185Arheološki vestnik 76, 2025, 185–194; DOI: https://doi.org/10.3986/AV.76.05; CC BY-NC-SA 4.0 As is well known, graves 14 and 7/8 from Idrija pri Bači yielded three objects with inscriptions in the Venetic alphabet. The two bronze vessels, a cup (Fig. 1: 1; 2) and a ladle with a strap-loop handle and bovine protomes (Fig. 1: 2; 3) from grave 14,1 both dated by Božič (2009) to Late Hallstatt, bear one and the same pre-Roman votive inscription (Is 1, Is 2, following the typology of Pellegrini and Prosdocimi), while the inscription on the fragment of a bronze 1 Kept by the Naturhistorisches Museum Wien, Prähis- torische Abteilung, inv. no. 13692 and 13693 respectively. plaque from grave 7/82 (Fig. 1: 3; 4) (with allotted siglum Is 3) was clearly produced in a Romanized linguistic context. Based on the chronologically latest objects in the graves, both burials are to be dated to the Augustan period (see Božič 2009, 78 for Grave 14, specifying the middle Augustan period as the terminus ante quem non, and Istenič 1985, 323, Guštin 1991, 73 for Grave 7/8). The inscriptions themselves, however, are significantly older, though not equally old. Based on palaeographic grounds, 2 Naturhistorisches Museum Wien, Prähistorische Abteilung, inv. no. 13671. The Venetic inscriptions from Idrija pri Bači revisited Venetski napisi z Idrije pri Bači: ponovna analiza Luka REPANŠEK Izvleček Prispevek ponuja izčrpno analizo paleografskih značilnosti vseh treh venetskih napisov z Idrije pri Bači (Is 1, Is 2 in Is 3), ki jo je omogočila ponovna avtopsija napisnih predmetov na Oddelku za prazgodovino Prirodoslovnega muzeja na Dunaju. Ob doslej nepojasnjenih in nepreverjenih pomembnih odstopanjih v detajlih dostopnih prerisov in prepisov je s tem tu prvikrat podano in utemeljeno pravilno branje Is 1 in Is 2 ter razrešene ter interpretirane dileme v zvezi z bistvenimi podrobnostmi paleografskih posebnosti vseh treh napisov. Ključne besede: prazgodovina; Posočje; Idrija pri Bači; paleografija; venetski alfabet Abstract The article offers a thorough palaeographical analysis of the three Venetic inscriptions from Idrija pri Bači (Is 1, Is 2, and Is 3), based on the renewed autopsy of the sources in the Prähistorische Abteilung of the Naturhistorisches Museum in Vienna. Given the significant discrepancies in the existing drawings and transcriptions of the three inscriptions, the gap for a much-needed authoritative reading of Is 1 and Is 2 is now filled, as is the need for a reinterpretation of the still existing dilemmas in relation to several significant palaeographic peculiarities of the three inscriptions. Key words: Prehistory; the Posočje / Isonzo region; Idrija pri Bači; palaeography; Venetic alphabet Luka REPANŠEK186 Fig. 1: Idrija pri Bači. Bronze vessels (Grave 14) and the fragment of a bronze plaque (Grave 7/8) with inscriptions in the Venetic alphabet. Scale 1, 2 = 1:2; 1a, 2a, 3 = 1:1. Sl. 1: Idrija pri Bači. Bronasti posodi (grob 14) in trak (grob 7/8) z napisi. M. 1,2 = 1:2; 1a, 2a, 3 = 1:1. The Venetic inscriptions from Idrija pri Bači revisited 187 Fig. 2: Idrija pri Bači. Bronze cup from Grave 14 with an inscription in the Venetic alphabet. Not to scale. a – first half of the inscription; b – second half of the inscription. Sl. 2: Idrija pri Bači. Bronasta skodelica iz groba 14 z napisom v venetskem alfabetu. Ni v merilu. a – prva polovica napisa; b – druga polovica napisa. Is  3 undoubtedly belongs to the same horizon as the bronze plaque from Colle Mazéit near Verzegnis (see Repanšek 2022, 615). Given its ornamental characteristics, workmanship and shape, the latter has been convincingly dated to the end of the 2nd