In his essay "Literature Before Letters" (Nineteenth Century, November, 1899), Friedrich Mueller discusses the interesting question of whether literature could have been in existence before the creation of an alphabet and a writing system. It seems that one's natural first response to such a question would be "Of course not, that's impossible." However, Mueller's discussion of the Indian mnemonic system makes a strong case for a non-written literary tradition.

Mueller first quotes several examples of "literature" passed down through generations by means other than written. He mentions records of American Indian tribes retained mentally by specific tribe members whose role was solely to be a keeper of the trial history. These records were sometimes recalled with the aid of wampum-shell mnemonic systems, which were arranged on strings to suggest ideas to the historian. Mnemonic literature also appears to have been prominent in many islands of the South Pacific. It is interesting to note that such "literature" often served a functional role, and was as acceptable as we would consider a written, legal document today. In the latter case, these remembered records were used to settle "any questions of inheritance, of disputed frontiers, of dynastic succession, &c." (p. 112, Last Essays)

Mueller goes on to discuss the mnemonic system in India, particularly among the caste known as the Brahmans. The Brahmans offer an excellent example of literature existing apart from a written system-- they insisted upon the oral transmission of their sacred "texts". The author notes that while retaining such vast quantities of information mentally may seem a daunting task to many people today, there was a time when such activities were expected and required of all students. He also claims that the lack of evidence of an abrupt appearance of a writing system in either Greece or India suggests that this "mnemonic literature" was gradually replaced by writing. "One can only suppose that people grew accustomed gradually to the idea of writing, of parler aux yeux, by inscriptions on coins, monuments, milestones, and by the marks on cattle, so that its use, it may at first be as a mere help to the memory and afterwards as taking the place of the ancient and sacred mnemonic literature of the country, was not so sudden a shock as it otherwise would have been." (p. 116, Last Essays)

One Indian "text" that Mueller mentions is the Rig-veda-Samhita, which contains around 1,000 hymns, and which (at the time he was writing) was still known by heart by thousands of Brahmans. There are several different mnemonic devices used to help remember such a text-- one example is the Krama, in which a line of independent words would be recited in the following sequence:

This and other much more elaborate systems of recitation were meant to ensure that each word would be remembered both in its modified and unmodified forms.

The author concludes his essay with a lament to the disappearance of such memorizational skills, which he attributes to our dependence on written materials. While one may not agree with his disdain for the written word, it is hard to argue that the elaborate system of memorization and recitation of such ancient texts as the Rig-veda is not in some way a form of literature, to be appreciated as such.