The Arabs, commanding most of the centres of ancient learning (Ptolemy's own Alexandria above all), riveted the pseudo-science of their predecessors on the learned world, along with the genuine knowledge which they handed down from the Greeks. In many details they corrected and amplified the Greek results. But most of their geographical theories were mere reproductions of Ptolemy's, and to his mistakes they added wilder though less important confusions or inventions of their own. The result of all this, by the tenth century a.d., was a geography, based not on knowledge, but on ideas of symmetry. It was a scheme fit for the Arabian Nights.
And how did Ptolemy lend himself to this?
His chief mistakes were only two;—but they were mistakes from which at any rate Strabo and most of the Greek geographers are free. He made the Indian Ocean an inland sea, and he filled up the Southern Hemisphere with Africa, or the unknown Antarctic land in which he extended Africa.[8] The Dark Continent, in his map, ran out on the one side to the south-east of China, and on the other to the indefinite west, though there was here no hint of America or an Atlantic continent. It was a triumph of learned imagination over humdrum research. Science under Hadrian was ambitious to have its world settled and known; it was not yet settled or fully known; and so a great student constructed a mélange of fact and fancy mainly based on a guess-work of imaginary astronomical reckonings. On the far east, Ptolemy joined China and Africa; and on this imaginary western coast, fronting Malacca and Further India, he placed various gratuitous towns and rivers. Coming to smaller matters, he cut away the whole of the Indian peninsula proper, though preserving the Further or "Golden" Chersonesus of the Malays, and he enlarged Taprobane, or Ceylon, to double the size of Asia Minor. Thus the southern coast of Asia from Arabia to the Ganges ran almost due east, with a strait of sea coming through the modern Carnatic, between the continent and the Great Spice Island, which included most of the Deccan. The Persian Gulf, much greater on this map than the Black Sea, was made equal in length and breadth; the shape of the Caspian was, so to say, turned inside out and its length given as from east to west, instead of from north to south; while the coast line, even of the familiar Euxine, Ægean, and Southern Mediterranean, was anything but true. Scandinavia was an island smaller than Ireland; Scotland represented a great eastern bend of Britain, with the Shetlands and Färoes (Thule) lying a short distance to the north, but on the left-hand side of the great island. The Sea of Azov, hardly inferior to the Euxine, stretched north half way across Russia. All Central Africa and the great Southern or Antarctic continent was described as pathless desert—"a land uninhabitable from the heat"; and the sources of the Nile were accounted for by the marshes and Mountains of the Moon.
Thus all the problems of ancient geography were explained: where Ptolemy's knowledge failed him altogether, no Western of that time had ever been, or was likely to go. The whole realised and unrealised world was described with such clearness and consistency, men thought, that what was lacking in Aristotle was now supplied.
Yet it is worth while observing how, centuries before Ptolemy, in the ages nearer to Aristotle himself, the geography of Eratosthenes and Strabo, by a more balanced use of knowledge and by a greater restraint of fancy, had composed a far more reliable chart.[9]
This earlier and discredited map avoided all the more serious perversions of Ptolemy. Africa was cut off at the limit of actual knowledge, about Cape Non on the west and Cape Guardafui on the east; and the "Cinnamon-bearing Coast," between these points, was fringed by the Mountains of Æthiopia, where the Nile rose. This was the theory which revived on the decline of the Ptolemaic, and which encouraged the Portuguese sailors with hopes of a quick approach to India round Africa, as the great eastern bend of the Guinea coast seemed to suggest. Further, on this pre-Ptolemaic map the Southern Ocean was left untouched by a supposed Southern Continent, and except for an undue shrinkage of the Old World in general as an island in the midst of the vast surrounding ocean, a reliable description of Western Asia and Central Europe and North Africa was in the hands of the learned world two hundred years before Christ.
It is true that Strabo's China is cramped and cut short; that his Ceylon (Taprobane) is even larger than Ptolemy's; that Ireland (Ierne) appears to the north of Britain; and that the Caspian joins the North Sea by a long and narrow channel; but the true shape of India, of the Persian Gulf and the Euxine, of the Sea of Azov and the Mediterranean, is marked rightly enough in general outline. This earlier chart has not the elaborate completeness of Ptolemy's, but it is free from his enormous errors, and it has all the advantage of science, however imperfect, over brilliant guessing.
Of course, even in Ptolemy, this guess-work pure and simple only comes in at intervals and does not so much affect the central and, for his day, far more important tracts of the Old World, but we have yet to see how, in the mediæval period and under Arabic imagination, all geography seemed likely to become an exercise of fancy.
The chief Greek descriptions of the world, we must clearly remember, were before the mediæval workers, Christian and Moslem, from the first; these men took their choice, and the point is that they, and specially the Arabs, chose with rare exceptions the last of these, the Ptolemaic system, because it was the more ambitious, symmetrical, and pretty.
Let us trace for a moment the gradual development of this geographical mythology.