century / beginning of the 1st century BC (Božič 2011, 258–259) based on the comparison with the bronze plaque from Gradič above Kobarid, which cannot have been minted before 114–113 BC (see Osmuk 1998). The two inarguably autochthonous votive inscriptions on the bronze vessels from Grave 14, however, bear all the characteristic traits of the local, Isonzian variety of the Venetic alphabet and are typologically fully comparable to the Venetic inscription *Is 4 on the situla from Grad near Reka, dated on typological grounds to the 4th century BC by Turk et al. 2009, 52–53. All three inscriptions make use of the Etruscan heta to spell .i. (that is, phonetically, a /j/) – an innovation of the Posočje region which is almost certainly connected with the change in the writing practice at Este, Padova and Vicenza that are to be dated to the end of the 4th century BC (see Lejeune 1974, 28; cf. Repanšek 2022, 604 and 607). The terminus ante quem non for the oldest Venetic inscription of the Posočje region (Is 1, Is 2 and Is 4) then seems to be late 4th / early 3rd century BC. The claim by Božič (2009, 82) that the significantly older date of the two vessels from Grave 14, which corresponds perfectly to the old- est possible date for the inscriptions Is 1 and Is 2, negates the evidence that the Isonzian type of the Venetic alphabet persisted well into the 1st century BC was indeed well founded and could later on be supported by the discovery of *Is 4 on the bronze situla from Grad near Reka. The discovery of the silver votive plaque from Vrh gradu near Pečine, however, now speaks in favour of the survival of the local Isonzian variety at least up to the end of the Luka REPANŠEK188 Fig. 3: Idrija pri Bači. Bronze ladle from Grave 14 with an inscription in the Venetic alphabet. Not to scale. a – first half of the inscription; b – second half of the inscription. Sl. 3: Idrija pri Bači. Bronasta zajemalka iz groba 14 z napisom v venetskem alfabetu. Ni v merilu. a – prva polovica napisa; b – druga polovica napisa. The Venetic inscriptions from Idrija pri Bači revisited 189 2nd / beginning of the 1st century BC (see Repanšek 2022, 607–709).3 Is 1, Is 24 and Is 35 were first published by Szombathy in 1901 (the drawings published in Guštin 1991 are also taken from Szombathy), then republished by Cordenons (1911), of which, however, only the facsimile are trustworthy, by Conway in 19336 (based entirely on the original publication by Szombathy and Kretschmer),7 by Lejeune in 1965 (with facsimilia), and studied again in 1967 by Pellegrini and Prosdocimi in their 3 For a full inventory of Venetic inscriptions discovered in the upper Posočje/Isonzo region see Repanšek 2022: 602–603 and 612. 4 Both represent an identical votive inscription ‘To Lajvna Vrota’. For the possibility of interpreting the letter wau in these inscriptions as see Prósper 2018, 462 and 2019, 33. Is 2 additionally contains a parascrip- tural sequence IXIXI (of unclear significance), which can be compared to IIXIIX and XIIX attested on the Negau B helmet, and probably to similar markings on the yet unpublished inscription from a more adjacent region. 5 The inscription luk.s. .m.elink.s. / ga.i.jo.s. kab[---] bears a Veneticized onomastic formula in the nominative: Luks Melinks Gajos = Lucius Melincius Gaius. The remain- ing three-letter sequence kab remains ultimately unclear. 6 Pisaniʼs transliteration laj.v.naj vrota.j. (1953, 253) is based on Conway (1933) and appears to take the in- terpunct between and in v.rot.a.j (Is 1) as an insignificant mistake. 