Post by Teuta1975 on Jul 9, 2008 20:14:29 GMT -5
The Inscription on the Spearhead of Kovel
Author(s): Gustav Must
Source: Language, Vol. 31, No. 4, (Oct. - Dec., 1955), pp. 493-498
Published by: Linguistic Society of America
Stable URL: www.jstor.org/stable/411361
In 1858, a farmer plowing a field in Suszyczno, near Kovel in the old adminis- trative unit Volhynia (now in the northwestern Ukraine), turned up an iron spearhead with an inscription and abundant ornamentation. After its publica- tion some seventeen years later, the inscription became, and has remained, an object of keen interest to scholars, who at first had been inclined to doubt its authenticity because of forgeries that had recently occurred. R. Henning (who saw the spearhead itself before it was lost), L. F. A. Wimmer, and others were convinced that it bore inscribed runes and that the inscription was East Ger- manic. This is still the general opinion; because of the final -s of the inscribed word, it is considered to be Gothic. It is believed to date from about the 3rd century A.D. The inscription and the ornamentation are inlaid with silver wire. The inscrip- tion runs in retrograde direction (from right to left). The following tracing is taken from Henning.
OIlliTr
Henning transliterated the inscription as tilarids;2 this reading has been accepted by most scholars. It is the transliteration of the penultimate letter (from right to left) that has presented the most difficulties, because it has no counterpart in the runic alphabets-a fact pointed out already by C. J. S. Marstrander.3 The sign has been interpreted as a variety of the runic letter for ^, either a simplification (Wimmer) or a transition form in the assumed development from the letter D (Henning).4 M. Hammarstrom, however, showed that the prototype of the runic letter for a is found in the ancient alphabets of Northern Italy.5 On the other hand, it has been assumed that the penultimate letter may have the value ,b, since the change of a to voiceless ] is to be expected in Gothic before s.
Its form also is believed to originate from Latin D and is presumed to have been changed into an angular form in the Germanic inscription.6 However, it is incomprehensible how a symbol for a voiced stop should have been taken over for a voiceless spirant; moreover, the runic alphabet already possessed a sign for this sound. As J. Whatmough has shown, the prototype of the form of runic ,b is found in a North-Etruscan alphabet (Magr6).7 It was stated already by Marstrander that the reading tilarids is impossible.8 He tried to solve the prob- lem by transliterating the penultimate letter as i, reading ik hariys.9 W.
Krause rejects this firmly for graphic reasons.'0 Since the penultimate letter of the inscription is not runic d or 1, the read- ings tilarids and tilarips are untenable. Therefore the interpretations of the inscription as 'the one who strives toward the goal'1 or 'Angreifer, Anreiter'l2 are also baseless. No entirely satisfactory interpretation can be achieved unless the transliteration is certain. In any investigation of the Kovel inscription, we are first confronted by two questions: Is this a runic inscription? Is it a Germanic inscription at all?
When we examine the inscription with these questions in mind and compare the letters of the inscription with runes, we see that the form of the first letter, t, with a horizontal stroke not broken into two slanting strokes, differs from runic t and therefore does not belong to the runic alphabets. That this cannot be explained as due to the engraving technique is evidenced by the third letter, whose upper stroke is slanting. The penultimate letter is not runic either. The remaining symbols may belong either to runic or to other alphabets.
As Marstrander has pointed out,'3 every letter of the inscription occurs in the same form in ancient Greek and Italic alphabets. Actually they are better explained by alphabets other than runic. The symbol most relevant for the interpretation of the inscription is the penultimate. It has exact graphical equivalents in Greek local alphabets (particularly in Elis, Arcadia, Korkyra), and there its value is o.14 Provided the symbol in our inscription has the same value, the ending of the inscribed word is -ios. Incidentally, these letters are the most clearly and legibly written part of the inscription.
This ending at once suggests a derivative suffix of personal names which is very frequent in Greek, e.g. Demetrios, Apoll6nios, Dionli- sios.'4" Let us examine each letter of the inscription. The first is obviously t. It is attested in this same form, with one half of the upper stroke considerably longer than the other, in Greek local alphabets.'5 The second letter is i, the third is I. The fourth letter reminds one of Greek digamma. But we know that digamma disappeared from use very early: in the east before the 7th century, in the west at the latest before the 3rd century B.c.;16 the spearhead itself indicates a later period. Moreover, digamma between two consonants would present an unusual accumulation of consonants. On the other hand, an identical symbol occurs in the pre-Italic alphabets of Northern Italy,'7 as also in the runic alphabet, with the value of a; it was probably developed by raising to a vertical position the hasta of the form of a which occurs in ancient alphabets of Italy and in Greek local alphabets as well. The next letter is r; incidentally, r occurs in the same open form, and with the prolonged addition to original P, in Greek local (pre-Milesian) usage, particularly in the Chalcidian alphabets.18 The remaining letters are, as already stated, ios. The letters r and o are written in the angular form which is generally characteristic of inscriptions on metal as well as on stone.
