(1) The first point to be considered is that verse 36 follows naturally after verse 33. As we have seen, in verses 30-3 we have an angelic announcement to Mary to the effect that she is to give birth to a son who is destined to become the Messiah. He will be called “the Son of the Most High”, and to him the Lord God will give “the throne of his father David”. To this message, it may be said, verses 36 and 37 form a fitting sequel. They add the assurance that “no word from God shall be void of power”, in proof of which it is declared that Mary's kinswoman, Elisabeth, is shortly to bear a son in her old age. The whole speech (Lk. i. 30-3, 36, 37) is a consistent passage, and in relation to it the words of Mary in verse 38—“Behold, the handmaid of the Lord; be it unto me according to thy word”—are a natural and fitting reply. Canon Box, in the article already cited, questions this view. “There would be nothing extraordinary”, he says “in Mary's conceiving a son as Joseph's wife”—nothing, that is to say, to require the sign offered. But surely it is not a question of “conceiving a son”, but of conceiving such a son, the long-promised Messiah; and, moreover, the ratification of so great a promise by means of a miracle is a commonplace of OT. thought. It is not suggested, of course, that this argument proves Lk. i. 34 f. to be an interpolation. That a section runs smoothly when a particular passage within it is excised, is no proof that that passage is not original. This last conclusion must be established on the ground of other arguments. If, however, in the present instance, other arguments carry weight, then the fact that verse 36 can be connected easily and naturally with verse 33 becomes of very great importance, and it is for that reason that we introduce it here.
(2) We take a really decisive step when we instance what already has been found, namely that verse 34 follows quite unnaturally upon Lk. i. 30-3. We have seen that Mary's question implies the announcement of an immediate conception, and we have failed to find any such announcement in the angel's words. There is thus a complete difference of point of view in the two passages. No possible ground is provided in the [pg 042] angelic announcement for the objection raised in verse 34. It is difficult, therefore, to deny the suggestion that Mary's question already implies a knowledge of what is told for the first time in verse 35. But this view, if we accept it, is to say that Mary's question could not possibly have been present to the mind of Mary in the connexion in which it stands; it was the last question she would have thought of asking. The question can only have been put into her mouth by one who already knew of the Virgin Birth, and wished to introduce that doctrine into a context in which originally it did not appear. On the interpretation which we have given to Lk. i. 30-3 and Lk. i. 34 f., this conclusion is inevitable, unless we prefer to find in St. Luke an utter inconsecutiveness of thought which does him no credit as writer, and which neither of his works justify us in attributing to him.
(3) We are unable to attach the same force to the contention that verse 35 is followed unnaturally by verses 36 and 37 (so Schmiedel), though this view has something to be said for it. Verse 35 announces the virgin birth of the promised Messiah, a doctrine which is not found in Jewish literature and tradition, and for which, therefore, the mind of Mary must have been utterly unprepared.[38] As the section now stands, the statement of verses 36 and 37 is added as a sign that what has just been promised will surely come to pass. This sign, we have already argued, would be quite natural, according to OT. modes of thought, as authenticating such a message as that given in Lk. i. 30-3. But can we say this in reference to the promise of a virgin birth? To the modern mind at least the argument seems faulty and unconvincing. Mary is bidden to accept as the divine promise what is so remarkable as to be otherwise unknown to her, on the ground of what is certainly remarkable but familiar to her mind and outlook. In truth, this seems a remarkable argument with which to credit an angel! At the same time, it has to be admitted that such an objection may be too stringent, and that it may not allow sufficiently for ancient modes of thought, according to which the argument from the less to the greater is by no means uncommon. For [pg 043] this reason the present writer would not feel confident in pressing the argument sketched above.[39]
(4) A much stronger argument calls attention to the similarity between Mary's question and that of Zacharias (Lk. i. 18), and the difference with which they are treated by the angel. “Mary's speech expresses doubt of the truth of the angel's message, and yet she is not so much as blamed, whilst Zacharias is actually punished for a like doubt (i. 20)”.[40] The presumption is that the two cases do not emanate from the same cycle of tradition. The force of this argument depends, of course, upon the way in which we interpret Lk. i. 34. It is true that we have no indication of the tone in which the question is asked, beyond the words themselves and the sentences which follow; but quite sufficient is given to indicate the presence of doubt. The point is not merely one of subjective valuation. This will appear if we consider Plummer's view, which is quite different from Schmiedel's. “She does not ask for proof, as Zacharias did (ver. 18); and only in the form of the words does she ask as to the mode of accomplishment. Her utterance is little more than an involuntary expression of amazement.... It is clear that she does not doubt the fact promised, nor for a moment suppose that her child is to be the child of Joseph” (op. cit., p. 24). In weighing this opinion, it should be noticed that it refers only to the words, Πῶς ἔσται τοῦτο? We may readily agree that if all that Mary had said were, “How shall this be?”, we should be unable to contest this view. But to divide Mary's question in this way is not permissible. The second part, “seeing I know not a man”, clearly determines the first, and debars us from viewing it as merely “an involuntary expression of amazement”. The presence of doubt, we think, must be conceded, though it is less marked than in the case of Zacharias.
