; Suddov zumbras ‘bison, wisent’; Latv subrs; Rus zubr; Prus wis-sambris ζουμβρος—Tomaschek considers zimbru as Thracian; Derksen *zǫbrь m. ‘wisent’; E Rus zubr ‘wisent’; Gsg zúbra; WCz. zubr ‘wisent’; Slk. zubor ‘wisent’, etc.; a connection with the root * ǵombh- of PSl. **zǫbrь, Latv. zùobs ‘tooth’ cannot be ruled out, but it is possible that we are dealing with a migratory term, cf. Osset. dombaj ‘bison’; DEX.RO Slv *zǫbrь > DRom zimbru ‘bison’ PIE IEW 700 *marko- ‘horse’; DRom n. măgar? ‘donkey, ass’, common with Alb Note: only Celt. and Gmc. Cogn.: Ir marc, Welsh etc. march ‘horse’, Gaul. μάρκαν acc. sg., Marcodurum PN; OHG marah, OE mearh, OIce marr ‘horse’ (Ger. in Mar-schall, -stall), fem. OHG meriha, OE mīere, OIce merr, Ger. Mähre. Maybe Alb (*mahar) magar, gomar ‘donkey’. PIE MA *hayuhx-n̥-k’ós ‘youth’ deriv. from *haóyus ‘stength’; DRom n. junc(ă) ‘young cow’ Cogn.: Lat iuvencus ‘young (cow)’; OIr ōac ‘youth’; Skt yuvaśá ‘young’ PIE MA *bolg(w)-o- ’bulb’; DRom n. baligă, balegă ‘dung’, n. bălegar ‘manure’ Cogn.: Gr. βολβός ’onion, bulb’; βόλβιτον ’cow-dung’; Arm. boɫk ’radish’; PAlb. *balga̅; Alb balgë, bajgë ‘cow-, horse-dung’ (Demiraj, 1997); Serb balega < DRom or Alb.
Appendix
275
PIE IEW 555 *kem-1 ‘to press, squeeze’; DRom n. ham ‘rein, harness’ Cogn.: Arm. k”amel ‘to press, squeeze, wring; to filter, make flow’; ON hemja (hamda) ‘curb, restrain’, hemill ‘hindered’, hamla f. ‘strabp’; MHG hemmen and hamen “hinder’; OFris hemma ‘hinder’, German dial. ham, hamen ‘horse colar’; MLG ham, OE hamm ‘frozen land’, NGer hamme ‘fenced field’, Bal.-Slav. *kama- m. ‘clump’; Ltv. kams m. ‘clump’; Lith kãmanos ‘lederner bridle, rein’, kẽmuras; “grape”, Russ kom ‘clump’, komítь ‘clump’ PIE IEW 725 *mēms(o-), mēs- mēmso-, mē(m)s-ro- ‘flesh, muscle mass’; DRom n. mânzat’veal meat, calf under one year old’, mânzare n.‘sheep with kids’ Cogn.: PIIr. *mānsá-,*mās- Skt mümsa- n. ‘flesh’, müms-pacana- ‘cooking meat’; müs n. ‘meat’ (Ved. māmsá-, māms°, mās-); Arm mis, gen. msoy ‘flesh’; Gk. μηνιγξ ‘skin, meat skin’ (*mēsno- or *mēmsno-); μηρός ‘thigh piece’; μηροί ‘leg’, μηρα, μηρία ‘thigh bone’ (μηρός = OIr mīr, IE *mē(m)s-ro-); Alb. mish ‘meat’ (first from *minsa) PIE MA *mú(k)skos ‘ass, donkey’; DRom n. m. muşcoi, mâşcoi, f. mâșcă ‘mule’ Cogn.: Lat mūlus ‘mule’; ORus mŭskŭ ‘mule’; Grk mukhlós ‘he-ass’; Alb musk [Istoria limbii române 1969 OCS mižgŭ, miška, Bg măška; Venetian musso, Friul muss]. Pan-Balkanic? PIE IEW 482-482 *gʷou-: cattle; DRom n. godac, godinac ‘young pig, piglet’ Cogn.: Ltv. gùovs ‘cow’, dimin. guõtina; Grk βοτόν ‘cattle’, βοτάνη ‘food’, βοτήρ, βώτωρ ‘herdsman, shepherd’; Lith gaujà ‘herd’, guju, gùiti ‘drive, push’, gúotas ‘herd’; if Macedonian γοτάν gotán (‘pig’) is related to *gwou (‘cattle’), this would indicate that the labiovelars were either intact, or merged with the velars, unlike the usual Greek treatment (Attic βοῦς boûs) Macedonian γοτάν gotán (‘pig’) is related to *gwou (‘cattle’) http://www.gutenberg.us/articles/ancient_macedonian_dialect DEX.RO Slv OCS godъ ‘year’ (one year old animal?) ‘time, suitable time, holiday, year’; Derksen (2008) E Ru. god ‘year’, Gsg. Góda, W Cz. hod
276 Appendix
‘religious holiday’; hody Npl. ‘feast’; Pl. gody Npl. ‘feast’, BSl. *godos; *gŏdos, Lith guõdas ‘honour, worship, hospitality’; Latv. gùods ‘honour, banquet, wedding’ < PIE *guodh-o-; IEW 423-424 *ghedh-, ghodh- ‘to join, make a bond’ OCS godъ ‘time, right time’, godina ‘ὥρα’, godьnъ ‘compliant’, Russ. gódnyj ‘suitable’, OCS u-goditi ‘please’ PIE MA *pipp- ‘young bird’ onomatopoeic; DRom n. bibilică (Numida meleagris); also possibly DRom n. pipotă ‘chicken gizzard’; Cogn.: Slovak pipa ‘hen’; Alb bibë, Grk pipos; Skt pippaka- all ‘young bird’; perhaps a compound formation with PGerm balika/ón- ‘coot’, OSaxon beliko m. ‘coot’ (Kroonen, 2013) PIE MA *h3or- ‘eagle’; DRom n. erete, herete, hârău [hăreț in document dated 1688] ‘pigeon hawk’ Cogn.: OIr irar, Lith erelis, Slv orël, Hit haras, Grk ornis ‘bird’; Arm oro, urur ‘seagall’; Alb ort ‘vulture’; German ara, aro, aru’vulture’; NE erne ‘eagle’; Hit haran ‘eagle’ (Kloekhorst, 2008) IEW 325-326 *er-1, or- ‘eagle’ OIr irar, MIr ilar; Welsh eryr, MBre. erer ‘eagle’; Goth ara, OIce ari, ǫrn (from *arnuz), OE earn, OHG aro, aru ‘eagle’, Lith erẽlis, dial. arẽlis, OPruss arelie, Ltv ḕrglis (from ḕrdlis) ‘eagle’; Balt basic form *ereliа-, compare Lith. ẽras, ãras ‘eagle’; OBulg orъlъ (*arila-) ‘eagle’, Russ orëɫ, gen. orɫá; Hit ḫa-a-ra-aš (ḫaras). PIE MA *g̑han-s ‘goose’; DRom n. masc. gânsac (fem. gâscă) ‘gander’ Cogn.: OIr gēis; Lat ānser; Lith žąsis; Rus gusī; OHG gans ‘goose’; ON. gás, OEng. gōs, MDutch gans, MLG. gōs, gūs. Derksen (2008) *gǫsь f. i (c) ‘goose’. PIE MA *h1epop- ‘hoopoe’; DRom n. pupăză ‘hoopoe’ (-ză associative as in spetează ‘back of a chair’: from spate ‘back’) Cogn.: Lat upupa; Lith pupútis; Pol hupek; Grk épops; Arm popup (irreg., expect *hohup, John Colarusso personal communication), NPers pŭpŭ; Alb pupë, pl. pupa, deriv. pupëzë ‘hoopoe’
Appendix
277
PIE MA *b(e)u, *b(h)eu; IEW97-98 b(e)u-1, bh(e)u- : expr. sound of hiting ‘owl’; DRom n. buhǎ, bufniţă ‘owl’ onomatoepoeic Cogn.: Lat bŭbō; Blg buk; Grk búas; Arm bu ~ bueč (irreg. expect *pu ~ pueč, John Colarusso personal communication); NPers bŭm all ‘owl’; Alb (*buph) buf ‘owl’ PIE MA *k̑er(s)- ‘black, dirty; a dark color’ IEW 583; DRom n. cioară ‘crow’ Cogn: Alb sorrë ‘crow’; Lith šárka ‘crow’; Friulian çore, Calabrian ciola, Neapolitan ciàula, Sicilian ciaula and Tarantino ciola. See also sur(ă), suriu PIE MA *ker- ‘caw’ ; DRom câr ‘caw’, v. cârâi ‘croak’, onomatopoeic? Cogn.: Lat corvus; Grk kóraks ’raven’; Skt karața ‘crow’ PIE MA *ulu- ‘owl’; DRom uliu ‘hawk’; possibly Oroles ‘Dacian prince’ (Tomaschek, 1883, p. 407). Cogn.: Lat ulu(c)us; Skt úlŭka- ‘owl’; IEW 1105 *ul ‘to howl’: ulach ds., Ir. olchobhchán, ulchobhchán, ulgadán ’owl’ probably to el-, ol-. DEX.RO Hun ölyü, ülü—not in the Etymological Hungarian Dictionary in this form, but as ölyv ‘accipiter elu’ a Turkish loan; in Clauson is ü:gi: ‘owl’; also u:hi:, but the correct form is u:gi: with ügi teg usuz ‘sleepless like an owl’; ugu (‘wlth -g-’) ‘a bird like the sparrow-hawk which can see in the dark’; ‘owl’ ugi; ‘the great owl’; ügü: ügi ‘owl’. The DRom form seems better explained as a cognate of the Latin form. PIE MA *ul- ‘howl, hoot’ < *ulu- ‘owl’; IEW 1105 *ul- ‘to howl’; DRom v. ului ‘astonish’, adj. uluitor ‘astonishing’ Cogn.: Lat uluc(c)us (possibly onomatopoeic), Skt úlŭka- ‘owl’; Latin ulula screech-owl; ululāre ‘howl’ (>DRom urla ‘holler, scream’); Skt ulūlú ‘ululating’; Lith uluoti ‘shout’; Alb ulerij ‘shout’. DEX.RO Rus ulovit ‘catch’? IEW 1178 *ulek-, *ulk- (*sulek-) ‘to shine; fiery’; DRom v. ului ‘astonish’, adj. uluitor ‘astonishing’ Note: It derived from Root se- : ‘reflexive pronoun’ + lek ‘shape, apparition’.
278 Appendix
Cogn.: OInd ulkü, ulkusī ‘meteor’, perhaps to Gk. ἄFλαξ ἀβλάξ (whether λαμπρως Hes.); incredible ἠλέκτωρ ‘gleaming, sun’, ἤλεκτρον ‘mixture of gold and silver, amber’; Toch. AB lek ‘shape, apparition’. PIE MA *bherhxĝ- ‘shine, white’; IEW139-140: bherǝg-, bhrēg- ‘shine, white, ‘ash wood, ashen, birch tree, elm’, extensions to bher- ‘bright, brown’; DRom n. barză, pl. berze ‘crane’, adj. barz ‘grey, bird with mixed feathers black and white’ bardzu, bardză ‘white, spotted black and white’; DRom bârsană ‘a kind of sheep with long hair’; adj. m. bre(a)z, fem. brează, ‘animal with white spot on forhead’; antroponym Barzea, Bărzănescu, Berzea; toponym Bărzuica; see also DRom brad ‘fir tree’ Cogn.: Alb bardhë < Proto Alb bardza’white’ (Orel, 1998: bajzë, also balzë ‘coot, kind of water-fowl)’; Bulg. brěz ‘white spotted’; Nor bjørk; Slovn breza ‘name of a white spotted cow or nanny goat. IEW 158 *bhles- ‘to shine’; DRom adj. m. bre(a)z, fem. brează, ‘animal with white spot on forhead’ Cogn.: (note: up to now only in the Gmc. provable extension from bhel‘shine’) OHG blas-ros ‘horse with with a bright spot’ (with a bright spot on the forehead); Eng blaze ‘blaze, glow; white forehead spot’ PIE MA *kerdheha ‘herd, series’; IEW 579 *k̑erdho-, k̑erdhā ‘troop, line’; DRom n. cârd ‘flock, herd’; v. cârdui ‘organize’; Cogn.: Gk. κόρθυς ‘heap’; M.Welsh cordd “troop, multitude, crowd, family’; maybe Alb. kordhë ‘sword (of soldiers)’. NE herd; Lith (s) ker̃džius ‘herdsman’; OCS črěda ‘series, herd’; Skt çárdhah ‘herd’. DEX.RO Serb krd < OCS črěda > DRom cireadă ‘herd’: good example of language stratification and coexistence between an older form and the newer Slavic influence. PIE MA *lóḱs ‘salmon’; IEW 653 *lak- *laĝh- ‘to be spotted, salmon’; DRom n. lostriţă < *lo(k)s +trutta? ‘salmo trutta’ Cogn.: OE leax; OHG lahs; OIce lax; OPruss. lasasso; Lith. lašiša, lāšis; Rus losósī, all ‘salmon’; Arm losdi; Toch B laks; Alb losë ‘grayling.’
Appendix
279
PIE MA *dhĝhuhx- ‘fish’; IEW 416-417 *ĝhðū- (or ĝhiū-?) ‘fish’; DRom n. sg. juvete ‘any small fish’ Cogn.: Lith žuvis, žuvytis, Grk ikhthŭs, Arm jukn, all ‘fish’. PIE IEW 767 *nē-tr, nǝ-tr ‘snake’; DRom n. năpârcă ‘snake, viper’ Cogn.: Alb nepërkë ‘viper’; Lat natrix, -icis ‘water snake, penis’, OIr nathir, gen. nathrach ‘natrix, serpens’; Cymr neidr ‘snake’ (*natrī), pl. nadroedd; Goth nadrē, OIsl naðr m., naðra f. ‘snake’; OHG. nātara, nātra f. ‘snake’ (*nē-trā). PIE MA *mel(h1)- ‘soft’; ? DRom n. melc ‘snail’ Cogn.: Arm melk’, meɫk ‘mushy, softish, delicate, mollycoddle, slack’ (*meldu-i-). PIE IEW106 *bhabhā ‘bean berry, small round thing’; DRom n. buburuză ‘lady bug, small round insect’; see also bumb ‘button’ Cogn.: Skt bimba-h -m ‘ball’, bimbī f. ‘momordica monadelpha’ (Cucurbitacee); bimba-m ‘red fruit of the bimbi’ PIE IEW 1117 *u̯eg- *ueg- ‘to weave, bind’; DRom n. fagur(e) ‘honeycomb’ Cogn.: OInd vāgurā “catching net, yarn “; Lat vēlum ‘sail’, OIr figim ‘weave’, fige ‘the weaving’; OIr wichengarn ‘cotton for wicks’; Nor vik ‘skein or yarn’; MHG wiht ‘wick’; OE wice, wecca, wekko ‘wick’; MLG wecke ‘wick,’, MHG wicke ‘wick, lint’, wicke ‘spindle wrapped flax’, OHG wickilī(n) ‘wrap for spinning off’; OE wōcig ‘loop, noose, snare, pitfall’; ˙ IEW 1114 *u̯ebh-1 ‘to weave, plait’; *uebh-2: ‘to wander, roam, swarm’ *uobhsü ‘wasp’. *u̯okso- ‘wax’; DRom n. fagur(e) ‘honeycomb’ based on the assumption b > g in DRom with only couple of examples: fagur(e), negură Cogn.: OHG waba ‘honeycomb’; Lat favus ‘honeycomb’, Meyer-Lubke 3228. favus ‘honetcomb’ (uncertain etymology, Germ waba?), same opinion in de Vaan, 2008.
280 Appendix
Flora PIE MA *dóru, gen. dréu̯-s ‘tree’; IEW 214-217 *deru-, dōru-, dr(e)u-, drou-; dreuǝ- : drū- ‘tree’; DRom n. druete (+ suf -ete) drui, droi ‘thick cut wood’; n. codru ‘forest’, n. m. drug, fem drugă ‘thick wooden post’ < PSl. *drągъ ‘bar, pole’, see bellow; Cogn.: OIr daur ‘oak’; Grk doru ‘tree trunk, wood, spear; Hit taru ‘tree, wood; Av dauru ‘tree, tree trunk’; Skt daru ‘wood’, gen. dróș ‘wood, timber’; NE tree is a derived form as are Grk drus ‘tree, oak, OCS drŭva ‘wood’, drěvo ‘tree’; Alb dru ‘wood, tree’, kodrë ‘hills’; PAlb. *dru(n)- {1} Alb. drỹ [m] (g) ‘lock, door-bolt (of wood)’ {2} (AE 146); in Irish and Greek tents to mean oak; Ir druid ‘priest’ that would always use oak leaves in service. Matasovic (2013b) gives it as Blato-Slavic substratum without reference to drū- ‘tree’: PSl. *drągъ ‘bar, pole’ (OCS drągy ‘sticks’, Pol. drąg, Croat. drȗg ‘rail’), Lith. drañgas ‘pole (used as a lever)’, ESSJa V:129–130; OIc. drangr ‘detached pillar of rock’, OIc. drengr ‘heavy stick’; a variant with *-k- is PSl. *drąkъ ‘bar, pole’ (Russ. druk, Cz. drouk, Croat. druk), ESSJa V:130–131. The alternation of voiced and voiceless root-final stops might point to the substratum origin (thus Derksen 121) “*droga f. ‘bar, pole’ E Ru. drogá ‘wooden bar or metal strip uniting the front and the rear axis of a cart, centre pole’; drogá (dial.) ‘pole’; dróga (dial.) ‘cart for transporting wood’”. For the DRom codru ‘[big] forest’ I offer two solutions: co associative’+dru; or, *kaito- ‘forest’ + dru (Falileyev et al., 2010 *dru ‘oak’), ‘forest + (oak) trees’?—cogn. OWelsh coit, Welsh coed ‘wood, forest’, OCorn cuit, MCorn coys, cos ds, Bret coet, coat ‘wood, forest, spinney’ (IEW 521); Blažek, 2018: *dru ‘oak’. PIE MA *sokwós ‘sap, resin’; DRom n. soc ‘small tree with leaves used for tea, and the roots, bark and black fruits for coloring’; Latin name— Sambucus nigra. Cogn.: Lith sakai pl. ‘resin’; Russ sok ‘juice, sap, sapwood’ Grk opos ‘sap, resin’ Dreksen: OCS sokomъ, Blg sok, SCr sok, Ukr sik all ‘juice’, which may indicate a borrowing- contamination with the DRom form, even though
Appendix
281
the Slavic form does not relate to the tree; in DRom there is the later borrowing ‘suc’ from It, or Fr. PIE IEW 139-140: *bherǝg-, bhrēg- ‘shine; white, *ash wood, ashen, birch tree, elm’ MA *bherhxĝos ‘birch’; DRom n. berc ‘small forest’, bercar ‘wild honey gatherer’; Cogn.: Northwest Germanic *berkō > OIce bjǫrk (*bherǝgü) ‘birch’, OE beorc(e); *berkiō > OE bierce, OSaxon birka, berka, OHG bircha ‘birch’ (Blažek, 2018) DEX.RO Hun berc > DRom; but this form, Hun berc ‘salary, wage’, has no etymology in A Magyar Nyelv Törteneti-Etimologiai SzotaratEtimologiai Szotara. PIE IEW 109-110: *bhrozdh- < *bhares- : bhores- ‘point, stubble’, formant bhrezdh-, bhrozdh-; Vasmer I:109 *bhers- ‘point’; DRom brad ‘fir-tree’ Dacian PN Bersovia Cogn.: Alb. bredh (< earlier *berdh) breth, bredhi ‘spruce, fir-tree’ PAlb. *braδ- (AE 107) (common Alb. -ĝ- > -dh- phonetic mutation); OHG burst ‘bristle’; OIr brot ‘sting, prick’; OCorn bros, Bret broud; OHG brart ‘edge, border, stem, stem bar, stem post’, OIce barr- ‘needle, conifer’; Swe dial. bradd; Lith. béržas, Latv. bērzs ‘birch’; Rus berëza ‘birch’; Oss bärz ‘birch’; Skt bŭrjá ‘birch’; also Lithuanian place-names Beržorai; Lat farnus/fraxinus ‘ash’; *tilium > DRom tei, yet the other cognates plead for a PIE etymon, candidate for contamination? PIE MA *keh2po- ‘land, garden’, 1. IEW 529 kāp, kǝp ‘piece of land’; DRom n. copac, pl. copaci ‘forest tree, fruitless tree’ Cogn.: Alb kop(ë)sht ‘garden, orchard’, PAlb. *ka̅p-(śt-) Grk (Dor.) κα̃πος ‘garden’ Note: Enlargement with the suffix -(ë)sht(ë) (Albanian inherited lexicon) PIE MA 1. *pérkwus ‘oak’; DRom [Moş] Crăciun, (arch. Cerciun—pronounced Kerciun) ‘Old (Father) Christmas/Old Nick, Christmas; ARom Cărciunu, MRom Cărciun; also found as a surname.
Appendix
283
Cogn.: Gaulish érkos ‘oak-forest’; Lat quercus ‘oak [Quercus robur]’; ON fjor ‘tree’; [Celto-Latin hyrcania—per John Colarusso personal communication]; Irish ceirt (quiert) ‘apple tree’, (Proto-Celtic perkunia, later ercunia), Welsh perth, ‘bush, hedge’ (assuming a Proto-Celtic form kwerkwti), Italian form quercia, ‘oak’. 2. *kwrésnos ‘tree, brush (wood)’ Cogn.: OIr crann ‘tree’; Grk prînos ‘holm-oak’ (Quercus ilex) For more on quercus ‘oak-tree’ see De Vaan (2008, p. 506) PIE MA *dhergh- ‘sloetree, blackthorn’; DRom n. zârnă, zărnă, dzărnă ‘poisonous plant, euphemism for a black sheep, Solanum nigrum’, v. zârni ‘to darken’ Cogn.: OIr draigen ‘sloetree’; OHG dirn-baum ‘cornel cherry’; Rus derën ‘cornel cherry’ Dacian προ–διαρνα Veratrum nigrum; toponyms Dierna, Tierna, Zerna, today Cerna; Today toponyms Zărna, Zărnești, etc. (Istoria limbii române, 1969) PIE MA 1. *sre/ohags- ‘berry, fruit’; DRom n. sg. strugure, pl struguri ‘grapes’ [PIE sr- > DRom str-] Cogn.: Lat frāga ‘strawberries’; Grk hrốks ~ hrā́ks ‘berry, grape’; Alb shtrud, shtrydh ‘strawberry’ 2. *ster-, sterə-, strē-, steru, streu ‘to stretch, spread’ Rădulescu (1975, p. 389) reconstructs *(s)trugh for Dacian, Albanian and Greek, relating it with German Traube ‘grapes.’ PIE IEW 749 *moro- ‘blackberry’ and 910-911 ser-3, sor- ‘red’; DRom n smeur, smeuriș ‘the raspberry bush’, smeură, zmeură ‘raspberry’ Cogn.: OIr smer, Welsh mwyar ‘blackberry’ (Hamp, 1967, p. 523); smeur, smiar ‘ brambleberry’, Irish smeur, Early Irish smér, Welsh mwyaren, Breton mouar (pl.): smeuraich grope; from meur (Alexander McBain) The DRom smeur(ă) can be the result of PIE *ser ‘red’ and the PIE *moro ‘blackberry’, other examples: formants -bho- Lat sorbum ‘red berry of rowan tree’, sorbus ‘this tree’, and Lat mōrum ‘blackberry’ > DRom mură ‘blackberry’
284 Appendix
PIE MA *wrehagh- ‘thorn’; DRom n. vrej, vreg ‘stalk, as of vine stalk’ Cogn: MIr fraig ‘needle’; Lith rāžas ‘dry stalk’; Grk hrākhós ‘thornhedge’; hrákhis ‘spine, backbone’; PSl. *versъ, *verskъ ‘heather’ (Russ. véresk, Cz. vřes, Croat. vrı ̏jes, Lith vìržis, Latv virzis, vìrsis; (Derksen 516); the Slavic form points to *werk’-; the Baltic forms to *wrk’- or *wrg’-. T. Pronk (p. c.) thinks that Baltic -ž- may be due to influence from Lith. veržti ‘string’, ‘tighten’, ‘squeeze’ < *uerǵh-. Undoubtedly related are Gr. ereíkē (< *wer-eyk’-o-), OIr. froech, W grug (< *wroyk’o-), but no common prototype can be posited (Snoj assumes *werk’-, *wrg’-, *wereyk’-, *wroyk’-). All things considered, this is probably a loanword from some Non-IE language ((thus also Beekes 452). PIE IEW 868-871 *reu-2, reuǝ- : rū- ‘to tear out, dig out, open, acquire; DRom n. rug, pl. ruji ‘bramble, blackberry stalk’ Cogn.: Lat rubus ‘blackberry perennial, blackberry’ (‘*shrub, bush’) Meyer Lubke 7414; Lat ruō ‘tear open, dig, scratch’, OHG roufen, MHG roufen, reufen, raufen ‘pluck’; *reuk- (partly probably also reug-, reugh-?) ‘pluck’ DEX.RO Lat rubus > DRom rug with b > g (see above) PIE MA *g̑embh, IEW 367 *g̑mbh ‘bite, tooth’; DRom n. ghimpe ‘thorn’ Cogn.: Alb gjemp, gjemb ‘needle’; OHG kamb, OE comb? IEW 470-471 *gʷel-1 ‘to stick; pain, death’ Alb. Gk. glimp (gjëmp, gjëmbi) ‘thorn’ PIE IEW 595 *ḱeub- ‘thorn’; DRom n. sâmbure ‘stone, pit’ Cogn.: Alb thumbull ‘button, etc.’; Maybe Alb (*ḱeub-) thumb ‘thorn’ common Alb. s- > th- phonetic mutation; compound of IEW 18-22 ḱemen-, ḱemel-, ḱōmen- ‘stone, skies’ +Grk βρυω ‘swell up, blow’ > έμβρυον? Grk έμβρυον? No etymology in Chantraine, Beekes. PIE IEW 948-49 *(s)kerb(h)-, (s)kreb(h)- ‘to curve, turn’; or, IEW 631 *ku̯erp-, ku̯erb- ‘turn, wind’; DRom n. curpăn, curpen ‘plant tendrils, name of a plant with tendrils’; the main semantism is ‘curve, turn’. cârcel, pl. cârcei ‘tendrils, plant with spiral tendrils as vine’, and ‘the ring used to hook pieces in wood wagon;’ also ‘Charlie horse spasm’; v. cârci ‘to squat’
Appendix
285
Cogn.: Alb kulpër, kulpën ‘clematis’; Lat curvus ‘curve’; Grk kurtós ‘curved’; karpos ‘hand root, carpus’; Ir cor ‘circuit’; Lth kreīvas, OCS krivŭ ‘curve’; Toch A kārp- ‘climb down, go down’, B k̑ārp ‘turn around’. PIE MA *haer- generally ‘reed, rush’; DRom n. sg. arac, harac; ARom hărac ‘stick or rush to support plants, vine, tomatoes, etc.’ Cogn.: Lat harundō-arum ‘reed’; Grk áron ‘arum-plant from Aracea family’; Khotanese [East Iranian] arā ‘reed, rush’ PIE MA *nedós ‘reed’; DRom n. nai ‘the musical instrument as the Pan’s flute’ Cogn.: NPers nai ‘reed’; Arm net ‘arrow’; Lith néndré ‘reed’; Luv ‘nātatta‘reed’; Skt nadá ‘reed’ PIE MA *tṛ́nu- ‘thorn’; DRom n. târn ‘broom made of twigs’; n. târnă ‘wicker basket’, toponym Târnava, hydronim Târnava Mare, și Târnava Mică, forms before the Slavic metathesis. Cogn.: NE thorn; OCS trьnь ‘thorn’, strьnь ‘straw’; Bg trăvna ‘grass’; Rus Хоть тарра ‘grass’; Skt tṛ́rna-m ‘grass, herb’; Grk τερναξ ‘cactus stem, artichoke’; Goth þaumus, OIce, OE þorn, OHG dorn ‘thorn’; Finish tarna ‘sedge, grass’ (old loan) DEX.RO OCS trьnь ‘thorn’; contamination? PIE MA *h2/3(e)lĝ(h)- ‘grain, barley, millet?’; DRom alac ‘spelt’ Cogn.: Hit halki- ‘barley, grain’; NPers arzan ‘milet’ Grk áliks ‘spelt’; Lat alica ‘spelt’ > DRom alac? (merger?); Toch B lyekśye ‘barley’; Alb lakër ‘green leafy vegetable, cabbage’. DEX.RO Hun alakor ‘spelt’ > DRom alac, but of unknown etymology in A Magyar Nyelv Törteneti-Etimologiai Szotarati-Etimologiai Szotara, 1970, possibly from DRom. PIE IEW 3 ˙*ades-, *ados- (*hegh-) ‘sort of cereal, spelt’; DRom n. hat ‘strip of uncultivated land marking the edge between neighboring fields, border’; hotár (accent on á which explains the phonetic devel. a > o) n. ‘border’
286 Appendix
Cogn.: Lat ador, -ōris n. ‘a kind of grain, spelt’, maybe in Goth. atisk (*ades-ko-) ‘sowing field’, probably OHG ezzisca pl. ‘sowing’, MHG dial. Esch, Swiss dial. Aesch ‘field entrance of a village’; Toch. AB üti ‘grass’; Toch B atiyo ‘grass’ (Adams 9); Grk ἀθήρ ‘an ear of corn’; Perhaps Arm.hat ‘grain’, Hit. hattar n. ‘cereal’. Note: It seems root *ades-, *ados- ‘sort of cereal’ evolved from an older root *heĝh- ‘a kind of grain’. This root was suffixed with common -ska formant in Gmc. branch Gmc. *at-isk-a-, while in Anatolian branch the root was suffixed with common PIE -tar formant. The old laryngeal (Centum ḫ- > a-, e- : Satem ḫ- > s-) was lost except in Hit. and Arm, to which we can add the DRomanian hat, hotár (accent on á). Clearly Gmc. languages borrowed the cognate from a reduced Lat (*hattar-) adŏris > Gmc. *at-isk-a-.; Finally zero grade in Alb (< adōris) *dris, drizë ‘thorny plant’, (< dris) drithë ‘grain’ where the Lat. -is ending has been solidified. The surprise is the phonetic mutation -ĝh- > -d- found only in Av.—Illyr.- Balt languages. DEX.RO: Ukr hat ‘barrier, fence’ (Not in Derksen); for DRom hotar— Hun hátar ‘between regions, border’, in: Latin document from 1130-40 ‘inde ad caput quercus et ibi est cumulus hoc est hotar’ (sic) etymology probably derived word from hat ‘back’ A Magyar Nyelv TörtenetiEtimologiai Szotarati-Etimologiai Szotara, 1970 p. 72–73; more likely borrowed from DRom. PIE MA *wélsu- ‘meadow’ < *wel- ‘grass’; DRom n. vâlcea ‘meadow’ Cogn.: NWels gwellt ‘grass; OPrus woltis ‘head of grain’; Hit wēllu(want) ‘grass as grasssy plain’; Grk Elysian (ēlúsion) fields; NE swale. PIE MA *haek̑—‘point, sharp’; a derivative *haek̑sti- ‘bristle’; DRom adj. ascuţit, v. a ascuți, pers I (eu) ascut ‘I sharpen’; DRom ac ‘needle’ < Lat acus (Dacian *akmon ‘stone’) Cogn.: NWels eithis ‘furze’; Lith akstis ‘spit (for roasting)’; Rus ostı ̆ ‘awn, bristle’; Toch B asce ‘head’; DEX.RO Lat *excotire < cos, cotis ‘quern’ > DRom ascut
Appendix
287
PIE MA *lendh- ‘open land’; DRom n. lan ‘(grain) field’ Cogn.: OIr lann ‘open land’; OPrus lindan ‘valley’; NE land; Rus ljadá ‘overgrown field’ PIE MA *melh2- ‘grind’; IEW 716-719 *mel- /*mol- /*ml- ‘to rub, to crush, to grind’; DRom mălai ‘flour’ (today only ‘corn flour’) Cogn.: Lith malù; OIr meilid; Lat molō; NE meal; OCS meljǫ; Grk múlē ‘mill’; Arm malem, Hit mall(a); Skt mṛnāti; Toch B mely- ‘grind’, Alb mjel ‘meal, flour’; OHG malan, OIce mala ‘grind’. PIE MA *hamerg- ‘gather, harvest’; IEW 738 merĝ-1 ‘strip off, wipe’; DRom v. a merge ‘to walk’ Cogn.: Lat mergae ‘reaping boards’ cutting fork’, deriv. merges, -itis ‘sheaf of corn’; Grk amérgō ‘gather, harvest’; Arm meržem ‘expulsion’; de Vaan (2008, p. 375) mergae ‘reaping-board’, derivatives: merges -Ms * sheaf of corn (Verg.+); merges can be understood as ‘what one can take with the mergae;’ many scholars doubt the connection with Gr. άμέργω ‘to pluck (flowers), squeeze olives1 , but the two are quite similar. It is a different question whether these two forms stem from PIE *h2/3mer/lg—‘to wipe’ (Skt marj-, Av. marz-, Gr, όμόργνΰμι ‘to wipe’ maybe Gr. άμέλγω ‘to milk’), since the semantic connection is not straightforward. Lat mergae and Gr. άμέργω may continue a separate PIE root *h2merg -, or they might go back to a non-IE loanword of agricultural terminology. DEX.RO Lat. mergere ‘dive, drown, going to sink’. PIE MA *(s)kerb(h)-, (s)kreb(h)-, nasalized (s)kremb- ‘to curve, turn’; IEW 938-947 *skrēbh-, skrōbh-, with r-suffix, or, IEW 948-49; 958 *sku̯erb(h)- ‘to pierce, stick (of thorns)’; DRom n. scorbură ‘tree hollow’, *scorubura < *(s)kerb(h) Cogn.: Blg koruba; Alb korubë ‘tree hollow’; Russ. skorbnutь ‘crooked’, kórob m. ‘box, basket’; Lith. karbas m. ‘basket’; Lat. corbis f. ‘basket’; OHG korb m. ‘basket’; Fi. karpas m. ‘basket’; Derksen (2008) Slav *korbь; *korba m. f. a ‘basket’ possibly an early (Balto-Slavic?) borrowing from Germanic. The Germanic word was in turn borrowed from Latin: de Vaan PIE *skrobh -i- ‘hollow, pit’ Lat scrobis ‘hole in the ground, pit’; Alb. shko-zë (*skrēbh-r-), O.Pruss. scober-wis (*skrōbher-), Lith.