To sum up, it is most probable that our inscription reads tilarios. The termination -ios points to a personal name. It occurs in Greek names, but because of the non-Greek form of the letter a this cannot be a Greek inscription. It cannot be a Germanic inscription either, because in Germanic the ending would be -as (not -os).
A further analysis of the word reveals that it contains the ending -arios. There are personal names in -arios, Latinized -arius, in Ill- yrian, e.g. Battfrios (Batdrios), Dindarius, Samiarius.'9 This suggests the as- sumption that tilarios is an Illyrian personal name. In that case the inscription is Illyrian. Most of what we know about Illyrian proper names we owe to Hans Krahe; I have made extensive use of his material in the discussion which follows.
The ending -arios, -arius of Illyrian personal names is composed of the derivative suffix -ios and the suffix -aro- of old proper names.20 Thus, the personal name Dindarius is derived from a tribal name Dindari,2' and the gentile name Samia- rius is formed from the personal name Samiarus.22 Both suffixes, -io- and -ar-, also occur independently in Illyrian, e.g. -io- in Phalios for native Illyrian *Balios; Bldtios, Blattius, Blassius, Laidius, Voltius;23 similarly -ar- in personal names like Leukaros, Mazaros, Audarus,24 Deivarus,25 in tribal names like Din- daroi, Separi, and in geographical names like Decadaron, Phedgaron, Kit- taros.26
The personal name Tilarios is similarly formed with the suffix -ios. Consequently it must be derived from a proper name *Tilaro- containing the suffix -aro-. Whether *Tilaro- was originally a personal, a tribal, or a geographical name is not immediately clear. The stem to which -ario-s has been added, til.
is actually recorded in Illyrian. We find it in a word whose formation is strik- ingly similar to that of Tilarios: an Illyrian river name Tilurius.Y A city on that river, Tilurium, was called after it.28 This name can be analyzed as til- ur-io-.29 Both names, Tilarios as well as Tilurius, appear to have been derived from the same stem til-. It is easiest to understand the etymology if we start out from the name of the river, Tilurius.
We know that river names often go back to roots which mean 'moist, wet, etc.', e.g. Illyr. Savos < *souos (PIE root *seu- 'Saft, Feuchte'), cf. OHG sou 'Saft', Gk. huiei 'es regnet', Illyr. Volcos from PIE *uelq- 'feucht, nass',30 cf. also the Russian river name Volga, which belongs to Russ. wologa, OCS wlaga 'Feuchtigkeit'.31 Among the IE roots of this semantic group there is also *"t-, tai- 'schmelzen, sich auflosen (fliessen)', cf. the ON river name pZTa f. (to J'Sa 'schmelzen, tauen'), OE ]znan 'feucht werden' and Gk. tilos 'duinner Stuhlgang, Abfuiihren', tiphos 'sumpfige Stelle, feuchter Grund'.32 It is evident that Illyrian til-, which contains a suffix -1- (like the Illyr. river name Dri-l-on33), goes back to the same root. According to this etymology it is conceivable that the proper name from which the name Tilarios was derived, *Ti-l-aro-, was a place name.
The formation Tilarios from *Tilaro- is parallel to the Caravantius from the place name Caravantis, Naronia from Narona, etc.34 The present analysis leads to the conclusion that the Kovel inscription is not runic, and not Germanic at all. Most probably it represents the name of the owner of the weapon, and this name appears to be typically Illyrian. Since the alphabet used in the inscription agrees with Greek local alphabets, except for the letter a (which occurs in Sub-Alpine alphabets), it is conceivable that such an alphabet was used in a peripheral area of the Greek alphabets between Greece itself and Italy-in the area of Illyricum.35
We know that the Balkan-Illyrian area belonged to the sphere of Greek influence and, after conquest by the Ro- mans, to the eastern half of the Empire where the Greek script (as well as lan- guage) prevailed. The place where the spearhead was found, however, is a considerable distance northeast of the Balkan-Illyrian area. As is well known, archeological and linguistic study, chiefly of place names, has shown that between the ancient Germanic and Slavic tribes there dwelt a people who occupied a large territory between the rivers Elbe and Vistula and from the Baltic Sea to the Balkans. They were the carriers of the Lausitz culture (ca. 1500-500 B.C.); scholars differ
concerning their nationality. Recent research is mostly inclined to regard them as Kelts (Proto-Kelts) or as Illyrians.
