=tλ

This form is attested on a dedicatory plate found in the mid-Kaystros valley and a small fragment found in Sardis. The former text mentions the twelfth regnal year of Alexander and can thus be dated to the early Hellenistic period.     

[I.Y.]

indecl.	=tλ	LW 50.5 (Tire)
		LW 107 (Sardis)

The form cited above is not a stem but an inflected form, if this clitic was indeed inflected. Yet, since the other case form of this stem are not attested, while its grammatical function is not definitely established, we would gain little from positing a questionable stem-form =t(i)-, or anything similar.   

[I.Y.]

The marmor plate LW 50 must have accompanied a votive architectural installation (taacn) dedicated to Zeus, but the clitic under discussion is attached to the name of Zeus, functioning as the subject of the respective clause (this is also true of the fragmentary passage LW 107). This opens a possibility that the clause under discussion (LW 50.5 lefs=tλ faτaν rawal) refers to a background event that provided the rational for erecting the monument. This hypothesis derives further support of the etymology of Lyd. rawa-, which indicates that it denotes a favorable action. Since this background information follows the key clause of the inscription, referring to the dedication of the monument, it is likely to be expressed by a dependent clause. The clitic =tλ is the only candidate for a subordination marker in the clause under discussion, and given its inflectional ending, it appears to be functionally equivalent to the form of a relative pronoun. If so, the sentence can be interpreted along the lines ‘To whom Zeus rawa-ed the faτa-’, where ‘whom’ refers back to the author of the inscription. A different syntactic analysis offered in Hajnal 2004a:198 (‘(It is) Zeus, to whom (he = author) granted the faτa-.’) does not change the functional interpretation of =tλ.

A substantially different analysis of =tλ, found in Brandenstein 1931b:35 and Gusmani 1964a: 213f., implies that this is a form of a demonstrative pronoun. This analysis is contextually less attractive, since the information that requires backgrounding on discourse functional grounds is no longer rendered by a dependent clause, and yields no obvious etymological gain, since a demonstrative pronoun beginning with t- is not attested in Anatolian. Yet, since the solution preferred here is not based on a solid etymology either (see below), this alternative cannot be formally refuted. Definitely unacceptable is the proposal of Carruba 1959a:37 to analyze =tλ as =t=λ, since the particle =t cannot occupy the position before an inflected form within a clitic chain.        

[I.Y.]

On the assumption that =tλ means ‘whom’ it can be analyzed as a secondary cliticized form of qλ, the dat. sg. form of the Lydian relative pronoun qiš. The change q- > t- would find a precise parallel in the evolution of the relative pronoun in Lycian A, but there it reflects a phonetically regular palatalization, whereas in this case it could only reflect a precocious change in a clitic formation. Since this phonetic hypothesis lacks independent support, it must remain a hypothesis, which in turn undermines to an extent the synchronic analysis given above.  

[I.Y.]