288 Appendix
skrúoblas (*skrōbh-ro-), newer skroblùs, but Ltv. (with secondary ü) skübardis, skübarde (*skrōbhar-) ‘beech’, s. Jokl WuS. 12, 71 ff.
Anatomical Parts PIE MA *k̑ṛrēh2 ‘head’; IEW 574-77 1: k̑er-, k̑erə- : k̑rā-, k̑erei-, k̑ereu‘head, horn, upper part of body’; DRom n. creier ‘brain’; ARom n. crier ‘brain’; Cogn.: ON hjarsi ‘crown of the head’; Lat cerebrum ‘brain’; Alb krye ‘head’, Gheg krëja ‘head’, Tosc krye- ‘head’; Grk kárē, karárā ‘head’, krānion ‘crown of the head’; Av sāra, sarah ‘head’; Skt śiras ‘head’, Toch B kraiñye ‘neck’; Hit kitkar ‘headlong’; Medieval Latin cranium ‘cranium, skull’ DEX.RO Lat cerebrum > *cerbrum > DRom n. creier. PIE MA *ghebhōl ‘head’; DRom n. ceafă ‘nape, back of the neck’; IstrRom čâfa ‘nape’ Cogn.: ON gafl ‘gable, gable-side’; NE gable; Grk kephálē ‘head’; Toch śpāl ‘head’; Welsh gwdff; Alb. qafë (neck), Gheg klafë; [Alb *qafë, kjafa < Balkan-Trk kafa ‘back of the neck’, Holm: JIES vol. 39, no. 1–2; 2011, p. 45] PIE MA *gwer(h3) ‘swallow’ > *gwrihxw- eha-; DRom n. gură ‘mouth’ Cogn.: Latv griva ‘river mouth’; Rus griva ‘mane’; Av grīvā ‘neck [of demonic being]’, Skt grīvā́- ‘neck’; Alb gōjë, a ‘mouth’ Tosk golë; Ital gola ‘neck’ [Holm: JIES vol. 39, no. 1–2, 2011, p. 45] IEW 352 *g̑ā̆r-, ger- ‘to call, cry’ Cogn.: Lat garrio, ire ‘chatter, prattle’, Grk gêrus ‘voice, call’, OIr do-gair ‘call’, OE cearu ‘care, sorrow, mourning’; possibly DRom n. câr, v. cârâi ‘croak’ (onomatopoeic?), see below. DEX.RO Lat gula ‘throat, gullet, pharynx’, for which de Vaan: “The traditional etymology *gwl-h2- > gula is morphologically unlikely, and will phonetically not work: such a preform would yield *gla *gala *guala or *vala maybe *vo/a; cf glans, gravis from roots with a labiovelar, A preform *gel- is impossible too, since this would yield Lat. **gola. Hence, LIV
Appendix
289
posits a root *guel-, with a zero grade *gul-. The schwebe-ablaut is conspicuous, as is the absence of direct evidence for *guel-. Hence, the root may be onomatopoeic, having only the shape *gw/- In fact, all forms may have arisen in the separate branches. Relatedness of gula to the stem glutis possible, but cannot be demonstrated;” de Vaan (2008, p. 690) voro, -are ‘to swallow, devour’—PIE *gwo/erhj-o/h2—‘devouring, devoured’. Probably a denominative verb to a noun/adj. *gwora- ‘food’ or ‘devouring’, cf. Gr. βορά. In theory, a preform *gwera- is also possible. PIE IEW 103 *bu- ‘lip, kiss’; DRom n. buză ‘lip’ Cogn.: Alb buzë ‘lip’; Pers bŏsīdan ‘kiss’; MIr bus, pus ‘lip’, busóc, pusíc ‘kiss’; Ger Buss ‘kiss’; E buss, Swe puss ‘kiss’; Thracian Byzas, Byzos, Dacian Beusas; Illyrian Buzos, Buzetius; Byzantion ‘town on the shore’. PIE MA *ker- ‘caw’; DRom câr, v. cârâi ‘caw’ (onomatopoeic) possible developments: ciocârlan masc. for ciocârlie ‘lark’ (cioc ‘beak’ + câr ?) Cogn.: Lat corvus; Grk kóraks ‘raven’; Skt karața ‘crow’; Ir goir ‘call, cry, crow’, Old Irish adgaur *garô, speak; Lat garrio ire ‘chatter, prattle’ PIE MA *gutṛ ‘neck, throat’; DRom n. gât ‘neck’, n. gâtlej ‘trachea’ Cogn.: Lat guttur ‘gullet, throat, neck’ (de Vaan, 2008: the –ur stem is difficult to explain; p. 276); Hit kuttar ‘nape of neck’; PCelt *gutus -ους /-us; Old Irish < Goidelic guth ‘voice’, Scottish Gaelic guth ‘voice’ (Matasović, 2011) PIE IEW 474-476 *gʷer-1, gʷerǝ- ‘devour; throat’, *gʷrīuü ‘neck’; DRom n. grumaz ‘neck, nape’; v. sugruma ‘strangle’ Cogn.: OInd grīvü f., Av grīvü ‘nape’, Av garǝman- ‘throat, neck’; OCS griva ‘mane’, grivьna ‘collar, neckband’; Alb *griva, krifa ‘mane’? Grk ἔβροξε, Aor ‘devoured’, Hom. ἀνα-, κατα-βρόξειε, ἀνα-βροχείς, βρόξαι ῥοφησαι, Hes., βρόχθος m. ‘gullet, throat’; MHG krage ‘neck, throat, nape’; Ger Kragen, ON kragi neck, collar’, MEng.crawe, Eng craw; Orel (1998) Alb gurmaz ‘gullet’ > DRom grumaz ‘nape’
290 Appendix
PIE IEW 379-380 *geng-, gong- ‘lump’; DRom n. gușă ‘goiter’ Cogn.: Lith gùnga ‘hump, hunchback, ball, clump’; gùzas, gùzikas ‘hump, hunchback, gland, knag’; Maybe Alb (*gunsa) gushë “neck, throat, Adam’s apple’—considered a loan from DRom; Serb gûz ‘buttock’, Russ. gúzka ‘backside, tail , rump’, Sloven goza f. ‘buttock, backside’. PIE IEW 138-139 *bhereg- ‘sound, roar, cry, *sharp voice’; IEW 145 *bherug-, bhrug-, bhorg- ‘throat’ (Russu, 1981); DRom n. beregată ‘trachea’ Cogn.: Grk φάρυξ, -υγος, later (after λάρυγξ) φάρυγξ, -υγγος ‘windpipe, gullet’,close to φάραγξ ‘cleft, gap, abyss’; perhaps related to *bherug-, bhrug-, bhorg- ‘throat’ (IEW 145); Cogn.: Grk βάραθρον, Hom. βέρεθρον, ark. ζέρεθρον, δέρεθρον ‘neck’, perhaps Grk βρόγχος, Ion. βρογχίη “ windpipe “, βράγχια, βαράγχια ‘fish gills’; OIr brügae ‘neck, nape’; MWelsh breuant ‘windpipe’; OCorn briansen ‘neck’. PIE IEW 554 *kelǝu̯o-, *keleu̯o- ‘bald’; DRom chel n. ‘bald’ Cogn.: Pers. kal ‘naked, bald, bleak’; Lat. calvus ‘naked, bald, bleak, haarlos’ (basic form Ital. kalouos; Osc kalúvieís, kalaviis ‘bald’; Lat. calva ‘cranium’; OInd khalatí- ‘baldheaded’, khalvüta- ds (: Arm. xalam ‘cranium’?). PIE IEW 624 *krūs- ‘shank, lerg’; ‘*kr(o)k-sko-’, -u- ‘arm’; DRom n. crac n. ‘leg, one leg of pants’; n. cracă ‘branch’ (Grk brachiôn; but cf. Sanscr. bühu; Albanian krahu, krahi ‘arm’ (ph- > h- in Arm.) : Rom. cracă branch, branching, limb, arm IEW 739-740) Cogn.: OInd kisku- m. ‘forearm’; Alb krakë ‘upper arm, shoulder’, (*krok-sko-?); maybe Alb. krah ‘arm’ [common Alb. -k- > -h- phonetic mutation]; probably Lith. kárka ‘pig’s foot, foreleg of pigs’, ablaut. E.Lith. kirkãlis m.” ‘stilt’; Slav *korka f.: in Sloven kráka, kráča ‘pigs foot’; besides Slav *korkъ in Bulg. krak ‘leg, foot’, Russ. ókorok ‘ham’; in addition ablaut. Slav. *krokъ m. in Sr.-Cr. krôk (gen. krȍka), Cz. krok ‘footstep’ DEX.RO Bg krak ‘leg’ > DRom crac , but not in Derksen. Contamination?
Appendix
291
PIE MA *smók’wṛ ‘beard’; DRom n. smoc ‘turf of hair’; v. smuci ‘pull, jiggle’ Cogn.: Alb mjekër ‘beard, chin’; Hit z(a)munkur ‘beard’; Skt śmáśru ‘beard’; Lith smakrà ‘chin’; OE smæras [pl.] ‘lips’; Ir smech ‘chin’. Kortlandt #305 (2016) The preservation of the palatovelar in Skt śmáśru ‘beard’, but not in Arm mawruk’ < *smoḱru-, Lith smãkras ‘chin’, Alb mjekrë, shows that it represents an ancient derivation of the root *smeḱattested in Old Irish smech ‘chin’, Latin māla ‘cheeks’ < *smḱslā PIE MA *kert- ‘weave, plait, braid’; IEW 584-585 *kert-, kerǝt-, krāt- ‘to turn, roll, wind’; DRom adj. creţ ‘curly hair’; v. încreți ‘to curl’ Cogn.: Lat crātis ‘wickerwork, hurdle, honeycomb’; Grk kartallos ‘basket’ IEW 583 *k̑er(es)- ± ‘rough hair, bristle’ Cogn.: NE hair; Lith šr̃ys ‘bristle, animal hair’; Rus šerstı ̑ ‘wool, animal hair’ PIE MA 1. *(s)kel- ‘crooked’; IEW 928 (s)kel-4 ‘to bend; crooked’; IEW 935-938 (s)ker- ‘to bend, turn’; DRom cârcă n. ‘(uper) back, piggyback’, n. cârjă ‘crutch’, n. cârlig ‘hook’, n. cârlionţi ‘ringlets’, n. cârnat ‘sausage’, v. cârni ‘turn’; ARom cârligu ‘hook’ Cogn.: OE scēolh ‘crooked’; OPrus culczi ‘thigh’; Bulg kúlka [most probably from Lat pulpa DRom cârnat ‘sausage’. PIE MA *h3elek ‘elbow, forearm’; DRom n. olog ‘crippled of one leg’, v. ologi ‘to criple’ Cogn.: Arm olok ‘shin, leg’; Lith ùolektis ‘ell’, alkū́ne ‘elbow’; Bg lакътъ ‘ell’; Rus lokótı ̆ ‘elbow, ell’; Grk álaks ‘forearm’
292 Appendix
PIE MA *twéks ‘skin’; IEW 1099 tu̯akos ‘skin’; DRom n. toval ‘cow skin for shoes’ Cogn.: Ukr tovar ‘cow skin’; Skt tvák- ‘skin, fell, fur’; Grk σάκος ‘shield’ (leather); Hit tuekkas ‘body, person’ PIE MA *(s)k̑up ‘shoulder’; IEW 627 k̑up- ‘shoulder’ DRom n. supţioară, subsuoară ? ‘armpit’ Cogn.: Alb sup ‘shoulder’; Av supti ‘shoulder’; Skt supti ‘shoulder’. DEX.RO Lat *sub + axilla ‘armpit’ < *haeksleha ‘armpit’, Lat axis, āla ‘shoulder’. PIE MA *g̑hés-r- ‘hand’; IEW 441-442 *g̑herzd(h) ‘to scrape, scratch, slit’; *g̑her- *g̑herdh- ‘to grab, grip, seize’; DRom n. gheară, ghiară ‘claw’, and possibly n. hârleţ ‘hoe’ Cogn.: Lat hīr ‘hollow of the hand’; Alb dorë ‘hand’ < *ghēhrā (IEW 447); Grk kheir ‘hand’; Arm jer`n ‘hand’; Hit kissar ‘hand’; Toch B şar ‘hand’ DEX.RO n. hârleţ ‘mattock’ < OCS rylici? (Derksen, 2008: rydlo ‘spade, snout’, OCS rylo ‘spade’, Blg rilo ‘snout’) (Pușcariu: Lat *garra; Ciorănescu Lat *ung(u)laris) PIE MA *bulis ‘rump’; DRom n buric ‘navel’ Cogn.: Lith bulis ‘rump’; Skt buli- ‘vulva, anus’; DEX.RO Lat. umbō, -ōnis ‘navel of the shield’; umbilı ̑cus ‘navel’ PIE MA *gheug̑h ‘protect, hide; DRom găoază n. ‘ass hole’; Cogn.: Lith gūžti ‘cover with something warm’; Av gūzra ‘hidden, secret’; Skt gū́hati ‘conceals’; Av gaoz—‘hide’ (Matasovoč IEW 450 *ghewgh-) PIE MA 1. *udero- ‘uterus/womb’; DRom n. uger; ARom udzire, IstrRom uger ‘udder’ Cogn.: Lat uterus ‘abdomen, womb’; Grk húderos ‘dropsy DRom uger udder’ PIE MA *telp- ‘space, have room’; IEW 1062 *telp ‘space, spacious’; DRom n. taplă ‘sole of foot or shoe’, [metaphorical use for any base] ex. talpa casei ‘base of house’; Cogn.: OIr -tella ‘have room for something’; Lith telpù, talpà ‘spacious, have enough room’; Skt tálpa ‘bed’; Toch B tälp- ‘be emptied of, purge’; Hit patalha ‘floor, sole’; Skt tala ‘plain area, palm, sole’; OPruss. Talus ‘ankle, ankle bone, heel’; Bg talpa ‘wooden board’ IEW 1061 *tel-, telə-, telu- flat (ground, board); PIE *tlāp- ‘hand’ (Duridanov) DRom labă ‘pow’; Friuli tapla ‘paw’ Lat talpa ‘mole’ [f. a] (Varro+) ? de Vaan (2008)—no etymology. DEX.RO Hun talp ‘base, sole’, presumed from Italian, of unknown etymology in A Magyar Nyelv Törteneti-Etimologiai Szotara, 1970 > DRom? PIE MA *g̑hóln- ~ *g̑hólos- ( DRom galben ‘yellow’
Vital Functions PIE MA *bher- ‘to bear (a child)’); DRom n. burtă; (old dialectal) borţ ‘belly’; IEW 128-132 *bher- ‘well, yeast, bubble’; IEW 125-126 *bhelĝh- ‘swell’ > *bhólĝhis ‘skin, belly’; DRom v. borâ ‘to throw
294 Appendix
up, vomit’, n. borhot ‘marc, fermented fruit for alcohol, or other fermented food’; n. borhan ’animal stomac’; n. borfei ‘intestin’; n. bulz ‘ball of polenta’, (n. burduf arch burduh ‘leather sac; stomach of an animal to carry cheese,’ although Vasmer considered the source to be Azerbaijan burduk “fur for wine.” (Radlov 4, 1832) (Persian influence?) Cogn.: Lat fermentum ‘ferment, leaven’; MIr fobar ‘well’; OE beorma ‘yeast, leaven’; Grk porphū́rō ‘bubble’; Skt bhuráti ‘moves rapidly, quivers’; OIr berbaid ‘boils, seethes’; Lat ferveō; Alb brumë ‘dough’; Skt bhurváni- ‘restless, excited’; Thracian brûtos ‘a kind of beer’—DRom bragă ‘millet beer’ < *bhŗg- ‘cook,’ (Welsh Corn. brag ‘malt’ IEW 739-740); NE brew; Alb (*brauk) barku ‘belly’ (IEW 170). bërsi {2} [fnp] (tg) ‘rest from wine-, olive-, plum-press’, PAlb. *britśia̅ {1} (AE 98) PIE IEW 125-126 *bhelĝh- ‘swell’ > *bhólĝhis ‘skin, belly’; DRom n. bulgăr ‘ball of matter: earth, snow, etc.’ Cogn.: OIr bolgr ‘sack’; Gaul bulga ‘leather sack’; Ir bolgaim ‘swell’, bolg ‘bubble’, bolg ‘sack, bag, belly, husk, trouser’, Mir bolgach ‘swelling, blister, bubble, blister; pox’, bolgamm “gulp’; OE bel(i)g ‘bag’ < NE belly; OHG balg ‘skin’; Skt upa-barhani ‘cover, bolster’. PIE MA *k̑er- ‘grow’; DRom n. cârlan ‘lamb under 1 year, foal under 2 years’; ciocârlan masc. for ciocârlie ‘lark’; toponyms Cârlani, Cârlănești. Cogn.: Lat creō ‘create’, goddess Cerēs; Grk korénnūmi ‘satisfy’, koȗros ‘adolescent’; Arm sirem ‘bring forth’. PIE MA *gheha ‘yawn’; DRom căsca v. ‘yawn’ Cogn.: Grk kháskō ‘yawn’ > DRom? PIE MA*losiwos ‘weak’; IEW 680 *lēs: lǝs- ‘weak, feeble’; DRom n. leşin ‘faint’, v. leșina ‘to faint’, n. leş ‘corpse’, lihnit? ‘famished’ Cogn.: Goth lasiws ‘weak’; MHG er-leswen ‘become weak’; OIce lasinn ‘weak, destructed’, las-meyrr ‘weak, woeful, wretched, miserable’, MLG
Appendix
295
lasich = lasch, las ‘slack, faint, languid’; Toch B leswi ‘attack of weakness’; NE lazy; Alb. lasht ‘old, weak’; Slav *lošь in Bulg. loš ‘evil, bad, evil, ugly’, Ser-Cr lȍš ‘unlucky, evil, bad’. PIE MA *leh1d- ‘tired, sluggish; DRom n. lălâi, fem. lălâie ‘slow, stupid, untidy’; Cogn.: Lat lassus ‘tired’, OE lœt ‘sluggish’; Lith lė́nas ‘lazy, gentle’; OCS lěnŭ ‘lazy’; Alb lodhet ‘becomes tired’; Grk lēdeîn ‘be tried’; Toch B lāl ‘exert oneself ’ PIE MA *h1ermen- ‘sickness’; DRom adj. sărman (s-mobil) ‘man in distress, poor’ Cogn.: OE earm ‘weak’; Alb jerm ‘stupor’; Arm olorm ‘pitty’; Hit arman~ ērman ‘sickness’; Blg сиромах (siromah) ‘poor, pittiful’, Macedonian сиромав (siromav) ‘poor, pittiful’ IEW 615 *kormo- suffering, pain? Cogn.: OIce harmr ‘sorrow, distress’, OE hearm ‘distress, pain, damage, pity’ MLG harm ‘pain’, OHG har(a)m ‘affliction’ DEX.RO Cr siromah ‘poor’, Serb siromašan, cognates? PIE MA *(s)keng- ‘limp’; IEW 930 *skak-thi-, IE *skeng-ti-; -thi- after Skt ásthi- ‘bone’; DRom n. ciung, ciump ‘one armed, crippled’ v. ciumpăvi ‘cut randomly’, ciunt, unihorned, one armed’ v. ciunti ‘cut short’, in the same family: ciot ‘stump, stub’; ciut, fem. ciută ‘hornless’ (of deer) PanBalkanic—Alb shut(ë) Cogn.: OHG hinkan ‘limp’; Grk skázō ‘limp’; Skt kañj- ‘limp’, sákthi- n (= Av. haxti-) ‘shank’ PIE IEW 928 *(s)kel-4 (extended klā-, klō-) ‘to bend; crooked, rotate, crooked (morally wrong); body parts: flexible joint, heels, knees, hips’; DRom n. ciul m., ciulă f. ‘animal with small ears, or ripped, without one ear or both, maily of sheep’; ciuli v.’lift up the ears to listen, hark’ Cogn.: Grk κυλλός ‘writhed, crooked, humped, lamed’, κῶλον ‘limb, member’, Grk κυλλος ‘deformed, crippled, crooked’ (Beekes: Pre-Greek?)
296 Appendix
PIE IEW 922 *skēi-bh-, -p-, nasalized ski-m-bh- ‘slant; to limp’; DRom n. șchiop ‘lame’ Cogn.: Maybe Alb shqeponj ‘walk with a limp; to lean upon a walkingstick’; nasalized Grk σκιμβός ‘lame’ PIE MA *(s)kerb-~(s)kerbh- ‘shrink, shrivel’; DRom v. scoroji ‘shrink, shrivel’ Cogn.: Lith skur̃bti ‘suffer a decline, wither, mourn’; Rus skórblyj ‘shrivelled’ DEX.RO OCS skora ‘bast, skin’ (Derksen: Czech skura ‘skin, bast, leather’; Sloven skǫrja ‘bark, crust’, Pl skóra ‘skin’) PIE MA *ster- ‘barren’; IEW 1031 *ster-6 ‘sterile’; DRom n. m. sterp, f. stearpă [arch] ştiră, stiră ‘steril, barren’ Cogn.: Lat sterilis ‘barren’; NIce stirla ‘barren cow’; Bulg sterica ‘barren cow’; Grk steîra ‘barren cow’, stériphos ‘barren’; Arm ster ‘barren’ Skt starī́ ‘barren cow’; Alb shtjerrë ‘lamb’, shterpë ‘steril’ IEW 1022-1027 *(s)ter-1, (s)terǝ- : (s)trē- ‘stiff’, immovable; ‘solid’, etc.; labial extensions: (s)terp-: Cogn.: Lat stirps ‘progeny, origin, source, beginning, ancestor’; Alb shterpë ‘barren’ PIE MA *wedh- ‘push, strike’; DRom v. a vătăma,’to hurt, injure, damage; 1st sg. eu va’tăm (first vowel accentuated ‘) ‘I hurt you/ someone’; IEW 1108 u̯ā-, u̯ō-, u̯ǝ- ‘to hit, wound’ [Note: also with t- forms] Cogn.: OIr fāiscid ‘presses’; Lith vedega ‘a kind of axe’; Grk éthei ‘destroys’; Hit wezz- ‘strike, urge’; Skt vadh- ‘strikes, pushes, slays’; Toch B wät‘fight’; Ltv vâts ‘wound’, Lith votis ‘open wound’ DEX.RO Lat victimare ‘to give offering, sacrifice, slaughtering’ > DRom vatăm ‘I hurt (you)’, possibly a merger.