From linguistic evidence alone it is diffi- cult to solve the problem, because of the similarity of Keltic and Illyrian at such an early period. It is primarily because of the archeological evidence that this nation is assumed to have belonged to the Illyrians. G. Kossinna was the first to express this view;36 he called them North-Illyrians, in order to distinguish them from the Illyrians of the Balkan peninsula and from a third group of Illyrians, the Messapians, who lived at the southern end of the Apennine peninsula. Espe- cially significant is the work of Max Vasmer, who has discovered a considerable number of Illyrian place names in the area of the so-called North-Illyrians.37
From about the 3rd century B.C., Germanic tribes infiltrated into this area, until finally they occupied it.38 But this did not mean the immediate extinction of the Illyrians;39 thus, the Osi, a central European tribe mentioned by Tacitus, were Illyrians living in the area occupied by Germanic peoples.4u It is not known how long the last remnants of the Illyrians remained there, and whether they were already extinct before the Slavs invaded the area from the east and, in the 6th century A.D., occupied it westwards to the Elbe river. The spearhead of Kovel was unearthed in the Slavic area. But we must bear in mind that it is an isolated find, which need not be connected with the people resident in the area where it was discovered. It has been connected with the Goths who migrated to the Black Sea through the Pripet swamps, and even with those who returned from there, and hence has been assumed to date from the 2nd or 3rd (or 4th) century A.D.
It is scarcely conceivable, however, that the loss of the spearhead coincided in time with its inscribing; it is far more plausible that the spearhead was made and inscribed considerably earlier. It is true of inscribed objects in general, and especially of single scattered finds, that the time at which the object is assumed to have been brought to the place of its eventual discovery proves nothing about the time when it was made; nor need the place where it is found be near the place of its origin. A better clue is the alphabet used for the inscription.
Since the Kovel alphabet is probably Illyrian, it may bear witness to Illyrian battles with Slavs. It may have been carried to the place where it was later found, either by the Illyrians themselves during a campaign against the Slavs, or by the Slavs as booty of war brought from the Illyrian area. Since we possess no exact information concerning the prehistory of this area, we cannot precisely determine the date of the inscription. Even the technique of the inscription does not reveal the date: the technique of inlaid precious metal was in use for a long time, beginning in the bronze age:4' it was known also to the Lausitz culture. Nor do the ornaments (swastika, sun disc, crescent moon) offer a criterion for determining the date of the inscription, as Henning knew; for these symbols were used from early times in a wide area. Of oriental origin, they spread from the Near East to southeastern and central Europe.42 In short, the physical spearhead itself offers no evidence of its owner's na- tionality. Only the language of the inscription points to it.
The retrograde direction of the script is a feature that attracts one's attention, and suggests an early date. But although the retrograde order disappeared early from Greece, in Italy its use prevailed longer, and in runic inscriptions longer still. The direction is therefore not a definite criterion, for archaic features were preserved longer in peripheral areas. Another inscribed spearhead, that of Dahmsdorf (Miincheberg), is ornamented with symbols which partly resemble those on the spearhead of Kovel: sun disc, crescent moon, and swastika. On the other hand, the Dahmsdorf spearhead differs somewhat in form from that of Kovel, and the technique of its orna- mentation, according to Henning,43 is more complicated and developed; he therefore concludes that the spearhead of Dahmsdorf must be younger.
Because of the resemblance of the ornamentation on these two spearheads, they belong, in HIenning's words, to the same artistic tradition.44 They probably stem from the same civilization-one from the eastern, the other from the western edge of its area. This of course does not mean that they must come from the area of a single language.
The alphabets of the inscriptions are different, the letters on the Dahmsdorf spearhead being identical with runes. We have then no definite criteria for determining the date of the Kovel in- scription. From the evidence of the inscription itself it is likely that it comes from a considerably earlier period than the presumed 2nd or 3rd century A.D. More than two hundred Messapic inscriptions are extant. Up to the present only one inscription is known from the Balkan-Illyrian area: an inscription on a ring, ANA OH H ICER, which Krahe has interpreted as 'sacred to ana Oethe (or Etheo)'.46 If the views expressed in the present discussion are correct, the Illyrian name Tilarios on the spearhead of Kovel is another Illyrian inscrip- tion.