This view, moreover, is supported by the fact that, in the narrative as it stands, an explanation follows, which is also confirmed by a sign. Since, as Plummer says, Mary, unlike Zacharias, does not ask for proof, we need not object that she is not “punished.” And it is just possible that we make the parallelism too rigid if we lay stress on the fact that “she is not so much as blamed”. It is rather the “eulogium” of Lk. i. 45 (“Blessed is she that believed”) which presents the difficulty. It is true that, in the narrative as we have it now, Mary believes ultimately (verse 38), but Lk. i. 45 seems rather to belong to a narrative in which Mary believes from the first. We conclude that the present argument gives real support to the view that Lk. i. 34 f. belongs to a source distinct from its context.
(5) A fifth argument dwells on the different senses in which Divine Sonship is predicated of the promised child in verse 32 as compared with verse 35. As we have seen, the term υἱὸς ᾿Υψίστου in verse 32 is purely Messianic. But in verse 35 the expression υἱὸς θεοῦ must be given a very different meaning. It is in consequence of (διὸ καί) the divine overshadowing that the child is to be called “Son of God”. Here, to quote Dr. J. Estlin Carpenter, the term denotes “not official adoption, but actual origin”, and, with the same writer, we must conclude that verse 35 “is thus a doublet of verses 31, 32 on another plane” (op. cit., p. 487).[41] It is more difficult to decide whether the difference supports the theory of interpolation. We cannot shut out the possibility that two diverse types of Sonship might have been attributed by St. Luke to the same speaker at the same time of writing. But, having said this, we may observe that it is certainly much easier to suppose, and is much more probable, that they belong to different periods of reflection, and are the product (or deposit) of different traditions. This argument, then, may be said to lean in the direction of the theory of interpolation, but, for the reason given above, we should hesitate to urge it, if it stood alone.
(6) We have lastly to look at the vexed question of the Davidic descent. It is safe to say that, if we had not Lk. i. 34 f. in the Gospel as it stands to-day, we should have no ground for [pg 045] regarding Mary as of Davidic descent. It is the presence of these verses that makes possible that inference in verse 32, where, in addressing Mary, the angel speaks of David as the forefather of the promised child. It is surely a remarkable fact that a point so vital to St. Luke's narrative as the Davidic descent of Mary should be introduced in so incidental a manner. Our wonder is increased when we observe that St. Luke is at great pains to assure Theophilus of the Davidic descent of Joseph. In ii. 4 it is said that Joseph was “of the house and family of David”; not a word is said of Mary's descent. It is true that the Sin. Syr. reads, “because they were both of the house of David”; but this does not naturally fit into the structure of the sentence, is unsupported elsewhere, and is accepted by no one; it clearly represents an attempt to remove a difficulty. In. i. 27 it is also said that Joseph was “of the house of David”. The phrase cannot be construed with the word “virgin”, which occurs earlier in the sentence, in view of the fact that after ἐξ οἴκου Δανείδ St. Luke resumes the thread of the story by saying “and the virgin's name was Mary”; otherwise, he would have continued (so Schmiedel, op. cit., col. 2957), “and her name was Mary”. It is not easy indeed to resist Schmiedel's further contention that the phrasing of the sentence expressly forbids our ascribing the Davidic descent to Mary, though the opinion is put forward with greater confidence than seems justified. The one passage in which St. Luke directly refers to the family of Mary is dubious. In i. 36 Elisabeth is said to be “the kinswoman” of Mary, and we know from i. 5 that Elisabeth was “of the daughters of Aaron”, which seems to imply that Mary too was of Levitical descent. But as the precise nature of the relationship is not stated, we cannot say, with Schmiedel, Usener, and others, that this is so. Nevertheless, the broad fact remains that apart from an inference, which itself depends on Lk. i. 34 f., we have no grounds for believing Mary to be a descendant of David. St. Luke undoubtedly believes Jesus to be of Davidic descent; he carefully shows Joseph to be of that descent; he gives us no reason to suppose, that, like the author of the First Gospel, he traced the descent of Jesus through Joseph as His legal father; and yet, in spite of all this, he has left the vital question of the Davidic descent of Mary at the mercy of an inference! If he knows Mary to be [pg 046] a descendant of David, why does he not say so explicitly? We have a right to ask the question, which is neither captious nor unfair. No one has yet answered it satisfactorily, except in the answer that St. Luke had no tradition of the Davidic descent of Mary at his disposal, that he traced the descent of Jesus through Joseph as His real father, that this is the true interpretation of verse 32, and that Lk. i. 34 f. is a later insertion, which has imposed on verse 32 a sense which originally it did not bear. Regard Lk. i. 34 f. as a later insertion, and all the facts alleged by St. Luke about the Davidic descent fall into intelligible order; refuse to do this, and they remain in inexplicable confusion.
When we consider the cumulative force of the preceding arguments, it becomes impossible for us to think that Lk. i. 34 f. was written at the same time, and from the same point of view, as the context in which it now stands; it is clearly a later insertion. With some reason we may hesitate to say that verse 36 does not follow naturally upon verse 35, and we may speculate whether two diverse conceptions of Sonship may not be held in the same mind at the same time of writing. But when we ponder the question of the Davidic descent; when we compare verse 34 with Lk. i. 18 ff.; when we observe the natural coherence of Lk. i. 30-3 and Lk. i. 36-8, and the radical difference in point of view between verses 34, 35 and the angelic announcement; when, in short, we have a narrative, which, if Lk. i. 34 f. was present from the first, ought to be dominated by those verses, but on the contrary does not seem to be influenced by them; we are compelled to conclude that the suspected verses represent a later insertion in the Gospel.