Folia archeologica 25.

János Harmatta: Two inscriptions scratched on fragments of pottery from Pannónia

102 J. H ARM ATT A the Lepontic inscription PID No. 292 the left-side stroke of the r is similar to that of Tokod, while its right-side stroke does not bend inside backward, but curving only a little inside it terminates on the same level as the left-side one, without touching it. The r of the Lepontic inscription PID No. 262 is also similar, with the only difference that in this the left-side stroke crosses below the right-side stroke. It can be mentioned that as from the 2nd century B.C. the rho with an open head appears also in the Greek cursive script, and its form is very close to the r of the inscription from Tokod. Thus in connection with this letter we can eventually also think of a Greek influence. The 8th letter is an /, the right-side stroke of which is considerably shorter than its left-side stroke. Otherwise, if the two strokes were equal in lenght, we could also think of a reading u. This form of the / can only be found again in the Sondrio alphabet (as a Greek influence), while the /in the Lepontic alphabet differs from it. Thus, in final conclusion, the alphabet of the inscription from Tokod can be put between the Lepontic alphabet and that of Sondrio. Of the two it can perhaps still be connected closer to the Lepontic alphabet, because among the alphabets of Northern Italy this form of the m developed only in the Lepontic one. At any rate it must also be taken into consideration that here we have to do with a considerably later variant of the Lepontic alphabet, which could be influenced to a certain extent by the Raetic alphabet or even by the Greek and the Latin ones.

7 On the basis of all these the inscription can be read as follows: ^a^mattral. This form is undoubtedly a patronymic with the suffix -л/,

8 which therefore had to be preceded by a personal name. Thus the inscription from Tokod, as regards its structure, could be similar to the Lepontic inscriptions PID No. 255 (raneni ualaunal

) and PID No. 301 (koimi latunal)

. Unfortunately even the suffix -я/does not decide the origin of the inscription quite unequivocally from the linguistic point of view, because this suffix is also known in Raetic territory and it occurs in the inscriptions of Magrè and Sondrio, and it is possible that it was borrowed by Lepontic from the Raetic language.

9 The form of name amattral is obviously the derivative with the suffix -tra of the Etruscan name *am(a

) (cf. Etruscan amnei, amanas, amre

, etc.

1

0). Thus it is a gentile name with a formation like sectra, %actra, Santra, Caetra, etc.

1

1 In this case the geminate -tt- ought to be understood as an orthographic phenomenon. However, if we attribute to the geminate a significance from the phonetic point of view, then we can also think that the basic form of the name was *a?nat- or 7 For Greek and Latin influence on North Italian alphabets see PID II. 506, 515, 517. 8 For the suffix -al see Pedersen, H., The Lepontian Personal Names in -alo-s and some Remarks on the Lydian Inscriptions. Philologica. JCPh 1(1921—22) 38ff.; PID II. 70, 552.; Tbitrneysen, R., Glo. 21(1933) 6f.; Krabe, H., Ligurisch und Indogermanisch. Germanen und Indogermanen. Festschrift für Herman Hirt. II. (Heidelberg 1936) 247. a As presumed by Tktriieysen, R., Glo. 21(1933) 6ff. For the suffix -al in Raetic inscriptions see PID II. 548. 1

0 Schulze, IT., ZGLE 121. 1

1 Ibid. 342, 393. For the Etruscan suffix -tra see Pfiffig, Л. J., Die etruskische Sprache. (Graz 1969) 169.