CO·48

edit with form

Inscription
Reading in transliteration:	uvamokozis : plialeθu : uvltiauiopos : ariuonepos : siteś : tetu
Reading in original script:
Object:	CO·48 Prestino (slab)
Position:	front
Orientation:	0°
Frame:	(left: feet, middle: top and bottom, right: feet)
Direction of writing:	sinistroverse
Script:	North Italic script (Lepontic alphabet)
Letter height:	4–5 cm1.575 in <br />1.969 in <br />
Number of letters:	49
Number of words:	6
Number of lines:	1
Workmanship:	carved
Condition:	complete
Archaeological culture:	Golasecca III A 1 [from object]
Date of inscription:	second quarter of the 5th c. BC [from object]
Type:	dedicatory
Language:	Celtic
Meaning:	'Uvamokozis Plialeθu gave/dedicated/put up (the) seats (?) to/for (the) U/uvltiauioi A/ariuones'
Alternative sigla:	Tibiletti Bruno 1981: 23 Solinas 1995: 65 Motta 2000: 2 Morandi 2004: 180
Sources:	Morandi 2004: 638–640

Images

Drawing of inscription CO·48 from Gambari & Colonna 1988	Slab CO·48 Prestino with inscription CO·48 from Lejeune 1971	Drawing of inscription CO·48 from Morandi 2004	First half of inscription CO·48 from Morandi 2004	Second half of inscription CO·48 from Morandi 2004
Slab CO·48 Prestino with inscription CO·48 from Tibiletti Bruno 1966b	First part of inscription CO·48 from Tibiletti Bruno 1966b	Middle part of inscription CO·48 from Tibiletti Bruno 1966b	Last part of inscription CO·48 from Tibiletti Bruno 1966b	Drawing of inscription CO·48 from Tibiletti Bruno 1966b
Detail of inscription CO·48 from Tibiletti Bruno 1969b	Detail of inscription CO·48 from Tibiletti Bruno 1969b

Commentary

First published in Tibiletti Bruno 1966b.

Images in Mirabella Roberti & Rittatore Vonwiller 1966: tav. LXXXIV (drawing by Rittatore Vonwiller, cf. Tibiletti Bruno 1968c: 351 with n. 40, = Mirabella Roberti 1966: 115 f. [in two parts] = Gambari & Colonna 1988: 132, fig. 9), Mirabella Roberti 1966: 115 (photo), Tibiletti Bruno 1966b: 289, fig. 1 (photo), 290, fig. 2 (photo of first part), 291, fig. 3 (photo of middle part) and 4 (photo of last part), 293, fig. 5 and 6 (drawing), 294, fig. 7 (detail photo), 295, fig. 8 (detail photo) and 9 (detail photo = Tibiletti Bruno 1968c: 354), Tibiletti Bruno 1969b: 298, tav. XXVI.77a (photo = Lejeune 1971: 98, pl. XII = Markey & Mees 2003: 129, fig. 3 = Eska 2024d: 78, fig. 1) and 77b (detail photo), 299, tav. XXVII.77c (detail photo), Morandi 1982: 188 (drawing) and tav. XLI.1 (photo), Tibiletti Bruno 1990b: 281 (drawing), Morandi 2004: 642, fig. 22.180 (drawing) and 801, tav. XXIV.180b and c (photo in two parts), Morandi 2017b: 425 (drawing) and tav. XCVII (photos), Stifter 2020b: 357, fig. 8 (drawing).

Description.
Inscribed sinistroverse in a single line in neat and evenly spaced letters along the narrow side of the slab (length ca. 187 cm), closer to the top edge (distance from the top edge ca. 5 cm, distance from the bottom edge ca. 7 cm) and somewhat off-centre toward the left (distance from the left edge ca. 75 cm, distance from the right edge ca. 100 cm). The text is written between two frame lines (distance ca. 5–6 cm) which end on both sides in "feet" reminiscent of VA·6 Vergiate and the pala-stelae. According to Mirabella Roberti 1966: 114, the frame lines and letters were incised with different instruments. Words are separated by vertical rows of three dots. The grooves of the letters were filled with talcum soon after the discovery (Tibiletti Bruno 1966b, 281, n. 5), but the filling was removed by the making of two casts, one of paper and one of plastic (Tibiletti Bruno 1968c: 350, n. 36, Tibiletti Bruno 1968: 386). Morandi 2004: 638 notes traces of red paint in the letters of siteś.