-------------------------------
Notes:
1 Die deutschen Runendenkmaler, Table 1 (Strassburg, 1881). 2op. cit. 3. a Norsk tidsskrift for sprogvidenskap 3.27 (1929). Loc. cit. Om runskriftens harkomst, Studier i nordisk filologi 20.42 (1930). 6 H. Arntz and H. Zeiss, Die einheimischen Runendenkmaler 30 (Leipzig, 1939). 7 J. Whatmough, The Prae-Italic dialects of Italy, 2:3.505 fn. 1 (Cambridge, Mass., 1933) Abbrev. PID.
8 'Tilarids ist ... eine unmogliche Lesung' in his article Germanische Waffennamen aus romischer Zeit, Norsk tidsskrift for sprogvidenskap 3.233 (1929). 9 Norsk tidsskr. f. sprogv. 3.35, 234. G. Stephens, The Old Northern runic monuments of Scandinavia and England 3.268 (Copenhagen, 1884), had already transliterated the sign in question as p, reading tilkerings. 10 W. Krause, Beitrage zur Runenkunde 55 (Halle [Saale], 1932). "0. v. Friesen, Ro-stenen ..., Uppsala Universitets Arsskrift 1.128 (1924). 12 W. Krause, Framea, Germanen und Indogermanen: Festschrift fur Herman Hirt 2.587 (Heidelberg, 1936); Arntz-Zeiss 40 f. 13 Norsk tidsskr. f. sprogv. 3.27. 14 W. Larfeld, Handbuch der griechischen Epigraphik 1, Table 3 (Leipzig, 1907). 14a Forms written in Greek letters are here shown in boldface transliteration, accord- ing to the system set forth by Martinet, Word 9.152-61 (1953). 15 See W. Schubart, Palaeographie I: Griechische Palaeographie 14 (Munchen, 1925); E. S. Roberts, An introduction to Greek epigraphy 1.288, 291 (Cambridge, 1887).
16 Larfeld, Handb. d. griech. Epigr. 1.397. 17 See Whatmough, PID, table of alphabets. 18 Handbuch der klassischen Altertumswissenschaft ..., ed. I. Muller: Vol. 1, G. Hinrichs, Einleitende und Hilfsdisziplinen, Tabelle der griechischen Alphabete (Nordlingen, 1886). 19 Hans Krahe, Lexikon altillyrischer Personennamen 20, 43 (Heidelberg, 1929); IF 59.69. 20 Krahe, IF 58.131. 21 Krahe, Lex. altill. Pn. 43; Wurzburger Jahrbucher fur die Altertumswissenschaft 1.192 (1946). 22 Krahe, IF 59.69. 23 Krahe, Wi6rzb. Jb. f. d. Altert. 1.186; Lex. altill. Pn. 22, 61, 130. 24 Krahe, IF 58.131 f. 25 Krahe, Wurzb. Jb. f. d. Altert. 3.83 (1948). 26 Krahe, Die alten balkanillyrischen geographischen Namen 57 f. (Heidelberg, 1925).
27 Krahe, Balkanill. geogr. N. 39. 28 Krahe, Balkanill. geogr. N. 60, 101. 29 Krahe, Balkanill. geogr. N. 60. 30 Krahe, Wurzb. Jb. f. d. Altert. 1.208. 31 F. Solmsen and E. Fraenkel, Indogermanische Eigennamen als Spiegel der Kultur- geschichte 45 (Heidelberg, 1922). 32 Walde-Pokorny, Vgl. IVb. d. idg. Spr. 1.101 f.; F. Holthausen, Etym. Wb. d. Altwest- nord. s.v. 33 Krahe, Balkanill. geogr. N. 61. 34 Krahe, Wurzb. Jb. f. d. Altert. 1.192. 36 Cf. Krahe, Die Illyrier in ihren sprachlichen Beziehungen zu Italikern und Griechen, Die Welt als Geschichte 3.133, 291 (1937).
36 G. Kossinna, Mannus 4.183, 287 if., 293 f. (1912). 37 Vasmer, Zeitschrift fur slavische Philologie 5.360 ff. (1929), 6.149 (1929), Namn och bygd 21.121 ff. (1933). 38 Vasmer, Namn och bygd 21.124. 39 Vasmer, Zs. f. slav. Phil. 5.370. 40 Whatmough, The Osi of Tacitus-Germanic or Illyrian?, Harvard studies in classical philology 42.140 (1931). 41 Ebert, Reallex. d. Vorgesch. sub 'Tauschierung'.
42Henning 6, 19; Ebert, Reallex. s.v.; J. de Vries, Altgermanische Religionsgeschichte 1.154 etc. (Berlin and Leipzig, 1935). 43 Die deutschen Runendenkrn. 8. 44 Loc. cit. 45 IF 46.183 ff.