7 Although not explicitly stated, I assume this on the basis of his observation that the ‘bottom’ of in Is 2 is broken away, whereupon he draws a square (that, in his words, could potentially be read as an were it not for the parallel vrot.a.j in Is 1) which exactly corresponds to the shape of the remaining part of the heta on the drawing published by Szombathy. authoritative La lingua Venetica, in 1974 by Lejeune in his Manuel de la langue Vénète (which does not contain drawings), and, finally, by Prosdocimi in 1976 (with no alterations to the drawings and read- ing). The drawings themselves as well as the exact interpretation of the palaeographic traits, however, differ in several important details (marked here in bold and additionally singled out in brackets), especially as far as Is 1 and Is 2 are concerned:8 Transcription based on the drawings9 in Szom- bathy (1901, 311, Fig. 74a–75b):10 Is 1: laj.v.naj v.rot.a.j = la.i..v.na.i. v.rot.a..i. (no interpunct after n)11 8 Note that I will adapt the obsolete reading to = <.i.> throughout. Since corresponds to what in normal (i.e. non-Isonzian) Venetic inscriptions would have been graphically represented as <.i.>, the transcriptions are given here with both and <.i.>. 9 And partly on P. Kretschmer’s interpretation of the sequences within the same publication (Kretschmer apud Szombathy, 1901, 352–353). Note that Szombathy’s own transcription/reading of the inscriptions (Is 1: hia.tor.v han.v(t)hap and Is 2: hia.torvhan.vhap) is erroneous. 10 Cordenons (1911, 234–236) obviously bases his drawings on Szombathy, i.e. with no interpunct marked in before v in Is 2. The same goes for the transcriptions published in Pellegrini 1955, 116–117. 11 Note that the which Szombathy tentatively reads between and in laj.v.n.aj (1901, 311: ‘Ich glaube nicht, dass dieses t zur Inschrift gehört’) is a mirage produced by the vertical interpunct and an unintentional horizontal stroke above it, which, indeed, does not belong to the inscription. Fig. 4: Idrija pri Bači. Fragment of a bronze plaque from Grave 7/8 with an inscription in the Venetic alphabet. Not to scale. Sl. 4: Idrija pri Bači. Fragment bronaste ploščice iz groba 7/8 z napisom v venetskem alfabetu. Ni v merilu. Luka REPANŠEK190 Is 2: lajv.naj vrot.aij = la.i.v.na.i. vrot.ai.i. (no interpunct before v; an interpunct after v; an interpunct after final a, but interpreted as )12 Conway, The Venetic inscriptions (1933, 165–166):13 Is 1: laj.v.naj v.rot.a.j = la.i..v.na.i. v.rot.a..i. (no interpunct after n) Is 2: laj.v.naj vrot.a.j = la.i..v.na.i. vrot.a..i. (an interpunct before and after v; an interpunct after final a) Pellegrini and Prosdocimi, La lingua venetica (1967, 592–595): Is 1: laj.v.n.aj v.rot.a.j = la.i..v.n.a.i. v.rot.a..i. (an interpunct after n) Is 2: laj.vnaj vrot.a[---]j = la.i..vna.i. vrot.a[--- ].i. (no interpunct after v; lacuna after final a, but transcribed as vrot.a.j = vrot.a..i.) Lejeune, Manuel de la langue Vénète (1974, 307, No. 253 and No. 252):14 Is 1: laj.v.naj v.rot.a.j = la.i..v.na.i. v.rot.a..i. (no interpunct after n) Is 2: laj.v.naj vrot.a[.]j = la.i..v.na.i. vrot.a[.].i. (an interpunct before and after v; an interpunct after final a) In the case of Is 1, the left hasta of the final a is tentatively traced in La lingua venetica, so that its shape appears as < >, while the drawing in Szombathy’s publication seems to favour < >. As far as Is 2 is concerned, Pellegrini and Prosdocimi reproduce it as a sinistroverse sequence, which in Szombathy’s representation is clearly understood to be dextroverse.15 In addition, when comparing the 12 Later correctly recognized as an interpunct by Vetter (1931, 71: ‘nicht vrotaih, sondern vrotah mit punktiertem a zu lesen sein dürfte’), who must have seen the inscrip- tion prior to conservation of the vessel. In a similar vein, Conway (1933, 166) comments: ‘the last [interpunct] before the h of vrot.a.h has slipped into a full-length hasta which Kretschmer needlessly read as i’. 