Reading.
Detailed descriptions of the letters are provided in Mirabella Roberti 1966: 115 and Tibiletti Bruno 1966b: 290–296. The reading is largely unproblematic; only in two cases have there been doubts about the identification of letters. A vertical crack between letters 28 and 29 has caused some early commentators to misread letter no. 28 as rho rather than pi (Mirabella Roberti & Rittatore Vonwiller 1966: 408, Mirabella Roberti 1966: 114, Prosdocimi 1967: 201 with consequent analysis of uvitiauioros as another nominative PN); the issue was settled by Tibiletti Bruno 1968c: 352–354 (also Tibiletti Bruno 1968: 387 f., Lejeune 1971: 98). Tibiletti herself (Tibiletti Bruno 1966b: 292–294 and in all later publications, especially 1968c: 272, 1968: 386 f., 1981: 181, n. 4) read the third letter of the third word as iota rather than lambda, arguing that the bar looked different than the other bars of lambda in the inscription, and was an unintentional mark caused by the removal of dirt from the inscription. Her assessment was originally accepted by Prosdocimi 1967: 200 f., but rejected by Lejeune 1971: 97, n. 338 (also Eska 1998c: 67, Motta 2000: 197 and Morandi 2004: 639, who note that the distance between putative iota and tau would be irregularly large, and Markey & Mees 2003: 130, n. 5 after autopsy; cf. also Dupraz 2015: 36, n. 14). Prosdocimi 1986: 233 concluded, also after an autopsy, that the bar was intentional, but suggested that it was an aborted misspelling, and that iota was intended (also 1987: 569). Solinas 2010: 137–140 retains Tibiletti's reading (without discussion). The persistence of Tibiletti's iota in the discussion is conceivably due to the problems posed by the phonetically implausible sequence ⟨uvlti⟩ (cf. Lejeune 1971: 100, n. 348, Eska 2024d: 83, 85 f.) – unless ⟨uv⟩ is taken to represent /u/ (hardly, see below), or /l/ in the form is admitted to be syllabic (see uvltiauiopos), a writing error seems to be the only solution. Transposition of upsilon and waw was suggested by Morandi 2004: 639 (also 2017b: 425), Lambert 2008: 256 (dittography with uvamokozis) and De Bernardo Stempel 2014: 95. Isaac 2007: 14 proposed that a vowel letter was left out, and the intended sequence was ⟨uvlat⟩. Stifter 2003: 239, n. 1 opts for transposition of tau and iota: ⟨uvlti⟩ for ⟨uvlit⟩; his suggestion is pursued by Eska 2024d: 86. Against the assumption of a writing error Mees 2008: 197, n. 5.

Letter forms.
The inscription features a few rare or unusual letter forms: framed theta with a dot (cf. VA·3 and maybe the damaged VA·4.1), tau in the shape of a Greek cross (cf. TI·36.3) rather than the standard St. Andrew's cross variant (though the latter may in fact be a form of theta, see Θ), zeta in an otherwise unattested form with horizontal bars , and san with prolonged hastae rather than the standard Lepontic . The horizontal bars of zeta and orientation of tau have been interpreted as a cohesive stylistic choice (Tibiletti Bruno 1966b: 297 and 1969b: 234, Prosdocimi 1967: 201 and 1987: 573). Tibiletti Bruno 1966b: 292 also considers the varying orientation of four-bar sigma to be intentional for aesthetic purposes (also 1990b: 95; Markey & Mees 2003: 136 compare GR·3).