Author(s): Gustav Must
Source: Language, Vol. 31, No. 4, (Oct. - Dec., 1955), pp. 493-498
Published by: Linguistic Society of America
Stable URL: www.jstor.org/stable/411361
In 1858, a farmer plowing a field in Suszyczno, near Kovel in the old adminis- trative unit Volhynia (now in the northwestern Ukraine), turned up an iron spearhead with an inscription and abundant ornamentation. After its publica- tion some seventeen years later, the inscription became, and has remained, an object of keen interest to scholars, who at first had been inclined to doubt its authenticity because of forgeries that had recently occurred. R. Henning (who saw the spearhead itself before it was lost), L. F. A. Wimmer, and others were convinced that it bore inscribed runes and that the inscription was East Ger- manic. This is still the general opinion; because of the final -s of the inscribed word, it is considered to be Gothic. It is believed to date from about the 3rd century A.D. The inscription and the ornamentation are inlaid with silver wire. The inscrip- tion runs in retrograde direction (from right to left). The following tracing is taken from Henning.
OIlliTr
Henning transliterated the inscription as tilarids;2 this reading has been accepted by most scholars. It is the transliteration of the penultimate letter (from right to left) that has presented the most difficulties, because it has no counterpart in the runic alphabets-a fact pointed out already by C. J. S. Marstrander.3 The sign has been interpreted as a variety of the runic letter for ^, either a simplification (Wimmer) or a transition form in the assumed development from the letter D (Henning).4 M. Hammarstrom, however, showed that the prototype of the runic letter for a is found in the ancient alphabets of Northern Italy.5 On the other hand, it has been assumed that the penultimate letter may have the value ,b, since the change of a to voiceless ] is to be expected in Gothic before s.
Its form also is believed to originate from Latin D and is presumed to have been changed into an angular form in the Germanic inscription.6 However, it is incomprehensible how a symbol for a voiced stop should have been taken over for a voiceless spirant; moreover, the runic alphabet already possessed a sign for this sound. As J. Whatmough has shown, the prototype of the form of runic ,b is found in a North-Etruscan alphabet (Magr6).7 It was stated already by Marstrander that the reading tilarids is impossible.8 He tried to solve the prob- lem by transliterating the penultimate letter as i, reading ik hariys.9 W.
Krause rejects this firmly for graphic reasons.'0 Since the penultimate letter of the inscription is not runic d or 1, the read- ings tilarids and tilarips are untenable. Therefore the interpretations of the inscription as 'the one who strives toward the goal'1 or 'Angreifer, Anreiter'l2 are also baseless. No entirely satisfactory interpretation can be achieved unless the transliteration is certain. In any investigation of the Kovel inscription, we are first confronted by two questions: Is this a runic inscription? Is it a Germanic inscription at all?
When we examine the inscription with these questions in mind and compare the letters of the inscription with runes, we see that the form of the first letter, t, with a horizontal stroke not broken into two slanting strokes, differs from runic t and therefore does not belong to the runic alphabets. That this cannot be explained as due to the engraving technique is evidenced by the third letter, whose upper stroke is slanting. The penultimate letter is not runic either. The remaining symbols may belong either to runic or to other alphabets.
As Marstrander has pointed out,'3 every letter of the inscription occurs in the same form in ancient Greek and Italic alphabets. Actually they are better explained by alphabets other than runic. The symbol most relevant for the interpretation of the inscription is the penultimate. It has exact graphical equivalents in Greek local alphabets (particularly in Elis, Arcadia, Korkyra), and there its value is o.14 Provided the symbol in our inscription has the same value, the ending of the inscribed word is -ios. Incidentally, these letters are the most clearly and legibly written part of the inscription.
This ending at once suggests a derivative suffix of personal names which is very frequent in Greek, e.g. Demetrios, Apoll6nios, Dionli- sios.'4" Let us examine each letter of the inscription. The first is obviously t. It is attested in this same form, with one half of the upper stroke considerably longer than the other, in Greek local alphabets.'5 The second letter is i, the third is I. The fourth letter reminds one of Greek digamma. But we know that digamma disappeared from use very early: in the east before the 7th century, in the west at the latest before the 3rd century B.c.;16 the spearhead itself indicates a later period. Moreover, digamma between two consonants would present an unusual accumulation of consonants. On the other hand, an identical symbol occurs in the pre-Italic alphabets of Northern Italy,'7 as also in the runic alphabet, with the value of a; it was probably developed by raising to a vertical position the hasta of the form of a which occurs in ancient alphabets of Italy and in Greek local alphabets as well. The next letter is r; incidentally, r occurs in the same open form, and with the prolonged addition to original P, in Greek local (pre-Milesian) usage, particularly in the Chalcidian alphabets.18 The remaining letters are, as already stated, ios. The letters r and o are written in the angular form which is generally characteristic of inscriptions on metal as well as on stone.
