Phrygian etymological database (in progress) [Lubotsky] :

   The inscription is written on a small alabaster falcon. It was analysed by Lejeune (e.g. Lejeune 1969b: 293, Lejeune 1979) as dat.sg. of the dedicatee + nom.sg. of the dedicator + acc.sg. of the dedicated object. He further took bagun to be "un nom ge´ne´rique du `don' " and reconstructed *b<sup>h</sup>agom.
This interpretation is formally possible, but not very probable. The original meaning of the IE root *b<sup>h</sup>ag- (*b<sup>h</sup>eh<sub>2</sub>g-) was `to (give) share', and derivatives of this root hardly ever get the meaning `gift' (Lejeune further mentions in this connection the Hesych gloss Baga‹oj ZeÝj FrÚgioj and interprets the name as "dothr ˜£wn", but this is of course very uncertain). If bagun rather means `idol', then a borrowing from Iranian becomes likely.

   nom.sg. baj 48 99 111, ba[j] 86;
acc.sg. batan 33 36, batan T-02b (?), batant·e· (Mys.6) (?)

   48. ... Mitrafata / ke maj temroge/ioj ke pountaj / baj ke ...
86. ... ba[j] / ioi bekoj mebere[t] / a·t tih ke ti tetikm[e]/noj eitou
99. ... me ke oi totosseiti baj bekoj
111. ... baj ioi b(e)koj meberet (---)/
33. ... autoj ke oua k eroka gegaritmenoj aj batan teutouj
36. ... autoj ke ou/a k oroka gegaritmeno/j a<j> batan t/eutouj

   We come across Baj in malediction formulae with bekoj `bread', cf. 86, 111 Baj ioi bekoj meberet `Bas will take his bread away' {1}, similarly 99 me ke oi totosseiti Baj bekoj. The acc.sg. of the same theonym is probably Batan (cf. Lubotsky 1989b: 149), found in the curses of 33. autoj ke oua k eroka gegaritmenoj aj Batan teutouj and 36. autoj ke oua k oroka gegaritmenoj a<j> Batan teutouj `and may he himself and his progeny (?) become cursed by Bas' (cf. syntactically similar ti(t) tetikmenoj aj Tian eitou, attested in 14, 53, 99). We may thus reconstruct the paradigm: nom.sg. Baj < *bat-j, acc.sg. Batan. It is remarkable that Baj and Batan only occur in inscriptions found in the northern part of the New Phrygian area, approximately at the 39th parallel and to the north of it. This fact can be seen as an indication of the northern provenance of this deity.{2}
Baj, Batan may reflect PIE *b^heh₂-t-, if it is identical with Fatum, cf. frequent appearance of mÒira in Greek inscriptions of Phrygia.

   33, 76, 108. ... ak ke oi bekoj akkaloj ti dregroun eitou ...
86, 111 ... baj ioi bekoj me beret ...
99. ... me ke oi totosseiti baj bekoj ...
18. ... b·e<k>o/j ioi me totoss' eugisarnan
120. ----]u·o·i·j· g·e·n·t·i·b·e·p·a·i· ke p·arthj b·e·koj· (?)

   The Phrygian word is also mentioned by Herodot and Hipponax and is given as a gloss by Hesychius. The only reasonable Indo-European etymology connects Gr. fègw, ON baka, OHG bahhan, OE bacan `to bake' < PIE *b^h(o)h₁g´- ( Panagl - Kowal 1983: 186f), but it presupposes Lautverschiebung in Phrygian. As indicated by Lejeune 1979: 223, a word with this semantics can easily be borrowed.

    Lejeune 1979: 224 proposed to analyse this compound PN as *g^wenƒ `woman' + *g´onh<sub>1</sub>o- `born', which seems rather unlikely (`woman-born' is not very suitable for a name, there is no evidence for the development *g<sup>w</sup> > b in Phrygian, the Phrygian word for `woman' is now found in 116 knaik-).
In general, it is useless to speculate about the etymological connections of the names. The second member may as well be compared with Greek compounds in -cqonoj (see s.v. pokgonion).

   3sg.act. *ad-beret: abberet 6 13; ab·b·e·ret 114; abbe[ret] 11; [a]beret 103;
3sg. med. *ad-beretor: abberetor 73 75; [abb]eretor? 79; [a]bberetoi (read -r? {1}) 91; abberetoi (read -r? {1}) 113; abbireto(<r>?) 25;
3sg.act. meberet 111, mebere[t] 86 `to bring away, deprive of'

   6. [io]j ni semoun knoumane [kakon] / abberet atnou [-]mon[--]n ...
113. [ioj] semon ti knoumani k[ak/on a]beret{i·t·e·t·i·k·-}partan
114. ioj ni [s]emon knoumanh kakon ab·b·e/ret atnoukton mrossaj ...
73. ioj ni semon knoumani kakon abberetor ai·ni· saj mdu·ei ---
25. ioj ni simoun / knoumani [k]akoun ab/bireto aini mmura ...
86. ... ba[j] / ioi bekoj mebere[t] / a·t tih ke ti tetikm[e]/noj eitou
111. ... baj ioi be·koj meberet (---)

    Orel 1997: 139f. takes bevdos to be a proper name in the nominative. He refers to Zgusta 1984: 121, who mentions several Phrygian place names like PalaiÕn Beu~doj, Beudou O‡koj. Zgusta further connects the gloss found at EM 195.52, viz. beu~doj ... ¥galma (at Hermione) `statue of a god' and writes: "es kann sich um eine phrygische Glosse handeln, und das ¥galma konnte das Bild einer Gottheit sein". I would add that Gr. beu~doj n. `sumptuous woman's dress' (Sappho, Kall., etc.) presumably is the same word. The Greek may have borrowed this word from Phrygian in the meaning `statue of a goddess', but since these statues were lavishly adorned and dressed, beu~doj was used in the narrower meaning of a specific sumptuous woman's dress.
It is then much more probable that OPhr. bevdos is not a name, but the word for the statue (of a goddess). The Germanos inscription (B-01) is written on a rock immediately beneath a niche which most probably served for a statue of Kybele. The Old Phrygian inscriptions often start with the accusative (e.g. M-04 a·kinanogavan· : tiyes / mod·rov·anak : avara·), so that si bevdos may mean `this statue', bevdos being acc.sg.n. of an s-stem. For si acc.sg.n. of a demonstrative pronoun cf. si keneman in M-01b.
As already surmised by Orel 1997: 140, this word is derived from IE *b^heud^h- `to perceive'. I take it as a regular s-stem *b<sup>h</sup>eud<sup>h</sup>-os- (cf. Gr. ¢-peuq-»j `ignorant', Av. bao‹ah- n. `perception'). The original meaning of this formation must have been `perception, image', which seems to be a suitable term for a god's image.

   dat.sg. bratere 31;
dat.pl./acc.pl. (?) braterais Mys.8