Appendix
297
Family Relations PIE MA *méryos ‘young man’ and *meriha- ‘young woman’; IEW 738-739 *merio- ‘young man, woman’; DRom n. mire ‘bridegroom’ Cogn.: Lat marītus ‘husband, lover, suitor’; Alb shemër (MA notation) ‘co-wife, concubine, mistress’; Grk meîraks ‘young man or woman; Av mairya ‘young man’; Skt márya- ‘young man, lover, suitor’; MPers mērak), Gk. μεῖραξ m. f. ‘knave, boy, girl’; Lith martì ‘bride, maid, OPruss martin acc. sg. ‘bride’. PIE IEW 687 *leugh-2. *lugh- ‘oath’; DRom v. a se logodi ‘to get engaged’ Cogn.: OIr lu(i)ge n., Welsh llw m., Bret. le ‘oath, vow, pledge’ (*lughiom); Goth liugan, -aida ‘marry’, liuga ‘matrimony’, (*oath), OHG urliugi (*uz-liugja) ‘war, fight’ (*state, status), zero grade MLG orloge, orloch, OS orlag, -logi, OFris orloch ds (therefrom also MHG urlage ‘fate, destiny’ OFris logia ‘marry’ DEX.RO Slv lagoditi ? not in Derksen; Pol lagoditi ‘to reflect well’ PIE IEW 447 *ĝhesor-1, ĝhesr- ‘hand’ *ĝhesto-2 ‘hand, arm’; DRom n. zestre n. ‘dowry’, meaning that the dowry was sewed by the bride’s hand? Note: both roots *ĝhesor-1, ĝhesr- ‘hand’ and *ĝhesto-2 ‘hand, arm’ derived from an extended archaic root ĝhes + reduced form of the common PIE suffix variants -tar, -ter, -tra, tre-. Cogn.: OInd hásta-h m. ‘hand’, Av zasta-, OPers dasta- ds.; Lith. pa-žaste, pa-žastis f. ‘place under the arm, armpit’; common Balt-Illyr. ĝh- > z phonetic mutatIon; Alb dorë f. ‘hand’, pl. duar; *ĝhēr- > *ĝhēsr- ‘would have’ *dostrë ‘yeld, result in, devoted’; Hit, Toch metathesis from -rs- to -sr-; with ĝher- ‘grasp’ development. DEX.RO Lat dextrae ‘right (hand)’ includes the same semantism, ‘take hand in marriage’
298 Appendix
PIE MA *h2euh2os ‘grandfather’; IEW 89: *auo-s (*ḫuḫḫaš) ‘grandfather’; DRom arch. n. auş ‘grandfather, old man’; Cogn.: Hit hūhhas; Lat avus, ON afi; Arm haw, Toch B āwe [uncle] all ‘grandfather’; OIr aue ‘grandson’; OPr awis, Lith avynas ‘uncle on mother side’; Lyc χuga ‘motherly grandfather’; Skt sūh ‘progenitor’; Alb gjysh ‘grandfather’(< *sū-s-io-) DEX Lat avus ‘grandfather’ (a+ suffux uș), except that Lat av > au is rare, while Lat au- is well preserved in DRom ‘aurum > aur, audire > auzi’; the DRom is closer to the PIE. PIE MA *tat- ‘father’; IEW 1056 tata-, tēta- daddy; child word, father’; DRom n. tată ‘father’ Cogn.: NWels tad; Proto-Celtic: *tato- ‘father, foster father’; Middle Welsh: tat; Middle Breton: tat; Cornish: tat ‘pater’; Lat tata (child word); Grk tatâ; Luv tātis; Skt tatá, all ‘father’; Skt tatà ‘father’; Alb tatë ‘father’; Welsh tad, Corn tat ‘father’; Ltv tẽta; Lth tẽtis, tẽtë ‘father’; OInd tatá‘father’; Gk. τέττα (Hom.) voc., τατᾶ voc. ‘o father!’ The general use pleads for a common source. De Vaan (2008) PIE *h2et-o- ‘daddy’. IE cognates: Hit. atta-; CLuw. tata/i-; HLuw. tati-; Lyd. taada’ father\Gr. δττα ‘little father’, Alb. at, Go. atta ‘father’; a nursery word for ‘father’. It usually has the form TaTa or aTTa, with Τ being a dental or labial stop. PIE MA *népōts ‘descendant’; IEW 764 nepōt- (*nepotis) ‘uncle; nephew, *descendant’; DRom n. nepot, ‘grandchild, nephew’ Cogn.: Lat nepōs; OLith nepotis, nepuotis ‘grandchild, grandson’; OCS netijī ‘nephew’; Alb nip ‘grandson, nephew’; Grk népodes ‘descentant’, Hom. νέποδες (δηοη) ‘young ones, children’; Skt nápāt ‘grandson, descendat’; Av. napüt-, naptar-, ар. napütDEX.RO Lat nepōs, nepōtis; widespread form, could indicate a merger. PIE IEW 699 *māk̑- ‘long, slender’; DRom n. moș ‘old man,’ moașă ‘old woman, midwife’ Cogn.: Alb moš ‘old(er)’; Lith móša ‘sister of man’; OPr moazo ‘Muhme’, Latv mãsa ‘sister’; Toch A mok, B moko ‘old’, mokone ‘age’; Pasto məšər ‘elder, patriarchs chief of village’ (Rădulescu, 1984); cognates with the
Appendix
299
meaning ‘our relatives [elders]’—personal pronouns and pronominal adjectives such as*asma- ‘us’ OAv. ahm,YAv.ahma, Khotan. maha, Parāči mâ, with derivative*asmākam ‘our’ YAv. ahmākəm; OPers. amāxam; Bactr. αμαχο, μαχο; Oss.max and *asmāʧia- Wanetsi moš; Ishk. mьš, miš; Sanglechimič (Handbook, 2017, p. 584) PIE MA *n-h4en-, h4en- ‘grandmother’; DRom n. nană [arhaic respectful form] ‘aunt, old woman’ Cogn.: NWels nain ‘bunică’; Late Lat nonnus ‘nurse’; It nana; Alb nënë, nanë ‘mother’; Rus njánja ‘nurse; Grk nánnē ‘cousin fem, aunt’; NPers nana ‘mother’; Skt nanā ‘mother’ PIE MA *per- ‘offspring’ (of an animal); IEW 818 *per-2: D. *per- ‘to bear (child)’; DRom n. prunc ‘baby, small child (only of humans)’; prichindel ‘small child’ Cogn.: OE fear ‘bullock, steer’; Grk póris ~ pórtis ‘calf, heifer’; Skt pṛthuka- ‘child, young of an animal’; Lat. pariō, -ere, peperī, partum, paritūrus ‘to give birth to children’, partus, -ūs ‘to give birth to children, birth’; Lat puerculus ‘little son’ (de Vaan, 2008) DEX.RO Lat puerunculus > DRom prunc, prichindel ‘small child’?
Habitat PIE MA *stéh2-no- ‘standing, place’; IEW 1004-1010 *stā- : stǝ- ‘to stand’; IE n- present *stǝ- nă- ; DRom n. stână ‘summer sheepfold’ Cogn.: Skt sthãna ‘place, abode’; Grk ástēnos ‘unfortunate’; Lith stónas ‘place’; Alb stan ‘winter shelter for sheep, stall’ from med.L stantia f. L stant- pres. ppl stem of stare ‘to stand’; OCS stanŭ ‘stand’, stanь ‘camp’, Rus stan ‘figure, torso, camp’, Cz, Slk stan ‘tent’, SCr stān ‘habitation, loom’, Bg stan ‘loom, camp’ (Derksen, 2008) Considered Thracian-Dacian by Philippide 1928, Russu, 1981, Brâncuși 1983 and borrowed by pastoralism practices in Serbian, Slovenian, etc.
300 Appendix
PIE IEW 1036-37 *strenk-, streng- ‘narrow, tight, stiff’; DRom n. strungă ‘narrow path for milking sheep, gorge’; diminutive development n. strungăreaţă ‘midline, teeth gap’. Cogn.: στραγγγσ ‘twiddled, twisted’ < στραγγουρια ‘stricture’; Ir strengim ‘pull, drag’; Lth stringt ‘shrink up’; OIce strengr ‘cord, stripe’, strangi ‘tree trunk’; OE streng ‘strand’; OHG stranc ‘lash’; Alb shtrungë ‘milking enclosure’. PIE IEW 553-554 *kel- ‘to conceal’; DRom n. cuib ‘nest’; n. colibă ‘cottage, hut’ pan-Balkanic possibly a merger with a form from substrat? Cogn.: OInd śaraná- ‘protectively’, n. ‘protective roof, cottage’, śármann. ‘umbrella, protective cover, cover, custody’; śüla f. ‘cottage, house, etc. (: Ger Helm), lengthened grade (as Lat. cēlō, (= OIr celim, OHG helan) in occulō, -ere “conceal”; cēlō, -üre “conceal”, nominal cella ‘chamber, cell’; Grk καλῑά ‘cottage, barn, nest’; κόλυθρος m. ‘sack, bag, pouch’; Hom. κολεόν ‘stretched metric’?; κουλεόν ‘butler’, Att. κολεός “vagina” (*κολεFός; Lat collegia < collegium ‘shepherds huts’ (conf. Lat lingua > DRom limba); unclear Lat. culleus ‘leather sack’, from which Russ. kulь, Pol. kul ‘sack, bag’, out of it again Lith. kulìs ds., kulìkas, O.Pruss. kuliks ‘bag’); with labial extension Grk καλύπτω ‘wrap, concealed’, καλύβη ‘shelter, cottage’; DEX.RO Bg колиба, Serb koliba > DRom colibă, but not found in Derksen, perhaps an old Balkan lexeme from Thracian/Greek; Lat cubō ‘I lie down’ > cubile ‘bed’ > DRom cuib ‘nest’; also Matasovič (2013): “PSl *kolyba ,*koliba ‘hut’ (Slov kolíba, Cz. dial. koliba ‘tent’, Bulg. kolíba), Gr. kalýbē; possibly related to PSl. *xalupa ‘hut, cottage’ (Croat. halupa (Kastav), Slov. halúpa, Russ. dial. xalúpa, Pol. chałupa). PSl. kolyba may have been borrowed from Greek at a relatively late stage (after the Slavic migrations), but this does not solve the problem of the ultimate origin of this word.” PIE MA *wérhxus ‘broad, wide’; DRom n. uriaș ‘giant’; n. oraș ‘city’; toponyms Orșova, Oradea, Orăștie, etc.
Appendix
301
Cogn: Grk eurús ‘broad, wide’; Av vouru- ‘broad, wide’; Skt urú- ‘broad, wide’; Toch B wartse ‘wide’; Alb varosh ‘town-area outside a medieval castle’; IEW 1165 *u̯er-9; *u̯eru-s (*su̯er-) ‘wide, broad’: OInd urú-, Av vouru(from *varu-, IE *u̯eru-), Av uru- ‘wide’; comparative várīyas-, superlativ váristha-, in addition OInd urūcī f. ‘earth’; uruvyáñc: urūc- ‘far-reaching’, etc.; in addition ulūkhala- ‘mortar’ from *urū-khara- ‘with wide bottom’; OInd uras- (from *vuras), Av varō ‘width, circomference’; PIE *wriyo/eha- ‘citadel, city, town built of a hill’; DRom n. oraş? ‘city’ Cogn.: Thrac bria ‘city’; Messapic Uria—city name; Celtic and English place names: Wrekin, Wroxeter; Toch Brīye ‘city’; Grk hrion ‘promontory’; Toch Brye, Ari ‘town’, Thracian ßpia, probably /uria/, mentioned by Strabo 7,6,1 as a Thracian word for πόλις, τείχος (Lubotsky) DEX.RO Hun város ‘city’ > DRom oraș ‘town’, of unknown etymology according to A Magyar Nyelv Törteneti-Etimologiai Szotara, presents phonetic difficulties since the Slavic form for Lat Volcae was in ninth century Vlahu designating the Italians, speakers of Romance language as in DRom Vlah/Valah that is found in Hungarian as Olasz for an Italian and Olah for a Romanian; Gábor, 2006: vár2 ‘castle’ [1055 tn., 1193] ‘fortress’ Iranian, Avesta vāra‘cover, defensive’, Middle Persian war ’castle, fortification’, neo-Persian bār’mound, hill; castle wall, castle’, Afghan bāra’fortress, embankment’. PIE IEW 658-59 *legh- ‘to put down; to lie down’ (Russu 337); DRom n. leagăn ‘cradle, swing’, v. legăna ‘to cradle, swing’ Cogn.: Grk. λέχεται κοιμᾶται Hes., λέξομαι, λέκτο, ἐλέξατο ‘to lay, place to sleep’, ἔλεξα “lulled”, λελουχυῖα λεχὼ γενομένη Hes., λέχος n. ‘bed’, λέκτρον ‘lair’; Cyren. λέχ- (f.) ‘one who has just given birth’ (LSJ. 1043, λεχός); Gr. λέχ- (aoM.) ‘lay down’ (GEW 2: 110–112, λέκτο [3sg]) (Pyysalo, 2011); OHG lehtar ‘womb, uterus, placenta, afterbirth’; OIr lige ‘bed, grave’; OHG leg(g)en, OS leggian, OFris ledza, OE lecgan, OIce leggja; Goth ligrs ‘lair’; OHG, OS legar n.; OIce lag n. ‘place, position’ DEX.RO Lat *liginare?
302 Appendix
PIE MA *ghórdhos or *ghórtos ‘fence, enclosure’; IEW 444 *g̑herdh*gherdh- ‘encircle, enclose’; DRom n. gard ‘fence’; n. gardină ‘round notch at the bottom of a barrel to put it together, notch, croze’ Cogn.: NWels garth ‘pen, fold’; Rus górod ‘town’; Hit gurtas ‘citadel’; OCS gradŭ ‘city’; NE yard; Alb gardh ‘fence’, gardhën, gardhnia ‘croze’; Lith gar̃das ‘fold, pen’; OIc garðr ‘hedge, fence, court’ Some scholars consider it an Old Germanic loan (Poruciuc, 2009) PIE MA *del- ‘cut’; DRom n. daltă ‘instrument for hew, chisel’ PIE *dʰelbʰ- ‘to dig’? Cogn.: OIr dello ‘form’; Lat dolō ‘hew’; ON telgja ‘carve’; Lith dalti ‘divide’; Alb dalloj ‘cut’; Grk daidállō ‘work cunningly’. Alb. daltë ‘chisel’ (Orel) PGm. as *delbaną ‘to dig’ OE delfen ‘to dig’ DEX.RO OCS dlato, dlěto (?) ‘chisel’ PIE IEW 405 *gred- : grod- ‘to scratch’; DRom v. cresta ‘nick, snick, jab’, n. crestătură ‘scratch, notch, slot, indenture, nick, croze, score, slash’ Note: Only Alb. and Gmc. Cogn.: Alb gërrusë, gërresë, krūs(ë) ‘rasper’ (from its first grade derives Lat. grosa ds.), to lengthened grade present gërruanj, kruanj, kruj, also gërruëj, gërũj ‘scratch, scrape’ (from *grōd-, IE grēd-); ON krota (*grd-) “engrave”, ablaut (with intensive gemination) O.S. kratta ‘scratch, scrape”, OHG krazzōn, MHG kratzen ds (Gmc. *krattōn), besides j-verbs mnl. cretten (and cretsen), MHG kretzen, kretze ‘scabies’; here with expressive vocalism OHG krizzōn, MHG kritzen “ carve, scribble’. DEX.RO Lat crista ‘crest on bird’s head’ > DRom creastă (de cocoș) ‘rooster crest’, later development? PIE MA 1. *terh1- ‘pierce’; DRom n. ţăruş ‘stake used for gardening, enclosures’, -uș diminutival suffix, even though there is no DRom n. țar, perhaps a development as in n. țarc ‘pen, wicker fence—țăruși? Cogn.: OIr tarathar ‘instrument for drilling’; Lat terō ‘rub, wear away’; Lith trinú ‘rub’; OCS tīrǫ ‘rub’; Alb tjerr ‘spin’ DRom grindă; this solution could be considered but the Dacian PN Γράνδετον (Grandeton) sends us to a Pan Balkanic reflex. PIE IEW 543 *keipo-, koipo- ‘peg, sharp piece of wood or stone’ (Note: also *skeipo-); DRom n. țeapă. f., țep n. m. ‘spike, thorn’; v. înțepa ‘sting’, anthroponym Tepeluș, Vlad Tepeș Cogn.: OInd śḗpa-, śēpha- m. ‘tail, penis’ (with sk- prakr. cheppa- ds.); Lat cippus ‘a pale, stake, post, pillar’ (*keipos); Alb thep m. ‘sharp cliff’ (*koipos), tsep ‘prick, sting, point, edge, angle’, metath step ‘edge, cusp, peak’; Alb çapua ‘spur’< cape ‘step’ (Orel, 1998); also Lat scīpio, Grk σκῑπων and root skēip- ‘cut, clip’ is probably. skēi-p-: Gk. σκοῖπος m. ‘the basic beams on which the bricks rest; wall-plate of a building’; σκίπων ‘staff, stick, a staff or stick to lean upon, a walking-stick, crutch’
304 Appendix
IEW 930-933 *(s)kēp-2, *(s)kōp- and *(s)kāp-; *(s)kēb(h)-, *skob(h)and skāb(h)- ‘to work with a sharp instrument’; DRom n. țeapă. f., țep n. m. ‘spike, thorn’; v. înțepa ‘sting’, anthroponym Tepeluș, Vlad Tepeș Cogn.: Lith skãpsnė f. ‘stick’; Ltv. škeps ‘spear, javelin, spit, pike’, škepele f. ‘cut piece of wood’; OCS štapъ (*skēpos), Slovn ščap ‘stick’, Russ ščepa ‘chip of wood’, ščepatь, ščepitь ‘split’, OCS skopьcь ‘eunuch’ (Ger. Lw. Schöps), skopiti ‘castrate’; Lith skãplis, maybe Alb (*skop) shkop ‘stick’ PIE MA *keus- ‘hollow out’; DRom n. căuş, căuc, ‘wooden spoon, any cavity as in cup, hand, or ladle’; developments: descauc, dezgauc ‘scoop out’. Cogn.: Lith kaūšti ‘hollow out’, káušas ‘skull, ladle’; ON hauss ‘skull’; Skt koşa ‘vessel’. Arm xuc’ ‘room’; Toch B kusā- ‘village’. In Douglas Q. Adams A dictionary of Tocharian B: kauko* (n.) kauko//-, -, kaukom (?) If the meaning is something like ‘container’ we might compare this word to Latvian kaûss ‘ladle, scoop; skull,’ Lithuanian káušas ‘scoop; skull,’ and Sanskrit kóśa- (m.) ‘vessel, tub.’ DEX.RO Lat *cau (< cavus) + suf. -uș. A merging example? De Vaan (2008): “cavus ‘hollow, excavated, concave, deep (of water)’ PIE * kʽοuH-ó ‘hollow’. IE cognates: Olr. cua ‘hollow1, Bret keo, kev ‘cave’ < PIE *k‘οuio-; Gr. pl. κόοι ‘hollows, excavations, gaps, crevices, gorges’ (Hsch.) < *k‘óu̯o-, Gr. κοίλος ‘hollow’ < *κόριλος; Ann. soyl ‘cavity’ < *keulo-; maybe Alb. thettë ‘deep’.” PIE IEW 455-456 *ghrebh- ‘dig’; DRom n. hrubă ‘undergroud room, gallery, cellar’; Cogn.: NE grave; Lith grebti ‘rake’; Goth grōba ‘pit, pothole’; OHG gruoba ‘pit, pothole cave’; OCS pogrebǫ ‘bury’; OCS grobъ ‘grave’; Alb gropë ‘pit, ditch’; grep ‘hook, fishing rod’; Ukr Груба ‘rough, oven, stove, fireplace; moubh of an oven;’ from MHG. gruobc ‘pit, cavity, mine’, perhaps via Pol gruba ‘pit, cave’, see also DRom groapă ‘pit’ DEX.RO Ukr Груба ‘rough’. PIE IEW 586-587 *ket- *kot- ‘enclosure, chamber, dweling space’; DRom n. cătun ‘small village’; Romanians in Serbia n. cătuitor
Appendix
305
‘cottage’; toponyms Cătunu, Cătuna, Cătunaș; family names Cătuneanu, Cătunescu; n. cat ‘flat in a house, floor’; ARom & MeglRom cătun ‘small village’; possibly n. cat ‘first floor’. Dacian PN Κάττουζα? Cogn.: OE heaðor ‘enclosure, prison’; OCS cotı ̆cı ̆ ‘chamber’; Av kata ‘chamber’; Finnish kota ‘dwelling, tent, hut’; as storage pit; Av čāiti ‘in a hole’; Skt cā́tvāla- ‘hole for the sacrificial fire’; Toch B kotai- ‘hole’; Alb (*kotu-) katua ‘stable, basement, cellar’, nasalized katun, katund ‘house, village’; katund(i) ‘village’, katund ‘villa’; Grk κτιζω ‘au sens de fonder, établir une colonie’ (Chantraine: 592 ... largement attesté en mycénien sous la forme kotona ou kotoina, qui désigne un mode d’occupation de la terre, […] une parcelle de terrain’; Skt ksé-ti, pl. ks- y-ánti = , Av šaēiti, šgeinti ‘habiter’) Alb katun.. DEX.RO Turkish kat ‘floor’ < possibly early PIE borrowing PIE MA *trēbs ‘dwelling’; DRom n. treabă ‘[woman’s] work around the house’, pl. treburi ‘chores’ Cogn.: OIr treb ‘habitation’. atreba (*ad-treb-) ‘dwells’; Lat trabs ‘wooden beam’; ON þorp ‘farm, estate’ [whence NE place names in—thorp]; Lith trobà ‘house, building’; Grk téramna~téremna ‘house, dwelling’; Osc triibùm ‘domum’, tribarakkiuf ‘aedificium’; Um trebeit ‘versatur’ (Buck Grammar, 1904) PIE IEW 146-150 *bhueh2 ‘to be, thrive/dwell’; ~ *bur- ; *buro- ; DRom n. bordei ‘underground house’ (Holm: JIES v 39, no. 1–2, 2011, p. 71). Cogn.: Germ Bauer; Messapic βύριον οἴκημα ‘house, dwelling; Alb *bhur- ‘house’; OFris bold, bōdel ‘house, household utensil, household appliance, property’, OIce. bōl n. ‘dwelling; den (of animals)’; MIr baile ‘home, place’; (*bhuǝ-liio-); OIce būr ‘pantry’; OE būr ‘cottage, room’, OHG būr m. ‘house, cage’, (?) PIE MA *stup- ‘stump, broom, club’; IEW 1011-1013 *steb(h)- and stēb(h)- : stǝbh-, nasalized stemb(h)-; step- (also stēp-?), nasalized stemp-; nominal stǝbho-s, stemb(h)ro-s, tomb(h)o-s ‘post, pillar, stump; to support’; IEW1020 *stelb-, stelp-; stelbo- ‘post’; DRom n. stâlp ‘post, pillar; n. stup ‘beehive’.
306 Appendix
Cogn.: Gmc.*stapp-: OIce stǫpull m. ‘tower’, OE stapol m. ‘stem, column’, MLG stapel m. ‘column’; Watkins: *stel ‘to put, stand, standing object of place’: II. O-grade *stol- + no ‘stall’, IV, extended form *steld ‘stilt’; Alb shtalb [m] (g) ‘post, stud’, PAlb *śtelP- < QIE *stolP- (AE 376); OIce stolpi, MEng stulpe, MLG stolpe ‘balk, beam, post’; MLG stũlpen ‘column, pot cover’; Ltv. stùlps, OCS stlъpъ ‘column’, Rus stolp; Derksen: *stьblò n. ‘stem, stalk, trunk’; SerbCr stьblo ‘stem, stalk’; E Ru. stebló (dial.) ‘stem, stalk’; SerbCr stáblo ‘tree, tree trunk’; DEX.RO Slv stlŭpŭ? PIE IEW 860 *reid- ‘to lean on, support’; DRom v. rezema, n. reazem, razem, ‘prop, sustain, lean on’ Cogn.: Grk ἐρείδω ‘rest, support, urge, strain me’ (Hom. ἐρηρέδαται for -ριδ-), ἔρεισμα ‘pad’, ἀντ-ηρίς -ηρίδος ‘buttress, pad’; Lat. ridica ‘through fissures’ PIE IEW 626 *kÞei- ‘to settle’; DRom n. sat, pl. sate ‘village, settlement’ (e > a in DRom); Cogn.: OInd ksḗti, ksiyáti ‘stays, dwells’, Av šaēiti ds; OInd ksití-, Av šiti- ‘residence, settlement’, OInd ksḗtra-, Av šōiϑra- n. ‘estate, residence’, OInd ksēma- m. ‘quiet, peaceful staying’; Maybe Alb (*ksati) fshati ‘village’; Rom sat ‘village, countryside.’ See above page 155 for more details. Another solution: MA *sed- ‘sit’; IEW 884-889 *sed ‘to sit’, *sed-to ‘sit, settlement, chair’; sed-ter ‘seat, settlement?’; sed-ti ‘gathering’; DRom n. sat, pl. sate ‘village, settlement’ (e > a in DRom); Cogn.: NWels sedd ‘seat’; OIr saidid, said ‘to sit’, and OIr sid ‘peace’: Grk hédos ‘seat’; Av hadiš ‘home’; Skt sádas ‘place’, sad- ; sādayati ‘to sit/place down’; satti, sadas ‘seat, place to stay’; sattá- ‘sitted’; ppp. satta’to ‘sit down, settle down’; sat ‘existence’? Av pasuš-hasta- ‘hurdle’ (*settlement); Av had- ‘I sit’; Lat sedeo, ere, sedi ‘to sit’; OIsl set ‘raised on ground’; Etruscan sath-, śat- ‘to put, establish’ (B-B 2002, p. 218). Unknown origin. Sounds Indo-European *sed. Cf. śuth-, sut- as well with apparent metaphony (Fournet, 2012, p. 9)
Appendix
307
As per Lehmann ‘A Gothic Etymological Dicitonary’ 1986, p. 296: PIE *sod- sed- ‘sit’; Skt sadayati ‘set’; Av ni- šāδayeiti ‘set down’; OPers niu-āšādayam ‘I set back’; OIr suid- as in suidigir ‘place’; Lith sodinti, OCS saditi ‘set, plant’; Goth: ‘S27 *satjian ‘set, place, determine’; at-satjan ‘present’; bi-satjan ‘set around’; *ga-satjan ‘place, establish’ DEX.RO Lat (based on Byzantine Greek φουσσάτον ‘citadel, army establishment’ in use by the Roman army in fifth century) > Lat fossa ‘ditch’ < PIE *bhedhe ‘to dig, ditch’ > Alb fshat ‘settlement’ > DRom fsat (present only in Psaltirea Scheiana, sixteenth century; the Albanian linguists are divided on this etymology, some considering it from PIE *k̑þei‘to settle’, Alb *ksati > fshati ‘village’, pl. fshatra ‘villages’; maybe from Skt víś- (RV+) ‘settlement, community, tribe’ derived Alb (*víś) fis ‘settlement community, tribe’ which advance the possibility of a compound word of fis + sat; Alb vis ‘territory, spot, place’; Rom. sat (IEW), Av šaēiti, ‘stays, dwells’, Skt ksḗtra ‘estate, residence’, Grk κτίσις ‘a founding, foundation, a doing, an act, a creating’, Arm šēn ‘inhabited farm, village’. Note: ARom uses for’village’ hoară < Grk χωρα, χωριό ‘village’; in ARom we have fusatu meaning ‘ditch’ showing phonetic form f-u-sat, since both DRom and ARom do not accept the sifflant group fs-, and no other examples are available. The solution *fusat > DRom sat becomes very difficult to accept. PIE MA *plut- [*pléu̯eti ‘float’] ‘plank’; DRom n. plută ‘rafter on rapid waters’ v. pluti ‘float’ Cogn.: Lat plŭtĕus ‘movable penthouse, shed’; Lith plaùtas ‘plank’. DEX.RO OCS plovetŭ; Serb plutati ‘float’, pan-Balkanik? PIE IEW 128-132 *bher-1 ‘to bear, carry’ DRom n. zbor ‘fly’, v. zbura ‘to fly’ Cogn.: Thrac zbel- (from an initial *zibel- [z(i)burul in Harvey E. Mayer) ‘shining; a thunderbolt, a lightning’ Zibelthiurdos (Zibelthurdos or Zbelsurdos) was a Thracian god of storm and lightning; Latv zibele ‘shining’, zibelêt ‘to flash, to twinkle, to shine’; Lith ziburys ‘spark’; IEW: “Alb abbreviated (*nde-bor) zborë ‘snow’: Rom v. zbura ‘fly, fly out, flight, flit, flash, dash, flee,’ n. zbor ‘flight, soar, soaring, fly, flying”.
