Although set in the Carolingian era, The Song of Roland was written much later. There is a single extant manuscript of the Song of Roland in Old French, held at the Bodleian Library at Oxford. This copy dates between 1129 and 1165 and was written in Anglo-Norman. There are also eight further manuscripts, and three fragments, of other poems on the subject of Roland.
Some scholars estimate that the poem was written, possibly by a poet named Turold (Turoldus in the manuscript itself), between approximately 1040 and 1115, and most of the alterations were performed by about 1098. Some favour an earlier dating, because it allows one to say that the narrative was inspired by the Castilian campaigns of the 1030s, and that the poem went on to be a major influence in the First Crusade. Those who prefer a later dating do so based upon what they interpret as brief references made to events of the First Crusade.
In the text, the term d'oltre mer or l'oltremarin comes up three times in reference to named Muslims who came from oltre mer to fight in Spain and France. Oltre mer, modern French Outremer, literally "oversea, beyond sea, other side of the sea", is a native French term from the classical Latin roots ultra = "beyond" and mare = "sea". The name was commonly used by contemporary chroniclers to refer to the Latin Levant.
The occurrence of this term cannot be interpreted as showing influence from the Crusades; on the contrary, the way it is used, refer simply to a Muslim land, indicates that the author was unacquainted with the Crusades, and that the term was in French before the Crusades was held to refer to the far side of the Mediterranean. The bulk of the poem is adjudged to date from before the Crusades (which started in 1098), but there are a few items where questions remain about these items being late additions shortly after the Crusades started.
After two manuscripts were found in 1832 and 1835, the Song of Roland became recognized as France's national epic when an edition was published in 1837.
Certain lines of the Oxford manuscript end with the letters "AOI". The meaning of this word or annotation is unclear. Many scholars have hypothesized that the marking may have played a role in public performances of the text, such as indicating a place where a jongleur would change the tempo. An alternative hypothesis by Nathan Love is that AOI indicates locations where the scribe or copyist deviated from the primary manuscript.