A comprehensive theory concerning the origin of writing was developed by Denise Schmandt-Besserat, University of Texas.
1.1 Three dimensional clay tokens
Already from the 9th millennium onwards clay tokens (Lat. calculi) where used to depict objects and abstract numbers and was widely spread: from present day Sudan to Iran. The clay tokens in various forms and shapes were used as counters. Each type of counter represent e.g. a bull's head, a sheep, a basket, a bar of gold etc. They were, in many cases at least, pictographicallly used: that is, they depicted concrete objects. They have meaning in any language. Further specification was obtained with diacritical marks, such as scratches and strokes. It was the first steps towards an abstract notation.
In the second half of the 4th millennium clay bullae developed. (see a later example in this external image, or this external description): envelopes of clay used to contain the tokens. The bullae were used in transactions, as trafel documents, waybill, contract of carriage etc. To certify the content cylinder seals where used (see example in this external image, or this external description) that were roled over the wet clay.
It seems that a further step was to imprint the tokens on the outside of the
clay bulla. The content could thus easily be identified. It was the first step
in the development of a writing system. Numbers were given as simple strokes:
one stroke for 'one', two strokes for 'two' etc. Note that the same is true for
our currentthe symbols for the number 1, 2 and 3. The writing of the number 2 is
merely two horizontal strokes, in cursive writing written with 'pen down' in
between. Our symbol for the number 3 developed from three horizontal strokes on
top of each other in cursive writing.
Larger units are obtained as an impression of the round rear side of a reed stylus.
Hold vertically the impression would be
A skew impression would give
Still larger units could be a combination
The actual meaning of these number symbols may depend on the units measured: one, six, ten, sixty, ten times sixty etc. In the famous stratification layer Ur-IVa (around 3200 BC) in the ancient city of Ur also symbols that represent the unit 100 are found. The symbols were collected in sections. The order of the signs within a compartment was not important.
A diacritical mark on a three dimensional token was often not clearly copied on the outside of a clay bulla and had to be inscribed by hand. In the two dimensional writing a symbol like could stand for 'sheep', not a pictogram anymore. Further diacritical marks, like removing a segment could indicate 'ewe' (female sheep) where as removing two segments could be an indication of a sheep in gestation.
At the end of the 4th millennium an explosion of new signs suddenly appeared, most of which have no counterpart as 3-dimensional token, of which there are only a dozen or so. (This, by the way, is also seen as sever criticism of the development from 3-dimensional tokens sketched above). This development took place simultaneously with the building of the first cities, large palaces and temples etc. In general with an economy that was more centralized and where the emphasis was shifted gradually from concentration of villages to larger units.
The writing system was born, basically as pictograms. The signs were
curvilinear in shape. Symbols that never appear in a 3-dimensional form are:
A hand, that showed like
A head looked like
A foot was
The pubic triangle of a women as and stands for 'woman' or 'female' in general.
The word 'day' was symbolized as the sun at the horizon
Since the drawing of curvilinear lines in clay is cumbersome, lines are
imprinted (stamped) with a reed stylus and curvilinear shapes gradually make
place for straight lines, and the shapes simplified. Specific meaning was
obtained with further diacritical marks, such as
and so cuneiform (In English and French, from Lat. cuneus 'wedge shaped', German 'Keilschrift') was born. In the Akkadian writing more than a millennium later one can still recognize (knowing the intermediate steps) the pictographic origin of the signs. The Akkadian logogram (word sign) for 'woman','female' is the sign munus developed from a straight line approximation and rotation over 90 degree from the pubic triangle
The logogram for 'mountain' kur was an actual drawing of a mountain peak. Mountains, however, are not present in Mesopotamia, which is in the alluvial plains around the rivers Eufraat and Tigris. The word 'mountains' (north of Mesopotamia) therefore also symbolizes 'abroad', 'foreign country'. The Sumerian word (and later the Akkadian logogram) for 'female slave' is represented by the composite logogram, in Akkadian:
munus.kur a combination of 'woman' and 'foreign/mountain', thus the word for 'female slave' is 'woman from the mountains/from abroad'
(Note that the English word 'slave' has a similar origin: Slavic people where brought in large numbers as slaves to western Europe in the 10th century, during an intense period of ''Christianization'')
Pictografic writing would have meaning in any language, but the development is usually ascribed to the Sumerians. The Sumerian language has no proven relation with any other language.
Some of the earliest tablets can be seen on the Web, see Catalogue.
3. Cuneiform, the physical appearance of the signs
Cuneiform (from Latin meaning 'wedge-shaped') is composed of a series of
short straight wedge-shaped strokes made with a stylus into a tablet of soft
clay. The strokes are thickest at the top, like ,
on harder material it more looks like .
At first, symbols were written from top to bottom; later, they were turned onto
their sides and written from left to right. In later periods harder materials
were also used.
Five basic orientations are applied: horizontal, two diagonals, a hook and a vertical stroke:
The up-diagonal stroke has limited use. These five components occur in two different sizes. A small hook often not being distinguishable from a short diagonal. The two diagonal strokes are not used as an individual sign, but the other types are. Reverse orientations (e.g. with the head of the wedge at the bottom) are hardly attested. The inverse vertical is rare on old tablets (sometimes the vertical in the sign for 'hand', which normally is . Already from the Old Babylonian period onwards this orientation is not seen anymore. A stroke from right to left does not occur at all, a characteristic on the basis of which one may recognize simple falsifications.
The signs underwent a significant evolution in the course of time (e.g. Old Babylonian in the 18th century BC, versus New Babylonian almost a millennium later), slightly different in the various dialects. Present day sign lists are sorted to New Assyrian shapes. The field of study is called Assyriology and originated in the previous century with the discovery of the Assyrian Library in Nineve. New Assyrian shapes are more quadratic. They void diagonal and hook shaped components as compared with the contemporary Babylonian scripts. E.g. the sign ni is in Assyrian form and in Babylonian form .
Students usually start learning cuneiform signs in New Assyrian orthography. For the grammar Old Babylonian is studied as a starting point. For educational purposes I have included an Old Babylonian text (the Codex Hammurabi) in New Assyrian transliteration, combining Old Babylonian grammar with easier-to-learn signs.
The phonetic values are listed in the sign lists. Some of the signs have identical values throughout the Mesopotamian history and all of the dialects. Others are limited to the first half of the second milllenium, designated as "old" in the sign list, and mainly in use in the Old Babylonian Period. Sign values in use after this period are designated as "new" in the lists. An indication of geographical differences in the use of the signs are in these lists restricted to Assyrian (Ass.) and Babylonian (Bab.).
Since the use of accents is cumbersome in the html-language of this document (and because accents are difficult to read at some servers) I will often use ni2 for ní etc. and ni3 for nì
Grammatical forms of a word (e.g. the case for a nomen or the conjugation for a verb) are not expressed in a logogram. A logogram could stand for a nomen or adjective, but the same logogram could also be used as a verb, e.g. in the present tense or the past conjugation.
Akkadian example: É + phonetic complement
Akkadian example: AN + phonetic complement
The sign AN as logogram could mean the god Anum and Šamû 'sky', 'heaven'. But with phonetic complement ú