13 Almost entirely based on Kretschmer (apud Szom- bathy 1901, 352–353), who provides a single, unified transcription laj.v.naj v.rotaij (based on the characteristic traits of both inscriptions). Pellegriniʼs account (1961, 77) is based on Conwayʼs reading, but for some reason omits the interpuncts before the final alphas: laj.v.naj v.rota.j (Is 1), laj.v.naj vrota.j (Is 2). 14 Note that Lejeune 1965 reads laj.v.n.aj v.rot.a.j for Is 1, i.e. with an interpunct after n, although his drawing of Is 1 does not actually reflect that. 15 Guštin 1991, 73 is to be corrected as follows: ‘Dabei wurde die Inschrift auf der Schale von links nach rechts two drawings, it remains entirely unclear to what extent the damaged section between the final a and heta allows one to see/assume an interpunct. The chronologically younger Is 3 is less problematic, though the issue remains whether the initial l is indeed to be interpreted as a three-bar letter form < > with an additional articulus in the upper part that points to the left (explicit in Szombathy, as well as in La lingua venetica, although not at all clear from the drawing, and not problematized by Lejeune (1974), who assumes a plain lambda < >, i.e. l1 in his typology). The unresolved questions therefore are: 1. Is 1 – Is there another interpunct after n / before a in the first sequence? 2. Is 1 – What is the status of the aberrant final a in the second sequence? 3. Is 2 – Is there an interpunct after v in the first sequence? 4. Is 2 – Is there any trace of an interpunct after the final a in the second sequence? 5. Is 2 – Is the inscription dextro- or sinistro- verse and what is the reason for the oscillating orientation of the individual letter forms? 6. Is 3 – What is the status of the initial aber- rantly shaped l? The renewed autopsy on 25th April 2024 at the Department for Prehistory (Prähistorische Abteilung) of the Naturhistorisches Museum in Vienna resolved the remaining issues thoroughly.16 The new, authoritative reading of Is 1 and Is 2 can now be established as follows: Is 1 (scriptio discontinua), see Fig. 1: 1a laj.v.n.aj v.rot.a.j = la.i..v.n.a.i. v.rot.a..i. Is 2 (scriptio continua), see Fig. 1: 2a laj.v.naj vrot.a[.]j = la.i..v.na.i. vrot.a[.].i. Is 1 has a very distinct trace (in the form of an uncharacteristically slanting incision) of another interpunct after n in laj.v.n.aj / la.i..v.n.a.i. as opposed to Is 2, which provides syllabic punctua- geschrieben (Fig. 33: 1), die Inschrift auf dem Henkelgefäß jedoch von rechts nach links (Fig. 33: 2).’ 16 Here, I wish to express sincere gratitude to the Prähis- torische Abteilung of the Naturhistorisches Museum Wien and specifically to the director Dr. Georg Tiefengraber for kindly providing the objects for reappraisal and redrawing, and to Ida Murgelj for offering her expertise in helping to provide the new authoritative facsimilia. The Venetic inscriptions from Idrija pri Bači revisited 191 tion (now indubitably established as certain) for the coda-final v, exactly as one would expect (cf. so.u.v.na from Este (Es 55), and not **so.u.vna). The obviously aberrant punctuation of Is 1 would only be defensible if one assumed that the ad- ditional interpunct actually ‘syllabifies’ a in the final diphthong -aj, given that that same overgen- eralization is employed in the same sequence in vrot.a.j, but then one would actually expect another interpunct after a, so *laj.v.n.a.j