To sum up, it is most probable that our inscription reads tilarios. The termination -ios points to a personal name. It occurs in Greek names, but because of the non-Greek form of the letter a this cannot be a Greek inscription. It cannot be a Germanic inscription either, because in Germanic the ending would be -as (not -os).
A further analysis of the word reveals that it contains the ending -arios. There are personal names in -arios, Latinized -arius, in Ill- yrian, e.g. Battfrios (Batdrios), Dindarius, Samiarius.'9 This suggests the as- sumption that tilarios is an Illyrian personal name. In that case the inscription is Illyrian. Most of what we know about Illyrian proper names we owe to Hans Krahe; I have made extensive use of his material in the discussion which follows.
The ending -arios, -arius of Illyrian personal names is composed of the derivative suffix -ios and the suffix -aro- of old proper names.20 Thus, the personal name Dindarius is derived from a tribal name Dindari,2' and the gentile name Samia- rius is formed from the personal name Samiarus.22 Both suffixes, -io- and -ar-, also occur independently in Illyrian, e.g. -io- in Phalios for native Illyrian *Balios; Bldtios, Blattius, Blassius, Laidius, Voltius;23 similarly -ar- in personal names like Leukaros, Mazaros, Audarus,24 Deivarus,25 in tribal names like Din- daroi, Separi, and in geographical names like Decadaron, Phedgaron, Kit- taros.26
The personal name Tilarios is similarly formed with the suffix -ios. Consequently it must be derived from a proper name *Tilaro- containing the suffix -aro-. Whether *Tilaro- was originally a personal, a tribal, or a geographical name is not immediately clear. The stem to which -ario-s has been added, til.
is actually recorded in Illyrian. We find it in a word whose formation is strik- ingly similar to that of Tilarios: an Illyrian river name Tilurius.Y A city on that river, Tilurium, was called after it.28 This name can be analyzed as til- ur-io-.29 Both names, Tilarios as well as Tilurius, appear to have been derived from the same stem til-. It is easiest to understand the etymology if we start out from the name of the river, Tilurius.
We know that river names often go back to roots which mean 'moist, wet, etc.', e.g. Illyr. Savos < *souos (PIE root *seu- 'Saft, Feuchte'), cf. OHG sou 'Saft', Gk. huiei 'es regnet', Illyr. Volcos from PIE *uelq- 'feucht, nass',30 cf. also the Russian river name Volga, which belongs to Russ. wologa, OCS wlaga 'Feuchtigkeit'.31 Among the IE roots of this semantic group there is also *"t-, tai- 'schmelzen, sich auflosen (fliessen)', cf. the ON river name pZTa f. (to J'Sa 'schmelzen, tauen'), OE ]znan 'feucht werden' and Gk. tilos 'duinner Stuhlgang, Abfuiihren', tiphos 'sumpfige Stelle, feuchter Grund'.32 It is evident that Illyrian til-, which contains a suffix -1- (like the Illyr. river name Dri-l-on33), goes back to the same root. According to this etymology it is conceivable that the proper name from which the name Tilarios was derived, *Ti-l-aro-, was a place name.
The formation Tilarios from *Tilaro- is parallel to the Caravantius from the place name Caravantis, Naronia from Narona, etc.34 The present analysis leads to the conclusion that the Kovel inscription is not runic, and not Germanic at all. Most probably it represents the name of the owner of the weapon, and this name appears to be typically Illyrian. Since the alphabet used in the inscription agrees with Greek local alphabets, except for the letter a (which occurs in Sub-Alpine alphabets), it is conceivable that such an alphabet was used in a peripheral area of the Greek alphabets between Greece itself and Italy-in the area of Illyricum.35
We know that the Balkan-Illyrian area belonged to the sphere of Greek influence and, after conquest by the Ro- mans, to the eastern half of the Empire where the Greek script (as well as lan- guage) prevailed. The place where the spearhead was found, however, is a considerable distance northeast of the Balkan-Illyrian area. As is well known, archeological and linguistic study, chiefly of place names, has shown that between the ancient Germanic and Slavic tribes there dwelt a people who occupied a large territory between the rivers Elbe and Vistula and from the Baltic Sea to the Balkans. They were the carriers of the Lausitz culture (ca. 1500-500 B.C.); scholars differ
concerning their nationality. Recent research is mostly inclined to regard them as Kelts (Proto-Kelts) or as Illyrians.
