Charudes - Croatians

This is what Gol/a,b has to say in The origins of the Slavs about the
origin of the etnonym of the Croatians:

'Since the Iranian etymology of this ethnonym is only a hypothesis, on
equal footing with other hypotheses about its Slavic and Germanic
origin, I feel obliged to devote a separate excursus to this problem.
Excursus 1: The Puzzle of the Ethnicon Hrváti

1.) First, we should reconstruct the primary, i.e., PS1. form of the
ethnicon. On the basis of the following facts:

a.) S-C H`rva:t, gen. Hrváta, plur. Hrváti (old Hrvate);
b.) ORuss. Xorvate, Xrovate, Xrvate, XrUvate;
c.) OCz. (from OPol.) Charvaty (as a toponym);
d.) Greek (Constantine Porphyrogenitus) Xro:bátoi;
e.) Latin Chrouati (10th-11th cent.), also Chruvati vicus (in Thietmar
referring to a Lusatian territory, known now as Klein-Corbetha) from
North West Slavic *Xrovat- < PS1. *Xorvat- (?);
f.) Greek toponyms: Xarbáti and so forth (in Vasmer's Slaven in
Griechenland, where 'ar' seems to stand for an unmetathesized Slav.
tart, cf. Gardíki for *gardIkI, etc.); we can establish two PS1.
variants as basic: *XUrvaté and *Xorvaté, with the primary
consonantal-stem plural in a stressed -e.

In all early medieval sources various forms of this ethnicon refer
first to the South Slavic Croatians (Greek Xro:bátoi, Lat. Chrouati),
then to the sub-Carpathian Croats in the Upper Dniestr basin (an East
Slavic tribe in the Povest' vremennyx let: Xorvate, etc.) or to the
Croats in the Polish-Bohemian borderland in the region of the Sudeten
(OCz. Charvaty, etc.), and once to a tribe on the Upper Sorbian
territory (Thietmar's Chruvati). These facts, supported by a clear
statement in Porphyrogenitus, seem to indicate that the South Slavic
Croatians represent the main branch of a prehistorical Proto-Slavic
tribe whose primary habitat, before their migration to the Balkans,
was located north of the western Carpathians.


2.) For any linguist acquainted with comparative historical Slavic
grammar the very form of the ethnicon *XUrvaté provokes the suspicion
that it is of non-Slavic origin. First, the original Slavic character
of the initial x- would be justified only in the case of PIE *ks-,
which is quite a rare case. Second, the derivational form with a
consonant-stem suffix is also rather exceptional in the Slavic
material (among Slavic ethnica I would quote here as possible
parallels *Venete/i and Veleti). Therefore it is no wonder that most
Slavists have proposed a non-Slavic etymology for *XUvaté.


3.) Here I will briefly discuss the existing etymologies, reject them,
and propose a new one which seems to meet the requirements of exact
historical-comparative methods.

a.) Iranian etymology, recently supported by Trubac^ev (Ètimologija
1965:32).
Variant i) Sarmatian personal name Xoróaþos Xorouaþos (in Tanais),
2nd-3rd cent. A.D., i.e., *Hurvaþa 'bonus amicus' — but then we should
expect *XUrvot- or rather a patronymic form *XUrvot-itji, i.e., S-C
*Xrvotic´i.
Variant ii) Av. haurvata:t- // haurva:t-, i.e., *harvata:t //
*harva:t- 'totality,' etc. (from *harva- - Lat. salvus) — but then the
only form in Slavic would be *Xorvat-, although semantically there are
interesting parallels in Germ. Allemannen and, with a reversed
semantic order, in Oscan touto 'civitas,' Umbrian tota 'civitas' as
opposed to Lat. totus, -a, -urn.