308 Appendix
PIE MA *sphaen ‘flat-shaped piece of wood’; DRom n. pană ‘shim’ (homonym pană ‘feather’) Cogn.: Lat sponda ‘frame of bed, bedstead’; NE spoon/spade. PIE MA *k̑súlom ‘worked shaped wood; post, stake’; DRom n. sul ‘cylindric wooden tube used for rolling up any material’ Cogn.: Grk ksúlom ‘wood’; OHG sūl ‘pillar’; Lith šùlas ‘wooden post, stake’; Bur *súli ‘tube, pipe’ DEX.RO Lat sub(u)lum ? PIE MA *ĝhalgheha ‘pole, stake’; IEW 411 *ĝhalg(h)- (flexible) ‘twig’; DRom n. ghioagă ‘club, mace for fighting, pole’ Cogn.: NE gallows, Lith žalgà ‘long thin pole’; Goth galga m. ‘picket, pole’, OIce galgi ‘gallows’; OE gealga, OFris galga, OS, OHG galgo ‘gallows’, in addition further formations OIce gelgia ‘twig, branch, shaft, pole, stick’ PIE IEW 206-211 *der-, heavy basis *derǝ-, *drē- ‘to cut, split, skin’ (*the tree) derived from root *deru-, dōru, *dr(e)u-, *drou-; *dreuǝ- : *drū- ‘tree’; DRom dărâma v. ‘to demolish, destroy’ (Russu, 1981, p. 307) Cogn.: Skt dáríman- ‘destruction’, darmán ‘breaker’. Grk δέρμα n., next to which from the heavy basis dárīman- ‘destruction’; maybe Alb (*düras) dërrasë “board, plank (cut wood)”, dërrmonj “destroy, break, exhaust”. DEX.RO Lat *deramare?
Clothing and Textile PIE IEW 111 *bhasko- (*bhedh-sko) ‘bundle, heap’; DRom n. bască ‘hat’ Cogn.: Maked βάσκιοι δεσμοὶ φρυγάνων and βασκευταί φασκίδες (these genuine Gk. vowel form), ἀγκάλαι Hes.; perhaps here Gk. φάσκωλος ‘leather sack’; Lat fascia ‘bandage, band, girdle, girth, strap, land stripe’, fascis ‘alliance, bundle, parcel; the fasces with excellent hatchet as a token of the imperious power’; maybe Alb bashkë ‘together, bound’, bashkonj ‘put together, unite’, bashkë ‘fleece (a bundle of wool)’ PAlb *baśka̅; MIr basc ‘collar, neck-chain’ (AE 93)
Appendix
309
PIE IEW 516 *kadh- ‘to guard; to cover’; DRom n. căciulă ‘cap, hat’ (Russu, 1981, p. 286) Cogn.: Lat cassis, -idis ‘helmet’ (if echtLat., s. WH. I 177, then from *kadh-tis); OHG huota ‘hat, guard, watch’, OE hōd; OFris hōde ‘care’; OHG huoten, OE hēdan ‘look after, watch over, keep, guard, watch’; OHG huot ‘hat, hood, helmet’, OE hōd ‘cap’; OIce hǫttr, hattr, OE hætt, Eng hat, OIce hetta ‘cap’; maybe Alb. kas-, kasolle, kësulë (diminutive) ‘cap, hat’ PIE MA *p(e)h2no/eha ‘linen cloth’; IEW 788 pān- ‘fabric’; DRom n. pânză ‘linen’ (pân+ză? − pan + ză associative suffix) Cogn.: MIr anan ‘linen cloth’; Lat pannus ‘piece of cloth, garment’; Grk pė́ne ‘thread on the shuttle’; de Vaan 2008: Grk πηνος [a], ττήνη [f.] ‘woven cloth’ (Hsch.); Grk πηνίζομαι ‘weave.’ PIE IEW 951-953 *(s)keu-2, *(s)keuǝ, *(s)kū- ‘to cover, wrap’; DRom n. scutec ‘cloth diaper’ Cogn.: OInd skunüti, skunōti, sküuti ‘covered’; Lat. obscūrus ‘dark’; OE scu(w)a m. ‘shadow, darkness, protection’; OIce skaun f (or skaunn m.) ‘shield’; Nor skūme ‘dark’; Grk σκῦτος n. ‘skin, leather’, κύτος n. ‘skin’; Lat cutis ‘skin’; Balt *keutü ‘skin’ DEX.RO Blg skutak? PIE MA *andher-, ņdher- ‘stem, spike’; IEW 41 *andher-, ndher- ‘stem, spike’; DRom n. andrea, undrea ‘knitting needle’ Cogn.: Grk ανθεριξ ‘stalk point, stalk’, ἀνθρήνη, ἀνθρηδών ‘wasp, forest bee’. PIE IEW 520 *kais- ‘hair’, compare above kéśa- under kaik-; (520 kaikor koik-: to scratch, itch, comb, *dress (the hair), cut the hair about; DRom n. caier ‘distaff’ Cogn.: Lat. caesariēs ‘hair of the head’ (Rhotacism did not occur to avoid two r); OInd kēsara- n. ‘hair, mane” (s instead of s from a form *kēsra); Toch. A śiśäk, В secake ‘lion’; Lat caesariěs ‘long hair; plume’ (de Vaan, 2008, p. 81—Probably formed on the basis of an r-stem *caesary which might be identical to the name Caesar. DEX.RO Lat *caiulus?
310 Appendix
PIE MA *bhṛw ‘(bolt of ) cloth’ OE byrst; OHG bursta fem. ‘bristle’; IEW (108-9): *bhar-, bhor-, bhr- ‘bristle, stubble, stcking out’; Kluge: Barte ‘broad axe’; MidHG barte, OHG barta fem.; ODu and OSax barda, OIc barða, OFr barde ‘hatchet’ borrowed from Teut., OSl fem. brady ‘axe’. To be noted: the Old Norse skegg + ox ‘bearded axe’, OIc skeggja ‘broad axe’; there could be a metaphorical connection with the Lat barba ‘beard’, since the Old North called this tool a bearded axe, or just some other unclear connection, shaving tool? MA *bhardh-eha‘beard’, IEW 110: bhardhā ‘beard’ PIE MA 1. *k̑os-trom/dhrom ‘knife, dagger’ < *k̑es- ‘cut’; DRom n. custură ‘knife, any-blade, plow blade, sickle blade, and the mountain peak’; possibly n. cuţit ‘knife’ Cogn.: Lat castrō, (castrāre with depalatalization) ‘I prune’; Alb thadër ‘adze’; Skt śástra- ‘knife, dagger’; Skt kuţhāra- ‘axe’; Lat culter ‘(butcher’s) knife’; Derksen (2008, p. 238) *kos- Slk. kostura ‘big knife’, Ukr kostúra ‘knife for slaughtering animals’ perhaps a loan to Slavic or a merger. PIE IEW 111-112 *bhāt- : bhət- ‘to hit, strike’; DRom n. băţ ‘stick’, n. bâtă ‘club’; possibly v. a bate ‘to hit, strike’ < Lat battuo ? < OIr bathach ?; developments: n. bătaie ‘fight’, n. bătălie ‘battle’ Cogn.: Gaul Lw Lat battuō, -ere, more recently battō ‘to beat, knock’, Illyr Batto ‘appellation for rebellion leaders’, Gaul anda-bata ‘blind combatant, gladiator fights with a helmet without openings’; OE būtel ‘hammer’, MLG botel ‘hammer’, MHG bæzel ‘beetle, hammer’; OIce beysta ‘knock, hit’ (*bhaud-sti-, compare Lat. fūstis (*bhūd-sti-s) ‘a knobbed stick, cudgel, staff, club’); with –sk- suffix, perhaps MHG būsch ‘cudgel, club, blow, knock’ (*bhūd-sko-) (Fournet) OIr bacc ‘staff’; Lat baculum ‘staff’; Grk báktron ‘staff’; IEW 93 Pisani (REtIE. 3, 53) places baculum as *batlom to battuō, that he considers as Osc –ur ModBret Lw (from *bakt-). PIE MA *gwréhx-w-on- ‘quern’; IEW 405 *gred-, grod- ‘to scratch’; DRom n. gresie, ARom grease, greasă ‘quern’, MglRom, IstrRom ‘id’
312 Appendix
Cogn.: OIr brāu ‘quern’; Lith girna ‘millstone’, pl. grinos ‘quern’; OCS žrŭny ‘quern’; Arm erkan ‘quern’; Skt grā́van- ‘stone for pressing soma’; Toch B kärweñe ‘stone’; Alb gërrusë, gërresë, krūs(ë) ‘rasper’. PIE MA *dhúbhos ‘wedge, peg’; DRom n. dop ‘cork’; v. îndopa ‘to stuff’; n. tapă ‘spigot’ Cogn.: NE dowel, [tap (as keg), as per JColarusso pers comm]; Grk dial. túphos ‘wedge’; Lat tappum > Alb tapë ‘cork’; It tappa; Catalan tap; perhaps cognates with Germanic: Swed. Norw., tap; Danish tap; OIsl tappi ‘cork’. for DRom n. tapă ‘scratch, small hollow’ see Cevelova-Blažek (2009, p. 158): “Goth. tappa ‘bung, spigot’; cf. WGmc. *tappōn > OEng. tæppa, MDutch, MLG. tappe, OHG. zapfo ‘spigot’ (EWD 903); Rum tapă ‘scratch, small hollow’, Kat., Sp. tapa, Pg. tampa ‘cover, flap’; derivatives: Kat., Sp, Pg. tapar ‘clog, cover up’, Pg. tapadura ‘enclosure’, Baros. atupir ‘to choke the fire with earth or stones’ (ML 8565)”. PIE MA *kert- ‘wickerwork’; DRom n. cursă ‘snare, trap (made of wicker basket) < Grk kurtia ‘wattle’? Cogn.: Lat crātis ‘wickerwork, hurdle, honeycomb’; NE hurdle; OPrus corto ‘hedge’; Grk kártallos ‘basket’, kurtia ‘wattle’; Alb kurth, pl. kurthe ‘snare, trap’. PIE MA *ko(n)gos ‘hook’; IEW 537-538 *keg-, keng- and kek-, kenk‘hook, grappling hook, handle’; DRom n. cange ‘hook, gaff’ Cogn.: MIr alchaing ‘weapon rack’; NE hook; Rus kógotı ̆ ‘claw’; Hit kagas ‘tooth’; Pers čang ‘claw, nail, fist’ (*kengo-); MIr ail-cheng f. ‘rake’; Lith. kéngė ‘hook’. PIE IEW 930-933 *(s)kēp-2, (s)kōp- and (s)kāp-; (s)kēb(h)-, skob(h)and skāb(h)- ‘work with a sharp instrument’; DRom n. scoabă ‘clamp’, v. scobi ‘to hollow, scoop’ Cogn.: Lat. scabō, -ere, scübī ‘scrape, scratch, rub’, scůbiēs f. ‘scab, mange, itch’; Goth. skaban ‘scrape, shave, shear’; OCS skoblь ‘grater’; Engl coppice ‘crang’ < medieval latin colaphus m (genitive colaphī); a blow with the fist, cuff < Grk κολάπτω (koláptō, ‘to peck’), a Pre-Greek
Appendix
313
word according to Beekes, who doubts connections to Proto-IndoEuropean *kelh2- (‘to beat’). PIE IEW 1058-1059 *tekÞ- (*tesḱ-) ‘to plait; woodwork; carpenter’; DRom n. țesală ‘animals comb, especially for horses’ Cogn.: OHG dehsa, dehsala ‘hatchet, hack, mattock, hoe’; OIce þexla ‘axe’; dehsala ‘axe, adze’ (Derksen: *tesàti v (b) ‘hew’; OCS tešo; tesati ; Lith. tašýti ‘hew’; Skt. tákșati ‘cut, manufacture’; RuCS *tesla; *teslò ‘adze’; Lat. texō, -ere, -ui -tum “flax, wattle, braid, to weave, to build” (*tekslü = Slav. tesla, OHG dehsala); DEX.RO Srb Cr česlo > DRom țesală, a possibility, but the Rom form has accent on țes’ală, which could indicate the OHG dehsala as a better solution; a good case of contamination also with Lat Lat. texō, -ere, -ui –tum > DRom v. țese ‘to weave’ PIE IEW 555 *kem- ‘to press, squeeze’; DRom n. ham ‘rein, harness’ Cogn.: Goth *haman- ‘rod, shaft’—ham n. ‘horse collar’ shaft’; E hame ‘horse collar’; MDu hame, haem ‘id.’, Du haam ‘id.’; MHG ham(e) m. ‘fishing rod’ ~ *Jiom-o- (IE)—Skt samyci- ‘yoke pin, plug, wedge’ < *Ke/ om-feh2-; YAv. sfmci- 1. ‘yoke pin’ < *Km-feh2-; Grk Kaμa~, -aKo ?Rom zară ’butter-milk’; Proto-Alb *dzalā- (< PIE *g̑lh2-éh2 ‘whiteness’); Grk γάλα ’milk’ PIE IEW 251-252 *dher-1, dherǝ- ‘dreg, cloudy sediment of a liquid, verbal: stir up sediment and mud, cloud’; Lat serum ‘whey; any similar fluid’ [n. o] (Cat.+) Pit. *sero-[n.] ‘liquid’; PIE *ser-o- [n.] ‘flowing, liquid’; IE cognates: Skt sara- ‘flowing, liquid’ < *sero-,punah-sara- [adj.] ‘running back’, Pr sisarsi ‘flows, runs’; Grk ορός [m.] ‘the watery part of curdled milk, whey’ < *soro- cflower, runner’; Lat serum could be cognate with Skt sara- and Skt si-sar- ‘to flow, run’, < derived from PIE *sel- ?
Appendix
317
PIE MA 1. *bher- ‘well, yeast, bubble’; IEW 132-133 *bher- *bh(e)reu‘boil, swell, well up’; IEW 167-168 *bhrendh- ‘swell, sprout’; DRom n. brânză (+ză associative particle) ‘cheese’. Cogn.: OIr probably brenn- (*bhrendh-uā-) `spring up, bubble, effervesce’; maybe Alb brenda, përbrenda ‘inside, inward (*inward inflow)’, brendësi ‘inside, entrails’; OPrus pobrendints ‘weighted’; Lith brę́stu, bréndau, brę́sti ‘swell, ripen’, participle bréndęs `ripening’, brįstu, brìndau, brį́sti `gush, well up (e.g., from peas)’, brandà `ripeness, rich harvest’, brandùs `grainy’; Lett briêstu, briêžu, briêst `gush, well up, to swell, ripen’, bruôžs `thick, strong’; PIE MA 1. *sem- ‘draw water’; DRom n. zeamă, zamă ‘liquid (squeezed from), juice, broth’ Cogn.: Lat sen-tīna ‘bilge-water’; OIr do-essim ‘poors’; Grk ámē ‘bucket’; Lith sémti ‘draw water’, 2. *ĝheumn- ‘libation’; DRom n. zeamă, zamă ‘liquid, juice’ Cogn.: Phryg zeumán ‘libation’; Grk kheûma ‘that which is poored’; Skt hóman- ‘libation’, 3. *deh2 ‘to flow’ (with an m- derivation only in Balkan languages); DRom n. zeamă, zamă ‘juice, liquid, broth’ Cogn.: Grk δημός ‘fat of creatures’, ζυμη ‘leaven’, ζωμός ‘soup’; Arm tamuk ‘wet’; Alb djamë ‘fat’ [Holm: JIES vol 39, no 1–2, 2011, p.45); and AE dhjamë [m/n] (tg) ‘fat’, PAlb. *δem(-) (AE 161) _−_ (Pok. 175) Gr. δημός ‘fat’] PIE IEW1031-1032 *(s)terk- *(s)ter-8 ‘dirty water , mud, smear’; DRom n. terci ‘hominy, mush, gruel, mash of corn flour’ Cogn.: Welsh troeth ‘lye, urine’; Bret. troaz ‘urine’ (*troktü or *tronktü); Lith. teršiu, teršti ‘smudge’, apteršti ‘smudge’, tiršti ‘become thick’; Lith. trąšà ‘manure’, trèšti ‘rotten’, Nor. traa, OE ðrōh ‘rancor’; most likely develop. from Lat terō, -ere, trīvī, trītum ‘rub, grind’.
318 Appendix
ocial Organization, Getting and Giving, Law, Order, S Strife, Warfare PIE IEW 235 *dhau- ‘to press, squeeze, strangle’; DRom n. Dacian collective ethnonym tribe name, Lat Daci; Phryg δάος; DRom dulău n.m. ‘big aggressive dog, shepherd dog’; dolcă n. f. ‘bitch’ Cogn.: Phryg δάος ... ὑπο Υρυγων λύκος Hes. therefrom the people’s name Δᾶοι, Dāci), Lyd Καν-δαύλης (κυν-άγχης ‘Indian Hemp, dogbane, plant poisonous to dogs’), compare Καν-δάων, name of Thrac god of war, Illyr PN Can-davia; dhaunos ‘wolf ’ as ‘shrike’; Lat Faunus; Grk θαῦνον θηρίον (Hesykhios); Illyr Daunus there from Δαύνιοι, inhabitant of Daunia; compare Thrac Δαύνιον τεῖχος); Gk. Zεὺς Θαύλιος i.e. ‘shrike’; with ablaut Grk θώς, θω(F)ός ‘jackal’, maybe Alb dac ‘cat’; Phryg. δάος; Goth af-dauiÞs ‘rended, mangled, afflicted’; OCS davljǫ, daviti ‘embroider, choke, strangle’, Russ davítь ‘pressure, press, choke, crush’, dávka ‘crush’ Georgiev (1960): Dacian plant name δαχινα (Diosc. 4, 16, nv [W. 2, 183]) λυχου χαρδια ‘wolf heart’ IE *dhău-k-ino adj ‘of wolf ’ < *dhăwo-s ‘wolf ’ > tribe’s name Δαχοι, IE. *dh ~u-ko—‘wolf ’; older name Δαοι; περσον∋σ ναμε Dăvos, Daus < IE *dhăwo-s ‘wolf ’ > town Δαουα–δαυα in Moesia Inferior. Another solution: IEW 247 dhelg- ‘to stick; needle’; DRom dolcă ‘bitch’, dulău ‘big dog’ Cogn.: OIr delg n. (es-stem) ‘thorn, cloth needle’, Corn delc (i.e., delch) ‘a necklace, collar [for horses and other animals]’, MWelsh dala, dal ‘bite, prick, sting’; ON dalkr ‘needle to fasten the mantle about the right shoulder; spinal column of fish; dagger, knife’, OE dalc m.’clasp, hairpin’ (Germ Dolch, older Tolch, NGerm dolk, after Mikkola BB. 25, 74 the origin of Cz. Pol. tulich, Sloven. tolih, is namely borrowed at first from Lat. dolō ‘a pike, sword-stick; a small foresail sword-cane’, but perhaps reshaped after Gmc words as OE dalc) DEX.RO Polish dolow ‘pits, down’? > DRom dulău ‘big dog’? PIE IEW 632-633 *k̑uon-, *k̑un- dog (*animal with a strong sense of smell) Thrak-Phryg striking Lyd. Κανδάυλης “κυν-άγχης”, Thrac. GN Κανδάων (2nd part *dhāu- ‘strangle, throttle, choke’) with DRom n. câne ‘dog’;< Lat cane
Appendix
319
PIE MA *bheh2- ‘shine’ IEW104-105: bhā-, bhō-, bhə- ‘shine; DRom n. ban ‘small coin, money’ Cogn.: OIr bān ‘white”; OE bōnian ‘ornament, polish’; Goth bandwa, -wō ‘mark, token, sign’, bandwjan, OIce benda ‘give a mark, token, sign’; Alb bej ‘do’[< * ‘bring to light’], Grk phaínō ‘bring to light’; Av bā ‘shine’. PIE IEW 96-96 *bend-, bnd-no- ‘spike, needle, summit’; DRom n. ban ‘governor of a region’ Cogn.: al: MIr benn ‘horn, summit’ (*bnd-no- or *bend-no-?), bennach ‘pointed’, Welsh bann m. ‘hill, summit, horn (*bnd-no-)’, MBret ban ‘eminence, overhang, haughtiness, pride’; Lat. penna ‘feather’; Srb ban, Hung bán (unkown etym.) ‘governor’; Delamarre (2018): “Le thème Banno-/ Benno- ‘pique, pointe, corne’ est bien attesté dans l’onomastique galloromaine: Bannus, Banna, Banno, peut-être le Deus Cobanus (AE 2000, 1847), c.-à-d. *Co-banno-s ‘dieu à la corne’ *bhendh-no- / *bh dh-no- > benno- / banno- et il se continue dans l’irl. benn ‘sommet, corne, pointe’, gallois bann ‘id.’ De son côté, l’Augustus Deus Cobannus, c’est-à-dire *Co-banno-, plutôt qu’une variété de forgeron du type Goibniu (*Gobann-i on-) 9, serait une divinité priapique, ‘doté d’une corne’, de sémantisme comparable au Jupiter Bussumarus de Dacie, c’està-dire *Buđđu-māro-s ‘au grand pénis’” Poruciuc (2008): OGerm bann- and band- ‘proclamation, prohibition, decree’, Croatian title ban; It bano; DRom developments: v. bănat ‘accusation’, v. bănui ‘to suspect’ and v. bântui ‘to punish’; Romanian family names Ban and Banu. IEW 105-106 bhā-2 ‘to speak’; DRom n. ban ‘governor of a region’ Cogn.: Arm. ban (*bhü-nis), gen. -i ‘word, speech, reason, judgement, thing’, bay, gen. bayi ‘word, verbalism’ (*bhǝ-ti-s = Gk. φάτις); bay particle ‘(he, she) says’ (= φησί, also bam = φημί, bas = Lesb. φαι from *bhüsi); OIce bōn, bøn ‘request, prayer’, OE bäen ‘request, soccage’ (‘*bhü-ni-s; or with ō gradation as Gk. φωνή’?); OE bōian ‘brag, boast’ (as Lat. fōr from *füiōr, Slav. bajǫ); At a present *bh-en- based on O.Ind. bhánati ‘speaks’; auf *bhǝn-u- (or auf Gmc. reshuffling after spannan)
320 Appendix
O.H.G. bannan redupl. verb. ‘summon by proclamation (esp. to arms); curse or damn; pronounce an ecclesiastical curse upon’, OE bannan redupl. verb ‘summon, order’, OIce banna schw. Verb. ‘forbid’, whereof O.H.G. ban, PL banna ‘order under penal threat’ (Ger. Bann, Bannwald), OE gebann, OIce bann n. ‘forbid, ban’. PIE IEW128-132 *bher- ‘to carry, to move’; DRom n. bir ‘tax, burden’ Cogn.: Grk φέρω, Lat ferō, OIr 1 sg. biru, -biur, 3 sg. berid ‘bear, carry’; Alb bie, Goth. baira, OCS bero; OInd bhára-h ‘acquiring, carrying off, profit, gain, booty; burden’; Maybe Alb. barrë ‘burden’; OInd bhüra-h ‘bundle, work, load’. Considered by some Romanian linguists of Turkish etymology, but in Clauson An Etymological Dictionary of Pre-Thirteenth-centuryTurkish: bir anqa …. bir anqa ‘some . . .some’ 129; bir anqa ‘a certain amount, for some period’ 234, 9.54: XIII(?) At. bir anqa ‘a certain number of people’; bir anqa ‘a certain number of ’; bir means ‘one’ (wahid) and anqa ‘like it, some’; bir kiși ‘every single man’: DEX.RO Hun bér ‘wage’—of uncertain etymology in the A Magyar Nyelv Törteneti-Etimologiai Szotarat-Etimologiai Szotara. PIE MA *wet- ‘see (truly); IEW 1113 *u̯āt-, u̯ōt- ‘spiritually excited’; DRom n. m. vătaf ‘leader of a group of lads, governor over a region, land’ Cogn.: OIr fethid ‘sees, pays attention to’; Lat vātēs ‘seer, prophet’; Skt ápi vatati ‘is familiar with, is aware of ’; derived forms include *wō’to(true) knowledge, shamanic wisdom’, OIr fāth ‘profetic wisdom’, OE wōþ ‘song, poetry’, wōtó ‘having (true) knowledge’ > OE wōd ‘furious, frenzied’ (> NE wood ‘mad’) and wōtonó ‘who incarnates’; *wōto seen in Germanic divine names of OE Woden, ON Oðin. PIE MA *h1ep- ‘take, seize’; DRom v. apuca ‘grab, seize, take’ Cogn.: Alb jap ‘give’; Arm unim ‘possess’; Hit epzi ‘takes’; Av apayeiti ‘obtains’; Skt āpnóti ‘obtains’; Toch yapoy ‘land’, [*’dominion’]; Lat co-ēpī ‘seized’
Appendix
321
IEW 50-51 ap-1 (proper ǝp-) : ēp- to take, grab, reach’; DRom v. apuca ‘to grab, seize’ Cogn.: Grk ἅπτω ‘give a hand. touch, pick a quarrel, light, kindle’, ἅπτεσθαι ‘touch’, ἁφή ‘touch, adherence etc.’; Hom. ἀφάω (ἁφάω) ‘touch, palpate, feel, finger’, Lat. apīscor ‘touch, reach, attain, come to, come by’, adipīscor ‘to come up to, overtake; hence to obtain. perf. partic. adeptus, used passively, = obtained’, coēpi ‘has begun, commenced’, later coepī; OInd äpnōti ‘achieved, attained’, more recently äpta-h ‘clever, suitable, trusted’; Av apayeiti ‘achieved, reached’; Arm ap ‘hollow hand’ DEX.RO Lat aucupare? PIE IEW 107 *bhag- ‘divide, apportion’; DRom v. a băga ‘to put in, insert’, sg. 1st pers. eu bag ‘I insert in bag or any container, only of solids, not liquids’ Cogn.: Phryg Βαγαῖος Ζεὺς Φρύγιος Hes (perhaps of Iranian origin), bagun ‘gift’ < *bhagom; Skt bhága ‘apportion’, bhaaga ‘part’; Av baga‘good fortune’; Slv bogŭ ‘god’; Toch B pāke ‘share, part’; Grk phageı ̑n ‘eat’; Av bag- ‘distribute’; Skt bhajati ‘divide, distribute, place’; known as a divinity: Rus bog ‘God’, Av baga, ‘god’; Skt bhága- ‘lord’; Toch pāke ‘apportion, portion’; OInd bhajati ‘allocates, apportions, divides’, bhaga-h ‘property, luck’, Av baga-, baγa- n. ‘favorable interest, attractive lot’; OInd bhaga-h ‘allocator, master, mister’, Av baγa- ‘master, mister, god’, Pers baɣ ‘god’ : OCS bogъ ‘god’ (formal also Gk -φάγος) PIE MA *kap- ‘have, seize’; DRom n. hapcă ‘fishing rod with big hook’; v. hăpăi ‘gulp dwon’; interj. hap (onomatopoeic?) ‘catch with mouth’ Cogn.: Alb kap ‘catch, grab, seize’; Lat capiō ‘takes’; Skt kapatī [dual] ‘two handfulls’; Latv kàmpju ‘seize’; Grk káptō ‘gulp down’; Goth hafian ‘take, get, obtain’; Bg hapka ‘small bite’. DEX.RO Bg hapka ‘small bite of food’; Derksen (2008): OCS xapъjǫšte ‘biting’; RuCS xāpati ‘seize, sting’
322 Appendix
PIE IEW 407-409 *ghabh- ‘take, seize’; DRom v. găbji, găbui ‘seize’ Cogn.: OIr gaibid ‘takes’, *gabie-, Gaul *gʰabʰ- > *gab- ‘take’; Goth gabei ‘richness’; Lat habeo ‘have’; Lith gabenu ‘present’; Pol gabač ‘seize’; Skt gabhastin- ‘hand’. DEX.RO Ukr habaty ‘seize’ merger?; in Derksen (2008): gabati, habati ‘seize’; Bulg gábam ‘grieve’; IEW (LIV) classifies *gabati under *ghebh- ‘take, give’, from which it also derives Skt. gábhasti‘hand’; Goth giban ‘give’; Lith. geběti ‘be able’, gabénti ‘transport, remove’.also *ǵheHb- ‘take’, which is claimed to be represented in ItaloCeltic only, e.g., Lat. habēre, OIr. gaibid ‘take seize’. PIE MA *nem- ‘apportion, count, take, accept legally’; IEW 763-764 *nem-1 ‘to take; to put in order, count’; DRom n. neam ‘relatives, nation, people obeying the same laws’ Cogn.: OIr nem ‘gift’; Lith núoma ‘rent’, numai ‘house’; Av namah ‘loan’; Toch B ñemek ‘harvest’; Grk némō ‘distribute, possess’; nomos ‘law, custom’, νομεύς ‘herdsman, shepherd’; NHG nehmen ‘take’. IEW 764 *nem- 2 ‘to bend’ > ‘sacred grove, veneration’; DRom neam ‘nation’ meaning ‘people venerating the same gods, ancestors, sharing the same laws on the same pasture’ Cogn.: OInd námas- n. = Av. nǝmah- n. bowing, bending = worship, veneration, homage; Grk νέμος n. ‘grove’ = Lat. nemus, -oris ‘pasture’, etc.; Gaul. νεμητον “ sacre pasture’, PN Nemeto-brigü, VN Nemetes, OIr nemed ‘sanctuary’, nem ‘sky, heaven’, OIr neimed, Lat nemus, Fris nimidas, Grk nemos assuming ‘bend, bow in reverence, place to honor gods’ DEX.RO Hun neme ‘sex, gender,’ nemes ‘nobil’: of uncertain etymology, possibly of Indo-Iranian influence (A Magyar Nyelv Törteneti-Etimologiai Szotara, 1970) PIE MA *dergh- ‘grasp’; IEW 257 dherāgh ‘pull, drag’; DRom n. targă ‘stretcher’ (old loan from Got? or Germ. Trage ‘stretcher’); devel. v. tărăgăna ‘dally, drag one’s feet’ Cogn.: Alb tërhek ‘pull, drag’; MIr dremm ‘troop, band of people’; ON targa ‘shield’; NE targe ‘light shield’; Grk drássomai ‘lay hold of, grasp with hands’; Arm trc’ak ‘bundle of brushwood’.