From linguistic evidence alone it is diffi- cult to solve the problem, because of the similarity of Keltic and Illyrian at such an early period. It is primarily because of the archeological evidence that this nation is assumed to have belonged to the Illyrians. G. Kossinna was the first to express this view;36 he called them North-Illyrians, in order to distinguish them from the Illyrians of the Balkan peninsula and from a third group of Illyrians, the Messapians, who lived at the southern end of the Apennine peninsula. Espe- cially significant is the work of Max Vasmer, who has discovered a considerable number of Illyrian place names in the area of the so-called North-Illyrians.37
From about the 3rd century B.C., Germanic tribes infiltrated into this area, until finally they occupied it.38 But this did not mean the immediate extinction of the Illyrians;39 thus, the Osi, a central European tribe mentioned by Tacitus, were Illyrians living in the area occupied by Germanic peoples.4u It is not known how long the last remnants of the Illyrians remained there, and whether they were already extinct before the Slavs invaded the area from the east and, in the 6th century A.D., occupied it westwards to the Elbe river. The spearhead of Kovel was unearthed in the Slavic area. But we must bear in mind that it is an isolated find, which need not be connected with the people resident in the area where it was discovered. It has been connected with the Goths who migrated to the Black Sea through the Pripet swamps, and even with those who returned from there, and hence has been assumed to date from the 2nd or 3rd (or 4th) century A.D.
It is scarcely conceivable, however, that the loss of the spearhead coincided in time with its inscribing; it is far more plausible that the spearhead was made and inscribed considerably earlier. It is true of inscribed objects in general, and especially of single scattered finds, that the time at which the object is assumed to have been brought to the place of its eventual discovery proves nothing about the time when it was made; nor need the place where it is found be near the place of its origin. A better clue is the alphabet used for the inscription.
Since the Kovel alphabet is probably Illyrian, it may bear witness to Illyrian battles with Slavs. It may have been carried to the place where it was later found, either by the Illyrians themselves during a campaign against the Slavs, or by the Slavs as booty of war brought from the Illyrian area. Since we possess no exact information concerning the prehistory of this area, we cannot precisely determine the date of the inscription. Even the technique of the inscription does not reveal the date: the technique of inlaid precious metal was in use for a long time, beginning in the bronze age:4' it was known also to the Lausitz culture. Nor do the ornaments (swastika, sun disc, crescent moon) offer a criterion for determining the date of the inscription, as Henning knew; for these symbols were used from early times in a wide area. Of oriental origin, they spread from the Near East to southeastern and central Europe.42 In short, the physical spearhead itself offers no evidence of its owner's na- tionality. Only the language of the inscription points to it.
The retrograde direction of the script is a feature that attracts one's attention, and suggests an early date. But although the retrograde order disappeared early from Greece, in Italy its use prevailed longer, and in runic inscriptions longer still. The direction is therefore not a definite criterion, for archaic features were preserved longer in peripheral areas. Another inscribed spearhead, that of Dahmsdorf (Miincheberg), is ornamented with symbols which partly resemble those on the spearhead of Kovel: sun disc, crescent moon, and swastika. On the other hand, the Dahmsdorf spearhead differs somewhat in form from that of Kovel, and the technique of its orna- mentation, according to Henning,43 is more complicated and developed; he therefore concludes that the spearhead of Dahmsdorf must be younger.
Because of the resemblance of the ornamentation on these two spearheads, they belong, in HIenning's words, to the same artistic tradition.44 They probably stem from the same civilization-one from the eastern, the other from the western edge of its area. This of course does not mean that they must come from the area of a single language.
The alphabets of the inscriptions are different, the letters on the Dahmsdorf spearhead being identical with runes. We have then no definite criteria for determining the date of the Kovel in- scription. From the evidence of the inscription itself it is likely that it comes from a considerably earlier period than the presumed 2nd or 3rd century A.D. More than two hundred Messapic inscriptions are extant. Up to the present only one inscription is known from the Balkan-Illyrian area: an inscription on a ring, ANA OH H ICER, which Krahe has interpreted as 'sacred to ana Oethe (or Etheo)'.46 If the views expressed in the present discussion are correct, the Illyrian name Tilarios on the spearhead of Kovel is another Illyrian inscrip- tion.