rather than the attested laj.v.n.aj. As it stands, however, laj.v.n.aj instead of laj.v.naj clearly goes hand in hand with another anomaly within the same inscription, i.e. the long-recognized superfluous interpunct between v and r in the initial cluster vr (v.rot.a.j). Given that the two inscriptions were almost certainly produced by the same scribe, probably one after the other, the number of mistakes in Is 1 over Is 2 could potentially speak in favour of Is 2 being an already improved and corrected version of the same inscription. As far as the interpunctuation of the final diphthong in Is 2 is concerned, it can be assumed with certainty that the sequence was originally .a.j (vrot.a.j) and therefore identical to the one in Is 1. Measuring over the lacuna caused by the damage to the ladle (now filled in in the process of conservation), the distance between the right articulus of and the following is 13 mm. Since the rest of the alphas contained in Is 2 measure 6–7 mm on average, this leaves 6–7 mm between the final and , which can be understood to have once hosted an additional vertical incision. This can be further supported by the drawing reproduced in Szombathy (1901, 311) and his reading hia (as well as Kretschmer’s (Kretschmer apud Szombathy, 1901, 353) own reading vrot.aih with an additional read be- tween a and h), which strongly imply that before the lacuna was filled in, the upper part of the vertical interpunct was still plainly detectable. The final alpha in Is 1 is of the normal archaic type < >, such as is generally typical of eastern periphery (cf. Monte Sorantri, Findenig-Thörl, Würmlach/Bumlje, partly Gurina, Stramare/Štramar, Škocjan, Parti near Stara Sušica, Is 1, Is 2, *Is 4,17 as well as Vače, Negau B, and Negau A18), and corresponds fully to the other occurrences of the same letter form in the two inscriptions. 17 I.e. the autochthonous Isonzian inscriptions. 18 In the linguistically Gaulish inscription zuφniφanuaφi = dubnibanuabi, though not in the remaining three. As far as the question of the orientation of Is 2 is concerned, Lejeune (1974, 252) remarks: ‘L’orientation fantaisiste des lettres ne permet pas de préciser si le texte est sinistroverse (avec bas des lettres vers le haut du vase) ou dextroverse (avec haut des lettres vers le haut du vase).’ Szombathy’s own transcription (1901, 312) implies that he cor- rectly assumed the inscription to run from right to left, as does Kretschmer (apud Szombathy, 1901, 353; cf. also Pellegrini, Prosdocimi 1961, 594–95). Contrary to Lejeune’s view, the only letter form that really stands out is the digamma (), which in relation to the orientation of the rest of the letters (all clearly sinistroverse)19 appears to be upturned. The reason for its distinctive positioning becomes immediately clear, however, if one considers that the upper part of a digamma would be rather difficult to realise perfectly on the convex part of the body of the ladle. The inscription seems to have been incised using the rim as the base, i.e. with the vessel held bottom-up, progressing from right to left to the right of the handle, but for both digammas the vessel was then apparently turned upside down (i.e. the right way up), in order not to risk any slips in the incision of the parallel hastae. Several issues have been raised in relation to the shape of the lambda in Is 3 (Fig. 1: 3). The problem of reading the 8th letter as a third in the inscription (Szombathy 1901, 317, 353–354, cf. Kretschmer’s own views on p. 352) can be considered obsolete since the republication of the inscription by Pellegrini and Prosdocimi, who correctly read an (see Lejeune 1974, 