CHAP. VIII. THK RITUALS AXD THE AXNALS. 229 such beliefs we can easily understand that thei-e would be much involuntary error — a result of credulity, of a love for the marvellous, and of faith in the nation- al gods ; but voluntary falsehood is not to be thought of; for that would have been impious; it would have violated the sanctity of the annals, and corrupted the religion. We can believe, therefore, that in these books, if all was not true, there was nothing at least that the priests did not believe. Now, for the his- torian who seeks to pierce the obscurity of those early times, it is a great source of confidence to know that, if he has to deal with errors, he has not to deal with imposture. These errors even, having still the advan- tage of being contemporary with those ancient ages that he is studying, may reveal to him, if not the de- tails of events, at least the sincere convictions of men. These annals, it is true, were kept secret ; neither Herodotus nor Livy read them. But several passages of ancient authors prove that some parts became pub- lic, and that fragments of them came to the knowl- edge of historians. There were, moreover, besides the annals, — these written and authentic documents, — oral traditions, wdiich were perpetuated among the people of a city ; not vague and indifferent traditions, like ours, but tra- ditions dear to the cities, such as did not vary to please the imagination, such as men were not at liberty to modify ; for they formed a part of the wor- ship, and were composed of narrations and songs that were repeated from year to year in the religious festi- vals. These sacred and unchangeable hymns fixed the memory of events, and perpetually revived the tra-« dilions. Doubtless we should be wrong in believing that these traditions had the exactitude of the annals^
