The Assyrians derived their historical writing, as they did so many other cultural elements, from the Babylonians. In that country, there had existed from the earliest times two types of historical inscriptions. The more common form developed from the desire of the kings to commemorate, not their deeds in war, but their building operations, and more especially the buildings erected in honor of the gods. Now and then we have an incidental reference to military activities, but rarely indeed do we find a document devoted primarily to the narration of warlike deeds. Side by side with these building inscriptions were to be found dry lists of kings, sometimes with the length of their reigns, but, save for an occasional legend, there seem to have been no detailed histories. It was from the former type that the earliest Assyrian inscriptions were derived. In actual fact, we have no right to call them historical in any sense of the word, even though they are our only sources for the few facts we know about this early period. A typical inscription of this type will have the form "Irishum the vice gerent of the god Ashur, the son of Ilushuma the vice gerent of the god Ashur, unto the god Ashur, his Lord, for his own life and for the life of his son has dedicated". Thus there was as yet little difference in form from their Babylonian models and the historical data were of the slightest. This type persisted until the latest days of the Assyrian empire in the inscriptions placed on the bricks, or, in slightly more developed form, in the inscriptions written on the slabs of stone used for the adornment of palace or temple. For these later periods, they rarely have a value other than for the architectural history, and so demand no further study in this place. Nevertheless, the architectural origin of the historical inscription should not be forgotten. Even to the end, it is a rare document which does not have as its conclusion a more or less full account of the building operations carried on by the monarch who erected it.
It was not long until the inscriptions were incised on limestone. These slabs, giving more surface for the writing, easily induced the addition of other data, including naturally some account of the monarch's exploits in war. The typical inscription of this type, take, for example that of Adad nirari I, [Footnote: BM. 90,978; IV. R. 44 f.; G. Smith, Assyr. Discoveries, 1875, 242 ff.; Pognon, JA. 1884, 293 ff.; Peiser, KB. I. 4 ff.; Budge-King, 4 ff.; duplicate Scheil, RT. XV. 138 ff.; Jastrow, ZA. X. 35 ff.; AJSL. XII 143 ff.] has a brief titulary, then a slightly longer sketch of the campaigns, but the greater portion by far is devoted to the narration of his buildings. This type also continued until the latest days of the empire, and, like the former, is of no value where we have the fuller documents.
When the German excavations were begun at Ashur, the earliest capital of the Assyrian empire, it was hoped that the scanty data with which we were forced to content ourselves in writing the early history would soon be much amplified. In part, our expectations have been gratified. We now know the names of many new rulers and the number of new inscriptions has been enormously increased. But not a single annals inscription from this earlier period has been discovered, and it is now becoming clear that such documents are not to be expected. Only the so-called "Display" inscriptions, and those with the scantiest content, have been found, and it is not probable that any will be hereafter discovered.
It was not until the end of the fourteenth century B. C. with the reign of Arik den ilu, that we have the appearance of actual annalistic inscriptions. That we are at the very beginning of annalistic writing is clear, even from the fragmentary remains. The work is in annals form, in so far as the events of the various years are separated by lines, but it is hardly more than a list of places captured and of booty taken, strung together by a few formulae. [Footnote: Scheil, OLZ. VII. 216. Now in the Morgan collection, Johns, Cuneiform Inscriptions, 33.]
With this one exception, we do not have a strictly historical document nor do we have any source problem worthy of our study until the time of Tiglath Pileser I, about 1100 B.C. To be sure, we have a good plenty of inscriptions before this time, [Footnote: L. Messerschmidt, Keilschrifttexte aus Assur. I. Berlin 1911; Mittheilungen der Deutschen Orient Gesellschaft; cf, D. D. Luckenbill, AJSL. XXVIII. 153 ff.] and the problems they present are serious enough, but they are not of the sort that can be solved by source study. Accordingly, we shall begin our detailed study with the inscriptions from this reign. Then, after a gap in our knowledge, caused by the temporary decline of Assyrian power, we shall take up the many problems presented by the numerous inscriptions of Ashur nasir apal (885-860 B.C.) and of his son Shalmaneser III (860-825 B.C.). In the case of the latter, especially, we shall see how a proper evaluation of the documents secures a proper appreciation of the events in the reign. With these we shall discuss their less important successors until the downfall of the dynasty. The revival of Assyrian power under Tiglath Pileser IV (745-728 B.C.) means a revival of history writing and our problems begin again. The Sargonidae, the most important of the various Assyrian dynasties, comprising Sargon (722-705 B.C.), Sennacherib (705-686 B.C.), Esarhaddon (686-668 B.C.), and Ashur bani apal (668-626 B.C.), furnish us a most embarrassing wealth of historical material, while the problems, especially as to priority of date and as to consequent authority, become most complicated.