Sound value of letters.
Pi, tau and kappa, where clearly interpretable, denote /b/ (in the dat. pl. endings -opos = -obos, -epos = -ebos), /d/ (in siteś = sīdents, tetu = dedū), and /g/ (in -kozis = -gođis). Tibiletti Bruno 1968: 392 f. and Prosdocimi 1987: 571 f. assumed that the letters were bivalent in the Prestino inscription as they are usually in the Lepontic alphabet (with different theories about the function of theta, also Tibiletti Bruno 1990b: 100–102, Markey & Mees 2003: 135, 146, Mees 2008: 197). Prosdocimi 1967: 204 (also Prosdocimi 1987: 571–573) took the presence of tau and theta as evidence for influence from Venetic writing practice, where the two letters appear together in the Padova alphabet, but – as pointed out by Tibiletti Bruno 1968: 390 (also Lejeune 1971: 100 f.) – no other Patavinian elements are present in the inscription (e.g. spelling of /b/ and /g/ with pi and kappa, not phi and chi as in Venetic). That pi, tau and kappa were in fact reserved for the mediae, and theta represented /t/, was proposed by Lejeune 1971: 99 f. While Lejeune, in accordance with the document's low dating at the time, assumed that the introduction of theta (and implicitly chi) was part of an orthography reform which revived lettres mortes, Gambari & Colonna 1988: 134 proposed that the Prestino inscription's letter use was representative of archaic Lepontic orthography (with pi, tau, kappa for the mediae, theta and chi for the tenues, cf. NO·1 and see North Italic Script). Pro theta = /t/ also Eska 1998c: 67, Morandi 2004: 639 and Morandi 2017b: 425.

Zeta denotes the reflex of */st/ in -kozis < *-gʰostis, which according to established wisdom should be /ts/ or /t^s/; this agrees well with the letter's Etruscan sound value, though the same sound is spelled with san in the roughly contemporary VA·6. In the Prestino inscription, san is assumed to be used to denote /(n)ts/ with epenthetic /t/ < /ns/ in siteś, which should in essence be the same sound/cluster, but the employment of different letters suggests that they were pronounced differently. The logic behind the distribution of the two letters is not evident, but note that san seems to be used in the same function as in Prestino in the considerably later PG·1.4 (cf. Tibiletti Bruno 1966b: 303 f.). Cf. also Prosdocimi 1967: 215 (who assumed that san denoted a sibilant dissimilated from word-initial /s/), Tibiletti Bruno 1968: 392 f., Lejeune 1971: 100, Markey & Mees 2003: 155.

The function of waw, which occurs twice in a similar position, viz. in an anlaut ⟨uv⟩, is under debate. After the etymology of the first element of uvamokozis < *(H)upm̥mo- was established by Prosdocimi 1967: 213, different solutions for the relation of waw to a phonetically reduced/lost */p/ were proposed.

1. waw = /u̯/. The letter waw would a priori be expected to denote /u̯/ as it does in the Etruscan alphabet, but this function is fulfilled by upsilon in the Lepontic alphabet overall, and also in the Prestino inscription in ariuonepos and the second instance in uvltiauiopos. Prosdocimi 1967: 213, 202 f. initially suggested that waw was inserted to reflect a glide in hiatus [u^u̯a] (or [^uu̯a]) after the loss of */p/, to avoid a geminate spelling ⟨uu⟩. He subsequently modified his theory, considering waw a dead letter which was retained in the theoretical alphabet to be reactivated at pleasure, in the present inscription – ad hoc or an otherwise unattested writing school's practice – representing [h] < */p/ "realizzato come" /u̯/ (without quite committing to the exact sound value) (1986: 241, 1987: 571). See also Solinas 2010: 138 f., Markey & Mees 2003: 134, 145, Isaac 2007: 12, 14. While a hiatus stage /u.a/ per se cannot be excluded (Stifter 2011: 2), this theory is neither linguistically not epigraphically plausible, for multiple reasons. A) neither prothetic /u/ nor insertion of a glide /u̯/ in a comparable sequence has parallels in Celtic, or would be expected typologically (Eska 1998c: 69–71, Eska 2013b: 39 f.); B) in an intermediate stage /uu̯/, /u̯/ would then not be lost, but treated the same as in an inherited sequence (Schmidt 1980: 182, mistakenly aimed at Lejeune, Stifter 2011: 2 f., Eska 2013: 35). Cf. also the roughly contemporary VA·6 teu dēu̯ū, where /u̯ū/ is regularly spelled ⟨u⟩ (for illicit geminated ⟨uu⟩) (Eska 1998c: 71, Eska 2013b: 35).