-------------------------------
Notes:
1 Die deutschen Runendenkmaler, Table 1 (Strassburg, 1881). 2op. cit. 3. a Norsk tidsskrift for sprogvidenskap 3.27 (1929). Loc. cit. Om runskriftens harkomst, Studier i nordisk filologi 20.42 (1930). 6 H. Arntz and H. Zeiss, Die einheimischen Runendenkmaler 30 (Leipzig, 1939). 7 J. Whatmough, The Prae-Italic dialects of Italy, 2:3.505 fn. 1 (Cambridge, Mass., 1933) Abbrev. PID.
8 'Tilarids ist ... eine unmogliche Lesung' in his article Germanische Waffennamen aus romischer Zeit, Norsk tidsskrift for sprogvidenskap 3.233 (1929). 9 Norsk tidsskr. f. sprogv. 3.35, 234. G. Stephens, The Old Northern runic monuments of Scandinavia and England 3.268 (Copenhagen, 1884), had already transliterated the sign in question as p, reading tilkerings. 10 W. Krause, Beitrage zur Runenkunde 55 (Halle [Saale], 1932). "0. v. Friesen, Ro-stenen ..., Uppsala Universitets Arsskrift 1.128 (1924). 12 W. Krause, Framea, Germanen und Indogermanen: Festschrift fur Herman Hirt 2.587 (Heidelberg, 1936); Arntz-Zeiss 40 f. 13 Norsk tidsskr. f. sprogv. 3.27. 14 W. Larfeld, Handbuch der griechischen Epigraphik 1, Table 3 (Leipzig, 1907). 14a Forms written in Greek letters are here shown in boldface transliteration, accord- ing to the system set forth by Martinet, Word 9.152-61 (1953). 15 See W. Schubart, Palaeographie I: Griechische Palaeographie 14 (Munchen, 1925); E. S. Roberts, An introduction to Greek epigraphy 1.288, 291 (Cambridge, 1887).
16 Larfeld, Handb. d. griech. Epigr. 1.397. 17 See Whatmough, PID, table of alphabets. 18 Handbuch der klassischen Altertumswissenschaft ..., ed. I. Muller: Vol. 1, G. Hinrichs, Einleitende und Hilfsdisziplinen, Tabelle der griechischen Alphabete (Nordlingen, 1886). 19 Hans Krahe, Lexikon altillyrischer Personennamen 20, 43 (Heidelberg, 1929); IF 59.69. 20 Krahe, IF 58.131. 21 Krahe, Lex. altill. Pn. 43; Wurzburger Jahrbucher fur die Altertumswissenschaft 1.192 (1946). 22 Krahe, IF 59.69. 23 Krahe, Wi6rzb. Jb. f. d. Altert. 1.186; Lex. altill. Pn. 22, 61, 130. 24 Krahe, IF 58.131 f. 25 Krahe, Wurzb. Jb. f. d. Altert. 3.83 (1948). 26 Krahe, Die alten balkanillyrischen geographischen Namen 57 f. (Heidelberg, 1925).
27 Krahe, Balkanill. geogr. N. 39. 28 Krahe, Balkanill. geogr. N. 60, 101. 29 Krahe, Balkanill. geogr. N. 60. 30 Krahe, Wurzb. Jb. f. d. Altert. 1.208. 31 F. Solmsen and E. Fraenkel, Indogermanische Eigennamen als Spiegel der Kultur- geschichte 45 (Heidelberg, 1922). 32 Walde-Pokorny, Vgl. IVb. d. idg. Spr. 1.101 f.; F. Holthausen, Etym. Wb. d. Altwest- nord. s.v. 33 Krahe, Balkanill. geogr. N. 61. 34 Krahe, Wurzb. Jb. f. d. Altert. 1.192. 36 Cf. Krahe, Die Illyrier in ihren sprachlichen Beziehungen zu Italikern und Griechen, Die Welt als Geschichte 3.133, 291 (1937).
36 G. Kossinna, Mannus 4.183, 287 if., 293 f. (1912). 37 Vasmer, Zeitschrift fur slavische Philologie 5.360 ff. (1929), 6.149 (1929), Namn och bygd 21.121 ff. (1933). 38 Vasmer, Namn och bygd 21.124. 39 Vasmer, Zs. f. slav. Phil. 5.370. 40 Whatmough, The Osi of Tacitus-Germanic or Illyrian?, Harvard studies in classical philology 42.140 (1931). 41 Ebert, Reallex. d. Vorgesch. sub 'Tauschierung'.
42Henning 6, 19; Ebert, Reallex. s.v.; J. de Vries, Altgermanische Religionsgeschichte 1.154 etc. (Berlin and Leipzig, 1935). 43 Die deutschen Runendenkrn. 8. 44 Loc. cit. 45 IF 46.183 ff.