Exegesis:
archomenos ‘when beginning.’ No object stated but to be supplied from what follows in Ch. 4, i.e. his service.
ēn … hōsei etōn triakonta ‘was about thirty years (old).’ The main clause, to which is attached the long genealogy in the form of a participial clause, cf. next note.
hōsei with numbers ‘about,’ ‘approximately.’
ōn huios, hōs enomizeto, Iōsēph ‘being the son, as was supposed, of Joseph.’ The participial clause introduces the subsequent genealogy. hōs enomizeto has no logical subject but a subject like ‘people’ may be supplied if necessary. nomizō, cf. on 2.44.
Vv. 23b-38 present a genealogy of Jesus in the form of a series of genitives successively dependent upon one another. For a discussion of the theological and historical problems involved in these names the reader is referred to the commentaries. For translational purposes it will be sufficient to list the names in their Greek and Hebrew form. For the Greek form the text of GOOD NEWS BIBLE is followed. Variant spellings in the Greek are mentioned only when they have influenced the transliteration in English. Latin transliterations are given only when they have influenced the transliteration in modern languages. Column 5 lists (a) the passages where the persons named in Luke are first mentioned in the Old Testament., and (b) the passages containing the name in question but referring to another person than the one meant in Lk. 3; category (b) is given in brackets.
Translation:
Jesus, when he began … was…, or, ‘When Jesus began … he was…,’ or, ‘Jesus was … when he began….’
When he began his ministry, or, ‘his work,’ ‘his teaching,’ ‘to preach (lit. to transmit the way, Chinese Union Version, L),’ ‘to act-as-guru’ (Tae’). Some languages possess an expression for a person’s (first) appearance in a certain role, which requires no further qualification, e.g. ‘raised his head’ (Chinese BT), ‘made-his-appearance’ (Dutch, Zürcher Bibel), cf. also lors de ses débuts (Bible de Jérusalem).
Jesus … was about thirty years of age, or, ‘Jesus’/his age was about thirty (years),’ ‘Jesus/he was about thirty years old.’ See also 2.42.
The position of as was supposed, or, ‘as people saw it (i.e. according to people)’ (Chinese Union Version), must be such that it qualifies the relationship between Jesus and Joseph only, and does not suggest a supposed son, or even a supposed Joseph (as has been the case in one older version); in many cases the phrase is better placed at the head of the sentence, e.g. ‘people thought he was the child of Joseph’ (Manobo). For the verb to suppose see on 2.44. — Several versions make some kind of incision after Joseph, e.g. ‘people regarded him as the son of Joseph. Joseph (was) Heli’s, he (was) Matthat’s … Adam’s, he (was) God’s son’ (Marathi), cf. also Javanese, quoted below.
The long series of proper names in vv. 23-38, connected with each other by the simple device of the genitive case, may have to be rendered less concisely, e.g. by appositional phrases, or relative clauses, of the type of, ‘(who was) the son/child of,’ in some cases even duplicating the names, e.g. ‘Joseph, he the son of Heli, Heli, he the son of…’ (Manobo). In honorific languages a high level honorific equivalent of ‘son’ may have to be used in the very last phrase, ‘son of God’ (e.g. in Balinese). Some versions have made a successful attempt to express the structure of the original by other means than genitives, cf. e.g. “Joseph whose line went back through Eli, Matthat, … to Adam, the son of God” (The Four Gospels – a New Translation), .’.. was reckoned the son of Joseph, whose coming-forth (was) from H., from M., …, from Adam, from Allah’ (Javanese).
For general remarks on transliteration of proper names see above on the name “Luke” (pp. 3f), and references. The proper names in these verses present an additional problem in that most of them are themselves transliterations of Hebrew originals, cf. categories (a) and (b), as given in Exegesis. In the case of the names in category (a) it is clear that Luke wants to refer his readers to persons known from the Old Testament; hence it may be supposed that he intended to reproduce their Hebrew names as faithfully as the Greek permitted him to do. Consequently one should not transliterate the Greek forms of the names in question (as several older and some new versions do), but their Hebrew models. (For a comparable case see above on Gr. Kurēnios—Latin ‘Quirinius’ in 2.2.) In practice this means that the translator, by treating these names as they are treated in the Old Testament (or will presumably be treated when a translation of the O.T. passages concerned is made in the future) best fulfils Luke’s purpose, i.e. to help his readers to identify the Old Testament persons referred to. Names of category (b) can best be treated in analogy with (a). The remaining 14 names (probably transliterations also, but without known O.T. origin) should be treated as other New Testament names.
Quoted with permission from Reiling, J. and Swellengrebel, J.L. A Handbook on the Gospel of Luke. (UBS Handbook Series). New York: UBS, 1971. For this and other handbooks for translators see here . Make sure to also consult the Handbook on the Gospel of Mark for parallel or similar verses.