Appendix
323
PIE MA *reus- ‘anger, rage’; IEW 867-868 *reu-1, rēu-, rū- ‘to roar, murmur’ (expr.), nomatopoeic words; DRom adj. rău ‘bad’ Cogn.: MHG rūn ‘uproar, make noise, rage, rave’; Skt ros ‘displeases, takes offence at’; Toch B räs- ‘criticize, accuse, object to’; Lat. rūmor ‘noise, shout, call, rumor’; rāvus ‘hoarse’, ravis f. ‘hoarseness’, raucus ‘hoarse’ DEX.RO Lat reus, f. rea ‘party in a lawsuit, defendent’ PIE IEW 466-467 *gʷedh- ‘to push, hit, harm’; DRom n. ghiont, ghiolt ‘nudge, jog’, v. înghionti, înghioldi ‘to give a nudge’ Cogn.: O.Ind. gandh- ‘bump, poke, prick, injure, destroy; Gk. δέννος ‘vituperation, shame’ (*gʷendhno-); M.H.G. quetsen, quetschen ‘hit, bump, poke, squeeze’; Lith. gendu, gésti ‘damage, spoil, IEW 491-493 *gʷhen-2(ǝ)- ‘to hit’; DRom n. ghiont, ghiolt ‘nudge, jog’, v. înghionti, înghioldi ‘to give a nudge’ Cogn.: nominal formation gʷhono-s ‘hit’, gʷhn-tó-s ‘beaten’, gʷhn-tí-s and gʷhn-tiü (?) ‘hit’, gʷhen-tel- ‘bat’, gʷhen-tuo-s ‘killed’ PIE MA *plehak/g ‘strike, punish’; DRom v. plesni ‘strike’ ( *ĝhhawos (> Grk kháos); *ĝhóh1ros ‘free space, gap’; IEW 449 *ĝhēu-, *ĝhō(u)- , *ĝhəu- ‘yawn, gap’; DRom n. gaură ‘hole, cavity’; n. găună ‘hole in the ground’, adj. găunos ‘hollow’; n. găoace ‘eggshell, any small cavity’, n. văgăună ‘small mountain cave, lace in between hills, ravine’; adj. găunos ‘hollow’; n. genu-ne, arch. genoe ‘abis’; NOTE: both DRom n. hău ‘abyss, precipice’ and DRom n. găună (old dial.) ‘small hole in the ground, tree-hollow’ are in the same semantic group as, for example, Grk kháos ‘chaos’ fits well with the DRom n. hău ‘abyss, precipice’. Cogn.: Grk khôros ‘pit, hole’, kháos ‘chaos’, khaūnos ‘unraveling, lax’; Toch B kāre ‘pit, hole’, ko ‘mouth’; NE gorge? DEX.RO Lat cavus > *cavula > DRom gaură; Lat gyrus, -i ‘circle, ring’ > DRom genune ‘abis ‘ PIE MA 1. *pḷth2ús ‘broad’>*pleth2- ‘spread’ related to *pelhak ‘spread out, flat’; *pleh2- ‘field’; IEW 833-34 plā̆t-, plā̆d-, plē̆t-, plō̆t-, plətpləi- ‘broad, flat, wide spread out’; DRom n. plai ‘field’; n. platoșă ‘shield’, Cogn.: Lith platús, Grk platús ‘broad’; Av pərəθu ‘broad, wide’; and: OE flōh flagstone’; Lith plākanas ‘flat’; Grk pláks ‘flat surface’; Sw fala ‘plain’; OCS polje ‘field.’ PIE MA 1. *telhx-om ‘floor (of planks?); IEW 1061 *tel-, telǝ-, telu- ‘flat, flat ground, board’; DRom n. tărâm ‘region, area, poetic: ‘far away region’, tărâmul celălalt ‘underworld land’; Cogn.: OIr talam ‘earth’; Latin tellus, -uris ‘earth’; Skt tala- n. ‘surface, plain, area’, talimam- ‘floor’; Gk. τηλίᾱ ‘dining table’; θάλαμος m. ‘an
Appendix
325
inner room or chamber, surrounded by other buildings; store-room; the lowest, darkest part of the ship, the hold’. 2. *ter- *térmn̥ ‘cross over’ Cogn.: Lat termen ‘border’; Grk τέρμα, -ατος n. ‘border purpose, τέρμων m. “limit, boundary”, τέρμιος ‘end situated, lastly’; Arm t’arm ‘end’; Hit tarma- ‘stake’; Skt tárman ‘point of sacrificial post’. DEX.RO Tc tarum ‘abode’; Hun terem ‘large hall’ PIE MA *wet- ‘entire year’; IEW 1175 u̯et- (*su̯et-) ‘year’ DRom n. vătuiu ‘yearling goat, lamb, hare, cat’ Cogn.: Grk étos ‘year’; Hit witt- ‘year’ (Hit wett-/witt- & wettant-/wittant- c. ‘year’ in Doçkalová & Blažek); Skt vatsá- m (RV) ‘calf, juvenile animal’, vatsá- vatsará- ‘year’; Alb vetul, vetulë, ftulë, ftujë ‘yearling female goat’, PAlb. *ueteś- (AE 417); Thracian god Vetespios (Russu, 1981, p. 418); OInd vatsá, vatsará- m. ‘year’, vatsá- m. ‘yearling, calf, cattle’, vatsaká- m. ‘calf ’, sa-vütarüu NDu f. ‘identical calf ’ DEX.RO Lat *vituleus (vitulus) ‘calf ’ > DRom vițel ‘calf ’ PIE MA *h2enti ‘in front’; DRom adv. înainte, MglRom inte ‘ahead, in front of, before’ Cogn.: Lat ante ‘in front of ’; Lith añt ‘on, upon, at’; Grk anti ‘instead of, for’ Arm ənd ‘for’; Hit anti ‘facing, frontally, opposed, against’; hanza ‘in front of ’ DEX.RO Lat ante, in abante > DRom in-ante, in-ainte ? PIE MA *terh2- *ter ‘bring across, through’; IEW 1074-1075 *ter-4, terǝ: tr-, trā-, teru- ‘to cross, transgress, to stay’; IEW 842 *po-ti ‘against’ extended; Av paiti, ap. patiy preverb.and preposition ‘against’; Greek (Homeric) ποτί ‘to, toward, upon, against’; DRom v. a (se) potrivi ‘fit, match’ compound form of *po-ti ‘against’+ terh2 / trī-vi cross, transgress’; n. potrivit ‘fitted, matched, potriveală ‘fitness, evenness’; DRom v. trece ‘pass, go through’ the REW Meyer-Lübke offers: Lat traicĕre (?) DRom trece (pronounced treče) ‘pass’, Provence tragir, trazir, Waldensian treize ‘swallow’; developments: v. pe+trece
326 Appendix
‘accopmany, go through, pass’, n. petrecere ‘party’; DRom n treaptă ‘stair, step’; OPR trapt ‘step’ Cogn.: Lat trāns ‘across’; Av taro ‘over, to’; Skt tirás ‘over, across, apart’; Gaul trē-, tri ‘by, through’ The established solution for DRom a (se) potrivi ‘fit, match’ relates to Dreksen: *protivъ; *protivo; *protivo; *proti prep. ‘against,’ (see IEW 815) > DRom împotrivă ‘against’, possibly a semantic development. Cogn.: OCS protivъ (Supr.) adv. ‘against’; protivo (Euch.) prep. ‘against, in accordance with’; protivъ prep./adv. ‘against, towards’ WCz. proti prep. ‘against’; Slk. proti prep. ‘against’ Bulg. protív prep. ‘against’; PIE *proti Cogn. Gk προτι?? prep. ‘to, against’ (Dreksen, 2008) DEX.RO Lat trāiciō—trāicere (present infinitive) ‘through, overstep, transgress’ > DRom trece; Slv *protivъ; *protivo; *proti prep. ‘against’ > DRom adv. împotriva ‘against’, v. a (se) potrivi ‘to fit, match’ Lat trajecta (trajectus, a, um) ‘crossing, passage’ > DRom n. treapă, pl. trepte ‘step, stair’ PIE MA *h1énh1u ‘without’; DRom conj. însă < *in+*sen [apart]? ‘but, however’ Cogn.: NHG ohne; Grk áneu; Oss ænæ all ‘without’; and *sen-i-/u‘apart’: OIr sain ‘especially’; Lat sine ‘without’; Skt sanitúr ‘apart from’; Toch B snai ‘without’ DEX.RO Lat ipsa. PIE MA *bhē̆, bhō̆ (an emphatic particle; particles of the protestation and emphasis, bāδa “yea, in truth”, (IEW 113); DRom dial. ba ‘no’, [duplicated] ba nu ‘no, no’, ba da ‘yes yes’ Cogn.: Av. bā, bāt, bē, bōit (the latter, as Lith. beı ̃, probably with strengthening particle *id) particles of the protestation and emphasis, bāδa ‘yea, in truth’, maybe Alb bah ‘absolutely not’; Arm ba, bay emphasizing particle; Goth ba conditional particle: i-ba, i-bai ‘if ’, Conj ‘that not’, ni-ba, nibai ‘possibly not yet?’, Conj ‘if not’, ja-bai ‘if ’, OHG ibu, oba, MHG ob(e) ‘if, whether’
Appendix
327
PIE MA *h4épo ‘back, behind’; *h4ep-ér- ‘after’; IEW 53 *apo-, pō̆, ap-u, pu, pos ‘away from, out of; DRom adv. apoi ‘then, afterwards’, înapoi (în-apoi) ‘back, backwards, behind’; adv. afară outside, out’; fără ‘without, less’ [Gothic influence?] Cogn.: Lat ab ‘from’, Lat. af s. just there 1; abs = Gk. ἄψ ‘back, again’; out of ’; Umbr ap-ehtre ‘from without, from the outside; on the outside, outwardly’; Goth af ‘from, since’; af prefix and preposition m. dat. ‘from, away from, from here’, ON af adv. and preposition m. dat., OE æf, of, OS af,; Grk apó ‘from’; Hit appā ‘behind’; Av apa, Av. ape ‘after’ (m. acc.), compare apaya adv. ‘hereafter, prospectively’, -pe emphasizing particle; Skt ápa ‘away from’; PAlb *apa ‘without, before, then’; Toch A apta ‘earlyer, before’; Goth afar ‘after’; Av apara- ‘behind, following, other’; Skt ápara ‘later’; NE after; Alb pa ‘without’; Hit para ‘outside’, Alb afër ‘near’ DEX.RO Lat *ad+post > DRom apoi ‘then’; Lat ad foras ‘to the outside’ > DRom afară ‘outside’; Lat foras ‘outside’ > DRom fără ‘without, less’ PIE MA *h2erdus, *h2rhdhuos, ‘high’; DRom toponym Ardeal ‘Transylvania’ (compound from *h2erdus ‘high’ and *dhel- ‘valley’? the Latin Transylva could be a translation of the local toponym ‘the high valley, across the forest’) Cogn.: OIr ard ‘high’; Lat arduus ‘high, upright, lofty, difficult’; ON ǫrðugr ‘steep’; Hit harduppi-‘high’, Falileyev et al., 2010: ContIr ardu‘high’. Well attested in G. LNN; cf. also artuaš (Todi) and the Insular comparanda (OIr ard, OB ard, art LEIA A-87, DGVB: 72). 1. IEW 339 *er(ǝ)d- (er/ǝ/d-), er(ǝ)dh-; DRom toponym Ardeal ‘Transylvania’ Cogn.: Gaul. Arduenna silva, OIr ard (*rduo-) ‘high, big, large’; Welsh hardd ‘beautiful’; Kroonen, 2013: PGm *arduga- adj. ‘steep’; ON ǫrðugr adj. ‘id.’, Far. ørdugur adj. ‘id.’ < *h3erdhu-ko- (IE); Skt. ūrdhava- adj. ‘tending upwards, upright, high’, YAv araðβa-, ərəduua- adj. ‘risen, upright, erect’, Gr ορδοσ adj. ‘set upright’, Lat arduus adj. ‘high, steep’, Olr ard adj. ‘high’ < *h3rdhuo-; Probably an old compound of the root *h3er- ‘to rise’ and the root *dhh1- ‘to put’. 2. IEW 332 *er-4 (er-t-, er-u-) [*herĝʷhe] ‘earth’ Gmc. *erÞō in Goth. aírÞa, OHG (etc.) erda “earth”; maybe cogn to TN Illyr. Ardiaei (*er-ĝʷhe) [common Illyr.-Alb. -ĝh- > d- phonetic
328 Appendix
mutation] PGm *erÞo- f. ‘earth, soil, land’—OS ertha f. ‘id.’, Du aarde c. ‘id.’, OHG erda ‘earth’ < *h1er-eh2(?) (Kroonen, 2013) DEX.RO Hun erdély < erdő ‘forest’, < possibly loan from German, OHG erda ‘earth’, Germ ‘Erde’, Engl earth. PIE MA *worhxdhus ‘upright, high’; DRom n. ortoman ‘handsome, man with power, rich (standing out, tall man?)’ Cogn.: Grk (w)orthós ‘upright, standing’; Skt ūrdhvá- ‘upright, high’; Toch A orto ‘from above’; Lat ortus; Umbrian orto ‘rising, standing up’. PIE MA *wers- ‘peak’, IEW 1151-1152 u̯er- (*su̯er-) ‘highland, high place, top, high’; DRom n. vârf ‘peak’ Cogn.: OIr ferr ‘better’[ DRom (č) cingătoare ‘girdle; Alb qingël, qingje ‘girdle’. DEX.RO Lat *clinga ? (clingo, ere- , cingo, clug) *kleng ‘to bend’ PIE MA *haek̑- ‘sharp, pointed’; IEW 18-22 *ak-, ok- , (hekw-), ok̑tō(u) ‘sharp, stone’, IEW 541-542 *kēi-, kōi-, kǝi- ‘sharpen, whet’, *kū- ‘sharp, spear’; DRom n. cață ‘hook’; derivatives: v (a se) cățăra ‘climb’, v. acăța (voiced agăța)’to hang on 29or catch with a hook, to hook’; ARom cățari ‘snatch’; IstrRom (a)coț ‘snatch’
Appendix
329
Cogn.: Latin catus `acutus, sharp witted, shrewd’; Old Irish cath ‘wise’; Lat acer ‘sharp’, acus ‘needle’; Lith aš(t)rùs ‘sharp’; OCS ostrŭs ‘sharp’; Alb athët ‘sour’; Grk akế ‘point’; Arm aseln ‘needle’; NPres ās ‘grinding stone’; Skt áśri ‘[sharp] edge’, DEX.RO Latin *ad + captiāre ‘to snatch’, pt > t rare but attested (?); Lat captio ‘a taking, deceit’? (Alb omonym kacë ‘butt, barrel, bin’< Slv kadьca ‘small bath tub’, Blg kaca, Serb kaca ‘barrel’).
Placement Verbs PIE MA *h1ēs- ‘sit’; DRom v (se) aşeza ‘sit (oneself ) down, lay down (self or something’); n. aşezare ‘settlement’ Cogn.: Grk êsthai ‘sit’; Hit ēsa ‘sits’, āszi ‘stays, remains, is left’; Av āste ‘sits’; Skt āste ‘sits’; IEW 342-343 *ēs- ‘to sit’ Cogn.: OInd āstē, Av āste ‘he sits’ (= Gk. Att. ἧσται ds.), EIran ās-, 3. pl. OInd āsatē (== Grk Hom. εἵαται, lies ἥαται ); Hit Med e-ša (esa) ‘sits’, e-ša-ri (esari) ds., infin. a-ša-an-na (asanna) etc., perhaps zero grade a-ša-ši ‘places, sits’; hierogl.-Hit es- ‘sit’. DEX.RO Lat sed, sedere > *assediare > DRom aşeza ‘lay down’; the Latin form actually developed in DRom in (a) şedea ‘to sit;’ but Lat assidere ‘to sit by, dwell close to’ (P1.+) offers a better solution. De Vaan, 2008, p. 551; this may be an example of merger. PIE MA *som- ‘together’< an o-grade of *sem- ‘one’; IEW 902-905 *sem2 ‘one’; DRom Simziana [compus sim-ziana] ‘Fairy Queen’, adv. samă, seamă ‘same’ as in de-o seamă ‘of the same age’ (with the meaning ‘born in the same year’ it could be related to OI. sámá f.‘half-year, season, year.’ Cogn.: OHG samn ‘together’; Lith sam- ‘with’; OCS so- ‘with’; Skt sam ‘with’; Av ha(m)- ‘together’; Goth. sa sama ‘the same.’ 2. *sem- ‘put in order/together’ DRom samă, seamă ‘same’ (A Magyar Nyelv Törteneti-Etimologiai Szotara, 1970); Gábor, 2006: ‘number, shape, counting’ of Turkish origin. PIE MA *ser- ‘line up’ deriv. fr. terminology for textiles; IEW 911 *ser/*sor-/*sr- ‘to arrange in order’; DRom n. şir ‘row, line’, v. înşira ‘line up’; n. șira spinării ‘the spine’ Cogn.: OIr sernaid ‘arranges’; Lat serō ‘line up, join, link’; Lith sėris ‘thread’; Grk eirō ‘lineup’; Hit sarra- ‘break’; Skt sarat- ‘thread’. PIE MA *reik- ‘scratch, line’; DRom v. râcâi ‘scratch’; Cogn.: NWels rhwyg ‘break’; NE row; Lith riekĕ ‘slice [of bread]’; poss Grk ereikō ‘bend, bruise’; Skt rekhā́ ~ lekhā́ ‘line’ PIE MA *wórghs ‘chain, row, series’; IEW 1154-1155 *u̯er-3: E. *u̯er-ĝh(*su̯erĝʷh-) ‘to turn, press, strangle’; DRom n. vargă, vergea ‘rod, wand, twig’ Cogn.: Alb varg ‘chain, row, string, strand’; Grk órkhos ‘row of vines’; Toch B warke ‘chain, garland’; Lat virga ‘thin branch, rod’
Basic Numbers, Measure and Quantity A few number forms show close affiliations with the PIE roots. PIE MA *dwoh3(u) (neuter: IEW 228-232 *dwoih1); duō(u) ‘two’; DRom m. doi, f. douǎ Cogn.: OIr m. dāu, f. dī, nt. Dā, Ir do; Lat (m, nt.) duo, f. duae; Lith m. dù, f. dvi; OCS m. dŭva, f. dŭvě; Alb dy; Grk dùō; Lat duo > DRom doi;
Appendix
331
here is the list of all the Romance forms and the reconstructed ProtoRomance form: http://stella.atilf.fr/scripts/DERom.exe?LISTE_EXHAUSTIVE; BALISE=SSignifiant;NOMENCLATURE=1;PAGE=D;DEFILE MENT=1;ISIS=isis_DERom.txt;MENU=menu_base;OUVRIR_ MENU=2;OO1=-1;OO2=-1;OO3=-1;s=s10110298 */ˈdʊ-i/ num. card. pl. “un plus un” I. pl m. nom: */ˈdʊ-i/: */ˈdʊ-i/ > DRom doi; IstrRom doi; MglRom doi; ARom doı ̆. PIE MA *tréyes ‘three’; DRom trei (a form closer to Greek) Cogn.: Lat trēs, Grk treîs, OIr trı ̆, Lith trŷs Skt tráyas; OCS trije, Alb tre. DEX.RO Lat trēs > DRom trei PIE IEW 643-644 kʷetuer-, kʷetuōr-, kʷetur- m., kʷetes(o)r- f. ‘four’; DRom patru ‘four’ Cogn.: Lat quattuor; Osc pettiur; petora; OWelsh petguar, Welsh pedwar; OInd cátur-; Lat quadru-[pēs]; Umbr. petur[-pursus “quadrupedibus”]; Gaul petru ‘four’, petrudekan ‘fourteen’, petru-deca-metos ‘fourteenth’ DEX.RO Lat quattuor > DRom patru - kw > p in Osc and Gaul etc., is worth mentioning here. PIE MA *ksek̑’s ‘six’; DRom şase Cogn.: Lith šeši, OCS šestı ̆, Skt şáş; DEX.RO Lat sex < *sek̑s > DRom şase ; the phonetic development in the DRom is closer to the Sanskrit. PIE IEW 909 *septm̥ (*sek̑ʷh-) ‘seven’; DRom șapte ‘seven’ Cogn.: Alb shtatë ‘seven’ (derived from a truncated *sa(p)tata ‘seven); later OInd saptáthah, also saptatí-, saptá, Av haptaϑa-, haptüiti-,. Hapta; Grk ἑπτά, Lat septem, OIr secht, Cymr saith, OCS sedmь (after dem ordinals); Toch. A spöt, В suk(t); Hit šipta; Lith. septiñtas; This numeral could be regarded as an example of merging between the Dacian and Latin
332 Appendix
PIE 1. IEW 905-906 *sēmi- (*ghemi-) ‘half ’; 2. IEW 505 *jem- ‘to hold’; *yemo- ‘twin’; DRom n. jumătate, jumate, juma ‘half ’ Cogn.: Alb gjymës, gjymësë; Alb.Gheg (*ghem-us) glymsë, gjymsë, Tosc gjysmë “half ” [common Alb. gh- > gl- > gj- : lith. gh- > dz- phonetic mutation] ; Lettish jumis, ‘double fruit’; Middle Irish emuin ‘twin’; Grk ήμισυς ‘half ’ *yemos? OIr emon; Lat geminus; Av yəma-; Skt yamá- ‘twin’. PIE MA *h2ent- ‘the first’; DRom adv. întâi (întân’iu archaic and dialectal) ‘first’ Cogn.: OIr ētan ‘forehead’; Lat ante ‘in front of, before’; Grk anti ‘in front of, opposite’; Hit hant ‘face, forehead, front part’; Skt ánti ‘in front of, opposite’, ánta- ‘end, limit’; Toch B ānte ‘surface, forehead’ DEX.RO Lat *antaneus? PIE MA *meh1ro-~*moh1ro ‘large’ < meh1- ‘grow’; IEW 704 *mē-4, mō‘big, important’; DRom adj. mare ‘big’ Cogn.: Grk -μωρος in ἐγχεσί-μωρος ‘big, large’; OIr mār ‘large’; mór ‘great’, Welsh mawr ‘big, large’, Bret. meur ds., Gaul. -müros ‘big, large in power’; OHG -mür in names as Volk-mür, Gothic -mêrs,’famed’, mêrian ‘proclaim’, OHG mâri ‘famed’, -mar in Germanic names; Slavic -meru; Latin merus, English mere; ON mærr ‘known, famous, great’; Thracian personal names Βαριμαρος, Zμερτομαρος, Kαρσιμαρπς; Celtic names Catumarus, Dannomarus, Germ. Volkmar and Hlodomar; OCS Vladi-mĕrŭ (Katičić & Mate, 1976) DEX.RO Lat mas, maris ‘male’ PIE IEW 1022-1027 (s)ter-, (s)terə-: (s)trē-6 : *ster- ‘stiff, immovable; solid, etc.’; DRom adj. tare ‘hard, tough, strong’, v. întări ‘to harden’; Cogn.: Grk στερεός (Att. also στερρός from *στερεός) ‘starr, tight, firm, hard’; OHG starc, starah ‘strong, big, large’ DEX.RO Lat talis ‘of such kind’?
Appendix
333
Emotions, Perception, Knowledge, and Thought PIE MA 1. *ghe(n)dh- ‘seize, take in’; DRom n. gând, gândire ‘thought’; v. gândi ‘think’; possibly v. găsi ‘to find’—cogn. With Alb gjënj, gjëj ‘find’ Cogn.: OIr ro-geinn ‘finds a place in’; Lat pre-(he)ndō ‘grasp’; Lith godóti ‘guess, suppose’; OCS gadati ‘imagine, guess’; Alb gjëndem ‘be found’, gjënj, gjëj ‘find’; Grk khandánō ‘take in, comprise’; Skt gan, ganati ‘count, consider, think’. 2. IEW 437 *ghend, *ghou- ‘to grab, grip, perceive, pay heed to’ Cogn.: Gk. χανδάνω (*ghend-) ‘take in, hold, contain, take; to be capable, able; catch’; Goth. bi-gitan ‘find’, OIce geta ‘reach; bring forth, assume’, OE be-gietan ‘receive, produce’, for-gietan ‘forgotten’; (Eng. get, beget, forget) 3. IEW 1125-1127 u̯(e)id-2 ‘to see; to know’: MWelsh 1. sg. gwnn, Corn gon, MBret goun ‘I know,’ Welsh gwys ‘knowledge’ from *uid-tu-s, M.Welsh gwyss, MBret gous ‘became known’; DEX.RO Hun gond ‘care, worry, concern, anxiety, trouble’; v. gondol ‘consider, believe, think’, of unknown etymology, Finno-Ugric, Iranian ? (A Magyar Nyelv Törteneti-Etimologiai Szotara, 1970) PIE MA *wet- ‘see (truly),’ IEW 1113 *u̯āt-, u̯ōt- ‘spiritually excited’; DRom v.a învăța (with prep. în- ) ‘to learn’, eu învăț ‘I learn’; ARom vețŭ, MeglRom anveț, IstrRom (an)mețu. Cogn.: Lat. vātēs, -is (probably Celt. Lw.?) ‘seer, prophet’, OIr fāith ‘seer, prophet’, Skt ápi vatati ‘is familiar with, is aware of ’, OE wōÞ ‘song, sound, voice’ DEX.RO Lat *invitiare ‘into vice’? (< vitium ‘vice) > DRom învăța ‘to learn’. PIE IEW 93 *bak- ‘stick, to hit’; DRom a (se) bucura v. ‘to be glad, joyful’, bucurie adj. ‘joyful’; toponym Bucuresti, Cogn.: ?Lat baca (bacca) -ae f. ‘a berry, fruit; a pearl’, bacalis, bacale ‘berry-bearing (designation of the female laurel), MFr (*bacale) Bacoule, n.f. Belette ‘weasel’: Alb. buklë ‘weasel’, bukur, bukurosh ‘good, pleasing,
334 Appendix
beautiful, slender (like a weasel)’: DRom bucuros ‘glad’; Note: In many lang. the name for weasel and good come from the same root. This note from Pokorny could give us a solution for a reflex that is generally considered unsolvable. IEW 153 *bheug-4 ‘to enjoy, consume, bite’; DRom a (se) bucura v. ‘to be glad, joyful,’ toponym București. Cogn.: Skt bhōga-ḥ ‘enjoyment’; Alb. bungë ‘kind of edible oak fruit’: with -u- grade Alb. (*beuka) buka ‘bread’ : Phryg. βεκός “bread”, actually ‘crumb’ prove that from an extended Root b(e)u-1, bh(e)u- : “expr. sound of hitting” derived Root bheg-, bheng- ‘to break’, bhenĝh-, bhnĝh- (adj. bhnĝhús) ‘thick, fat’, bheug-4 : ‘to enjoy, consume, bite’ as taboo words. PIE MA *teng- ‘feel, be of an opinion’; IEW 1088 tong-1 (*teng-) ‘to think, feel’; DRom v.tângui ‘complain, wail’ Cogn.: Lat. tongeō, -ēre, ‘to think’; Osc. tanginúd ‘sententia’; Alb tëngë ‘resentment, grudge’; OHG denchen, OE dencan ‘to think’, Goth. Þagkjan, Þühta ‘think’; NE thank; Toch B taṅkw ‘love’ and cäṅk- ‘please’ DEX.RO Slv tongovati ‘to weigh’; *tęgъ; *tęga ‘traction, weight’; The DRom v. tângui ‘complain, wail’ can be either a loan or a merger with the Slavic form *tǫga ‘sadness, melancholy’: OCS tǫga ‘confusion, melancholy, difficulties, misfortune’, ERu tugá ‘grief ’, WCz touha ‘longing, yearning, desire’; Slk. túha ‘melancholy’; Pl tęga ‘melancholy’ Bulg. tăgá ‘sorrow, sadness, hurt, desire’, which is semanticalyy more appropriate (Derksen) PIE IEW 2 *abhro- : ‘strong, mighty’ [the shift l>rabhro- (*hebhro-) related to IEW 52 apelo- ‘strength’; DRom adj. aprig ‘intens, fierce, bold, forceful, harsh’ Cogn.: MI. prefix abor-, Welsh afr- ‘very much’; Goth. abrs ‘get strong, violent’, adv. abraba ‘very much’, bi-abrjan ‘before were astonished, beside oneself ’, OIce.prefix afar- ‘very much’; Illyr VN A῎βροι, Thrac PN Α᾽ βρο; Alb (*a῎βροι) afronj ‘bring close, squeeze’, afër ‘near’ similar to formations of Ltv blaîzît ‘squeeze, clash, hit’; OCS blizъ adv.’nigh, near, adjacent; maybe Goth. aba (n- stem) ‘husband’,
Appendix
335
Note: the root abhro- : ‘strong, mighty’ is related to the cult of fertility, hence the goddess of love, beauty, and sexual rapture ‘Αφροδίτη (Aphrodite)’name derived from Grk ἀφρός ‘sea foam; + Σῑτανες ‘titaness’; the name of Aphrodite is related to PIE Root (enebh-2): nebh-, embh-, mbh- : (wet, damp; water; clouds); OInd abhrá- m. [ < mbhros), Av awra- n.] (Pokorny) PIE IEW 1070-1071 *ter-2, teru- ‘feeble, fragile, weak’; DRom v. întrema (în+trema)’recover, heal, enforce, arms’ Cogn.: Alb trim ‘valiant, gamy; m. young man’, pl. trima, armed henchmen’ (trmo-), if ‘young, youthful, war; Alb trim ‘brave, not scared’ is related to Tosk trëmp, Gheg trem ‘I scare’; Lat. tremō, -ere ‘tremble’ from root trem-, trems- ‘to thump; to tremble’; Grk τε’ρην ‘soft, delicate’ IEW 1071-1074 *ter-, *terə-, *teri-, *trēi-, *trī-, also *teru-, ‘rub’; DRom v. întărâta ‘to provoke, incite’ (internal development). Cogn.: OInd turá- ‘wund’ or ‘sick’; Grk τείρω ‘rub, troubling, tormenting, grieving’, Lat intertrīgō ‘chafing, fretting, or galling of the skin in riding, walking’; Al. tjer ‘spinne’ (*terō); DEX.RO Lat *interritare < interus ? PIE MA *men(s)-dh(e)h1- ‘mind-place/put, learn, mind-making’; DRom v. mântui ? ‘redeam in religious sens’, n. mântuire ‘salvation’, Mântuitor ‘the Savior’ Cogn.: NWels mynnu ‘wish’, OHG mendōn ‘rejoice’, munter ‘lively’, Lith mandras ‘lively, awake’, OCS mǫdro, ‘wise’, Alb mund ‘be able’, Grk manthánȏ ‘learn’, Av ma,z-dā ‘sharp in the memory’, mazdā ‘wisdom’; Skt medhā ‘wisdom’, supin mántum, participle perf. matá- ‘thought’;Skt mánman- ‘sense, mind, thought, notion’; mantár- ‘thinker’, mántra- ‘religiose formula’; Av. mąϑrō; Grk Μέντωρ; Skt matí-, máti-, Av -maiti‘sense, mind, thought, notion, opinion’, manti- ‘think’, Lith mintìs, OBulg. pa-mętь, Goth. ga-munds, Lat. mēns, (IEW 726-728); Grk μαντις, −εως; Iονιχ −ιος in Homer m. and f. ‘prophet, seer, predict the future’, v. ‘consult an oracle’; found in compound proper names Mantídeos (Pierre Chantraine: Dictionnaire étymologique de la langue grecque). Herodotus 9.93-5: manteis ‘seers’.
