Invented Scripts

Writing Systems of the World

Course 1 · Chapter 8 · Invented Scripts

Most scripts grew slowly through borrowing and change. A few were designed by a person on a particular occasion. This chapter looks at four of them: the Korean Hangul, the Canadian syllabics, and two scripts built for fiction, Tengwar and Klingon pIqaD. Cherokee (Chapter 4) is the other famous example.

The idea in a box
Wikipedia's Writing system article says a featural system "uses symbols representing sub-phonetic elements", with Hangul as the main example. If a designer builds the shapes of letters to show how sounds are made, the script can encode phonetic theory as well as sound.

Hangul

From Wikipedia's Hangul article (first 100,000 of 212,572 characters read).

  • Creation. King Sejong the Great (r. 1418 to 1450) was responsible for its creation. It was announced around late 1443 to early 1444 and officially published in 1446 in the Hunminjeongeum, which was also the script's original name.
  • Inventory. Mainstream orthographies use 24 basic letters (jamo): 14 consonants and 10 vowels. Modifications and combinations bring the total to 51.
  • Featural design. According to the Hunminjeongeum Haerye, the shapes of five basic consonants are based on human speech organs, and the other consonants are derived by adding strokes for "harsher" sounds. Vowels are linked to Chinese philosophy: ㆍ represents yin and earth, ㅡ yang and the sun, and ㅣ man (neutral).
  • The dispute. The "featural script" label was coined by Sampson in 1985, questioned by DeFrancis in 1989 and rebutted by Chin-Woo Kim in 1997. Pae, Winskel and Kim argue the designation is difficult to falsify and uncertain.
  • Syllable blocks. Letters are grouped into blocks of an initial consonant, a vowel and an optional final consonant, with a silent ㅇ when no initial sound is needed. Vowels with a long vertical stroke go to the right of the initial; those with a long horizontal stroke go below it; a final consonant goes at the bottom.
  • Count. 19 initial consonants, 21 vowels and 28 final consonants (including none) yield 11,172 syllables, though scholars estimate only about 2,000 see regular use.
  • Literacy. The article says hanja's difficulty contributed to high illiteracy, and that within a decade of the post-1945 push, a significant majority of Koreans in both Koreas became literate in at least Hangul.

So Hangul is a script designed to look like an alphabet and write like a syllabary: letters are phonemic, but they are stacked into syllable blocks. For learning it, the site has the Korean Alphabet 1 course.

Canadian Aboriginal Syllabics

From Wikipedia's Canadian Aboriginal syllabics article (first 100,000 of 198,188 characters read).

  • Creator. James Evans, a missionary, developed the script by 1840, probably with Indigenous language experts, working with Cree and Ojibwe speakers. Cree oral traditions give a different origin, crediting a Cree man named Calling Badger.
  • Principle. Each consonant has a fixed base shape, and the vowel is shown by rotating that shape. The consonant p is a chevron that points up, down, left or right to mark different vowels.
  • Classification. Despite its name, the article says it is an abugida rather than a true syllabary.
  • Use. All Cree languages and Naskapi, regionally Ojibwe and Blackfoot, and Inuktitut, which is co-official with Latin script in Nunavut.
  • Reception. Resisted by colonial and missionary authorities (the Hudson's Bay Company declined to import a printing press for Evans), but the Cree community quickly took to it; a Bible was published in Cree syllabics by 1861, and the article says Cree literacy rates were possibly among the highest in the world by the late 19th century.

Tengwar

From Wikipedia's Tengwar article (first 100,000 of 119,846 characters read).

  • Creator. J. R. R. Tolkien; probably developed in the late 1920s or early 1930s, with the first published sample in 1937. In his fiction the elf Fëanor invented it.
  • Featural design. "The shapes of the letters correspond to the distinctive features of the sounds they represent." Most letters combine a stem with one or two bows: doubling the bow voices a stop, raising the stem makes a fricative, and shortening it makes a nasal.
  • Grid. Four series (témar) by place of articulation: dental, labial, velar and labialized velar; six rows of manners of articulation.
  • Modes. The orthographies are called modes. Vowel-diacritic modes can be treated as abugidas, while the "full writing" mode writes vowels as letters; other modes exist for English, Spanish, German and more.
  • Encoding. Not officially encoded in Unicode; Michael Everson proposed it in 1997; the unofficial ConScript Unicode Registry maps it to U+E000 to U+E07F in the Private Use Area.

Klingon pIqaD

From Wikipedia's Klingon scripts article (whole answer as returned).

