the Ethiopians at Gebel Barkal (Napata). Among these, the narrative of his conquest of Egypt by Piankhi is one of the most remarkable documents of ancient Egypt (IV, 796-883).

23. The paucity of documents, so painfully evident during the Decadence, is even worse under the Restoration (Twenty-sixth Dynasty, 663-525 B. C.). Besides the great adoption stela of Psamtik I at Thebes (IV, 935 ff.), a few Serapeum stela, important for the chronology, a small number of statue inscriptions of noblemen of the time, and some miscellaneous stela of little importance, we possess almost nothing from the Restoration. Unhappily, the papyri, which are so plentiful during the Nineteenth, Twentieth, and Twenty-first Dynasties, are few and unimportant throughout the remainder of the Decadence and the whole of the Restoration. Fortunately, Herodotus, and the Greek historians after him, enter at this point with invaluable accounts of the history and civilization of the Restoration epoch; but these foreign sources do not fall within the province of these volumes.

24. Besides these contemporary native sources, we possess also a series of later native versions of important events in the history of the nation. These documents are either merely folk-tales, of course differing strikingly in form from the more formal contemporary records; or they are products of the later priesthoods, which, in the form of a tale, give an account of some earlier event, which they so interpret or so distort as to bring reputation, or even material gain, to their sanctuaries. Of the folk-tales we have three of importance: Papyrus Westcar, relating the prodigies attending the birth of the first three Fifth Dynasty kings; Papyrus Sallier I, narrating the cause of the war with the Hyksos; and Papyrus Harris 500, in which is told the story of the capture of

Joppa by one of Thutmose III's generals, named Thutiy. As tales these documents have no place in this series, although each is based on some actual historical incident, which may be obscurely discerned in the narrative. The priestly tales are likewise three in number: the Sehel inscription, recounting the gift of the Dodekaschoinos at the first cataract to Khnum by King Zoser of the Third Dynasty; the Sphinx Stela (II, 810 ff.), recording the accession of Thutmose IV to the kingship, because as prince he cleared the Sphinx of sand; and finally the Bentresh Stela, containing a tale in honor of one of the Theban Khonsus, by showing that he was carried to a distant Asiatic kingdom in order to heal its king's daughter, in the days of Ramses II (III, 429–47). The last two stories seemed of sufficient importance to be included here. It was with tales in common circulation like these that Herodotus' informants regaled him, and the narrative portions of Manetho's history were largely made

up

of just such stories, of which further examples from Ptolemaic times have survived in Demotic dress.

25. It will be seen that the great mass of the documents available are found in Upper Egypt, and but a scanty few in the Delta. This unfortunate fact makes all our knowledge one-sided, and the history of the Delta, the civilization. of which must have risen at a very remote date, remains for the most part unknown to us. Our loss is here like that in Greek history, in which we know almost nothing of the great civilization in the powerful cities of Asia Minor, from which the culture of the early states in Greece drew so much.

26. The documents thus briefly surveyed have reached us, with very few exceptions, in a state of sad mutilation. This mutilation and gradual destruction are a ceaseless. process, which, if not as rapid as formerly, nevertheless. proceeds without cessation at the present day. In Egypt,

the exposed monuments, like the great geographical list of Sheshonk I, are perishing with appalling rapidity, and many of them without ever having been properly copied or published. Even the portable stone monuments at present in the museums of Europe suffer more or less; and I have seen valuable stelæ so attacked by the moist air of northern Europe that whole layers might be blown from the inscribed surface by a whiff of the breath. Such an inscription is doomed to disappear in a few years. Papyri when mounted between hermetically sealed glass plates survive indefinitely. 27. These monuments, as employed in Egyptological science, are, for the most part, not accessible in the originals, but are consulted chiefly in publications. Such publications, to omit earlier and cruder attempts, began as far back as the colossal report issued in huge folios by the members of Napoleon I's expedition. Notable and useful as this great work was, its copies of the inscriptions are now quite unusable. To copy an inscription of any kind with accuracy is not easy. So close and fine an observer of material documents as Ruskin, could copy a short Latin inscription with surprising inaccuracy. In his incomparable Mornings in Florence he reproduces the brief inscription on the marble slab covering the tomb which he so admired in the church of Santa Croce; and in his copy of these eight short lines, which I compared with the original, he misspells one word, and omits two entire words ("et magister") of the medieval Latin.