336 Appendix
IEW 726-28 *men-3: ‘to think, mind; spiritual activity’; DRom v. mântui ‘redeem’, etc. Cogn.: Av. mąnayǝn ‘man could believe’; Supin. OInd mántum, participle perf. matá- ‘thought’ (=Gk. αὐτό-ματος, Lat. commentus, Lith. miñtas ‘idea’, OIr dermat); OInd mantár- ‘thinker’; Grk Μέντωρ, Lat commentor; múni- ‘seer’ (compare μάντις); mántra- m. “religios formula”, Av. mąϑrō ds. IEW 740–741 *mǝ-r, gen. mǝ-n-és, mntós ‘hand’; DRom mântui, etc. Cogn.: Lat mandare ‘to hand over, commit, command’; OIr muin protection, patronage’; Hit manii̯ahh-I ‘to distribute, entrust’ < *mn-ieh2- ; OHG munt, OIc mund ‘hand, protection, paternalism’; *mn-to- > OIc mundr ‘price payed by the groom for his bride’ (de Vaan, 2008). DEX.RO DRom mântui ‘redeam’ < Hun ment ‘excuze, forgive, escape, get out, enter’ < men-, menedek ‘refuge, shelter’; Gábor (2006): ment [12. end of the century] derivative word, the verb-forming form of the obsolete menik verb ‘flee’ [1508]. DRom Mântuitor (Hun Megmentő, Megváltó) found in the Bible translations from fourteenth to sixteenth centuries in Hungarian and Romanian is usually explained by the Hun ment1 ‘excuze, forgive, escape, get out’, ment2 ‘enter’, both forms in use in religious texts not to express ‘redeem’ but ‘to escape, exit’; the Hungarian form is used in religious context mostly with a prefix meg-: megmentett ‘salvation’, megmenekült ‘escaped’; the solutions offered here through IEW roots are controversial, and open for further discussion. PIE MA *swep- , pres. *swepti ‘sleeps, dreams’, and causative: *swopéyeti ~*swōpéyeti ‘puts to sleep’; DRom adj. svăpăiat, zvăpăiat ‘fidgety (of children)’ Cogn.: OE swefan; OCS sŭpati; Hit sup-, Av xvap-, Skt svápiti: all ‘sleep’; Skt svapáyati ~ svāpáyati ‘lulls to sleep’ Other possibility: DRom n. văpaie ‘shooting flame’ > DRom adj. svăpăiat ? PIE IEW 124 *bh(e)lāg-, *bhl_H2g-; or, *bhlāg- or *bhlōg- ‘ridiculous, silly, weak’; DRom adj. m. bleg, f. bleagă ‘weak, silly, flabby, sheepish, with floppy ears’ Cogn.: Sb bleka ‘stupid’; Grk amalos ~ ablekhros ~ blekhros ‘soft, weak’; in wRuss. bɫáhyj ‘evil, bad, nasty’ (hence borrowed Ltv. blāgs, Lith.
Appendix
337
blõgas ‘feeble, weak’), Russ. blagój ‘obstinate, nasty’, dial. blažnój ‘stupid’, Pol. bɫagi ‘bad, nothing worth’; barely to Gk. φελγύνει ἀσυνετεί, ληρεῖ Hes.; Lat. flaccus ‘flabby’ (Fournet Dictionary) Note: the Slavic reflexes have a negative meaning, whereas the Grk, Lat, DRom, and Lth are closer to the weak, silly, flabby meaning. PIE IEW 683-684 *leubh- ‘to care for, love’; DRom v. a iubi ‘to love’; n. iubit ‘lover’ Cogn.: Lat libet, older lubet, -ēre, -uit, -itum est ‘is beloved, is compliant’, lubens, libens ‘willing, be found of somebody’, lubīdo, lib+īdo ‘eagerness’; Osc loufir ‘well’ (compare OBlg ĺubo—ĺubo ‘well–well’); Goth liufs, OIce liūfr, OHG liob, OE lēof ‘dear’ (OBlg ĺubъ); therefrom derived *liubēn ‘be kind, please’, in OE lēofian, OHG, MHG liuben; *liubjan ‘willing’, lubīdo, libīdo ‘eagerness’;OBulg ĺubъ ‘dear’ (Russ. Ĺúbyj, etc.), whereof ĺubiti love’, ĺuby ‘love’; maybe Alb lyp, lip ‘beg, like, want’; Derksen: *ļūbìti v (b) ‘love’ ESSJa XV 174-176; OCS ljubiti ‘love, desire, worship’, Ru ljubít’ ‘love, like’; WCzech líbiti se ‘please’; OCzech líbiti ‘love, like, prefer’; SCr ljúbiti ‘love, desire, kiss’, Sln. ljúbiti ‘love, caress, promise’, DEX.RO OCS ljubiti, ljubljǫ, ljubiši (любити) ‘love, desire, worship’; the forms here suggest a merger. PIE IEW 488 *gʷhedh- ‘to beg, wish for’; DRom v. gudura ‘flatter, (toady)’ (Russu 328) Cogn.: OIr guidiu ‘please’ (= ποθέω), (*gʷhedh-ti-s); guide ‘prayer’ (*gʷhodhiü); foigde ‘beg’ (*uogediü); Welsh gweddi “prayer” (*uo-godīmü) DEX.RO Alb gudullis ? (not in Orel (1998)
Sight DRom offers uita with a double meaning ‘look’ and ‘forget’, traditionally explained by the Lat oblivio, which is problematic, and covers only one of the two meanings; the confusion may be older considering the other IE cognates.
338 Appendix
PIE MA *wet- ‘see (truly’; *weid ‘to see’ (Watkins, 2000); DRom v. uità ‘see, look’, uite! ‘look!, you look!’, and v. uità ‘forget’; ARom ultu, MglRom ul’tari, IstrRom uitu ‘forget’ Cogn.: OIr fethis ‘sees, pays attention to’; Lat vātēs ‘seer, prophet’; Skt ápi vatati ‘is familiar with, is aware of ’; Germanic *witan ‘look after, guard’; OProvencal guidar ‘to guide’; zero-grade *wid > OE wit, witt ‘knowledge, intelligence’ Possibly connected to IEW 1025-1027 *u(e)id-2 ‘to see; to know’; or, IEW 1136-1137 *u̯el-1 ‘to see ‘ Lat voltus, vultus, -ūs ‘facial expression, expression, appearance, shape’ DEX.RO Lat *oblitare; oblivio, oblivium—improbable etymology; de Vaan 2008, 422: obliviscor ‘to forget’ [v. III; pf. oblitus sum]
Colors, Hearing, Smell, Taste, Good, Evil, Ugly PIE MA *kwṛsnós ‘black’; IEW 573-574 *ker-6 and k̂er- ‘dark color, dirt’> IEW 583 *kers- ‘kind of color/black’; DRom v. cerni ‘to darken’; n. cioară ‘crow’; for river names Crişul—hydronym (in antiquity Κρισος, Crisia, Grisia, Gresia)rivulets Crişul Negru, Alb, Pietros Repede ‘Black, White, Stony, Rapid Criș’; derive. toponym Crișana; and river Cerna hydronym (in antique documents: Dierna, Tsierna in Latin or Greek transcription where the sound č was absent), see also chapter Water. Cogn.: OPrus krisnan ‘black’; OCS črŭnŭ ‘black’; OBg črъnъ, Russ čërenъ; Skt kṛsná ‘black’; Alb sorrë ‘crow’; Arm saṙn, gen. saṙin, saṙnum ‘ice’, Ven.-Illyr PN Carmō (Austrian province), rätorom ON carmún ‘weasel’; see under kor-men-; Alb thjer-më ‘gray’, per-thjerm ‘lazuline’ (*ker-uo- with secondary -më); i surme ‘ashen’ (*kor-mo-); s. Jokl Mel. Pedersen 153 ff.; Lith širvas ‘gray, greyish-blue’ (*kr-uo-s), širmas ds (*krmo-s), Ltv. sirms ‘gray’; Proto-Slav *sernъ: ORuss serenyj ‘white’ (of horses), Russ. serën ‘frozen snow’ PIE MA *h2ṛg̑(u)- ‘white’; DRom n. hârcă ‘skull’, pejorative for ‘old woman, witch’
Appendix
339
Cogn.: Hit harkis ‘white’; Grk árguros ‘silver’; Skt árjuna ‘light, white’; Toch B ārkwi ‘white’ DEX.RO Ukr hyrka? the proximity of the Moldova and Ukraine could suggest a common etymology. PIE MA *bhelh1 ‘white’: IEW 118-120 *bhel-1, *bhelǝ- ‘shining, white’; DRom n. băl(i) ‘blond’, bălai ‘having blond hair’, bălan ‘animals with white hair’, Dacian anthroponym Balius; DRom anthroponym, toponyms Bălașa; n. baltă ‘pond, puddle’, derive. băltoacă Cogn.: NWels bal ‘white’; Lith bālas ‘white, snowdrop’, báltas, Ltv balts ‘white’; Alb balë ‘badger’ (animal with white spots on the snout); Grk phalós ‘white’; Illyr *balta ‘swamp, marsh, white clay’; OCS bělŭ ‘white’; Lat flāvus ‘blond’; NE ball (horse with white blaze), Skt bhālam ‘gleam, forehead’; Lith n. bālas ‘white’, v. bálti ‘grow white, pale’; Derksen (2008, p. 53): *bòlto n. ‘swamp’ OCS blato ‘swamp, quagmire’; E Ru. bolóto ‘swamp’; W Cz. bláto ‘mud’; Pl. bloto ‘mud’; Bulg. bláto ‘mud, swamp’; Lith. báltas ‘white’; balà f. ‘swamp’; Latv. balts ‘white’; OPr. Namuynbalt [placename]; Lomb. palta, Piem. pauta (cf. Demiraj, 1997, pp. 87–88); NGrk βάλτη; Alb baltë ‘mud’< Illyrian (Orel); Matasovic (2013) PIE *bholHto- ‘white’. Lehmann (1986) B15 *bala ‘ ‘shning, gray’ Procopius De bello Gothico ‘the barbarians call the horse Bala’, and in Ennodius 355 ‘de equo badio at balane ‘of a gray and bala horse’; possibly a Germanic loan? DEX.RO Bulg. bláto ‘mud, swamp’? a regressive development not accepted in DRom PIE MA *h1roudhós adj. ‘red’; IEW 672-873 *reudh- ‘red’; DRom adj. roşu ‘red’, fem roşie, v. înroşi ‘to turn color red’ Cogn.: OIr rūad ‘red’; Lat rūfus ‘red’; Lith raūdas, Rus rúdyj ‘blood-red, red-haired’; Av raoidita- ‘red’; Skt róhita- ‘red’; Grk ερυδρος ‘red’; Goth rauþs ? MA *h1elu- ‘dull red’ Skt arușá, aruṇá ‘reddish, golden’ DEX.RO Lat russus ‘reddish’, or Lat rŏseus ?
340 Appendix
PIE MA *k̑er- ‘greyish blue, green’; DRom adj. sur(ă) ‘grey, grey hair’, adj. suriu ‘greyish’; n. suraia ‘name for a cow’, suran ‘name for a bull’; Cogn.: Lith šir̃vas ~ šir̃mas ‘blue-grey’; Alb thjermë ‘(blue) ash grey’, surmë ‘grey, black’, sorrë ‘crow’ ~ DRom n. cioară; Skt śārá- ‘colored’. DEX.RO Bg, Sb sur (not in Derksen) < DRom ? PIE MA *k̑yeh1- ‘blue, purple, green, azure, grey’; DRom n. sineală ‘blue paint’ Cogn.: Lith šývas ‘light grey’; OPrus sywan ‘grey’; OCS sivŭ ‘dark grey’; Alb thinjë ‘grey’; Lith šė́mas ‘blue-grey’; SerbCr sinji ‘sea green’ DEX.RO Bg sinilo ? (not in Derksen) PIE MA *g̑hel- ~ *ghel- ‘yellow, gold’; IEW 429-434 ̂*hel-1 (and ghel-?), also as i-, u- or n- stem; ĝhelǝ- : ĝhlē-, ĝhlō- : ĝhlǝ- (*ĝhwel-) ‘to shine; green, gold, blue, *sun’; DRom toponym Zarand [gold mines area]; possibly DRom gălbează / călbează ‘contagious liver disease caused by a worm in sheep, cows’: galb + ză associative suffix Cogn.: OIr gel ‘white’; NWels gell ‘yellow’; Lat helvus ‘honey yellow’; Lith geı ̃tas ‘yellow’; Av zairi ‘yellow’; Skt hári ‘yellow, green’; NE yellow, gold; Av zaranyam ‘gold’; Skt híranyam ‘gold’; Latv zèlts ‘gold’; OCS zlato; OIce. gall n. ‘gall, poison’ (*gallōn-, IE *ghol-n-), OE gealla m., OS galla ‘intense’ f., OHG galla ‘weak’ f. gall’; Alb gjelbër ‘green’ galbinus ‘green yellow’; DRom and Alb këlbazë, gëlbazë ‘liver disease caused by a worm’ PIE MA *mel-6, melǝ- ‘dark color, black, dirty’; IEW 720-721 mel-6, melǝ- ‘dark color’ (black, dirty, etc.); DRom adj. mieriu, mieru dialectal ‘blue, dark blue’ Cogn.: Lith. mélas, mélynas ‘blue’, Grk μέλᾱς, -αινα, -ᾰν ‘black’; Alb mel-еnë (Kollekt. *mel-inio-) ‘elm, color of wood’ , mel-ézë, mjerë ‘unlucky’ (*mel-ro- ‘black’); Lat. mulleus ‘reddish, purple’;
Appendix
341
Hot, Cold, and Other Qualities PIE MA *haes- ‘be(come) dry’; DRom v (se—reflexiv) usca, 1st pers. eu (mă) usuc ‘dry up (myself )’ Cogn.: Lat āreō ‘be dry’, āridus ‘dry’; Czech ozditi ‘dry malt’; Grk ázomai ‘become dry’; Toch AB ās- ‘become dry’; Lat uro, ussi, ustum ‘to burn’ (Grk ευω); Thrac usku- ‘water; aquatic, marshy’; OIr u(i)sce ‘water’, OCimmr uisc, Irish esc ‘water, bog, swamp’ . DEX.RO Lat *exsucare ? or, Lat uro, ustum ‘burn’, but Lat ūstŭlare > DRom ustura ‘sting’, Lecce uškare, Apul aška, Abruz aškya, Molfett aškere ‘sting’ (Meye-Lubke 9097) may show contaminations in forms and semantics Etrusc us- ‘to draw (water)’, useti ‘scoop, ladle’ (B-B 2002:220). Looks like a loanword of PIE *H2ews- ‘to draw (water)’ (Fournet, 2012) PIE IEW 894-895 *se(n)k- ‘dry up’; DRom adj. sec ‘dry’, also of nuts, hallow’ n. secetă ‘drought’, v. seca ‘dry out’ Cogn.: NE singe, Lith senkú ‘ebb, drain away, dry up [of water]’; OCS i-sęknǫti ‘dry up’; Reduplicated *siskus ‘dry’ Lat siccus ‘dry’ > DEX.RO > DRom sec, It secco, Sp seco –the DRom n. secetă ‘drought’, later loan? De Vaan: “siccus. Derivatives: siccdre ‘to dry (tr./intr.)’ (Cato+), siccitas ‘dryness’ (P1.+), siccoculus ‘having dry eyes’ (PL); exsiccare ‘to make dry, drain’ (Enn.+) Pit. *siskwo-l PIE *si-sk-u~ ‘dry’? IE cognates: Mir. sesc ‘dry, barren, infertile’, MW hysp ‘dry’ < *siskuV-; Skt. a-sascuși- ‘not drying up’, Av. hišku‘dry’, Khot skala ‘dry places’, Gr. έσκετο φωνή ‘the voice broke down’, ισχνός ‘dry, arid’; Lith. sekti ‘to lower oneself, sink, dry out’. DEX.RO offers a reconstructed Lat *siccitas > DRom secetă ‘drought’ PIE MA *gwr(e)ha(-u) ‘heavy’; *gwrehx-u- ‘heavy’; IEW 476-477 gʷer-2, gʷerǝ-, gʷerǝu-, gʷerī- ‘heavy’; DRom adj. masc. greu, fem. grea ‘heavy, hard’ also, euphemism for pregnant woman.’ Cogn.: Grk barús, MIr bair ‘heavy’; Lat gravis; Lith grūts ‘heavy’Toch B krāmär ‘weight, heaviness’; Skt guru ‘heavy, wise man’.
342 Appendix
DEX.RO Lat *grevis (?) < gravis ‘heavy, serious;’ possibly a merger? ‘PIE IEW 966-967 *smē-, smeī-, sm-ei- ‘smear, rub’ > *smē[i]k-: *smīk‘grated, tiny crumbs’; DRom adj. m. mic, f. mică ‘small’ Cogn.: Gkr σμῑκρός, (σ)μικρός ‘small, petty, short’, Dor. Ion. μικκός ‘small’; Lat. mīca ‘grain, crumb’, mīcidus ‘tiny’; OHG smühi ‘small, little, low’, OIce smür (*smüha-) ‘small’, OE smēalīc ‘fine, painstaking’, OHG smühen ‘make small’; De Vaan (2008): Lat mico, -are ‘to quiver, dart, flash’, mica ‘grain, crumb’ < PIE *mikH-(e)ie- ‘to blink’, *meik-h2- ‘blinking’ DEX.RO Lat *miccus ?
Speech, Interjections and Human Sounds PIE MA *ter- ‘say, noise’; DRom n. tărăboi ‘noise, uproar, riot’ Cogn.: MIr to(i)rm ‘noise, uproar’; OPrus tārin ‘noise’; Rus torotóritı ̆ ‘chatter, prattle’; Luv tātariya- ‘curse’; Toch B tär- ‘implore’ DEX.RO Grk thorybos, Alb terboz PIE IEW 350-351 *gal- ‘call (out)’ *ĝar- ‘shout, call’; DRom n. gălăgie ‘noise of loud voices, shouting’; n. zarvă ‘uproar, loud voices’ Cogn.: OIr gall, ‘swan’; NWels galw ‘call’; Lat gallus ‘cock’; OE ceallian ‘call’; NE call; Lith gaı ̃sas ‘echo’; OCS glasŭ ‘voice’; Oss yalas ‘sound’; OIr do-gair [Ir glao] ‘call’; Lat garriō ‘chatter, prattle’; OE cearu ‘care, sorrow, mourning’; Grk gêrus ‘voice, call’; OCS zovǫ ‘call’ (> DRom zvon ‘gossip’); Av zavaiti ‘calls’; Skt hávate ‘calls, invokes’. PIE MA *kelh1- ‘noisy call’; DRom n. chiot, v. chiui ‘yowl during dance, scream’ onomatopoeic? Cogn.: Lat calendae; Grk kaléō ‘call’, kalếtōr ‘herald’; ON hjala ‘chatter, talk’; Latv kaɭuôt ‘chatter’ PIE MA *ul- ‘howl’; DRom v. hăuli ‘long call over mountains, yodel, singing in long deep sounds’
Appendix
343
Cogn.: Lat uluc(c)us, Skt úlŭka- ‘owl’; Latin ulula screech-owl ; ululāre ‘howl’ (> DRom urla ‘holler, screem’); Skt ulūlú ‘ululating’ ; Lith uluoti ‘shout’ ; Alb ulerij ‘shout’. PIE MA 1.*(s)trep- ‘cry out, dispute’; 2. *(s)preg- ‘speak’; DRom v. striga ‘call out, shout, cry out’ 3. IEW 1036 (s)treig-, streid(h)- ‘to hiss’, Lat strı ̑deo creak, rattle, hiss’ Cogn.: 1. Lat strepō ‘cry loudly, make noise’; OE þrafian ‘restrain, reprove, urge, demand’; Cogn.: 2. NE speak, Alb shpreg ‘express, voice, utter’. DEX.RO Lat *strigare < strix, strigis ‘owl’, Grk ςτριγξ −γγος ‘night bird’, more probably the DRom development strigoi ‘spirit of the dead’< Lat striga ‘evil spirit, witch’ PIE MA *balba- , *balbal-, *barbar- ‘stammer’; IEW 91-92 *bal-bal- ‘barbaric search’; DRom v. bolborosi ‘mumble’ onomatopoeic?; n. bală, pl. bale ‘slobber, slabber’ adj. bălos, ‘slobbering, drooling’ Cogn.: Lat balbus, balbuttı ̑re > DRom v. bâlbâi (?) ‘stammer’; NE babble; Lith blebenti; Czech beblati; Grk βαμβαλυζω ‘to have a chattering tooth’, barbaros; Skt balbalā-karoti ‘stammer’, barbara- ‘barbarian’; Lith. balbasyti, blebenti, blebeti ‘to chatter’, Ru. bolobolitb, Bulg. blabol’v, ‘to blether’, SCr. blabositi, PIE MA *b(h)(o)mb(h) ‘muffled noise’; DRom v. bombăni ‘mumble’ Cogn.: Grk bombos ‘muffled noise’ > DRom late loan?; ON bumba ‘drum’; Lith bambeti ‘roar’ Alb bumbulit ‘it thunders’; Rus buben ‘drum’. PIE MA *dhren- ‘drone’; DRom v. trăncăni ‘speak endlessly, blabber’ Cogn.: Lith trānas, Grk thrônaks ‘drone’; MIr dresacht ‘creaking noise’; Lat drēnsō ‘cry’; Grk thrênos ‘funeral lamentation’; Arm drnč’im ‘toot, resound’; Skt dhránati ‘resounds’; Toch B treṅk- ‘speak’
344 Appendix
PIE IEW 1088 *tolkʷ- ‘to speak’; DRom n. tâlc ‘meaningful, wise (speech)’ Cogn.: OCS tlŭkŭ < Proto-Slavic tъlkъ ‘interpretation, explanation’; Latin loquor ‘to speak’, Old Irish do-tluchethar ‘to ask’, ad-tluchedar ‘to thank’; Old East Slavic тълкъ (tŭlkŭ), Belarusian толк (tolk); Russian толк (tolk) ‘interpretation, explanation’; Ukrainian толк (tolk); Danish tolk; Dutch tolk; Estonian tulk; Latvian tulks; Lithuanian tùlkas; Middle Low German; tolk; Norwegian tolk; Old Norse tulkr; Swedish tolk; Eng talk ‘speak’; possible merger ? PIE IEW120-122 *bhel-3, *bhlē-: ‘to grow, spread, swell’; DRom n. fluier ‘whisle’ Cogn.: Lat. flō, flüre ‘to blow; intransit., of winds, persons and instruments; transit., to blow, blow forth; to blow on an instrument’; Alb flojere ‘flute’ IEW158-159 *bhleu- ‘to blow; to swell, flow’; DRom n. fluier ‘whistle’ Cogn.: Gk. φλέ(F)ω ‘to be full of, to abound with, to be bursting with, to be bristling, be brimful’, Φλεύς (*Φληυς, lengthened grade), ephes. Φλέως (*ΦληFος) epithet of Dionysos as a vegetation God; presumably from the lushness of growth also Att. φλέως, jon. φλοῦς ‘reed plant’ φλοίω (*φλοFιω) ‘swell, to be full of, to abound with, to be bursting with, to be bristling, be in bloom, blossom’, ὑπέρφλοιος ‘growing excessively’ or ‘exceedingly succulent’, Φλοῖος, Φλοία ‘epithet of Dionysos and the Kore as vegetation divinities’ probably also φλοιός, φλόος ‘bark, husk’; Lat flo, flare ‘to blow’, flatus, -us ‘blowing, blast, breath’; sufflare ‘to puff > DRom v. sulfa, n. suflare ‘blow, breath’; Lat. fluō, -ere, flūxi, flūctum, newer flūxum ‘to flow; of a solid object, to flow’? Some onomatopoeic forms can be compared: for ‘pain’: PIE MA *wai ‘alas’; IEW 1110-1111 u̯ai ‘woe!’: DRom vai ‘woe’ [interjection], n. vaiet ‘groan, wailing, weep, lamentation for the dead’, v. a se vǎita ‘to complain’, vǎleu, aoleu ‘woe’ Cogn.: OIr fae ; Lat vae ; OE vā; Lith va; Grk ouai; Av vayōi; AE: PAlb. *uai Alb. vaj [m] {1} (tg) ‘wailing, weep, lamentation for the dead’ (AE 405) PIE *uai? ’woe!, alas!’ Grk ὀά ’woe!, alas!’; Arm vay ’pain’; Goth. wai,
Appendix
345
OIce vei, væ, OHG, OS wē; OE wü ‘wehe!’; Ltv. vaı ̃ ‘wehe, ach’, waı ̃dêt ‘lament, weep’; Serb. vâj ‘wehe!’ is neologism. Alb. vaj ‘lament’, vajtoj ‘to lament’. For laughing the DRom follow all the PIE dialects: PIE MA *ha ha : DRom haha Cogn.: Lat hahae; Grk há há; Skt ha ha; PIE MA *kha- ‘laugh’, duplicate *kha kha; DRom n. haz ‘fun’, n. hazliu ‘funny’, n. hohot ‘peal, guffaw’, v. hohoti ‘laugh out loud’; Cogn.: Lat cachinnō; OE ceahhettan ; OCS chochotati; Grk ka(g)kházō; Arm xaxank; Skt kákhati ~ khákhati, DEX.RO haz < NGrk χαξι < Turk hazz ? (Tk eğlence ‘fun’)
Activities PIE MA *dhers- ‘attempt’; DRom v. îndrăzni ‘dare’ (în+drăzni), îndrăsni, n. îndrăzneală ‘atempt, courage’ Cogn.: Lith dręsú ‘dare’; Skt dhŗşnóti ‘bold, dares’, dharșati ‘I dare’; Germanic *dorso- >NE dare; Grk thérsos ‘bravery’; Av darši- ‘brave’. PIE MA *hxópes- ‘work’; DRom v. apăs(a) ‘press, push’ Cogn.: Av –apah- ‘work’; OE æfnan ‘work, make’; Lat opus ‘work’; Skt ápas- ‘work’ DEX.RO Lat *appensare PIE IEW 868-871 *reu-2, reuǝ- : rū- ‘to tear out, dig out, open, acquire; DRom v. arunca ‘throw away’; ARom aruncarea, arcare, aruca ‘id’ Cogn.: Lat ruō ‘tear open, scratch’; Meyer Lubke 7414; Lat ruō ‘tear open, dig, scratch’, OHG roufen, MHG roufen, reufen, raufen ‘pluck’; de Vaan (2008): PIE *h3ru-n-k- ‘to dig out’. IE cognates:It runco, -are ‘to grub up (plants), weed1 [v. I] (Cato+) Pit, *runk-a-, Skt luncati ‘to pluck’ (ep.), Gr. όρύσσω, Att. όρύττω ‘to dig, grub’.