  • Origin. The Astra Image Corporation designed the letters for Star Trek: The Motion Picture, often wrongly credited to Michael Okuda. They were based on Klingon battlecruiser hull markings by Matt Jeffries and on Tibetan writing, with sharp forms evoking bladed weapons.
  • Size. 26 letters with a one-to-one correspondence between letters and sounds, plus ten digits.
  • In practice. The vast majority of Klingonists prefer Latin-based romanisation. The script occupies U+F8D0 to U+F8FF in the Private Use Area; Everson's 1997 proposal to encode it was rejected by the Unicode Technical Committee in May 2001, reasoning that the script saw almost no communicative use.

The site's Klingon courses use the romanisation, which fits the article's description of common practice.

ScriptDesignerDateTypeDesigned to
HangulKing Sejongpublished 1446Alphabet in syllable blocks; featural (disputed)Let ordinary Koreans read and write
Canadian syllabicsJames Evansby 1840AbugidaWrite Cree and Ojibwe
TengwarJ. R. R. Tolkienlate 1920s or early 1930sFeatural; modes vary (some abugida-like)Fiction (and later fans)
pIqaDAstra Image Corporationfor the 1979 film (date from my own memory, not the article)Alphabet-like, one letter per soundFilm props

The "Designed to" column is my own summary.

Seeing the Design in Python

My own example; I ran it and copied the output. It splits a Hangul syllable into its letters and reverses the arithmetic by which Unicode lays out all the syllables, then looks at Canadian syllabics and the Private Use Area.

import unicodedata s = "한" d = unicodedata.normalize("NFD", s) print(s, len(s), len(d), [unicodedata.name(c) for c in d]) n = ord(s) - 0xAC00 print(n // (21 * 28), (n % (21 * 28)) // 28, n % 28) print(0xD7A3 - 0xAC00 + 1, 19 * 21 * 28) for cp in [0x1431, 0x1432, 0x1433, 0x1434]: print(chr(cp), unicodedata.name(chr(cp))) print(unicodedata.category(chr(0xE000)), unicodedata.category(chr(0xF8D0)))
한 1 3 ['HANGUL CHOSEONG HIEUH', 'HANGUL JUNGSEONG A', 'HANGUL JONGSEONG NIEUN'] 18 0 4 11172 11172 ᐱ CANADIAN SYLLABICS PI ᐲ CANADIAN SYLLABICS PII ᐳ CANADIAN SYLLABICS PO ᐴ CANADIAN SYLLABICS POO Co Co

What the output shows (my reading): 한 is one precomposed code point that splits into three letters (initial ㅎ, vowel ㅏ, final ㄴ), and the block U+AC00 to U+D7A3 holds exactly 19 x 21 x 28 = 11,172 syllables, so the position of a syllable in Unicode is computed from its letters, which confirms the article's syllable count. The syllabics example shows the PI/PII/PO/POO series sitting side by side, consistent with one base shape in different orientations. The category Co means "private use": Tengwar and pIqaD have no standard characters, so a font must supply them.

Watch out: designed does not mean arbitrary
A designed script can borrow from existing ones (pIqaD from Tibetan writing, per the article) and works only if a community adopts it. Cree took to syllabics quickly; pIqaD saw almost no communicative use, according to the Unicode committee's reasoning in the article. And the featural claim for Hangul is debated by scholars, so treat sweeping praise for it with care.
Note on sources
This chapter uses Wikipedia's Hangul, Canadian Aboriginal syllabics, Tengwar and Klingon scripts articles (the first three read to 100,000 characters, the Klingon scripts one as returned). Dates and counts are the articles' and unchecked against primary scholarship. The Klingon article did not date the film or link the design to a later film, so the table notes the 1979 date as my own memory; the Hangul article's text I read did not say who coined the name "Hangul", so I give none. The comparison table's last column, the Python and its reading are mine.

Try It Yourself

Exercise 1

Use the article's figures to explain why only about 2,000 of Hangul's 11,172 possible syllables are in regular use, and why 11,172 is still an easy number for a computer to cover.

📄 View solution
Exercise 2

In Python, decompose the syllable 글 (as in 한글) into its letters with NFD normalisation. Name each letter and say which position (initial, vowel, final) it occupies.

📄 View solution
Exercise 3

Compare Canadian syllabics and Tengwar: how does each use the shape of a symbol to show something systematic? Which is closer to the "featural" idea from Chapter 1, and why?

📄 View solution

Chapter 8 Quick Reference

  • Hangul: King Sejong, published 1446; 24 basic letters (14 consonants, 10 vowels); syllable blocks; 11,172 possible syllables; featural status disputed
  • Canadian syllabics: James Evans, by 1840; vowels shown by rotating the consonant; classed as an abugida
  • Tengwar: Tolkien; letter shapes correspond to sound features; unofficial Private Use Area encoding
  • pIqaD: designed for the film; 26 letters; most Klingonists use Latin romanisation
Coming next
Writing Systems of the World 9 turns from the scripts to their rules: direction, spelling and orthography.