The translation in the Loeb edition of Josephus’ Antiquities 18.112 reads:
On his return after transacting his business in Rome, his wife, who had got wind of his compact with Herodias, before any information reached him that she had discovered everything, asked him to send her away to Machaerus, which was on the boundary between the territory of Aretas and that of Herod. She gave no hint, however, of her real purpose. Herod let her go, since he had no notion that the poor woman saw what was afoot. Some time earlier she herself had dispatched messengers to Machaerus, which was at that time subject to her father, so that when she arrived all preparations for her journey had been made by the governor. She was thus able to start for Arabia as soon as she arrived, being passed from one governor to the next as they provided transport. So she speedily reached her father and told him what Herod planned to do.
The footnote here says:
The reading of the mss. is “and to him who was subject to her father.”
The following notes are in the text here in the Loeb edition.
εἰς τὸν Μαχαιροῦντα τότε3 πατρὶ αὐτῆς ὑποτελῆ,4
3 τότε] ed. pr.: τῷ τε codd.
4 ed. pr.: ὑποτελεῖ codd.
The Neise edition simply reads, in agreement with the manuscripts:
εἰς τὸν Μαχαιροῦντα τῷ τε πατρὶ αὐτῆς ὑποτελεῖ
F. F. Bruce remarks in a footnote:
He appeals to the statement found in all the printed editions of Josephus, Ant. xviii 112, that Machaerus was subject to Aretas at the time of his daughter’s flight from Antipas (εἰς τὸν Μαχαιροῦντα τότε [τὸν τῷ Bekker] πατρὶ αὐτῆς ὑποτελῆ); but the manuscript tradition (εἰς τὸν Μαχαιροῦντα τῷ τε πατρὶ αὐτῆς ὑποτελεῖ) does not make this statement.
The reading of the printed texts and translations may be called a conjectural corruption of Josephus. Although it is to be found in the editio princeps of the Greek text of Josephus printed in 1544, it disagrees with the three extant manuscripts that attest to Antiquities 18 (AMW).
[Update May 2, 2015: The Latin version (and quotation of it) reads “apud Macherunta omnia pararentur quae itineris usus exposceret” and thus omits the phrase. A perceived inconcinnity in the original wording, “and to him who was subject to her father,” may explain this. On the other hand, if the original wording were the phrase “which was at that time subject to her father,” an omission by the Latin translators of the sixth century may be more difficult to explain. But if the original wording were the wording present in the Greek manuscripts, which could be considered a ‘difficult’ phrasing due to the oblique way of mentioning this person, both the omission in the Latin and the alteration in the Greek is explained.]
The first question might be to ask whether the reading of the manuscripts makes sense. Conjectures are often proposed when the manuscripts themselves do not present a text that makes sense. If the text does make sense, however, there is much less occasion for proposing a conjectural emendation.
τῷ – This is the singular masculine dative definite article that here means ‘to him’.
τε – This is an enclitic particle that is used here as a conjunction (‘and’).
πατρί – This is the singular masculine dative noun from πατήρ, ‘father.’
αὐτῆς – This is the singular feminine genitive adjective that means ‘her.’
ὑποτελεῖ – This is the singular masculine dative adjective that can be translated ‘subordinate’ or ’employed’ or ‘subject’ or ‘liable to taxation’ (from the adjective ὑποτελής) or the third person singular imperfect verb that is translated ‘was subject’ or ‘was paying’ or ‘was servicing’ (from the verb ὑποτελέω).
This is how the Loeb edition gets a translation reading, “and to him who was subject to her father.”
It is also possible to read it, “and to him subordinate to her father.”
There is an example of the adjectival form applied to an individual in Byzantine law: “Thus, for so many days he is called a visitor (παρεπίδηµος) and does not pay tax, but if he passes the fixed time, he becomes a metic thenceforth and liable to taxation (µέτοικος ἤδη γίνεται καὶ ὑποτελής).”
There is also this use in Lucian: “That is another thing that the women are keen about – to have men of education living in their households on a salary (μισθοῦ ὑποτελεῖς) and following their litters.” (Merc. Cond. 36)
Of the verb, there is this use in Lucian: “And it can not only solecize and barbarize, not only twaddle and forswear, call names and slander and lie – it can perform other services even at night (ἀλλὰ καὶ νύκτωρ τι ἄλλο ὑποτελεῖν), especially if your love affairs are too numerous.” (Rh. Pr., 23)
While both the verb and the adjective are also found in the context of paying tribute (and this may have indeed been the original reason that some have sought to correct Josephus here), that is not the only meaning that they can have. The rest of the construction naturally makes good sense as it is, so it is not very difficult to make good sense of the passage as it stands.
