MagyarNyelv

MagyarNyelv
SzentKorona  OsTudas

 
  • John Bowring
POETRY OF THE MAGYARS, Preceded by a Sketch of the Language and Literature of Hungary and Transylvania by John Bowring London : Printed for the Author, 1830. Preface 6.
 The Magyar language stands afar off and alone. The study of other tongues will be found of exceedingly little use towards its right understanding. It is molded in a form essentially its own, and its construction and composition may be safely referred to an epoch when most of the living tongues of Europe either had no existence, or no influence on the Hungarian region.



Variety 'from the street':
Would you be able forget this language?
The Hungarian language goes far back. It developed in a very peculiar manner, and its structure reaches back to times most of the spoken European languages did not even exist. It is a language in which there is a logic and mathematics with the adaptability and malleability of strength and chords. The Englishman should be proud that this language indicates an epic of human history. One can show fort his origin, and alien layers can be distinguished in it, which gathered together during the contacts with different nations. Whereas the Hungarian language is like a rubble stone, consisting of only one piece on which the storm of time left not scratch. It is not a calendar that adjusts to the changes of ages. This language is the oldest and most glorious monument of national sovereignty and mental independence. 
What scholars could not solve, they ignore. In philology it the same as in archeology. The floors of the old Egyptian temples which were made out of the a single rock cannot be explained. No one knows where they came from, from which mountain the wondrous mass was taken, or how they were transported and lifted in place in the temples. The genuineness of the Hungarian language is much more wondrous than this. He who solves it shall be analyzing the divine secret:
“In the beginning was the Word, and the Word was with God, and the Word was God.”

Sir John Bowring 
english philologist 
(1792-1872)

Feledni tudnád ezt a nyelvet?
A magyar nyelv a régmúltba vezet. Nagyon sajátos módon fejlôdött és szerkezete ama távoli idôkre nyúlik vissza, amikor a legtöbb európai nyelv még nem is létezett. Ez egy olyan nyelv, melynek logikája és matematikája a feszítet húr erejének kezelhetôségével és rugalmasságával bír. Az angol ember büszke lehet, hogy nyelve magában hordozza az emberiség történetét. Eredete kimutatható, láthatóvá tehetôk benne az idegen rétegek, melyek a különbözô népekkel való érintkezés során rakódtak egybe. Ellenben a magyar nyelv olyan, mint a terméskô, egy tömbbôl van, amin az idô vihara egyetlen karcolást sem hagyott. nem naptár ez, amely akorok változásához alkalmazkodik. E nyelv a legrégibb és legdicsôségesebb emlékmûve a nemzeti önállóságnak és szellemi függetlenségnek. 
Amit a tudósok képtelenek megfejteni azt mellôzik. A nyelvészetben csakúgy, mint a régészetben. A régi egyiptomi templomok alapjai, amik egyetlenegy kôbôl készültek, nem magyarázhatók. Senki sem tudja honnan származnak, melyik hegységbôl szerezték a csodálatos képzôdményeket, vagy miként szállították és emelték a helyükre  ôket a templomokban. A magyar nyelv kialakulása ennél sokkal bámulatosabb. Aki ennek titkát megoldja, az isteni titkot fogja kifejteni: 
“Kezdetben vala az Ige, és az Ige vala Istennél, és Isten vala az Ige.”
Sir John Bowring 
angol nyelvész, fordító 
(1792-1872)