2. ⟨uv⟩ = digraph for initial /u̯/. This theory, which assumes the complete loss of */p/, was floated and rejected by Tibiletti Bruno 1966b: 312 (after a suggestion from "un illustre studioso" – Vittore Pisani? – at a conference) and Prosdocimi 1967: 202, but argued by Lejeune 1971: 100, 102. Lejeune distinguished between intervocalic /u̯/ ⟨u⟩ and initial /u̯/ ⟨uv⟩, but could offer no convincing explanation for this spelling practice. Indeed, Eska 1998c: 69 observes that allography between initial and intervocalic /u̯/ would be ad hoc, noting examples for initial /u̯/ spelled with upsilon (cf. particularly uasekia, which may be formed with the same preverb). Dupraz 2015, however, notes that no evidence for upsilon denoting initial /u̯/ can be found in the oldest inscriptions, while the few, if uncertain/uninformative attestations of waw are all in the anlaut (VA·4.2 viχu, CO·31 v[). Dupraz concludes that there is allography between the an- and inlaut, and considers ⟨uv⟩ (among whose attestations he also counts the mark beside NO·1) a variant of ⟨v⟩ in the anlaut, an innovation introduced by Lepontic and Ligurian scribes, possibly because of allophony. (See further on the Ligurian attestations MS·2 uvezaruapus and MS·1 vemetuvis.) For this argument, Dupraz has to discount inlauting ⟨uv⟩ in VA·4.1 a??ouvi[ (as uncertain) and MS·1 vemetuvis (as deviant). The earliest attestations of initial upsilon appear on pala-stelae (TI·45.2 uesa?ai, arguably TI·54 )uikalai, maybe TI·38 uosiu(), any of which could potentially date from the 5^th c. BC. ⟨uv⟩ as a digraph was also preferred by Schmidt 1980: 182, Stifter 2003: 239, n. 1 (for [φ]), Markey & Mees 2003: 134 and Isaac 2007: 13 f. The latter two adduce Etruscan spelling variation as in ET Fa 3.1 auvilesi vs. regular avile, while Stifter 2011: 3 notes the Celtic loan in Etruscan script eluveitie, but cf. Eska 2013b: 36–38 on these forms.

3. waw = [φ]. That waw denoted a segmental reflex of */p/ before its loss was first asserted by Lejeune 1978: 111, who assumed that it represented a voiced spirant ([β] as per Eska 1998c: 72). Eska 1998c: 72–74 (also Eska 2013b) argues for [φ] (affirmative Schumacher KP: 133 f.). Isaac 2007: 12 f. contends that [φ] should have been written ⟨vh⟩ or , but cf. Stifter 2011: 2, who notes that the Etruscan fricative is [f]. Eska 1998c: 73 excludes that the reflex of */p/ is [h] under the assumption that the Lepontic alphabet had heta – while this is not the case, it can be argued that the Lepontic alphabet would have retained heta if it had had a [h] which to denote with it (cf. also Eska 2013b: 36). Eska 2024d provides an analysis of uvltiauiopos which agrees with his theory – waw is usually assumed to have the same function in this form as in uvamokozis, but had previously not been involved in the discussion due to its difficult reading.