b.) Germanic etymology, based upon a phrase Hárvaða or Hærvaþa fjöll,
i.e., 'Berge der Chorvaten' (according to R. Heinzel's
interpretation), in the Old Scandinavian Hervararsaga, a passage
referring to the Carpathians, with which the primary Croats, *XUrvate,
were closely connected, as we have mentioned. But what is the Germanic
source of Harvaða // Hærvaþa, i.e. PSl. *Xorvaté, cannot be
established: its interpretation as a Germanized form with
Lautverschiebung of the oronym Karpáte:s (in Ptolemy) is semantically
inexplicable and highly improbable in view of the bookish and learned
character of this oronym, which is unknown in the native linguistic
tradition of the Slavs dwelling in the Carpathian region.

c.) The third traditional etymology proposes a Slavic origin of the
ethnicon. According to this hypothesis *XUrvaté // *Xorvaté is related
to Lith. šarvúotas 'armored' ('geharnischt'), which in its turn comes
from šárvas 'armor' ('Harnisch, Panzer, Rüstung'); so *XUrvaté would
mean 'the armored.' But there are serious formal objections to this
etymology. First, Lith. šárvas does not continue any *ks- which would
result in Slavic x-, but rather represents primary IE *k'orHwos, a
substantivized adjective from the IE stem *k'erH-u- 'horn,' and it has
obvious correspondences in II'. languages, for which see below (cf.
here also Gr. kórus n. 'helmet'; Fraenkel, 965, and Pokorny, 574).
In this connection the primary meaning of Lith. šárvas would be
'horn-armor,' a type of armor well-known to the ancient East European
peoples. An exact Slavic correspondence representing the regular sat&m
treatment of *k'- should be *sorvU, then *Sorvaté, etc. Second, the
Lith. adjectives in -úotas have exact correspondences in Slavic
denominal adjectives in -atU (with the old acute), e.g., *bordatU,
Russ. borodátyj, etc. In addition, the primary word-final stress and
the consonant-stem declension of the ethnicon *Xorvaté // *XUrvaté
does not permit us to equate it with Lith. šarvúotas. Thus the
correspondence Lith. šarvúotas ~ PS1. *Xorvaté, etc., is illusory and
should be rejected.