346 Appendix
PIE MA *mer- ‘crush’; DRom adv. mărunt ‘small chopped, small pieces’, v. mărunţi ‘to crush’ Cogn.: Grk marainō ‘extinguish fire’; Hit maryattari ‘is smasked’; Skt mŗnā́ti ‘crushes’ And Lat minuere [minui, minutum] ‘to reduce in size, lessen’ (PL+), minutus ‘small, short’. DEX.RO Lat minutus, -a, -um > DRom mărunt ‘small pieces’—a merger? PIE MA *skebh- ‘scratch, shave’; DRom v. scăpăra ‘to sparkle with stones, make fire with flint’, ARom ascapir, MglRom scapir, Alb shkrep; Alb shkrepër ‘flint’, IEW 930-933 *(s)kēp-2, (s)kōp- and (s)kāp-; (s)kēb(h)-, skob(h)- and skāb(h)- ‘to work with a sharp instrument’; DRom v. scăpăra ‘sparkle (with flint), lighten’ Cogn.: Thrac. σκάλμη ‘sword, knife’ (*skolmü); Lat scabō, -ere, scübī ‘scrape, scratch, rub’, scůbiēs f. ‘scab, mange, itch; roughness; scaber ‘rough, scabby’; with o: scobis f. ‘filings, chips, shavings, sawdust’; scobīna’file, rasp’; M.Ir (s)cīp (with bb) ‘hand’; Goth.skaban ‘scrape, shave, shear’, OIce skafa ‘scrape, scratch, scrape’; OE scafan. scaban ‘scrape, scratch, cut (hair), clip’, OCS skoblь ‘rasper’; Russ skóbelь ‘slicer’; Ltv. skabrs (= Lat. scaber) ‘splitterig, sharp’, skabrums ‘sharpness, scabrousness’; Alb kep ‘hew stones, cut out’ (IE *kopō or *kapō) See also DRom scobi ‘to scoop, scratch’ PIE IEW 917 *(s)k(h)ai-, (s)k(h)ai-d-, (s)k(h)ai-t- ‘to hit, kick’; DRom n. scai, scaiete ‘thstle’ Cogn.: *khai-t- Arm. xait’ ‘prick, sting’, xait’em ‘prick’, xit’am ‘I’m scared’. Grk skállō ‘hoe, stir up’; Arm skalim ‘split, be splintered’; Hit iskalla ‘slit, slash, tear’; Thracian σκάλμη, μάχαιρα Θρακία, Greek σκόλυμος;, ‘artichoke’, PIE MA *(s)kel- ‘split (apart), cut’, IEW (923-927) *skel- ‘to cut’—with formants—mo*skolmeha- ‘sword, cutting tool’; see also IEW 938-947 *(s)ker-4, *(s)kerǝ-, *(s)krē- ‘to cut’; DRom v. scărmăna ‘teasel wool through sharp metal teeth’; phonetic form: zgărmana; v. scormoni, v.
Appendix
347
scurma ‘rumage, scratch’; DRom v. curma ‘to stop, to cut off (someone’s life)’, adv. curmeziș ‘across’ Cogn.: Thrac skálmē, σκάλμη, ‘sword, knife” (*skolmü); Hit karsmi ‘cut off, castrate’ Arm k’erem ‘scrape off, scratch off, Skt krmati ‘wounds, kills’; OIce skǫlm ‘fork, pod’ , pl. ‘scissors’ OHG scalm ‘navis’; also probably OHG scalmo ‘pest, corpse’; OE helma ‘helm, handle, grasp of ’; MDutch harst ‘rake’, Lith karšiu’comb or card (wool); hatchel (flax), OCS krasta ‘scab, itch’; Rus korósta ‘scab, itch’; OInd asati < (if ) karsati ‘rubs, scratches’. PIE IEW 938-947 *(s)ker-4, *(s)kerǝ-, *(s)krē- ‘to cut’; DRom v. zgâria, sgâria (phonetic form of *scâria?) ‘scratch’; v. scrijeli ‘scratch’ (rel. to Lat scriber); Georgiev 1960: Dacian σξιαρη ‘mulberry, Alb shqer ‘scratch’. Cogn.: OIce skera ‘cut, clip prick’, OHG sceran ‘shave, shear, cut off’, scero ‘mole’, scar, scaro m., scara f. ‘plow shear’, . scār, scāra, pl. scāri ‘scissors’; OE scieran, scēar ‘plow shear’; OS sker-sahs ‘shear knife”; Dental extensions: - (s)ker-d-DRom adj. scurt ‘short’, v. scurta ‘to shorten’; Cogn.: Illyr. Scordus (mons), Ʃκάρδον (ὄρος). TN Scordisci ‘men with shirts, kilts, like women’ cf. Alb (*skodra) kodra ‘hill’ (*short) low mountain, low hill’ [common drop of initial s- in Alb. sk > k]; Lith. skardùs ‘steep’, kertú ‘hew’; OHG scurz ‘short’; OE scort ‘short’; OIce skorta ‘lack, to be lacking’, skort n., skortr m. ‘lack’; Arm k’ert’em ‘skin’; Hit kartai- ‘cut off’; Av kərəntaiti ‘cuts’; Skt kŗntáti ‘cuts’; Alb (*scyrte) shkurt ‘short’, borrowed from Germ *skurtaz ‘short’, OHG scurz, OE sceort, (Orel, 1998); Labial extensions: - (s)kerb/h/-, (s)kreb(h)- DRom v. scărmăna ‘teasel wool through sharp metal teeth’; phonetic form: zgărmana; v. scormoni, v. scurma ‘rumage, scratch’ Cogn.: MIr cerb ‘sharp, incisive’, cer(b)aim ‘cut, bite’; OE sceorpan ‘scratch, scrape, gnaw’, OE scearp ‘sharp’; OIce skarpr ‘lean, strong, sharp’; OS skarp ‘sharp, rough, bitter’, OHG scarf, scarph, MHG scharf, -pf ‘rough, incisive’; Baltic-Slav *skirbü ‘chink’ (*skerbhü) in Ltv škirba ‘chink, gap’
348 Appendix
- (s)kerp-, (s)krep- DRom v. scărpina ‘scratch (when itching)’ < DEX.RO Lat scarpinare ‘to hike through the mountains’? Cogn.: OInd krpüna- m. ‘sword’, krpünī ‘scissors’; karpara- ‘shard, bowl, cranium’; OPruss kerpetis ‘cranium’, OCS črepъ ‘shard’, OHG scirbi ‘shard’); Alb. karpë, karmë (*korp-n-) ‘cliff’; Thrac. Καρπάτης ὄρος ‘Carpathians’; Grk καρπός ‘crop, fruit’, καρπίζομαι, καρπόομαι ‘harvest’; κρώπιον ‘sickle” (IE *krōp-); with s- probably σκορπίος ‘scorpion, seefish’; Lat. carpō, -ere ‘pluck’, originally ‘split off’, scalpo, -ere ‘to scratch, carve’; scarpo i.e. excarpo ‘pick’, scarpinat (OE ‘hen rales’); carpinus ‘hornbeam’; Hit karpina- ‘tree’; Nor. skorpa ‘crust’; - De Vaan (2008) Lat scalpo, -ere ‘to scratch, carve’; DRom v. crăpa ‘crack, split’ < DEX.RO Lat crepo, -are ‘to make a sharp loud noise’? Cogn.: Lat. scrobis ‘pit, pothole’; OE screpan ‘scratch, scrape’; OIce skrapa (*skrapōn) ‘scratch, scrape’; MHG schreffen ‘scratch, scrape’, schraffen ‘skin scratch’, schrapfe (*skrappō) “scratch tool’, MLG schrappen ‘scrape, scratch, scrape”; Welsh crafu ‘scratch, scrape, rub’ (P>F) PIE IEW 1029-1031 *ster-5, sterǝ- : strē-, steru- : streu- ‘to widen, to scatter’, PIE *st(e)r-n-h3- ‘to spread, strew’, *strh3-to- [ppp.]. DRom v. stârni ‘to stir up’ Cogn: De Vaan (2008, p. 586) sterno, -ere ‘to lay out, spread’; Olr. sernaid ‘to spread, build’, Skt. strnati ‘to spread out, extend, strew’; Gr. στόρνυμι, έστόρεσα [aor.] ‘to extend, strew out’, στρωτός ‘extended’; YAv fra-stdrsnaiti, fra-stsrvnata , ‘to spread out, extend’, Ilr *starH- ‘to extend, strew out’. PIE *(s)tergh- ‘crush’; DRom v. şterge ‘wipe out, clean, remove’, v. strica ‘damage’ Cogn.: In Hitite-Indic isogloss with arhaic ne- *(s)tṛ-ne-g̑h-ti > Hit istarninkzi ‘afflicts’; Skt tṛnédhi ‘crushes, bruises’. PIE IEW 1071-1074 *ter-3, terǝ- and teri-, trī- ‘to rub’; D. extensions from ter- and treu-: 2. terg-; DRom v. șterge ‘wipe out, clean, remove’
Appendix
349
Lat tergō, -ere, tergēo, -ēre ‘clean’, mantēlum, mantēle ‘towel’ (*manterg-sli-, to manus) DEX.RO Lat extergere ex- + tergeō ‘rub, wipe off, cleanse’; extricare ‘disentangle, extricate free’ PIE IEW 929 *(s)kel- ‘to spring’; *(škel-) *shkel ‘(*spring?), step’; DRom v. scula ‘rise, stand up, wake up’; Cogn.: MHG schel ‘jumping, starting up, upset’, schellec ‘jumping, angry, irate, wild’; OHG scelo ‘quick stallion’, Ger. schälen ‘spring, jump’; OIce skelkr ‘fear’ as *erschreckt Goth. skalks ‘servant' assumes man, basic meaning ‘springer, and errand boy dissimilation from *skal-[s]kas-; Lith. šuolys ‘gallop’; Ltv. suôlis ‘footstep’ DEX.RO Lat. *excubulare ? PIE IEW 455-456 *ghrebh- ‘dig, scratch’; DRom n. groapă ‘pit, ditch’; n. grapă ‘harrow’ Cogn.: NE grave; Lith grẽbti ‘rake’; OCS pogrebǫ ‘bury’; grobъ ‘grave, tomb’; Alb gropë ‘pit, ditch’; grep ‘hook, fishing rod’. Goth graban `dig’, graba f. ‘grave’; OHG grab n. ‘grave’; OHG graban s.v. ‘dig, carve’; MHG grũbeln ds.; Goth. Grōba f. ‘ditch, trench, channel’; OCS greti ‘to dig, scrape’ (Kroonen, 2013) Possible example of merger, with the DRom and Alb forms semantically generatized into ‘pit’, whereas the OCS limited meaning to ‘grave’. PIE IEW 678 *lep-2 ‘peel, flay’; DRom v. lepăda ‘throw, discard, (archaic about animals ‘to lose hair’) undress, disown, send away, abandon, abort earlier’ Cogn.: Grk λέπω `shell’, λέπος n., λοπός m. `bowl, bark, skin’, λεπίς, λοπίς f. `scale, husk, bowl, bark’, λοπάς `flaches Geschirr’, λεπάς `einschalige Muschel, Lat lapis ‘stone’, lapidāre ‘to stone’; Grk lépō ‘peel, worn out’; Alb lapë ‘dewlap of an ox, loose piece of skin’; OE lōf ‘headband’; Lith lāpas ‘leaf ’; Rus lápotı ̆ ‘bast shoe’ (Russu 340; IIR 1981). PIE MA*terh2- ‘bring across, overcome’; IEW 1074-1075 *ter-4, terǝ- : tr-, trā-, teru-‘to cross, transgress, to stay’; DRom v. trece ‘pass (by/ on)’, n. trecut ‘past
350 Appendix
time’; v. întrece (în+trece) ‘to beat in a competition’; v. pe+trece ‘pass time. party’; perhaps DRom tare adj. ‘hard, strong’, derive. v. întări ‘to harden’; Cogn.: Av taráni- ‘continuous, penetrating, rapid, helpful’, táras- n. ‘forward penetration, energy’, instr. tárasü adv. ‘hasty, rash’, tará- adj. ‘strong’; IE *trā- Gk. τρᾱνής, τρᾱνός ‘piercing = clear, bright, distinct’; Lat intrāre ‘enter’(DRom v. intra ‘enter’); Hit tarhzi “defeats’; Skt tárati ‘overcomes’; DEX.RO Lat trājı ̆cı ̆o ? ‘to throw over, across’. PIE IEW 1085 *teup- ‘to get down, conceal oneself; DRom v. tupila ‘get down, hide’ Cogn.: Lith. tūpiu, tũpti ‘squat down, on your knees’, tŭpiù, tupeti ‘squat down, on your knees, sit’; Ltv. tupt ‘squat’. PIE *sperĝh- ‘move energetically’; DRom v. zburda ‘kids romping’ Cogn.: Grk spérkhō ‘drive, press’; Skt spŗháyati ‘desires’; Av āspərəza‘excited’; Toch AB spärk- ‘perish’. PIE MA *krob- ‘hurry’; DRom n. grabă ‘rush’, v. grăbi ‘to rush, hurry’, Cogn.: OIr crip ‘quick’; ON hrapa ‘fall, hurry’; Toch AB kārpā ‘descent, come down’; Got *hrappa ‘fast’ (Kroonen, 2013). DEX.RO OCS grabiti ‘to rob’; Derksen (2008) “SCr grâbiti ‘seize, grab, rake’; Lith. gróbti ‘seize’; Latv. grâbt ‘seize’; In Balto-Slavic and Germanic, the roots *ghrebh- ‘dig, rake’ and *ghreb- ‘seize, grab’ were mixed up to a considerable degree (Kortlandt, 1988; Derksen, 1991, pp. 321–322).” PIE IEW 399 *ĝeu-, *ĝeu̯ǝ- (besides geu̯ǝ-) ‘to advance, to hurry’; DRom n. zor ‘haste, hurry’ Cogn.: OInd junüti ‘haste, hurry’, jávate ‘hurries’, jūta-h ‘hurrying’, jū‘quick, fast’, javín ‘hurrying’, jávana-h ‘driving, animating, inciting’; Av. zavah- n. ‘power, strength’, züvar- n. ‘(physical) power, strength’; Pers. zōr ‘power’, Bal. zūt ‘quick, fast’, Av. zǝvīštya ‘the hastiest, most fastest, the most conducive’, uzutay- ‘hurrying out, foaming’; Av. java ‘hurry’; Pers. zūd ‘quick, fast’,
Appendix
351
Derksen (2008): *žuriti v. ‘rage’; E Ru. žurít’ (coll.) ‘reprove, scold’, SCr. zúriti se ‘hurry, (dial.) complain’, Sln. žúriti se ‘hurry’ DEX.RO Trk zor ‘hard, difficult’ > DRom zor ‘hurry’ PIE IEW 801-802 *pel-, pelə-, plā-, pel- ‘to drive, set in motion’; DRom v. pleca ‘leave’ Cogn.: Lith liekù ‘leave’; Grk leipō ‘leave’; Delamarre (2018) IE 669-670 *leikw- ‘to leave, to quit’. DEX.RO Lat plico, plicare ‘to fold’ implying the Roman soldiers folding their tents (?), from PIE 802-803 *pel ‘to fold > Lat simplus, duplus, triplus, complicare ‘to fold together’, etc., PIE IEW 463-465 *gʷü-, gʷem *gwmto- ‘go, come’; DRom adv gata ‘finished, ready’; v. găti ‘be ready, prepare (cook)’; Cogn.: Proto-Indo-Irarian *gata- ‘gone’; Skt gatá-, (Skt jí-gā-ti ‘goes’); Av gata- ; Grk ἀνα-βατός; Alb gat, gatuaj ‘make ready, prepare’ < Slv готовати (gotovati < gotovu) ? PIE MA *meus- ‘move, remove’ [*meu(hx)- ‘move’] IEW 743 *meu-2, meuǝ- ‘to move’; DRom v. mișca ‘move’ Cogn.: Khot muśśa ‘robbers’; Skt mușnā́ti ‘steals’; Toch AB musnā- ‘lift, move aside’, musk ‘disappear’, mäs- ‘go’; s-extension seems *meu-s- in O.Ind. musnüti, mósati ‘steals’, mósa- ‘robber, thief ’ PIE MA *muskós ‘male/female sex organs; Grk múskhon ‘male/female sex organs’; Skt mușká ‘testicle, scrotum’ DRom v. mișca ‘move’ ? PIE MA *(s)ku(n)t- ‘shake, jolt’; IEW 957-958 *(s)kūt- ‘shake’; DRom v. scutura ‘shake’ Cogn.: Lith kuntú ‘recover, get better’; kustù and kuntu, kutau, kùsti ‘get upset’, kutrùs ‘quick’, kutà ‘tassel, fringe’; OHG scutten, OFris skedda ‘shake, upset’; NE shudder ; OCS skytati sę ‘wander’. 2. *(s)keud- ‘throw’: NE shoot; Rus kidátı ̆ ‘throuw’; Alb hedh ‘throw’; Skt códat ‘incite’; Toch B kaume ‘shoot of a plant’.
352 Appendix
DEX.RO Lat excutere ‘remove’, Vegl skutro ‘remove’, It scuodere ‘shake’ (Meyer-Lubke) > DRom scutura ‘shake’ PIE MA *sper- ‘strew, sow’; DRom v. sfărâma ‘to crash,’ sfarm ‘I crash’, n. fărâmă ‘crumb, bit’ Cogn.: Grk speirō, Hit ispāri ’strew, sow’; OHG sprāt ‘scattering’; Alb farë ‘slice’, Grk spérma ‘seed’; σπαρασσω ‘unravel’ (Chantraine, 1968) DEX.RO and Orel Alb n. thërrime ‘crumb’ > DRom n. fărâmă ‘crumb’, but it was suggested that the Alb form may have a different etymology[*ter‘rub’ IEW 1071] perhaps a borrowing from DRom. PIE MA *ghrendh- ‘grind’; DRom n. grunz, grunj ‘bran, crumbling material’; ARom grundă, grundză ‘bran’ IEW 459 (Russu 328) *ghren- *ghren-d-, *ghren-dh- ‘to rub, stroke roughly’ Cogn.: Lat frendō ‘gnash the teeth’; NE grind; Lith grendu ‘scrape, scratch (off)’; Grk khóndris ‘grain’ < *khrondrós; Alb grundë, krundë ‘bran’; OE grindan ‘grind, crunch’, Eng to grind; OE grindan; Ger grand ‘coarse sand, meal, flour, bran’, OHG in grente ‘in earth full of clay’, OIce grandi ‘sandbank, gravel’; Lith gréndu, gresti and gréndžiu, gresti ‘rub hard, scour, clean’; Russ. grjada. PIE MA *h3er- ‘set in motion vertically’ and *h1er- ‘set in motion horizontally’; DRom v. ridica, arch. ardica ‘lift up, rise’; v. urca ‘climb’; IEW 326 *er-, or-, r- ; 2: ereu- ‘to move, set in motion’; IEW 339 *erǝd-, rǝd-, *erǝdh- ‘to grow’; with dh Lat. arbor ‘tree’, wherefore kurd. ür- from *ard- ‘tree’ in ürzang ‘dark hue on the trees caused by wind and weather’, Alb rit ‘grow’, from rd- or rdh-; (zero grade) < DRom v. ridica “pitch, raise, elevate, lift, perk up, straighten, loop, kick up, pick up, hoist, take up, rise, arise, get up, mount, ascend, etc.’ IEW 1167 u̯erdh-, u̯redh- ‘to grow; high’ ? > DRom v. urca ‘climb’ Cogn.: Lat orior ‘rise’; and Grk érkhomai ‘set out; come; got’; Skt ṛccháti ‘goes towards, reaches’. DEX.RO Lat eradicare “uprooted, exterminate’; Russu (1981) Lat arduus, OIr ard ‘tall’; Lat oricare (oriri) > DRom v. urca
Appendix
353
PIE IEW 128-132 *bher-1 ‘bear, carry’; DRom n. burduhan, burdihan stomach of an animal to carry cheese’; v. burduşi ‘to stuff’, v. îmburda ‘to turn over under weight’ Cogn.: OIr beirid; Lat ferō; NE bear; Alb bie; Grk phérō, Arm berem; Skt bhárati; Toch AB pār, all ‘carry’; in Slavic, ‘take’, e.g., Rus berú; Alb burdë ‘sac’; ‘Nominal forms *bhérmn- ‘load’ OCS brĕmę ‘load’; Grk férma ‘fruit’; Skt bhárman ‘load’; *bhŗtis ‘carrying’ Lat fors ‘luck’; NE birth; Skt bhŗti ‘carrying’; Lat burdo, burdus < Gaulish burdus’mule’; Engl burden‘ The Romanian n. burduf, burduh, burduș ‘leather sac for water, wine’ seems related to DRom n. burtă ‘belly’—IEW 125-126 *bhelĝh- ‘swell’ > *bhólĝhis ‘skin, belly’ as both imply the meaning of ‘carry’ and ‘preganacy’, although Vasmer considered the source to be Azerbaijan burduk “fur for wine” (Radlov 4, 1832). Persian influence? PIE *h2mey-gw- ‘to change, exchange’; ?DRom v. amăgi ‘allure, trick’ Cogn.: Grk μαγευω ‘to charm, cast a spell’ (Anthony, 2007, p. 95); DEX.RO Lat magus ‘a learned man’? PIE MA *derha- ‘work’; IEW 212 *derǝ-, drā- ‘to work’; DRom v. deretica ‘housework, do one’s room’; ARom v. adarŭ ‘do, decorate’; Cogn.: Lith dar(i)aū, darẏti, Ltv darît ‘do, make’; Grk dráō ‘make, do’ DEX.RO Lat *deradicare ‘uproot’ ? PIE IEW 1138 *u̯el- ‘to press, push’; DRom v. vârâ ‘thrust, jam, push down (in a bag) Cogn.: OCS vryje ‘to press’ Derksen: *vьrěti v. ‘boil’; SCr. vrěti ‘boil’, Bulg. vrja ‘boil, seethe’ ? PIE IEW 1139-1140 *u̯el-4, u̯elǝ- ‘hair, wool; grass, forest’; DRom n. velință ‘cover, blanket, spread ‘, v. înveli ‘to cover’; (perhaps related to the kind of blanket typically from North of Romania, made of wool, processed so that it becomes very hairy, called cergă) Cogn.: Lat vellō, -ere ‘to pull out’, derivatives vellus, -eris ‘wool, fleece’; OE, OS wlōh ‘fibre, filament, fringe, flake’ (Gmc. *wlōha-); OIce lagðr
354 Appendix
‘tuft of wool or hair’ (*wlagaÞa-); OCS vlakno, Russ.voloknó ‘fibre, filament’; OCS vlasъ, Russ. volos ‘hair’; to and from both a root form Grk λάχνη ‘hair’ (*ulksnü), λάχνος m. ‘wool’; DEX.RO OCS valiti ? Not listed in Derksen
Pronouns PIE *h1eĝ ‘I’—DRom ieu; Lat ego; Grk egṓ(n); OCS (j)azŭ; PIE accusative *h1me—DRom mine formed with the unexplained particle (Dacian?) -ne; datif mie [short form mi-]; short form mă to form reflexive ‘myself ’ mă scol ‘I wake up’; PIE *túhx ‘you’; DRom tu; Cogn.: OIr tū; Lat tū ‘thou’, tē ‘thee’; OE ϸū; Lith tú; OCS ty; Alb ti, ty, (enclitic të); Doric Grk tú; Arm du ‘thou’; PIE accusative *tewe, enclitic *te—DRom tine (Dacian? –ne), short form te ‘yourself ’ te scoli ‘you wake up’; PIE *séwe ‘reflexive pronoun—self ’; DRom sine (Dacian? –ne), short form se, ‘self ’, examples: el se place pe sine ‘he likes himself ’; el/ea se scoală ‘him/herself wakes up’; impersonal: se întunecă ‘its getting dark; n. sine ‘self ’ Cogn.: Lat sē’, OHG sih ‘him-, her-, it-self ’; Lith savē ‘–self ’; Slavic -sę ‘-self ’; Alb u ‘him-, her-, it-self ’; Grk hé ~ heé ‘him-, her-, it-self ’; Skt svá ‘one’s own’ Toch B sañ ‘one’s own’. As a noun DRom sine ‘self ’, masc. articulated sinele, fem. sinea, ‘(the) self, itself ’, as in: în sinea mea ‘within myself, in myself ’; or, sinele omului ‘the human self; or, idea în sine ‘idea itself, in essence’. DRom plural reflexive short forms [noi] ne ‘ourselves’; [voi] vă ‘you yourselves’. DRom interrogative pronoun cine [pronounced č] ‘who’, usually explained through Lat *quem, echoes PIE *kwis + suffix -ne [Dacian?] as in mine, tine, sine; cogn.: Sard kini, Calabrese chine, Sp quien.