308). The apparent oblique incision, which Szombathy tentatively interpreted as part of an , is not intentional and therefore certainly not part of the letter form, so that one can – in spite of the fissure that now obscures it thoroughly – safely assume an . Concerning the first , however, whose two diagonal hastae make it stand out as almost unique among the available examples of Venetic lambda,20 it has now become clear that the deviant shape in fact results from a correction on part of the scribe. A slanting dextroverse lambda < > with the diagonal hasta positioned at the bottom was obviously incised first and in a single stroke, since there is no interruption to the line. 19 For the palaeographic significance of the opposing orientation of see Repanšek 2020, 168–169; 2022, 608–609. 20 The only other example is found in the inscrip- tion from Meggiaro, on which see Marinetti 2002, 181; Repanšek 2022, 608. Luka REPANŠEK192 The letter form was then corrected to correspond in its shape and orientation to the lambda in .m.elink.s. (i.e. the 7th letter), which becomes apparent from the fact that the upper hasta is not joined to the body of the letter form and actually roofs it. The resulting < >, then, is nothing but a genuine attempt to standardize the use of an inverted lambda, i.e. one that points opposite to the general direction of the inscription (i.e. sinistroverse in a dextroverse inscription in the case of Is 3). BOŽIČ, D. 2009, Late Hallstatt period bronze vessels in two late graves of the La Tène-Early Imperial Idrija group. – In: G. Tiefengraber, B. Kavur, A. Gaspari (ed.), Keltske študije II. Studies in Celtic Archaeology. Papers in honour of Mitja Guštin, Protohistoire Européenne 11, 77–84. BOŽIČ, D. 2011, Prazgodovinske najdbe s Tonov- covega gradu in železnodobna kultna mesta v Posočju / Prehistoric finds from Tonovcov grad and iron age cult places in the Posočje area. – In: Z. Modrijan, T. Milavec (ed.), Poznoantična utrjena naselbina Tonovcov grad pri Kobaridu. Najdbe / Late Antique fortified settlement Tonov- cov grad near Kobarid. Finds, Opera Instituti Archaeologici Sloveniae 24, 239–277. https:// doi.org/10.3986/9789612545871 CONWAY, R. S. 1933, The Prae-Italic dialects of Italy: Part 1: The Venetic inscriptions. – Cam- bridge, Mass. CORDENONS, F. 1911, Silloge delle iscrizioni ve- netiche. Con note sugli antichi alfabeti e sistemi di scrittura usati dagli italici e dagli etruschi. – Feltre. GUŠTIN, M. 1991, Posočje in der jüngeren Eisenzeit, Katalogi in monografije 27. ISTENIČ, J. 1985, Zapisi v venetski pisavi na Koroškem in v Sloveniji. – Zgodovinski časopis 39/4, 313–334. LEJEUNE, M. 1965, Les inscriptions vénètes du haut-Isonzo. – Révue des études latines 43, 152–164. LEJEUNE, M. 1974, Manuel de la langue vénète. – Heidelberg. MARINETTI, A. 2002, L’iscrizione votiva. – In: A. R. Serafini (ed.), Este preromana: una città e i suoi santuari, 180–184, Treviso. OSMUK, N. 1998, Plaquette votive de Kobarid (SI), inspirée d’un denier républicain. – Instru- mentum 7, 17. PELLEGRINI, G. B. 1955/56, Le iscrizioni venetiche. – Pisa. PELLEGRINI, G. B. 1961, Panorama di storia linguistica giuliano-carnica. Il periodo pre- romano. – Studi goriziani 29, 73–97. PELLEGRINI, G. B., A. L. PROSDOCIMI 1967, La lingua venetica I–II. – Padova, Firenze. PISANI, V. 1953, Le lingue dell’Italia antica oltre il Latino. – Torino. PROSDOCIMI, A. L. 1976, L’alfabeto (venetico) dell iscrizioni di Idria (Is 1, 2, 3) e gli alfabeti delle iscrizioni di Negau (A–B) e Vače. – In: V. Pisani, C. Santoro (ed.), Italia linguistica nuova ed antica. Studi linguistici in memoria di Oronzo Parlangeli, 203–229, Galatina. PRÓSPER, M. B. 