To me, I confess, it appears that Homer has nowhere done more, and rarely so much, in a single passage, as in this, with a view of declaring his intention. The island Ææa, irrespective of all geographical argument, is, as we have seen, directly bound and fastened to an eastern site by four separate cords. First, as the rising point of the Sun. Secondly, as the residence of Dawn. Thirdly, because Circe, its mistress, has the Sun, the most eastern of all mythological conceptions except the Dawn, for her father. Fourthly, because she has also Perse, whose name indicates a trans-Phœnician origin, for her mother. And further, I am convinced we cannot alter the place of Ææa without uprooting the whole Phœnician scheme of the Outer Geography.

The scope and range thus given to the adventures of Ulysses confines them without doubt to the northern semi-circle, but allows them to reach, within that semi-circle, to its eastern and to its western extremities, as they are imagined by the Poet. Æolus and the Læstrygonians are evidently placed by him in the north-west. The hypothesis, which has here been maintained for Ææa and Calypso, supplies an effectual counterpart, and properly fills up the eastern corner. But, independently of all other objections, the north-western hypothesis for these islands jumbles them, if I may so speak, in one heap with the others, and leaves the eastern quarter towards the North wholly unoccupied. And yet that East was, for a Greek, the source and the scene of the richest legendary and mythological representations. Such an incongruous view of the question would not, I think, be at all in keeping with Homer’s ordinary modes of conceiving, handling, and presenting his materials.

Construction of Od. v. 276, 7.

But I am aware that, up to this time, we have left out of view a passage, of which I freely admit that the prevailing, and in so far the most obvious, interpretation is against me. Ulysses sails over the sea from Ogygia, governing the rudder of his raft with art, and watching the stars, especially the Great Bear; which at that period, I believe, was nearer the Pole, and was a more conspicuous and splendid astronomical object, than it now is. It was with respect to this constellation that he had received a particular order from Calypso[623]:

τὴν γὰρ δή μιν ἄνωγε Καλυψὼ, δῖα θεάων,

ποντοπορευέμεναι ἐπ’ ἀριστερὰ χειρὸς ἔχοντα.

Or, according to the common construction of the words, he was to keep that constellation on the left during his voyage. But if his course lay in the direction of a right line drawn from St. Petersburgh to Corfu, it appears that Arctus, when visible to him, would be visible on the right, and not on the left.

I could not, however, accommodate myself to this passage at such a cost as that of oversetting an interpretation of the general scheme, which is so deeply rooted both in the letter and spirit of the poem, as is the eastern, and likewise somewhat north-eastern, hypothesis for Ææa, together with a northern site for Ogygia. These two, it may be observed, stand together. It is plain, from the times occupied by the several stages between Ææa and Ogygia, and from the language used where no precise time is stated, that the Poet conceived the distance between them to be limited, though very considerable. And indeed the north-western hypothesis for Ææa would do nothing for the passage I have quoted, unless we also carry Ogygia into the north-west, in order that Ulysses, on his way home from it, may have Arctus on his left. Inasmuch, however, as the admission of the received sense for the lines would involve us in a new series of the most complicated and hopeless contradictions, we must look for relief in some other direction.

On the genuineness of the passage.

I desire to eschew, as a general rule, the dangerous and seductive practice of questioning the genuineness of the text because it seems to stand in conflict with a favoured interpretation. I may however state, without unduly relying on them, one or two particulars which, drawn from the poem itself, may show that these two lines are not unjustly open to the suspicion of interpolation.