In summary: until the employment of ⟨v⟩ for [φ] or the use of a digraph ⟨uv⟩ for #u̯ can be shown to be unambiguously attested in other – ideally Lepontic – inscriptions than the present one, the issue cannot be resolved with certainty. See the forms mentioned in the discussion above and V on other instances of waw in the Lepontic alphabet. An entirely different approach (waw for */ps/ in *upsamo-) is taken by Tibiletti Bruno 1968: 391, 1990b: 98 f., De Bernardo Stempel 1990, De Hoz 1992: 225 (critical Eska 1998c: 71 f.).

Interpretation.
The syntactic analysis of the text as a nominative subject uvamokozis plialeθu, dative plural objects uvltiauiopos ariuonepos, accusative plural object siteś, and finite verb tetu, was already given by Tibiletti Bruno 1966b: 301. See the word pages for details about the linguistic analyses. The only significantly different analysis was forwarded by Prosdocimi 1967: 206–208, who interpreted plialeθu as another finite verb and uvitiauoros as another subject ('Uvamokozis plial-ed, Uvitiauioros gave the seat to the Ariuones'), comparing the structure of VA·6, but this is based on misreading (see above). The precise meaning of the indivdual forms is largely unclear, not least because the verb may mean 'gave' as well as 'put up' (see the word page and cf. Lejeune 1971: 105 f.; Eska 1998c: 66 give, also Markey & Mees 2003: 157). Cf. however, Eska 2024d: 79, n. 6.

The siteś are usually assumed to refer to the putative structure of which the inscribed and anepigraphic slabs were part, possibly a religious or public building (e.g. Tibiletti Bruno 1966b: 305–? 'set up/constructed a building/monument', Tibiletti Bruno 1969b: 234 "temenos (?)", Tibiletti Bruno 1978: 142 "recinto sacro, santuario (?)", Tibiletti Bruno 1981: 180 "dimora", Prosdocimi 1967: 214 f. 'palace, divine residence', Lejeune 1971: 104 f. plurale tantum referring to the monument, Motta 2000: 197; Prosdocimi 1986: 245, 1987: 573 assumes that the siteś are specifically the slabs; Markey & Mees 2003: 156). Mees 2024: 221 f., who argues that the text is funerary, links siteś with Etr. suθi 'tomb'; critical Eska 2024d: 79, who compares RIG L-30 canecosedlon. Cf. also the Messapic dedicatory inscription MLM 3 Ro, in which the accusative object hezzan is tentatively analysed as < *sedi̯am 'seating, seat' by De Simone 1988: 403 f.

The recipients of this gift, the uvltiauioi ariuones, have been suggested to be members of a family, minor divinities, magistrates or priests (e.g. Tibiletti Bruno 1968: 388, 1969b: 234, 1978: 141 f., 1981: 180, Prosdocimi 1986: 238, Motta 2000: 197 f.; the latter [also 1992: 314] considers ariuonepos to be a toponymic epithet, as in RIIG BDR-12-02 (RIG G-64) ματρεβο γλανεικαβο at Glanum). Eska 2024d: 79–81 compares Transalpine Gaulish inscriptions containing the verb dede, e.g. RIIG GAR-10-01 (RIG G-203) +αρταρ̣[ος] ι̣λλα̣νουιακ̣ος̣ δεδε ματρεϐο ναμαυσικαβο βρατουδε[καντεν] ']artar[os I]llanuiakos gave a tithe to the Matres of Nîmes in gratitude' and concludes that the uvltiauioi ariuones are also divine beings.

On the function of plialeθu (patronym? epithet? title?) see the word page. Complete interpretation in Markey & Mees 2003: 156 f., who translate 'Uvamokozis of the Pli° clan gave this territory to the fostered one's descendants, the Ariuones', a text commemorating and documenting the granting of land to a clan or tribe on a boundary stone, and compare the Vercelli bilingua. The authors also contend that the text is poetic/metrical; also Mees 2008: 196–199, critical Eska & Mercado 2005: 177 f.