2018, The Venetic agent nouns in -tōr-revisited. – In: J. M. Vallejo, I. Igartua, C. G. Castillero (ed.), Studia philologica et diachronica in honorem Joaquín Gorrochategui: Indoeuropaea et Palaeohispanica, Anejos de Veleia – Series minor 35, 543–471. PRÓSPER, M. B. 2019, Celtic and Venetic in Con- tact: The Dialectal Attribution of the Personal Names in the Venetic record. – Zeitschrift für celtische Philologie 66/1, 7–52. REPANŠEK, L. 2020, Towards the Interpretation of *Is 7. – In: Th. L. Markey, L. Repanšek (ed.), Revisiting Dispersions. Celtic and Germanic ca. 400 BC – ca. 400 AD. Proceedings of the Inter- national Interdisciplinary Conference held at Dolenjski muzej, Novo mesto, Slovenia; October 12 th – 14th, 2018, Journal of Indo-European Studies Monograph Series 67, 162–183. REPANŠEK, L. 2022, Posoškovenetski areal v luči novejših epigrafskih najdb (Isonzian Venetic inscriptions in the light of recent finds). – Arheološki vestnik 73, 601–615. SZOMBATHY, J. 1901, Das Grabfeld zu Idrija bei Bača. – Mittheilungen der Prähistorischen Commission der kaiserlichen Akademie der Wis- senschaften I/5, 291–363. TURK et al. 2009 = P. Turk, D. Božič, J. Istenič, N. Osmuk, Ž. Šmit 2009, New Pre-Roman Inscrip- tions from Western Slovenia: The Archaeological Evidence. – In: G. Tiefengraber, B. Kavur, A. Gaspari (ed.), Keltske študije II. Studies in Celtic Archaeology. Papers in honour of Mitja Guštin, Protohistoire Européenne 11, 47–64. VETTER, E. 1931, Messapische und venetische Wortdeutungen. – Glotta 20/1–2, 67–73. 193Venetski napisi z Idrije pri Bači: ponovna analiza Trije napisi v venetskem alfabetu z Idrije pri Bači, ki jih je skupaj z interpretacijo P. Kretschmerja prvi objavil Szombathy v svojem članku Das Grabfeld zu Idria bei Bača leta 1901, spadajo v posoško skupino venetskih napisnih spomenikov. Votivna napisa Is 1 (sl. 1: 1; 2) in Is 2 (sl. 1: 2; 3) na bro- nastih posodah iz groba 14 je po paleografskih značilnostih mogoče datirati v pozno četrto ali zgodnje tretje stoletje pr. n. št. (kot taka skupaj z napisom *Is 4 z Gradu pri Reki tvorita skupino najstarejših doslej odkritih venetskih napisov iz Posočja), medtem ko je napis Is 3 na fragmentu bronaste ploščice iz groba 7/8 (sl. 1: 3; 4) glede na način zapisovanja črke , ki ima vzporednico le na bronasti daritveni ploščici iz svetišča Colle Mazéit pri Verzegnisu, najverjetneje s konca drugega ali začetka prvega stoletja pr. n. št. Ponovna avtopsija in paleografska analiza vseh treh napisnih spomenikov sta se izkazali za nujni, saj so med doslej znanimi prepisi in prerisi napi- sov (Szombathy 1901; Cordenons 1911; Conway 1933; Pellegrini, Prosdocimi 1967; Lejeune 1974) precejšnje diskrepance (predvsem pri Is 1 in Is 2) v pomembnih paleografskih podrobnostih, zlasti ko gre za pozicije interpunktov. Nejasno je osta- jalo: a – ali je med n in a na napisu Is 1 ter med v in n na napisu Is 2 interpunkt, b – ali je mogoče predvidevati interpunkt pri Is 2 tudi med a in j v zaključni sekvenci (vzporedno z Is 1), c – ali zadnji a na Is 1 resnično izkazuje izstopajočo va- rianto, č – kakšen je status variantnega l z dvema poševnima hastama na napisu Is 3 ter d – kakšna je orientacija napisa Is 2. Za Is 1 in Is 2 kot avtoritativna odslej velja na- slednja interpretacija (za Is 3 kot v celoti pravilna še vedno velja transkripcija, objavljena v Pellegrini, Prosdocimi 1967, 596): Is 1 (scriptio discontinua), gl. sl. 1: 1a laj.v.n.aj v.rot.a.j = la.i..v.n.a.i. v.rot.a..i. Is 2 (scriptio continua), gl. sl. 1: 2a laj.v.naj vrot.a[.]j = la.i..v.na.i. vrot.a[.].i. Venetski napisi z Idrije pri Bači: ponovna analiza Povzetek V nasprotju z Is 2 ima Is 1 v prvi sekvenci napačno ozloženo in interpunktuirano soglasni- ško skupino -jvn-, in sicer kot -j.v.n.