  • Mario Alinei

Mario Alinei – Etrusco: Una Forma Arcaica di Ungherese [Etruscan: An Archaic Form
of Hungarian] (Il Mulino, Bologna – 2003)
Reviewed by Jonathan Morris1.
Etruscan words such as mi (I), eca/ita (this), maθ (honey), tin (day) and tur (give) have
long persuaded many scholars that Etruscan is a Eurasiatic language, perhaps even an
Anatolian language (Bomhard) that split from a common Indo-European stem at a very
early stage. The precise nature of its affiliations nevertheless remain obscure. In what is
probably the most interesting account of recent years, the Italian dialectologist, Mario
Alinei, suggests in his new book that Etruscan is nothing more than an archaic form of
Hungarian with extensive Turkic borrowings.
This linguistic proposition rests on two historical/archaeological propositions – an
uncontroversial one that the Etruscans came from the Carpathian basin, and a highly
controversial one that identifies them as a proto-Hungarian/Uralic people.
The first of these had already been demonstrated by the late 1960s by archaeologists such
as Hugh Hencken, who highlighted the cultural continuities between the Urnfeld cultures of
Central Europe and the proto-Villanovan cultures of Northern and Central Italy, suggesting
that the former culture had introduced a series of innovations to the latter, such as hydraulic
engineering, the horse, the sword. Hencken also pointed out that the Urnfelders had
probably left their signature among the Sea Peoples who attacked Mycenae and the Egypt
of Ramesses III towards the end of the second millennium B.C., in the form of ships with
prows in the form of horned birds’ heads, as well as a name cited by Egyptian sources, the
Tursha which agrees with the Greek name for the Etruscans, the Tyrsenoi, and as Alinei
tentatively suggests, with Türk.
Lawrence Barfield noted that Central Europe was the ‘industrial heartland’ of Bronze Age
Europe, whose inhabitants developed their metalworking skills and by extension, the
military technology that would have allowed them to become a colonial elite, capable of
seeking mineral resources elsewhere and subjugating other less technologically advanced
peoples. In this sense, their exploitation of Central Italy’s mineral wealth during the Bronze
Age is hardly surprising. Alinei nevertheless believes that this process of gradual
infiltration and scouring Europe for high quality mines may have begun as early as the
middle of the 3rd millennium, accelerating during the Polada culture. While the rule seems
to have been peaceful coexistence between these Central Europeans and the Italic locals of
the Palafitte/Terramare cultures, it appears that around 1250 B.C., migration from the
Carpathian basin led to conflict and the overthrow of these local cultures, after which the
proto-Etruscans moved into Central Italy and eventually carved out their own state that
became the locus of the Villanovan culture.
While the above sequence of events does not necessarily place a Hungarian label on these
Bronze Age Urnfeld peoples, it follows from Alinei’s continuity theory (see my review of
1 We would like to thank Mother Tongue for permission to reprint this article.
Origini delle Lingue d’Europa) that Italic speakers are the original occupants of Italy and
the Western Mediterranean. Hence, the Etruscans could only be an intrusive presence,
despite the claims to the contrary by the classical historian, Dionysius of Halicarnassus.
What has hidden the Uralic affiliations of Etruscan is its highly variable spelling, although
Alinei assures us that its latitude is no worse than in Mediaeval Florentine or Venetian
texts. If the Etruscans were a warrior aristocracy that was gradually absorbed by its
subjects, then it presumably recruited its scribes from its Italic-speaking subjects, who
wrote in a vowel-poor alphabet of Semitic origin, thus obscuring the open syllable,
agglutinative nature of a Uralic language with extensive vowel harmony.
These links nevertheless become clear when we consider the Etruscan vocabulary for its
offices of state. Writing in the 10th century, the Arab historian, Ibn Rusta, noted that
Hungarian tribes split their leadership between a warlord wielding de facto executive
power, the gyula, and a largely ceremonial but revered king, the kende. Alinei finds that the
main offices of the Etruscan state included the ZILA/ZILAΘ/ZILCI/ZILI/ZILX, identified
by Greek sources as the military commander, and the CANΘE/CAMΘI/CANΘCE, the
princes civitatis or leader of the Etruscan community. Then there is the knight, LUCUMO
(H. ló (horse) + Komi. kom (man)), the two-headed axe, PURΘ (H. balta (axe), Chuvash
purte), and the land surveyor, MARUNU (H. mérő (measure)), to cite but a few examples.
Once one overcomes this hurdle, the relationships become much clearer, the main
phonological differences being Etr. θ > H. t, Etr. c > H. k/h, Etr. z > H. gy/cs.
I have chosen the following examples from among the hundreds that Alinei provides to
give a flavour of his proposed correspondences, which demonstrate the phonological
conservatism of the Uralic languages. (NB H. = Hungarian, M. = Manty):
Etr. atranes > H. arany (gold) [Alinei points out that this was probably a general FUg
borrowing tharana, from Iranian saraña]; Etr. avil > H. év (year); Etr. calu > H. hal (die);
Etr. caθ/cat/caθinum/caθna > M. kot (sun); Etr. elśsi > H. első (first); Etr. fulu (smith) > H.
fűlő (stoker of fire); Etr. hus > H. hős (young); Etr. ilacve > H. elégvé/eléggé (sufficient);
Etr. iθal > H. ital (beverage); Etr. laukh/lux > H. ló (horse); Etr. mar- (measure) > H. mér-