4.) Now, after this criticism of existing etymologies I shall
propose a new one which seems to be better substantiated both formally
and semantically.
My contention is that PSl. *XUrvat- // *Xorvat- (a consonantal stem!)
was derived from a common noun *xUrvU // *xorvU 'armor' (primarily
'horn-armor'), which should be treated as a prehistorical loanword
from Germc. *hurwa- // *harwa-, the latter representing the PIE adj.
*k'r.Hwo- // *k'orHwo- (cf. Gr. keraós 'horned' and Lith. šárvas,
quoted above). The fact that the historical Germc. languages have not
preserved the hypothetical *hurwa- // *harwa- may be merely an
accident of history. We know of such cases in the history of
languages. For example, there are some Slavic loanwords in Rumanian
that have no correspondences in historical Slavic languages (e.g.,
zâpádâ f. 'snow,' from Slav. *zapada, etc.). Of course, other
derivatives of the root *k'erH- are known in Germanic, first of all
the noun *hurna = horn. The PIE adjective *k'erHwo- (full vocalism) //
*k'r.Hwo- (zero vocalism) is well attested in many IE languages; as is
well-known, Balto-Slav. *ka:rwa:- 'cow' with an exceptional kentum
treatment of *k'- belongs here too (see Pokorny, 576).
Some interesting traces of the borrowed PS1. *xUrvU can be found in
West Slavic, specifically in Polish. The most important is OPol. (1494
in the Poznan´ city-book) charwat 'miles in servitio civitatis,' which
seems to represent an adjective with the suffix -atU from the
hypothetical *xUrvU, but is not the exact counterpart of the ethnicon
because from a primary stem in -t we would rather expect *charwac´,
cf. paznokiec´, l/okiec´, dziesie,c´. Another one is Slovak (with the
Polish treatment of *tUrt) charvat' sa // charvit' sa 'defend
oneself,' charva // charvanie 'defense.' There are also two Kashubian
words: charwatynia 'abandoned cottage,' primarily 'sentry shelter'
(?), also a place name, and charwan´c (*xUrvanIcI) 'sheaf of grain
full of weeds,' probably a derivative from *charwac´ 'protect' and
primarily meaning 'a sheaf used to cover the top of a shock.'
All these lexical facts seem to indicate the existence in the
Northwest PSl. dialects of the noun *xUrvU 'armor' and its derivatives
*xUrvati se, // *xUrviti se, 'get armored' -> 'defend oneself.' In
this connection we should also mention some toponyms in Poland that
seem to continue the ethnicon *XUrvat- in its third variant *XUrvat-
(e.g., OPol. Chrwaty, etc.), which is well substantiated by the
phenomena of the historical phonetics of Germanic. Other toponyms
continuing the most frequent variant *XUrvat- are known in Northern
Bohemia and Northern Moravia. All of them, incidentally, represent the
Polish treatment of the primary *tUrt as tart: Charváty.
The geographical distribution of these linguistic facts would clearly
point towards the primary location of the prehistorical Croats
(*XUrvaté-) north of the Carpathians on the historical South Polish
territory (Mal/opolska) and towards a subsequent emigration of their
main bulk south through the Moravian Gate. We can even hypothesize
that the burrowing of the Germc. *hurwa- 'horn-armor' took place
somewhere in the sub-Carpathian region, and that its source was the
PGermc. dialect of the Bastarnians, who dwelt along the eastern
Carpathians in the first to third centuries A.D.
What still remains to be explained is the derivational type of the
ethnicon *XUrvat-e, etc., i.e., its consonant-stem suffix -at-. The
consonantal stems with the PIE suffix -et- (admitting the vowel
gradation -e(:)t- // -o(:)t-) are rare in Slavic: the only sure
example is pec^atI m., gen. pec^ate 'seal' Among the Slavic ethnica we
can quote *Venete/i and Veleti, both derived from verbal roots
*wen(H)- 'desire; win' and *wel- (Slav. vele^ti) 'command.' But
comparative IE evidence indicates that there were also denominal
derivatives with this suffix (-et-: Lat. equus -> eques (gen. equitis)
'horseman, horse-soldier' (*eque:t-s), pes (ped-) -> pedes (gen.
peditis) 'pedestrian, foot-soldier' (*pede:t-s), miles (gen. militis)
with no clear etymology; Gr. gumnós -> gumne:s (gen. gumno^tos) 'a
lightarmored foot soldier' can also be quoted here. So the derivation
of *XUrvat- // *Xorvat- from *xUrvU // *xorvU with the PIE suffix
-o:t- (PS1. -at-) seems to have sufficient substantiation. Of course,
it must have been an old and not very productive process.
As far as the primary semantics of *XUrvate is concerned, there seems
to be no doubt that it originally meant 'warriors clad with
horn-armor' und may not have been a name used by a certain tribe for
self-identification, but rather a description applied by the neighbors
of a tribe whose characteristic feature was the use of horn-armor. It
is also possible that *XUrvat-, etc., also denoted a warrior-class as
opposed to plain folk, the latter participating in war without armor.
These suppositions would explain why just on the territory where we
should locate the prehistorical *XUrvaté // *Xorvaté, i.e.. in
Southern Poland, there are no traces whatsoever of the ethnicon in the
local early medieval chronicles and other sources. In any case, a
primary descriptive and exogenous (i.e., used by the neighbors) name
of the tribe known later in history as C(h)rovati, Xro:bátoi, Hrváti,
in the course of time became an ethnicon, adopted ultimately by this
tribe itself, which is expressed clearly in the official title of the
first Croatian ruler: Trpimirus dux Chroatorum (852 A.D.).'


Now this unknown Germanic language would have been the para-Germanic
Bastarnian. If true, those Croatians were in contact with
Germanic-speakers early. So why shouldn't they be Ariovistus' Charudes?

One thing puzzled me about the story of Ariovistus giving away free
land to some tribe who had done nothing to win it. Perhaps they were
just too lazy to work the land and let Charudes/Croatians colonize the
land for them, in return for their products?


Torsten