- (laj.v.n.aj), medtem ko Is 2 izkazuje pričakovano interpunk- tuacijo sekvence laj.v.naj, tj. z v v predhodnem in n v naslednjem zlogu kot pri imenu so.u.v.na iz Est (Es 55). Is 2 ima ob tem med a in j v sekvenci vrot.a[---]j, kjer bi po analogiji z Is 1 pričakovali interpunkt, torej vrot.a[.]j, lacuno, ki je bila pri konservaciji zapolnjena, tako da so zdaj prekrite tri četrtine črke a in polovica sledečega j. Ker preostale alfe na istem napisu merijo v povprečju od 6 do 7 mm, razmika med desno hasto a in levo stranico j pa je prav toli- ko, ne more biti dileme, da je v poškodovanem delu nekdaj zagotovo stal interpunkt v obliki vertikalne zareze. To se neposredno potrjuje tudi na prerisu v Szombathyjevi objavi (1901, sl. 75b) in njegovi (sicer napačni) transliteraciji .aih, iz česar je mogoče sklepati, da je bil pred konservacijo zgornji del interpunkta še viden, ter ga je pozneje pravilno interpretiral že Vetter (1931, 71), ki je napis prav tako očitno videl še pred konservacijo. Dileme glede smeri poteka Is 2 od desne proti levi (pace Lejeune 1974, 307) ob ponovni analizi ne more biti, saj v resnici odstopa le črka v (digama), ki je orientirana zrcalno od preostalih. Razlog za to pa je očiten: pisar je za spodnjo podporno črto napisa namreč uporabil obod posode, tako da jo je pri vrezovanju držal obrnjeno na glavo, pri obeh digamah pa jo je nato očitno obrnil, in sicer zato, ker bi moral sicer zgornji del digame z dvema vzporednima poševnima hastama vrezati že na izbočeni del trupa, s tem pa bi tvegal zdrs stilusa in nenatančnost pri vrezovanju. Glede preostalih paleografskih posebnostih je bilo mogoče ugotoviti, da (kljub implikacijam prerisa, objavljenega v Pellegrini, Prosdocimi 1967, 592) zadnji a na napisu Is 1 po ničemer ne odstopa od običajne variante arhajskega tipa, kakršna je raz- širjena po celotnem vzhodnem perifernem loku, o prvi črki l na napisu Is 3 pa je treba zaključiti, da je njena odstopajoča oblika (ki ima doslej Luka REPANŠEK194 znano vzporednico le na napisu iz Meggiara, gl. Marinetti 2002, 181) posledica naknadnega po- pravka. Pisar je najprej v eni potezi vrezal nekoliko poševno desno orientirano lambdo z vertikalno hasto na spodnjem delu, nato pa jo je popravil v isto varianto, kot jo je uporabil pri sedmi črki (tj. v besedi .m.elink.s.), torej v levo orientirano lambdo z vertikalno hasto v zgornjem delu (o tej različici lambde gl. Repanšek 2022, 608–609). Kot se izkaže pri natančni povečavi, med zgornjo hasto in glavno vertikalno linijo lambde namreč ni stika (gl. sl. 1: 3). Luka REPANŠEK Oddelek za primerjalno in splošno jezikoslovje Filozofska fakulteta Univerze v Ljubljani Aškerčeva 2 SI-1000 Ljubljana luka.repansek@ff.uni-lj.si ORCHID: https://orcid.org/0000-0001-6530-7597 Illustrations: Fig. 1 (drawing: Ida Murgelj, NMS). – Fig. 2–4 (photo: Chloe Potter, Naturhistorisches Museum Wien). Slikovno gradivo: Sl. 1 (risba: Ida Murgelj, NMS). – Sl. 2–4 (foto: Chloe Potter, Naturhistorisches Museum Wien). The author acknowledges the financial support from the Slovenian Research and Innovation Agency (research core funding No. P6-0218) and the project Corpus Inscriptionum Veneticarum Sloveniae – CIVES (call RSF C.III.1), financed by the University of Ljubljana. Prispevek je nastal v okviru raziskovalnega programa P6-0218, ki ga financira Javna agencija za znanstvenoraziskovalno in inovacijsko dejavnost Republike Slovenije iz državnega proračuna, in v sklopu projekta Corpus Inscriptionum Vene- ticarum Sloveniae – CIVES (razpis RSF C.III.1), ki ga je financirala Univerza v Ljubljani.