Before taking up a more detailed study of these questions, it is necessary to secure a general view of the situation we must face. The types of inscriptions, especially in the later days of the empire, are numerous. In addition to the brick and slab inscriptions, rarely of value in this later period, we have numerous examples on a larger scale of the so called "Display" inscriptions. They are usually on slabs of stone and are intended for architectural adornment. In some cases, we have clay tablets with the original drafts prepared for the workmen. Still others are on clay prisms or cylinders. These latter do not differ in form from many actual annals, but this likeness in form should not blind us to the fact that their text is radically different in character.
All the display inscriptions are primarily of architectural character, whether intended to face the walls of the palace or to be deposited as a sort of corner stone under the gates or at the corners of the wall. We should not expect their value to be high, and indeed they are of but little worth when the corresponding annals on which they are based has been preserved. For example, we have four different recensions of a very long display inscription, as well as literally scores of minor ones, also of a display character, from the later years of Sargon. The minor inscriptions are merely more or less full abstracts of the greater and offer absolutely nothing new. The long display inscription might be equally well disregarded, had not the edition of the annals on which it is based come down to us in fragmentary condition. We may thus use the Display inscription to fill gaps in the Annals, but it has not the slightest authority when it disagrees with its original.
It is true that for many reigns, even at a fairly late date, the display inscriptions are of great value. For the very important reign of Adad nirari (812-785 B.C.), it is our only recourse as the annals which we may postulate for such a period of development are totally lost. The deliberate destruction of the greater portion of the annals of Tiglath Pileser IV forces us to study the display documents in greater detail and the loss of all but a fragment of the annals of Esarhaddon makes for this period, too, a fuller discussion of the display inscriptions than would be otherwise necessary. In addition, we may note that there are a few inscriptions from other reigns, for example, the Nimrud inscription of Sargon, which are seemingly based on an earlier edition of the annals than that which has come down to us and which therefore do give us a few new facts.
Since, then, it is necessary at times to use these display inscriptions, we must frankly recognize their inferior value. We must realize that their main purpose was not to give a connected history of the reign, but simply to list the various conquests for the greater glory of the monarch. Equally serious is it that they rarely have a chronological order. Instead, the survey generally follows a geographical sweep from east to west. That they are to be used with caution is obvious.
Much more fortunate is our position when we have to deal with the annalistic inscriptions. We have here a regular chronology, and if errors, intentional or otherwise, can sometimes be found, the relative chronology at least is generally correct. The narrative is fuller and interesting details not found in other sources are often given. But it would be a great mistake to assume that the annals are always trustworthy. Earlier historians have too generally accepted their statements unless they had definite proof of inaccuracy. In the last few years, there has been discovered a mass of new material which we may use for the criticism of the Sargonide documents. Most valuable are the letters, sometimes from the king himself, more often from others to the monarch. Some are from the generals in the field, others from the governors in the provinces, still others from palace officials. All are of course absolutely authentic documents, and the light they throw upon the annals is interesting. To these we may add the prayers at the oracle of the sun god, coming from the reigns of Esarhaddon and Ashur bani apal, and they show us the break up of the empire as we never should have suspected from the grandiloquent accounts of the monarchs themselves. Even the business documents occasionally yield us a slight help toward criticism. Add to this the references in foreign sources such as Hebrew or Babylonian, and we hardly need internal study to convince us that the annals are far from reliable.