The First Scripts

Chapter 1 listed Mesopotamia and Egypt among the places where writing was invented. This chapter looks at both systems, cuneiform and Egyptian hieroglyphs, because they show the same problem being solved in two ways: a picture can stand for a thing, but only a sound can stand for a name, a grammatical ending or an idea with no picture.

Cuneiform

Facts below are from Wikipedia's Cuneiform article, of which I read the first part.

• Origin. It grew out of pictographic proto-writing in the late 4th millennium BC, drawing on a Near Eastern clay-token accounting system. Writing is first recorded in Uruk, at the end of the 4th millennium BC. The article says the meaning and usage of the tokens is still a matter of debate.
• From pictures to wedges. Early signs were drawn with a sharpened reed stylus and lacked wedge shapes. In the mid-3rd millennium BC a wedge-tipped stylus pressed into clay produced the strokes that give the script its name.
• Shrinking inventory. Around 2800 BC syllabic use began, and the sign inventory fell from some 1,500 signs to some 600.
• Medium. Signs were mainly made on clay tablets. Most surviving tablets were baked by accidental fires.
• Logo-syllabic. Signs could stand for whole words (logograms) or for syllables, and many served both roles. Determinatives, such as signs marking gods, countries or cities, guided the reader to a word's category.
• Many languages. It began with Sumerian and was adapted for Akkadian, Elamite, Hittite and others. Old Persian used a simplified, separate set of signs.
• Decipherment. Old Persian was the first deciphered, beginning with Georg Friedrich Grotefend in 1802. Henry Rawlinson and Edward Hincks worked on the trilingual Behistun inscription, and the article says cuneiform was successfully deciphered by 1857.
• The end. The last known cuneiform inscription, an astronomical text, was written in 79/80 AD.

Egyptian Hieroglyphs

Facts below are from Wikipedia's Egyptian hieroglyphs article, of which I read the first part.

• Origin. They grew out of proto-literate symbol systems in the Early Bronze Age, around the 33rd century BC. The article dates the first decipherable sentence in the Egyptian language to the 28th century BC.
• Size. More than 1,000 distinct characters overall. About 900 were in use in the Middle Kingdom, and by the Greco-Roman period the known hieroglyphs exceeded 5,000.
• Three kinds of sign. Phonetic signs (24 single-consonant signs, plus signs for two or three consonants), logograms (the sun sign for "sun"), and determinatives (silent signs that clarify a word's meaning, since many words are homophones).
• Vowels. Vowels were normally not written, so some scholars call the script an abjad. The article says it is debated whether vowels were written at all.
• Direction. Usually read right to left, though left to right and alternating lines also appear; figures facing left or right show which way to read.
• Everyday forms. Hieratic and demotic are simplified cursive scripts developed from hieroglyphs and better suited to papyrus; hieroglyphs stayed in formal and monumental use.
• The end. The latest known inscription is the Graffito of Esmet-Akhom at Philae, dated 394 AD.

The Rebus Principle

The hieroglyphs article explains the key trick with an example: an eye picture could represent the word "eye" or the sound of the pronoun "I". Using a picture for its sound, ignoring its meaning, is the rebus principle. Once a script has it, any word can be written, including names and grammar. Here is the same idea in English, made up by me for illustration:

Comparison

Both Scripts in Unicode

Both scripts have Unicode blocks (the cuneiform block is U+12000 to U+123FF, the hieroglyph block U+13000 to U+1342F, per the articles). The example is my own; I ran it and copied the output.

Each sign takes 4 bytes in UTF-8 because both blocks are beyond the Basic Multilingual Plane. The hieroglyph names are catalogue numbers (A001, D010), a sign list used by Egyptologists; the cuneiform names use the sign's conventional reading (A, AN). That naming difference is my observation from the output.

Try It Yourself