(measure); Etr. nac/nacna > H. nagy (big); Etr. parliu (to cook) > H. párol (to boil/steam);
Etr. rasna (territory, region, country) > Old H. resz (region, territory) [from FUg räc3
(piece, part)]; Etr. tes/tez > H. tesz (do); Etr. uru (Sir, lord) > H. úr (landowner, lord); Etr.
zilacal (stars) > H. csillag (star).
Indeed, with such a key, the Etruscan phrase zilaθ mexl rasnal/s can be read as ‘magistrate
of the Etruscan country’. The word rasna which Dionysus of Halicarnassus misread as the
Etruscans’ name for themselves is merely the word for country, while Alinei identifies mex
as an archaic world for people, similar to magyar.
The origin of the Hungarian nation is traditionally dated to the conquest of its national
territory in the Carpathian basin by Arpad in 895 A.D. This view evidently obliges the
Hungarians to mill around on the steppes of Central Asia for millennia before they receive
a European ‘visa’, and may at first sight be reinforced by the fact that within the Uralic
family, Hungarian’s closest relatives are the Obugric languages, Mansi and Khanty, that
occupy lands around the upper Ob and Irtush rivers in Western Siberia.
What is highly suspect about this ban is that it does not apply to other Uralic peoples, such
as the Finns, Lapps, and Komi, who are thought to have spent the Ice Age in a watery
refuge in the Ukraine and Southern Russia before moving North to exploit the new hunting
opportunities provided by the retreating glaciers.
In addition, contemporary Arab sources from the 10th century onwards, most notably al-
Garnarti, writing around 1080, speak of two groups of Hungarians, one living on the
Danube and another 2000 km to the East in what is now the Bashkir republic, whose
aristocracy was bilingual in Turkish and Hungarian, and which shared the gyula/kende
model of kingship with the Khazars. Indeed, it is highly significant these words are of
Turkic origin, with H. gyula reflecting Bashk. yulaj and kende Tatar kündü [reverence,
profound respect].
Archaeological evidence (e.g. from cemeteries) has confirmed the cultural continuities
between the two groups. Furthermore, the Hungarian king, Géza I (1074-77) received a
crown from the Byzantine emperor inscribed with the legend ‘to Geza, the faithful king of
the Turks’. Indeed, the heavily Turkicized character of the Hungarians, as is apparent from
their music and mythology, makes it most likely that less discerning classical sources
would have labelled them with the hold-all description of Scythes.
On this point, the linguistic evidence is illuminating, in that Hungarian shares a vocabulary
with Mansi and Khanty for horses and wagons that is borrowed from Turkic (e.g. H. ló, M.
low [horse]; PUg. närk3, M. näwrä, H. nyerëg [saddle]); PUg. päkka, Kh. päk, H. fék
[bridle, rein]; PUg. säk3r3, Kh. iker, H. szekér [vehicle], but is unique among the Uralic
languages in also borrowing its agricultural vocabulary from Turkic (e.g. H. eke [plough],
H. árpa [barley], H. búza [wheat], H. sajt [cheese], H. tinó [ox]).
This suggests that the proto-Hungarians were still united with the Mansi and Khanty at a
stage when they were pre-agricultural nomadic pastoralists involved with horsebreeding,
but that the proto-Hungarians subsequently split away and were introduced to agriculture
by another Turkic people. We may also conclude that the Hungarians were not present in
Europe at the time they acquired their knowledge of agriculture, since if they had been, we
would expect them to have borrowed an Indo-European agricultural vocabulary.
Assuming that by the Neolithic, they were more or less located in the Obugric region, a
move South and West across the Urals would have brought them into contact with the
Sredny Stog culture, well known as the precursor to the Kurgan culture, which intruded
from the steppes into Europe, firstly into Eastern Hungary and Romania where its bearers
encountered the Bodrogkeresztúr culture towards the end of the 4th millennium, and later, in
greater numbers into the Carpathian basin itself, at the time of the Baden culture (around
2600 B.C.), which Alinei identifies as originally Slavic in origin, explaining the Slavic
toponomy of the area. Hence, far from announcing the proto-Balts of Gimbutas’ theory, the
Kurgans are actually a manifestation of a Hungarian invasion.
Alinei readily admits that there are areas of Etruscan that have not been explained by his
theory, such as its words for numbers. His main point about the Turkic origins of Etruscan
vocabulary for offices of state is nevertheless a powerful one. His theory also has the
distinct virtue of generating testable hypotheses, most notably regarding the separation of
the Hungarians from the Obugric group. If one accepts these, one is obliged to accept a
causal chain of events that projects the Hungarians back to a Bronze Age presence in the
Carpathian Basin, and by extension, to the Kurgan peoples. Alinei’s linguistic conclusions
may thus be as important for Uralic studies as Ventris’ decipherment of Linear B was for
Greek.
*Bomhard notes that various Russian scholars have tried to establish links between Etruscan and the North Caucasian languages.
Intriguingly, Starostin cites (Diakonoff-Starostin, 1986/46) Hurrian ki- in ki-ži (thirty) – which matches Etr. ci (three) closely).
BIBLIOGRAPHY
Bomhard/J. Kerns – The Nostratic Macrofamily, pp. 32-4 (Mouton de Gruyter, 1994)
Hencken, Hugh – Tarquinia, Villanova and Early Etruscans (Peabody Museum, Cambridge,
Mass., 1968).


Comments

Popular posts from this blog

Ősi magyar nyomok észak-amerikai indián földön (HU)

Etruscan/Magyar?