Mnemonic writing, as it is generally treated in the textbooks, includes all sorts of simple memory aids, and is generally, and probably rightly, regarded by writers of palaeography as preceding picture writing, although there is an element of abstractness even in the tally or knotted cord or pebble as compared with the actual imitation or representation of the picture, and in the evolution of human thinking, other things being equal, the abstract necessarily follows the concrete in time and in the order of evolution.

The most familiar examples of mnemonic books are the quipus or knotted cord books, the notch books, which include tallies and message sticks, the wampum belts of American Indians, and the abacus. Collections of any of these kept in the medicine tent or temple, or even the counting house, are, of course, true libraries, or at least true collections of written documents as generally understood by the historians of writing.

The knotted cord is best known under the name of quipu, which was the name for the Peruvian knot record. At bottom the idea does not differ from the simple tying of knots in a handkerchief as a reminder, or the sailor’s log line. It has been most commonly used for numerical records, but in many cases it preserved and transmitted very extensive historical records. One very simple use was the noting on different colored cords by knots the number of the different animals taken to market for sale, and again the price received for these at market.

It is still used among the Indians of Peru and some North American Indians, also in Hawaii and among various African tribes, and all over Eastern Asia and the Pacific.

It was the traditional method in China before the use of written characters, and the written characters themselves were, it is alleged, made up out of these combined with the pictures of bird tracks.

Among the ancient civilizations there are many remains or reminiscences of these knot books. They are found among the ancient Egyptian hieroglyphics (as in the sign for amulet and perhaps in several other signs); they appear also in the mnemonic knotted fringes to garments in the Jewish antiquities and, as Herodotus tells us, Darius made use of such knots to guide certain Ionians who remained behind to guard a bridge as to when it should be time for them to sail away. In 1680 the Pueblo Indians of North America marked the days to their uprising in the same way.

This use of the knotted cord for amulets is among the most widespread of uses, being found among the medicine men of nearly all primitive peoples. Juno wore such an amulet, and Ulysses carried one.

Among the ancient Peruvians and Mexicans there were many collections of quipus in charge of official recorders.

Traces of ancient use survive in the knots of a cardinal’s hat and perhaps most interestingly of all in the nautical knot used in casting the log or sounding. We may still travel so many knots an hour or sink mayhap so many fathoms deep. The knotted measuring line with fathom marks is probably the direct historical descendant of the Egyptian measuring line and by the same token probably of the Egyptian sign for one hundred, the fathom like one of the Egyptian units being at bottom the stretch of a man’s arm.

Most of the extant quipus have been found in graves. There is a “very extensive collection” of these in the American Museum of Natural History in New York, and a recent study of these (by L. L. Locke) concludes that they were used purely for numerical purposes and not for counting but for record keeping.

The best known notch books are the message sticks used in Australia and Africa and the tally used in the British Exchequer up to a recent date for the keeping of accounts. This is the method, famous in fiction for the recording on their knife-hilts by Indians and superhuman white scouts of the number of scalps taken in war. It is the essence of the so-called Clog Almanac, the nick-stick, and other ways of notching up accounts still often found in rural communities. The memory of it survives in the use of the word score or so many tallies, used until recently of the runs made in baseball.

Collections of notch records are found at least among the Australian aborigines—and it will be remembered that it was the burning of the huge collection of tallies in the early part of the last century which resulted in the setting fire to and burning up of the parliament houses.

It is possible that the notch method was preceded by a system of stripping off leaves or twigs from a branch, leaving a certain number. The early pictures of Seshait, goddess of writing among the Egyptians, who records the years of a king’s reign, suggests possibly this method, and in this case perhaps also the Egyptian sign for year with its single projection may refer to this method.

Wampum is one of the best known and most picturesque forms of mnemonic object writing. It was used by the American Indians for treaties, title deeds, memorials of events, etc., and considerable collections of these tribal records were not uncommon. Although in itself a later and more complex style, in essence it stands for a style still older than the knot writing which it resembles. Existing examples of wampum leave the simple mnemonic knot or notch far behind and have progressed even to figures or pictures often of an advanced or symbolic type, made in the beads, but the beads themselves stand for what may perhaps be the very earliest form of mnemonic record—that is the object record where each object is represented not by a pictorial object but by some sample object like a pebble or a twig. The heap of pebbles used for counting was possibly the very earliest mnemonic record.

An extremely interesting modern example of calculation in pebbles and the representation by them even of sums in addition, multiplication, and subtraction, turns up among the psychological investigations in the matter of mathematical prodigies. It appears that most of the famous lightning calculators have been the children of peasants, and a large part of these Italian shepherd boys, who apparently used pebbles for the counting of their sheep and amused themselves by making a plaything of these. Other lightning calculators (AmpÈre e.g.) used pebbles, and Bidder a bag of shot, while others have taught themselves by the use of marbles, peas, or the use of their fingers. (Bruce in McClure v. 39, 1912, pp. 593-4.) The counting by pebble heaps is found indeed generally in the playing of children. When it comes to transporting or making more permanent collections this was done by means of a pouch in the case of pebbles—one of the earliest forms of record holder and one of the most ancient forms even of phonetic writings, or tying together in bundles as in the case of twig bundles found among primitive peoples, or by stringing together as in trophy necklaces or some forms of the abacus.

With these mnemonic object writings is perhaps also to be classed the symbols formed with bits of wood used in the Indian game of caÑute described by J. P. Harrington. “The San Ildefonso caÑute figures present a symbolism so highly conventionalized and so complex that the term language might well be applied—a symbolism not essentially different in origin or practice from human speech, gesture language, African drum language, conventionalized graphic designs that have a commonly understood meaning, or writing whether executed in pictograms, ideograms, phonograms, or phonetic symbols” (AmAnthropol n.s. 14, 1912, p. 265). “These figures are, it is said, made much in the same fashion as children graphically represent certain ideas by arranging small objects.”