Comparanda.
As already noted by Tibiletti Bruno 1966b: 283 f., the shape of the object and layout of the inscription can be compared with that of the fragmentary CO·21 Rondineto. Tibiletti Bruno 1990b: 83 argues that the latter indicates the existence of another complex comparable to the one in the Fondo Giulini, but of a later date; the differences in ductus cited by her, however, are minor (size of omicron), and no reports exist of similar structures having been found at Rondineto. The height of the face side of CO·21 Rondineto agrees with that of the slabs in the Fondo Giulini, as does the letter/frame height (cf. Lejeune 1971: 97, n. 336). It thus seems likely that the small fragment comes originally from the Fondo Giulini structure and was at some point – conceivably after the delapidation of the structure – transported to its eventual find place ca. 600 m uphill. Tibiletti Bruno 1990b: 86 further compares the Feltre stones, two fragmentary sandstone slabs of similar dimensions and uncertain date which are inscribed in what is usually thought to be Etruscan language and alphabet (ET Pa 4.1); notably, one of the inscriptions also refers to a group of deities (ki aiser 'three gods').

Dating.
Tibiletti Bruno 1966b: 300 f. initially dated the inscription to the 2^nd c. BC – viz., according to the absolute datings at that time, between Lepontic in the 3^rd c. and the Gaulish monuments of the 1^st c.; she considered the frame with "feet" at both ends to be a late variant of the anthropomorphic frames of the pala-stelae of the Lugano area (p. 287–290, also 1981: 178, 1990b: 87–94, Morandi 1982: 188), but it is now considered an earlier variant (cf. VA·6). In tandem with the overall revision of Golaseccan chronologies in the early 1980s, the Prestino inscription's chronological position was shifted significantly backward. After Motta 1983 had pointed out that the inscription's letter inventory looked archaic rather than reformed (as assumed by Lejeune, see above), Prosdocimi 1987: 569–573 (also 1986: 232) argued for a palaeographical dating to late 6^th–early 5^th c. BC, viz. the "inizio assoluto della traditione scrittoria". His assessment was mainly based on comparison with the Etruscan alphabet, which he concluded to have been borrowed before the end of the 6^th c., and the presence of closed alpha beside waw, supported by a letter from De Marinis (1982), in which the latter expressed his conviction that the inscription dates to the 5^th c. BC (Golasecca III A) like the majority of inscription finds from the Como area (p. 569). See also CO·48 Prestino on the associated ceramic finds. Differently Tibiletti Bruno 1990b: 94–102, who considers the alphabet to be archaising.

Significance for the classification of Lepontic.
Tibiletti Bruno 1966b: 302 originally considered the inscription to be linguistically "Gaulish" (in quotation marks), later mixed Gaulish with Lepontic elements (Tibiletti Bruno 1969b: 234, 1978: 142). In a dedicated discussion, Prosdocimi 1967: 219–222 confronted linguistically Celtic features (*/st/ > /ts/ uel sim. [tau gallicum] spelled with zeta in uvamokozis, */ē/ > /ī/ in siteś, possibly the loss of */p/ reflected by waw in uvamokozis, the verbal ending -ū) with non-Gaulish features (the dat.pl. ending -bos, the personal name elements, especially -kozis), and concluded that the language of the inscription was not Gaulish, but a Lepontic "non antigallica". More explicity, Lejeune 1971: 110 f. classified the text as Lepontic "para-gauloise"; he argued (106–110) that */n̥/ > /en/ rather than /an/ in siteś is evidence for a divergent sound development in Lepontic and, together with the different distribution of the past endings -e and -ū in Lepontic and Gaulish, confirmed the status of Lepontic as a discrete branch of Celtic. Critical Solinas 1995: 344 f. See also Prosdocimi 1986e: 72–74.