Appendix
355
PIE IEW319-321 *eno- (probably e-no-): ono- : no- : -ne- ‘that’; DRom suffix –ne (Dacian?) forming pronouns in accusative mine, tine, sine: a căzut pe mine ‘it fell on me’ etc. Cogn.: Lat. ego-ne, tū-ne, dēnique (*dē-ne-que), dōnicum, dōnec (*dō-ne-kʷom, compare Umbr. Arnipo ‘as far as’ from *ad-ne-kʷom), quandō-ne, sīn (*sī-ne ‘if, however’), etc.; Av ciϑǝ-nă for the introduction of a question (= Lat quid-ne) Lat nē ‘yea, in truth’; Goth. -na in afta-na adv. ‘from behind’, hinda-na adv. ‘beyond’, etc.; OHG -na in obana ‘from here above’; ON Þēr-na ‘to yourself, to you’, etc. PIE MA *haénsus ‘god’; DRom n. îns, ins ‘a person, any person, an individ’, (see next under Religion); DRom pronouns m. însu-mi, fem însă-mi ‘myself ’, însu- însă-ţi ‘yourself ’, însu- însă-şi ‘him/herself ’, înşi-ne, înşi-vă, înşişi -însele, etc., Cogn.: Medieval Lat ens, entis ‘being’ < Grk ειμι ‘I am’, εις, εντος > Lat intus ‘within’; (*h2nsu-> Hit. haššu- ‘king’, OIc. áss ‘a kind of god’, Skr. ásura- ‘a kind of god’, Av. ahu- ‘lord’ in Matasovič, 2018); Grk ἑνός θεοῦ (henos theou), ‘of one god’. DEX.RO Lat in-ipse, but Lat ipse ‘self ’ is not found in Romanian by itself as in other Romance languages: It esso, Log issu, Ofr es, Sp ese, Port esse, Lomb is(a) ‘now’ (Meyer-Lubke 4541); in DRom the consonantal group ps did not develop in ss as in most Romance languages, Osc essuf, Umb esuf; the late Latin development for ipse, ipsum is is, thus in DRom in + ís with accent on í, thus DRom *ini’su ‘in self ’ with probably the syncope of accentuated i, could explain the DRom reflexive pronoun for enforcing the action of the person doing it: însu-mi, fem însă-mi ‘myself ’, însu- însă-ţi ‘yourself ’, însu- însă-şi ‘him/herself ’, înşi-ne, înşi-vă, înşişi -însele, etc., Polite forms dânsul ‘he’, dânsa ‘she’, dânşii, dânsele ‘they’ < de+insul etc. Lat idem ipse, It desso ‘himself ’, Comelico densu, denso, Friul zenso. PIE MA *yóti ‘as much’; *tóti ‘so much’; DRom atât (a-tât) ‘so much, this much’ Cogn.: Osco-Umbrian: e-tantu, e-tanto; It tanto; DEX.RO Lat eccum tanto
356 Appendix
Religious, Beliefs, and Related Vocabulary PIE MA *dhéĝhōm-, *dhĝhom- ‘earth’, IEW 414-416 *ĝhðem-, ĝhðom-, gen. ĝh(ð)m-és‚ earth’; Dacian god Zamolxis/Zalmoxis Cogn.: Skt kşam, Av za, zam, zme, Grk khthōn, Lat humus, Lith žēmė, OCS zemlja, Alb dhe, Hit tēkan, Toch A tkam, sends us to the Earth Goddess, found in the Thracian mythology as Ʃεμελη ‘Mother of Earth’ (West, 2007, p. 175), and Semele, διωσ ξεμελω, mother of Dionysos. The reflexes for ‘earth’ in Indo-European languages are: Thracian and Phirgian: ζεμελω, Old Slvavic: zemlja ‘earth’, Lithuanian Zemyna ‘Earth Goddess,’ Lat. Humus, Gr. Ξαμ−αι, Ξαμ−υνη, Demeter’s name, and possibly with the Cretan δηαί, Ionic ζηαί meaning “barley”. These forms lead us to relate the well-established divinity of the Earth with the male deity Zamolxes, the god who disappeared underground for a number of years, a connection that could be structural, etymological, or conjectural, due to an association made by the Greek speakers familiar with Thracian and Phirgian: ζεμελω ‘earth’, and the name of the Thracian goddess Ʃεμελη ‘Mother of Earth’. The same supposition could be made regarding the Greek writers transcribing the god’s name Zalmoxis, who probably were familiar to the Thracian form zalmo ‘bear skin, cover’, that could imply a relation to the German berserkers, even though there is no description of such sort in Herodotus or the others; yet, to interpret his underground sojourn as a cover under a bear skin seems less convincing. The appellative Gebeleizis or Beleizis, was linked to the same IndoEuropean root, *ǵhem-el- ‘earth’, as presumed base for Zamolxis. PIE *haén sus ‘god, spirit’; DRom îns, ins ‘a person, any person’, Aromanian nîsu; Megleno-Romanian ǫns; Istro-Romanian ăns; see above for pronoun forms. Cogn.: ON ōss ‘god’ [AEsir]; OE ōs (gen. pl. ēsa) ‘god’; Goth (as reported by Jordanes) anses ‘half-gods’, Av anhu ‘lord, overlord; life (period) of existance’, ahura- ( OIr. noíb ‘holy’, OPers. naiba- ‘good, beautiful’, cf. also MW nwyf ‘passion, joy’. OIr. níab ‘vital force’ (< *neybho-) shows that the semantic evolution was similar to the one in *(H)ish2ro- ‘provided with supernatural strength’. PIE IEW 243 *dheiǝ- : dhiā- : dhī- ‘see, show’; *dhēs- ‘religious’, dhísnya‘devout, religious’; DRom n. doina ‘elegiac song’ religious’; DRom n. doina ‘elegiac song’ Cogn.: Av daēna- ‘religion, internal being, spiritual ‘, dōiϑra- n. ‘eye’; Skt dhenā ‘prayer song’ (Oliphant, 1912); Lith dainá ‘poem song’, dainuoti ‘to sing’; Av daēnā ‘versified law, song’; Pers danah ‘woman voice’; Ir dan ‘song, poem’ (Istoria limbii române 1969) this etymology relates the DRom concept doina to ancient religious chanting. PIE *k̑ouh1ros ‘powerful, i.e., swollen’; DRom (only of men) n. însurat ‘married’, v. însura, (în+sura) ‘get married’; (REW 9107: Calabr, Abruzz nsura(re), Latr assurà; IstroRom. însurà ‘to get married’; Dalmat uzorizare) Cogn.: OIr cora(i)d, Skt śū́ra- ‘hero, powerful man’; Thracian proper name SouraDEX.RO Lat *(in)uxorare] ‘take a woman’ < IE ukʷsen- ‘bull, male animals’ in: Skt uksü m., Av uxšan- ‘bull’ (in addition? fem. *ukʷsōr ‘sprinkled’ > Lat uxor “wife”); Welsh ych ‘ox’; Goth aúhsus (gen. pl. auhsne), OIce. oxi, OE oxa, OHG, OS ohso ‘ox’; Toch В okso ‘cattle, bull’, etc. (IEW 1118); DRom did not preserve the Latin form uxor ‘wife’, only Lat soror > DRom soră ‘sister’. PIE MA *bhel- ‘blow, swell’, *bhleu- ‘swell, overflow, roar’; IEW 96 *bel-2 ‘strong’; also IEW 120-122 *bhel-3, bhlē- ‘to grow, spread,
358 Appendix
swell;’ DRom n. bală, balaur ‘monster, dragon’; Dacian anthroponyms Balas, Balius, Areibalos, Bazobalis, Decebalus, Dinibalis (I. I. Russu, 1967, p. 92), implying, strong man, king. Cogn.: Skt. bala ‘physical power’; Ir bale ‘strong’; Alb bollë ‘snake’; Skt bála-m n. ‘force, strength, power’; Phrig. βαλήν⋅ βασιλεύς. Φρυγιστί ‘βαλήν—king (Hsch); the Serb and Bulg forms blavor ‘hurricane, terrible event’ are considered loans from DRom; Lith bljaúju ‘spew’; OCS bljujǫ ‘spew’; Grk phleō ‘gush, teem, overflow’; Lat dēbilis ‘feeble, weak’ 2.*hxolu-, or *alu- ‘spell’; DRom n. hală, ală ‘monster, creator of storm’? OCS loan? Cogn.: Hit alwanzatar ‘witchcraft, spell’; Germanic/Runic alu ‘spell’. DEX.RO Blg Srb Mac ala ‘female demon’ DRom n. bală, balaur ‘monster, dragon’ may be related to negative divinities as Rus Velesu, Volosu; Lith Velnias; Skt Varuna, Gaul Vellaunos; Hit Walis (based on PIE *wel ‘see’; 2. *wel ‘die’ as per Jakobson, 1969 in Petrosyan, 2016) PIE MA *hayeu ‘young’; IEW 510-511 *ieu- ‘young’; DRom Iovan mythical character (see in this volume, chapter The Hero…) Cogn.: Lat iuvenis, Skt yuvan, ‘youthful’ yuvānā (RV 1. 117. 14) Av Yavan; Lith jáunas, Ltv jaûns; Gaul *yuwənko- ‘young’; OBulg junъ ‘young’—not Ivan. PIE MA *keudes- ‘magic force’; [IEW 955 *skeud-1 ‘to protest, grumble’]; DRom n. ciudă ‘envy, anger, distress, negative spell or feeling, spite’; n. ciudat ‘strange, unusual’; in expressions: în ciuda faptului ‘despite, in spite of ’; (se) înciuda v. ‘to get mad, upset’ Cogn.: NWels hud ‘magic’; ON seið ‘magic’; OCS čudo ‘wonder’; Grk kudos ‘renown, fame’ IEW 955 *skeud-1 Lith pra-skundù, -skudaũ, -skùsti ‘to ache, to exhaust begin’, skundà ‘accusation’, núoskunda ‘pity’, skùndžiu, skusti ‘bemoan’; Ltv skund-u, -êt ‘envious, be angry with’; Ltv skàužu, skàust ‘envious sein, injure, hurt’ DEX.RO OCS čudo ‘wonder, miracle’; a possible merger with a Slavic form that changed its meaning in DRom.
Appendix
359
PIE MA *dom(ha)u-no-s ‘master’*deiwós and *dyeu- ‘sky, day’ < *dei ‘shine, be bright’; DRom Dumnezeu, Dumnezău ‘God’, formed of Rom domn, Lat dominus; Skt damuna, word clearly derivative of ‘house’ *dom(ha)os with the suffix *-no-which was used to create words meaning ‘leader of ’, and zeu, zău (< Lat deus? Grk Ζεύς. vocative Ζεῡ (Zeu). Cogn.: OE dōm ‘fate’; Goth dōms ‘sentece, glory’; Av dāman ‘abode’; Skt dhā́man ‘law’; Grk θεμα ‘assertion’, αναθημα ‘votive offering’; OIr domnán ‘world’, domun ‘world’, domhan ‘universe’, dumno ‘house’, and also in proper names: Gaelic Domhnall, Dumnorix ‘worldking’; region Dumnonia in the western part of England; and OIr dīa; Lat deus, Lith diēvas; Hit sius; Skt deva; Thrac Saba-zios; Grk Zeus; Lat Jove; DEX.RO Medieval Lat dominus deus ‘master god’ > Rom Dumnezeu. PIE IEW 183-187 *dei-1, deiǝ-, dī-, diā- ‘shine; day; sun; sky god, god’ > Greek Ζεύς (= dyüu-h), acc. Zῆν (= dyüm), voc. Ζεῦ (*diĕu), may be cognate with DRom zeu ‘god’; Alb zot ‘god’; DRom zână, zeiţă, zeitate, ‘goddess’; Alb zana ‘nymph, goddess’. PIE *wihxrós ‘man’; DRom Fârtat, mythic character, creator of the earth, luminous, associated with light’ (see in this volume, the chapter Cosmogony) Fârtat formed with the suffix *-tah2t inherited from PIE semantically designating abstract nouns. Cogn.: OIr fer; Lat vir; OE wer; Lith výras ‘man, husband’; Av vīra ‘man’; Skt vīrá ‘hero, man, husband’; Toch A wir ‘young, fresh’; Alb Gheg burr, burrë (*buro-) ‘man, husband’, burrë ‘man, valiant man, proud man’; OHG baro ‘man, husband’ (Fournet, 2012); DEX.RO Lat frater ‘brother’. PIE IEW 473-474 *gʷēnā (*ĝhʷēnā) ‘woman, wife, *goddess. *gwénha ‘woman’; DRom n. zână ‘fairy’ Cogn.: OIr ben ‘woman, wife’; OE cwene ‘woman, female, prostitute’; OPrus genna ‘wife’; OCS žena ‘wife’; Grk gunḗ ‘woman, wife’; Arm kin ‘wife’; Av gənã- ‘woman, wife’; Skt gnā́ ‘goddess, divine female’; Toch B
360 Appendix
śana ‘woman’; Alb zonjë ‘mistress, wife, woman’ (*gʷeniü); gheg. grue, Tosc grua ‘wife, woman’ (*gʷn-ōn); Alb (*gʷaniü) zana ‘nymph, goddess’ 2. *ĝiena < *ĝeie- ‘subjugate, overpower, oppress’; DRom n. zână ‘fairy’ (Detschew, 1957) Cogn.: Skt. jyanay-, Av. zyanay-, zyana-, ‘to overcome, conquer, weaken’ 3. PIE IEW 183-187 *dei-1, deiǝ-, dī-, diā- ‘shine; day; sun; sky god, god’ > Greek Ζεύς (= dyüu-h), acc. Zῆν (= dyüm), voc. Ζεῦ (*diĕu), may be cognate with DRom zeu ‘god’; Alb zot ‘god’; DRom zână, zeiţă, zeitate, ‘goddess’; Alb zana ‘nymph, goddess’.
lphabetic List of DRomanian Words A Discussed Above abure n. ‘mist, vapor’, v. aburi *nbh(ro/ri) ‘rain,’ acăța, agăța v. ‘hang’ *haek̑- ‘sharp,’ ARom acățare, acats adia v. ‘to breeze, soft wind’; n. adiere ‘breeze’ *dheu(hx) ‘be stirred like dust or smoke,’ ARom adil’a, adil’are ‘breeze’ afară adv. ‘outside’ *apo-, pō̆, ap-u, pu, pos ‘away from’ alac n. ‘spelt’ *h2/3(e)lg̑(h)ameți v. ‘dizzy’ *médhu ‘mead’ Ampoi hydronym *nbh(ro/ri) ‘rain’ amurg n.’dusk’ *(ha)merhxgw- ‘dark,’ ARom n. amurgu andrea, undrea n. ‘knitting needle’ *andher-, ņdher- ‘stem, spike’ apă n. ‘water’ *h2eP- ‘living water’ apoi adv. ‘then, afterwards’, *h4épo ‘back, behind’ aprig adj. ‘intens, fierce, bold, forceful’ *abhro- : ‘strong, mighty” apuca v.’to grab’ *ap-! (proper ǝp-) : ēp- ‘to take, grab, reach’ arac, harac n. ‘stick or rush to support plants *haer- generally ‘reed or rush’ Ardeal ‘Transylvania’ *h2erdus ‘high’ ardica (archaic) v. ‘to lift up, raise’ see ridica argea (archaic) n. ‘underground room, niche’ *h1regw- es ‘place of darkness; *areg- ‘to lock’
Appendix
361
Argeș river, hydronym *h4ṛg̑ -es- ‘white, silvery’ arunca v. ‘to throw away’ ARom aruncarea, arcare, aruca *reu-2, reuǝ-, rū- ‘to tear out, dig out, open, acquire, *h3ru-n-k- ‘to dig out’. ascuţi, ascuţit, v. ‘to sharpen’, *haek̑ ‘sharp, pointed’ asemenea adv. ‘like’, v. asemănare *somos [o-grade nominal] ‘same’ aşeza (se) v. ‘sit (oneself ), set something down’ *h1ēs- ‘sit’ atât adv. ‘this much’ *yóti ‘as much’; *tóti ‘so much, this much’ auş (archaic) ‘old man, grandfather’ *h2euh2os ‘grandfather’ azi n. ‘today’ *haég̑hṛ ‘day’ ba ‘no’, [duplicated] ba nu ‘no, no’ *bhē̆, bhō̆ (an emphatic particle) băga v. ‘insert in a bag, container’ *bhag- ‘apportion’; ARom bag, băgare ‘id’ bală n. ‘dragon, monster’, *bhleu- ‘swell, overflow, roar’; *bel-2 ‘strong’; *wel ‘see; die’ bală n. pl. bale ‘slobber, drewling’ *balbabalaur n. reflex from bală ‘dragon’ băl(i) ‘ blond’, bălai ‘blond hair’, bălan ‘animals with white hair’ *bhelh1 ‘white’ bâlbâi v. ‘stammer’ *balbabaligă, balegă n. ‘dung’ *bolg(w)-o- ’bulb’; ARom balig, baligă baltă n. ‘puddle’, Bălașa anthroponym, toponym *bhelh1 ‘white’, ARom baltă ‘id’ ban n. ‘coin’ *bheh2- ‘shine’; ban n. ‘leader of a region’ *bend-, bnd-no- ‘spike, needle, summit’ bardă n. ‘hatchet’ *bhar-s ‘pointed’, *bhar-, bhor-, bhr- ‘bristle, stubble’ Bârsa toponym, Bârzava river name *bherg̑h ‘high, hill’ Bârzava, Dacian Bersobis, Bersovia, Țara Bârsei barză n. ‘crane’, adj. barz ‘grey, black and white bird’ *bherhxĝ‘shine, white’ bârzoi, adj., v. bârzoia ‘of hair, spiking, sticking up on the head, or an animal tail’ *bhares- bhores- bhrsti-, bhorsti- ‘point, stubble’ bască n. ‘hat’, *bhasko- (*bhedh-sko) ‘bundle, heap’, ARom bască, baști ‘id’ băţ n. ‘stick’; n. bâtă ‘club’ *bhāt- : bhət- ‘to hit, strike’ *bak ‘club’
362 Appendix
berc n.’small woods by water’; bercar ‘honey gatherer’ *bherhxĝos ‘birch’*smē[i]k-:smīk‘grated, tiny crumbs’ mied n. ‘mead, hodromel’ *médhu ‘mead’ miel n. ‘lamb’, [fem. mială, mioara] *meh1l- ‘small animal’ mieriu, mieru adj. dial. ‘blue’ *mel-6, melǝ- ‘dark colour, black, dirty’ mire n. ‘bridegroom’ *méryos and *meriha- ‘young woman’ mișca v. ‘move’ *meus- ‘move’ mocirlă n. ‘slush, slime’ *māk-1 ‘damp, to soak’ Moldova, old Moldava toponym, *mḷdho/eha ‘clay’ moș n. m., moașă f. n. ‘old man, midwife’ māk̑- ‘long’ Mureș River n. hydronym *mori, mōri ‘sea’ murg n. ‘colt’ *(ha)merhxgw- ‘dark’ murg(a) n. ‘evening’ *(ha)merhxgw- ‘dark’ mușca v. ‘bite’ *menth2- ‘stir’; IEW 732-733 menth-2 ‘to chew, mouth’
370 Appendix
nai n.’Pan’s flute’ *nedós ‘reed’ naiba n. ‘euphemism for devil, negatve force, negative holy’, *noibhos ‘holy’ nană n. [arhaic respectful form] ‘aunt, old woman’ *n-h4en-, h4en- ‘grandmother’ năpârcă n. ‘snake, viper’ *nē-tr, nǝ-tr ‘snake’ nărui v. ‘colapse of a wall, cliff, and such, ruin’ *ner ‘under’ neam n. ‘relatives, [own] people’ *nem- ‘apportion, count, take, accept legally’ negură n. ‘darkness’ *nébhos ‘clouds, sky’ nepot/nepoată n. ‘grandchild and nefew/niece’ *népōts ‘descendant’ oi pl., sg. oaie n. *h2ówis ‘sheep’ olog n. ‘cripple of one leg’ *h3elek ‘elbow, forearm’ oraș n. ‘city’ *wérhxus ‘broad, wide’; *wriyo/eha- ‘citadel, city, town built of a hill’ ortoman n. ‘man with power, rich’ *worhxdhus ‘upright, high’ pană n. ‘shim’ *sphaen ‘flat-shaped piece of wood’ pară n. ‘intense fire’*peh2ur ‘fire’ pânză n. ‘linen’ *p(e)h2no/eha ‘linen cloth’ pârâu, părău, pârău n. ‘brook, creek’ *per-1, perǝ- : prē-, preu- to drizzle, sprinkle, jet’ plai n. ‘field’, platoșă n. ‘shield’ *pḷth2ús ‘broad’ > *pleth2- related to *pelhak ‘spread out, flat’, *pleh2- ? pleca v. ‘leave’ *pel-, pelə-, plā-, pel- ‘to drive, set in motion’ plesni v. ‘strike’ (by hand) *ples+ne? *plehak/g ‘strike, punish’ plesni v. ‘rip, tear, burst’ *plēḱ- ‘break, peel, rip, tear off’ potecă n. ‘path’ *pent ‘find ones way’ *póntōh2s ‘(untraced) path’ potrivi v., (adv. împotrivă) ‘make even’ *terh2 ‘bring across, through’ prunc n. ‘baby, small child (onlt of humans)’ *per- ‘offspring (of an animal) pupăză n. ‘hoopoe’ *h1epop- ‘hoopoe’ râcâi v. ‘scratch’, n. *rei- ‘scratch’>*reik- ‘scratch, line’ rău m., rea f. adj. ‘bad’ *reus- ‘anger, rage’ rece adj. ‘cold’ *h1regw- es ‘place of darkness’
Appendix
371
rezema v. reazem, razem, n. ‘prop, sustain, lean on’ *reid- ‘to lean on, support’ ridica, rădica v.’rise, lift up’ *h3er- ‘set in motion vertically’ roșu adj. ‘red’ *h1roudhós adj. ‘red’ rouă n. f. ‘dew’ *rốs ‘dew’, *h1res- or h1ers- ‘dew’ IEW ere-s-2 (ers-, rs-, eres-), and rēs-, rōs- rug n. pl. ruji ‘bramble’ *reu-2, reuǝ- : rū- ‘to tear out, dig out, open, acquire’ ? samă, seamă, de-o seamă, asemenea, (se) aseamănă adv. *som- ‘together’< o-grade *sem- ‘one’ sâmbure n. ‘fruit stone, pit’ *ḱeub- ‘thorn’? Maybe Alb (*ḱeub-) thumb ‘thorn’ șapte ‘seven’ * septm̥ (*sek̑ʷh-) ‘seven’ sărman n. ‘man in distress, poor’ *h1ermen- ‘sickness’ şase *ksek̑’s ‘six’ sat n. ‘village’ *sed- ‘sit’ scai n. ‘thistle’ *(s)kel ‘cut’, *(s)k(h)ai-, (s)k(h)ai-d-, (s)k(h)ai-t- ‘to hit, kick’ scăpăra v. ‘to make a sparkle with flint’ *skebh- ‘scratch, shave’ scărmăna v. *skolmeha- ‘sword’ scărpina v. ‘scratch (of skin) *(s)ker-4, (s)kerǝ-, (s)krē- to cut’ (s)kerp-, (s)krepșchiop n. ‘lame’ *skēi-bh-, -p-, nasalized ski-m-bh- ‘slant; to limp’ scoabă n. ‘clamp’, scobi v. ‘to hollow, scoop’ *(s)kēp-2, (s)kōp- and (s) kāp-; (s)kēb(h)-, skob(h)- and skāb(h)- ‘work with a sharp instrument’ scorbură n. ‘tree hollow’ *(s)kerb(h)-, (s)kreb(h)-, nasalized (s)kremb- ‘to curve, turn’ scoroji v. ‘dry out, as of lether or skin’ *(s)kerb-~(s)kerbh- ‘shrink, shrivel’ scrijeli v. ‘scrape’ *(s)ker-, *(s)kerǝ-, *(s)krē- ‘to cut’ scrum n. ‘singe, cinder, ash’ *kr-em- ‘burn’ scula v. ‘rise, stand up, wake up’; *(s)kel- to spring; (škel-) *shkel (*spring?) ‘step’ scurma v. see scărmăna v. *skolmeha- ‘sword’; *(s)ker- ‘cut’ scurta n. ‘cut short’ *(s)ker-, extended form *(s)kert- ‘cut’ scutec n. ‘diaper’, swaddle’ *(s)keu-2, *(s)keuǝ : *(s)kū- ‘to cover, wrap’
372 Appendix
sec adj., seca v., secetă n. *se(n)k- ‘dry up’ sfărâma v. ‘to crumb’ *sper- ‘strew, sow’ Simziana ‘Fairy Queen’ *som- ‘together’< an o-grade of *sem- ‘one’ sine, se, pron. *séwe ‘reflexive pronoun—self ’ sineală n. ‘blue paint’ *k̑yeh1- ‘blue, purple, green, azure, grey’ şir n. ‘row, line’ v. înșira ‘line up’ *ser- ‘line up’ Siret (River Name) *sreu- *sr- *ser-1 (*ker- < *kel-) ‘to flow’ smeur, smeuriș n. ‘the raspberry bush’, smeură, zmeură ‘raspberry’ *moro‘blackberry’ and *ser3 *sor- ‘red’ smoc n. ‘turf of hair’; vb. smuci ‘pull, jiggle’ *smók’wṛ ‘beard’ soc n. ‘small tree used for tea from leaves *sokwós ‘sap, resin’ stâlp n. ‘post, pillar’ *stup- ‘stump, broom, club’ and *stḷneha ‘post’< *stel ‘stand’ stână n. ‘the sheep and shepherd dwelings, sheepfold’ *stéh2-no- ‘standing’ stâncă n. ‘rock’ *stegh-, nasal. *stengh- ‘to stick; pole, stalk, etc.’ stârni v. ‘stir up’ *ster-5, sterǝ- : strē-, steru- : streu- ‘to widen, to scatter’ steag n. ‘flag’ see stejar steajer, steajăr n.‘pole’ see stejar stei n. ‘peak of stone, sharp peak’ *(s)tei- ‘sharp, spike’ stejar n. ‘oak’; stinghie ; steag, steajer, steajăr ; stînjen 1,366 meters *(s)teg ‘pole, post’ şterg(e) v. ‘wipe out’ *(s)tergh- ‘crush’; tergō, -ere, tergēo, -ēre ‘clean’ sterp, stearpă, stiră n.’barren’ *ster- ‘barren’ stinghie n. ‘wooden beam’ see stejar stînjen n. ‘measurement—1.366 meters’ see stejar strica v. ‘damage’ *(s)tergh- ‘crush’ strugure, pl. struguri n. ‘grapes’ *sre/ohags- ‘berry, fruit’ strungă n. ‘narrow path for milking sheep, gorge’ *strenk, streng ‘narrow’ stup n. ‘beehive’ .*stup-, *stḷneha ‘post’< *stel ‘stand’ ‘stump, broom, club’ sul n. ‘hollow tube’ *k̑súlom ‘worked shaped wood; post, stake’ sup(b)suoară sup(b)ţioară n. ‘armpit’ *(s)k̑up ‘shoulder’ sur(ă) adj. ‘grey, grey hair’ *k̑er- ‘greyish blue/green’ surată n. ‘woman in sacred surority’*k̑ouh1ros ‘powerful, i.e. swollen’ talpă n. ‘sole of foot, [metaphorical] any base’ *telp- ‘have room’
Appendix
373
tângui v. ‘complain, whine’ *teng- ‘feel, be of an opinion’ tare adj. ‘hard, strong’, v. întări ‘to harden’ *(s)ter-, (s)terə- : (s)trē-6: ster‘stiff, immovable solid’ tărăboi n.’noise, uproar’ *ter- ‘say, noise’ tărâm n. ‘region far away, underworld’ *tel-, telǝ-, telu- ‘flat, flat ground’, *ter- ‘cross over’ țarc n. ‘pen’ *terh1- ‘pierce’; *serk- ‘repairing an enclosure’ targă n. ‘stretcher’ [old borrowing] *dergh- ‘grasp’ târnă n. ‘wicker basket’, hydronim Târnava *tṛ́nu- ‘thorn’ ţăruş n, ‘stake used for gardening, enclosing’, *terh1- ‘pierce’; tată n. ‘father’ *tatțeapă n. f. țep n. m. ‘spike, thorn’; v. înțepa ‘sting’, anthroponym Tepeluș, Vlad Tepeș *keipo-, koipo- ‘peg, sharp piece of wood or stone’; *(s)kēp-2, *(s)kōp- and *(s)kāp-; *(s)kēb(h)-, *skob(h)- ‘to work with a sharp instrument’ tei n. ‘linden’ *pteleyeha- pteleweha- ‘elm’ terci n. ‘hominy, mush, gruel, mash of corn flour’ *(s)terk- *(s)ter-8 ‘dirty water, mud, smear’ țesală n. ‘horse comb’ *tekÞ- (*tesḱ-) ‘to plait; woodwork; carpenter’ toval n. ‘cow skin for shoes’ *twéks ‘skin’ trăncăni v. ‘speak endlessly, blabber’ *dhren- ‘drone’ treabă n. ‘[woman’s] work around the house’ *trēbs ‘dwelling’ trece v. ‘pass, go through’ *terh2- *ter ‘bring across, through’ tupila v. ‘get down, hide’ *teup- ‘to get down, conceal oneself uger n. ‘udder’ *udero- ‘uterus/womb’, *h1óuhxdhŗ ‘breast, udder’ uità v. ‘see, look’ and ‘forget’ *wet- ‘see (truly) uliu n. ‘hawk’ *ulu- ‘owl’ ului v. ‘astonish’, uluitor n. ‘astonishing’ *ul- ‘howl, hoot’; *ulek-, *ulk(*sulek-) ‘to shine; fiery’ undrea see andrea ‘knitting needle’ urca v. ‘climb’ *h3er- ‘set in motion vertically’ urdă n. ‘sweet new cheese (similar to ricotta)’ *uerdh-, *uredh- ‘to grow; high’
374 Appendix
usca [se], eu usuc, se usucă v. ‘dry up’ *haes- ‘be(come) dry’ vai [interjection] ‘woe!, alas!’; n. vaiet ‘wailing, weep, lamentation for the dead’ vâlcea? n. ‘meadow’ *wélsu- ‘meadow’ < *wel- ‘grass’ văpaie n. ‘flame, shooting flame’ *u̯ep- ‘throw, throw out’) vârâ v. ‘to thrust, jab, push’ *u̯el- ‘thrust, press, push’ vârf n. ‘peak’ *wers- ‘peak’ vărsa v. ‘poor’, n. vărsare *h1wers-, or, *we/ohxr ‘rain, water on the move’ vătăma v. ‘to hurt, ingure, damage’ *wedh- ‘push, strike’ uā-, uō-, uǝ- ‘to hit, wound’ vatră n. ‘hearth’ *h2ehx-tr-eha ‘hearth’; *āt(e)r- ‘fire’ vătuiu n. ‘one year old lamb’ *wet- ‘entire year’ viscol n, ‘blizzard’ *au̯(e)-10, ˙*au̯ē(o)-, *u̯ē-; d. *u̯ē-s-: ‘to blow’ vrej, vreg n. ‘stalk, as of vine stalk’ *wrehagh- ‘thorn’ Zamolxes Dacian god *ĝhðem-, ĝhðom- ‘earth’ zână n. ‘fairy’ *gwénha ‘woman’ or, *ĝiena