28. This experience of the great art critic is not infrequently that of the schooled and careful paleographer as well. The best-known of the Politarch inscriptions appeared in eight different publications, each of which diverges in

aThird edition, 1889, 16.

bSee Burton, American Journal of Theology, II, 600-604.

some more or less important respect from all the rest, before a correct copy was obtained. The Greek and Latin inscriptions on the bronze crab from the base of the New York obelisk were long incorrectly read, and the mistake in the date led Mommsen to a false theory of the early Roman prefects of Egypt. In working on a mutilated inscription, the best of copyists will now and again overlook traces which his successors may discover and utilize, while now and then he will "nap," and be guilty of some egregious blunder of omission or misreading in a clear and perfectly preserved passage. Under these circumstances, an inexperienced or careless copyist will commit the most incredible blunders, and every line of his copy will contain many such. In the early days of Egyptology, when a reading knowledge of hieroglyphic was still impossible, it required a copyist of exceptional ability to produce a copy which can be used at the present day.

29. This difficulty was sorely felt by accurate and discerning scholars as far back as the days of Chabas, who in 1872 remarked, concerning the inscriptions of Ramses III: "Ces deux publications [Rosellini and Burton] sont très imparfaites; et les signes inexactement reproduits ajoutent à la difficulté causée par les lacunes." The introduction. of hieroglyphic type, while very useful in some respects, has also proved disastrous to accuracy, and the persistence of the old loose methods was bemoaned by Brugsch in the introduction to the last volume of his Thesaurus thirteen years ago. Brugsch already showed surprising appreciation of the necessity of modern methods in such work. He wrote:

The indispensable demands upon the publisher of known or unknown texts may be comprehended in a few words. In the first place, it is not a task to be undertaken by laymen and mere amateurs . . . . but

*See II, 632, note.

Études sur l'antiquité historique, 227 f.

only by the schooled specialist, who is thoroughly familiar with the language and writing of the ancient Egyptians, and with the researches and results in all departments of Egyptological investigation. How largely such a conviction is still lacking is proved. . . . by a number of publications by Egyptological tyros [Halbwisser] and laymen, who do not yet seem to have learned that Egyptology has ceased to be the pursuit of amateurs, and has become a very serious study, demanding a man's entire strength and entire time.a

C

30. Not long after this, Griffith called attention to the hurried, inaccurate, and insufficient methods still often observable, so that numerous publications could only be regarded as provisional. Two years later he referred to such work in these words: "Too often almost every third sign in the printed texts has had to be corrected according to probabilities by the would-be reader." This condition of things has gone so far that we have had publications issued at government expense, containing texts in vertical columns copied with the lines numbered backward, and even translated in this inverted order of the lines. It is safe to say that such a condition of things cannot be found in any other branch of paleographical science.

31. This is not the place to discuss the proper methods to be observed in the publication of ancient documents, but there is no doubt that better methods are constantly gaining ground. From decade to decade the publication of inscriptions has steadily improved, but it is only within the last ten or fifteen years that Egyptian documents on stone have in some cases appeared in a form which satisfies the demands of modern paleographic accuracy. With the exception of

aThesaurus, VI, vi.

bEgypt Exploration Fund Archæological Report, 1893–94, 10, 11.

cIbid., 1895-96, 21.

dIn the old publications plenty of examples of such inversion exist, especially in Mariette's books; nor are instances lacking in which modern scholars have employed such texts without discovering the inversion.