• Markey 2024
• Eska 2024d

Pellegrini 1969: 251 f., Lejeune 1974: 343, Pellegrini 1981: 58

See also Ferri 1967 (reply in Tibiletti Bruno 1968c: 354 f., n. 52), Campanile 1968 (sdegnato), Morandi 1982: 188 f., n. 69, Pellegrini 1983: 36 f., Prosdocimi 1991: 141, Prosdocimi 1991b: 52 f., Lambert 1994: 21, Meid 1996: 260, Meid 1999: 16 f., Morandi 2017b: 424 f., no. 553, Stifter 2020b: 345, 357.

Bibliography

Campanile 1968	Enrico Campanile, "Su due interpretazioni della iscrizione di Prestino", Studi e Saggi Linguistici 8 (1968), 207–213.
De Bernardo Stempel 1990	Patrizia De Bernardo Stempel, "Einige Beobachtungen zu indogermanischem /w/ im Keltischen", in: Ann T. E. Matonis, Daniel Frederick Melia, Celtic language, Celtic culture. Festschrift for Eric P. Hamp, Van Nuys: Ford & Bailie 1990, 26-46.
De Bernardo Stempel 2014	Patrizia de Bernardo Stempel, "Livelli di celticità linguistica nell'Italia settentrionale", in: Philippe Barral, Jean-Paul Guillaumet, Marie-Jeanne Roulière-Lambert, Massimo Saracino, Daniele Vitali (eds), Les Celtes et le Nord de l'Italie. Premier er Second Âges du fer. Actes du XXXVIe colloque international de l'AFEAF, Vérone, 17–20 mai 2012 [= Revue Archéologique de l'Est Suppl. 36], Dijon: 2014, 89–102.
De Hoz 1992	Javier de Hoz, "Lepontic, Celt-Iberian, Gaulish and the archaeological evidence", Études celtiques 29 (1992), 223–240.
De Simone 1978	Carlo De Simone, "I Galli in Italia: testimonianze linguistiche", in: Paola Santoro (ed.), I Galli e l'Italia, Roma: De Luca 1978, 261–269.
De Simone 1988	Carlo de Simone, "Iscrizioni messapiche della Grotta della Poesia (Melendugno, Lecce)", Annali della Scuola Normale Superiore di Pisa. Classe di lettere e filosofia 18 (1988), 325–415.
Dupraz 2015	Emmanuel Dupraz, "Nochmals zum lepontischen Digraphen uv-", Münchener Studien zur Sprachwissenschaft 69,1 (2015), 33–50.
Eska & Mercado 2005	Joseph Francis Eska, Angelo O. Mercado, "Observations on verbal art in ancient Vergiate", Historische Sprachforschung 118 (2005), 160-184.
Eska 1998c	Josef Francis Eska, "PIE *p (doesn't become) Ø in proto Celtic", Münchener Studien zur Sprachwissenschaft 58 (1998), 63-80.
Eska 2013	Joseph F. Eska, "A salvage grammar of Galatian", Zeitschrift für celtische Philologie 60 (2013), 51–63.
Eska 2013b	Joseph F. Eska, "In defense of Celtic /φ/", in: Adam I. Cooper, Jeremy Rau, Michael Weiss, Multi nominis grammaticus. Studies in classical and Indo-European linguistics in honor of Alan J. Nussbaum on the occasion of his sixty-fifth birthday, Ann Arbor: Beech Stave Press 2013, 32–43.
Eska 2024d	Joseph F. Eska, "Digamma and Prestino and related matters", Zeitschrift für celtische Philologie 71/1 (2024), 77–96.
ET²	Gerhard Meiser, Etruskische Texte. Editio minor, auf Grundlage der Erstausgabe von †Helmut Rix neu bearbeitet in Zusammenarbeit mit Valentina Belfiore und Sindy Kluge. Teil 1: Einleitung, Konkordanz, Indizes, Teil 2: Texte, 2nd, revised edition [= Studien zur historisch-vergleichenden Sprachwissenschaft 4], Hamburg: Baar 2014.
Ferri 1967	Silvio Ferri, "Esigenze archeologiche - VIII", Studi Classici e Orientali 16 (1967), 417–435.