In a 2013 post, the user named spin made this comment in explanation of the text:
It makes little sense that Herod should send her out of his kingdom, if he needed to keep control of her (such marriages were treaties of sorts). Machaerus as Josephus indicates was on the border with Aretas. At Machaerus there was an agent of Aretas (“him who was subject to her father”), who was informed and who prepared for her onward journey, so she went forth (from Machaerus) into Arabia (καὶ ἀφωρμᾶτο εἰς τὴν Ἀραβίαν), ie out of Herod’s territory. Machaerus was not under the control of Aretas.
The second question is to ask whether the conjectural emendation helps to solve problems in the text. On the contrary, however, not only does it fail to solve any real problems in the text, because there are none, but it also throws up some problems of its own.
This topographical map shows how the Arnon River and River Wala (to the north of the Arnon River) created a natural southern border for Perea.
Likewise this political map.
The difficulty involved in believing that the fortress was held by Aretas has led some to accuse Josephus of simply being in error here.
The Loeb edition has a footnote here that reads (“he is wrong”):
N. Glueck, “Explorations in the Land of Ammon,” Bull. of the Am. Sch. of Orient. Res. lxviii, Dec. 1937, p. 15, on the basis of an archaeological survey of the area, concludes that Josephus is approximately correct in placing Machaerus on the border between the territory of Aretas and that of Herod, but that he is wrong in placing it in the territory of Aretas, which was a few miles away.
Similarly a book on Herod Antipas, p. 286, by Harold Hoehner, states (“…incorrectly…”):
At the time Antipas’ wife, daughter of Aretas IV, was about to be divorced, she asked to be sent to Machaerus, which Josephus describes correctly as being on the border between the lands of Aretas IV and Antipas, but incorrectly places the fortress in the territory of Aretas.
Emil Schürer is ready to countenance “a misstatement through carelessness” by Josephus and notes that some have even seen grounds for “an interpolation” here on account of the apparent error:
As Schürer notes in passing, the difficulty involved in the conjectural corruption of Josephus is not merely that the conjectural reading claims that Machaerus was held by Aretas, although that is indeed “very remarkable” given that it “at all other periods, before and after, formed part of the Jewish territory.” The difficulty is further, and it is worth repeating, that:
It is equally remarkable that Antipas should have guilelessly allowed his wife to go to this fortress belonging to the Arabian king. Or did he consciously agree to it in order to smooth the way for her flight, wishing thus to be rid of her? Josephus did not so conceive of the matter, for according to his representation Herod Antipas knew nothing of the meditated flight.
Thus it is not only a difficulty involved here, objectively, in believing that Machaerus would have been held by Aretas, which thus leads many to conclude that Josephus is in error here. There is further a difficulty, subjectively, for Josephus, in that his own account makes very little sense if Antipas is viewed as sending his wife to a castle being held by her father, since Josephus says that Antipas had “no hint … of her real purpose” to flee. If Josephus believed the castle was held by Aretas, this point that he emphasizes three times (saying also that she asked Antipas to let her go to Machaerus “before any information reached him that she had discovered everything” and “he had no notion that the poor woman saw what was afoot”) makes no sense.
All of these difficulties simply disappear if we allow the text to say what it actually does say in the manuscripts. We are thus in a fairly unusual position here, in that the conjecture not only has no manuscript support but also makes much less sense in context than the original text does. There is only one way out of this strange situation, and that is to disavow the conjecture and let the text stand.
This then restores the translation of Josephus in the Loeb edition to agree with the footnote there and with the Greek text of the manuscripts.
On his return after transacting his business in Rome, his wife, who had got wind of his compact with Herodias, before any information reached him that she had discovered everything, asked him to send her away to Machaerus, which was on the boundary between the territory of Aretas and that of Herod. She gave no hint, however, of her real purpose. Herod let her go, since he had no notion that the poor woman saw what was afoot. Some time earlier she herself had dispatched messengers to Machaerus and to him who was subject to her father, so that when she arrived all preparations for her journey had been made by the officer. She was thus able to start for Arabia as soon as she arrived, being passed from one officer to the next as they provided transport. So she speedily reached her father and told him what Herod planned to do.
The word translated “governor” in the Loeb edition (στρατηγός) can also be translated “magistrate,” “general,” “commander,” “leader,” or “officer.” I have used the word “officer” here.
The original account makes sense, subjectively and objectively, as the text of Josephus. Machaerus is not said to belong to Aretas, as it did not. Machaerus is stated to be near the border of the territory of Aretas, as it was. There was a man who answered to Aretas IV inside Machaerus, who may be considered either a spy or a diplomat. Antipas’ wife sent messengers before her to have preparations for a journey into Arabia and into her father’s territory ready when she arrived. This was necessary so that she “was thus able to start for Arabia as soon as she arrived, being passed from one officer to the next as they provided transport.” Herod Antipas let her go to Machaerus without any suspicion that she knew anything or had any plan to escape. Everything in the passage makes sense if and only if we understand that the castle Machaerus was under the control of Herod Antipas and belonged to his territory.