Latin Carolingian Reforms — ~782 AD, Germany

Latin “m”

Emperor Charlemagne apparently tried to learn how to read and write, but with poor success.  Probably part of his difficulty was that he had to spend a bunch of time conquering countries, part of the difficulty was that he started learning later in life, but probably part of the problem was that it was very difficult.  There were a bunch of different variations on Latin script that were used, the letters were all crammed together, you couldn’t always tell where word boundaries were, it was even harder to tell where sentence boundaries were, and even abbreviations were different in different “dialects” of the script.

Being the emperor, Charlemagne could do something about it.  He called Alcuin, a hotshot scholar, teacher, and cleric of the day to come to his court and um, do something about it.  Alcuin looked at the different dialects of Latin script that were in use, and came up with a standardized version of the script.

They also instituted looser spacing: letters were to have one space between them, words two spaces, and sentences three.  (Obviously these spaces were smaller than the space characters we use.)

(Alcuin’s team popularized spaces between words, but they did not invent it.  The Lycians sometimes used spaces between words, but the Europeans forgot about the idea for a millenium.  Irish monks apparently came up with the idea independently, about a century before the Carolingian reforms, and this time it might have been done out of necessity.  While the Italian or French monks’ native language was a direct descendant of Latin, the Irish monks’ native language (Gaelic) was a much more distant cousin and thus much harder to for them to read.  Breaking the words up with spaces was probably a coping mechanism for the Gaelic monks, and probably helped Charlemagne a bit.)

In addition to spaces to divide letters, words, and sentences, the Carolingian reforms encouraged even more visual cues for sentence beginnings and endings.  Sentences were to start with a capital letter and punctuation was to show where the reader was supposed to stop or pause.  This was the first time that capital letters (majuscule) and lower case (minuscule) were mixed together.  Before, they were just thought of as essentially different fonts.  There are only a handful of other writing systems that have both upper and lower cases.

With Charlemagne’s backing, Alcuin’s own scholarly reputation, and a hefty dose of money (in the form of hiring lots of scribes to make lots of copies of lots and lots of manuscripts written with Alcuin’s specifications), Alcuin was able to promote the script effectively and succeed in getting people to standardize on one script: the so-called “Carolingian minuscule”.  People also started spacing letters, words, and sentences, and using punctuation.

In my opinion, this was a very big deal.

Links: Wikipedia

Posted in Alphabet, government-mandated, Rating: 5 "Whoa!!" | Leave a comment

Ogham – 300 AD, Ireland

Ogham "z"

Ogham is a runic script mostly used in Ireland, but to a lesser extent in the northern island of Britain.  While the earliest provable use dates from the 4th century AD, there are linguistic clues that it is older: there are two characters H and Z which are in the script but which have not been found in early inscriptions.  You might ask how they could tell that the characters were part of the original script if there weren’t examples in the early inscriptions, and the answer lies in Ogham’s regularity.

Ogham is written with a centerline running in the direction of the characters, and strokes crossing the centerline in various patterns.  There were  originally four patterns: to the left or above, to the right/below, diagonally across, and straight across.  For each pattern, there were characters for one, two, three, four, and five strokes.  (NB: five other characters were added later.)  The “H” was made by putting one stroke to the left or up; the “Z” was made by putting four strokes diagonally across.  If the “H” and “Z” were not in the original set, that would upset the tidy symmetry of the alphabet.  (“H” and “Z” were seen in later inscriptions.)

There are arguments about where Ogham came from.  I prefer the theory that it derived from Latin, since there were Romans running around Britain from 43 AD to 420 AD.  The alphabetical ordering is different from almost all ther alphabets, but that might be due to the groupings of the characters.  The vowels are all in the “straight-across” set of glyphs, and I can imagine that when the Ogham creators put the vowels into their own set, they rearranged everything.

As to why they needed a new script when Latin was there for the taking, the suggestions that I have seen talk about a desire for a script the Romans wouldn’t understand and/or Latin not suiting the language well.  However, there are enough bilingual Latin/Ogham inscriptions that it doesn’t seem like they tried very hard to keep it secret, so I have problems with the former.  There aren’t any characters in the original set which are not in Latin, so I have problems with the latter.  It seems to me that the answer might lie with the technology of the writing system.  Latin has lots of curves, which are hard to carve in wood, and harder still to carve on sticks.  Ogham is really easy to carve onto sticks, which Ireland had in abundance.

Sticks from 300 AD Ireland unfortunately have not survived.  What has survived are carvings on rock tablets.  Interestingly, the early writing on stone was not right-to-left or left-to-right, but rather around the edges of the stone.  The edge of the stone was used as the centerline of the script.

Interestingly, the names of trees are used for the names of the characters.  Legend has it that all the names came from local trees, but some names don’t correspond to any trees known, and some came from trees that are not found in Ireland.

Links: Wikipedia, Ancient Scripts, Omniglot

Posted in Alphabet, probably first in its area, Rating: 3 "I did not know that" | Leave a comment

Old Hungarian — 600? AD, Hungary

Hungarian Runes "e"

Hungary, despite being solidly in Europe, has had a long history of trade with and conquest by Central Asian peoples.  It is not entirely clear where Hungarians came from — or more specifically, where the people who brought the Hungarian language came from — but the Hungarian language has a number of Turkic loan words.

Hungarians also appeared to be influenced by Orkhon (“Turkic runes”).  An ancient writing system called either “Old Hungarian”, “Hungarian Runes”, or “Hungarian rovas” (from the Hungarian word for “to carve”) appeared at some poorly-understood point, probably around 600.  All the glyphs in Orkhon that represent similar sounds in Hungarian are similar in Orkhon.  For non-Turkic sounds, the Old Hungarian glyphs appear to be novel, or perhaps from Greek.  However, there are factions which contend that Old Hungarian came from Greek and others that contend that Orkhon derived from Old Hungarian.

There are three dialects of the Hungarian Runes: two in Hungary and one in Kazakhstan.  The existence of the Kazakh variant is another strong argument for the origin of the script to lie in Central Asia.

This script was quite successfully stamped out in Hungary (which perhaps accounts for why the origins of the Hungarian people are somewhat unclear?) in favour of Latin, starting around 1000 AD by Christian leaders.  (The Christian leaders associated the Old Hungarian with paganism.)  The eradication was not perfect, however, as some remote regions continued to use runes until the mid 1800s.  It remained popular perhaps in part because the Hungarian runes actually represent the Hungarian sounds better than Latin script does.  (Some people say that literacy in Hungary dropped with the coming of the Latin script.)  Old Hungarian has enjoyed a little bit of a revival in modern times.

Old Hungarian was written right to left, or occasionally boustrophedonically. While Old Hungarian is an alphabet, the writers would occasionally drop vowels if doing so did not make the word ambiguous.

Old Hungarian script used three vertical dots to denotes word boundaries.

Links: Wikipedia, Omniglot, Ancient Scripts, Upon Reflection, Hungarian Heritage, Origins of Hunnish Writing claims that Turkic runes descended from Hungarian runes, RovasPedia




Posted in Alphabet, now ceremonial, probably first in its area, Rating: 3 "I did not know that" | 2 Comments

Orkhon — ~700 AD, Mongolia

Orkhon "m"

Orkhon is also called Old Turkic or Göktürk script.  It was used mostly in Mongolia and Western China, but there are dialects that were used in Siberia (Yenisei) and Kazakstan.

Orkhon is sometimes called Turkic Runes because of  their angular shape, and there are enough similarities to Futhark that there are a number of people who argue passionately that Orkhon is the ancestor of Futhark.  While I acknowledge that there is a familial relationship, I tend to agree with the faction which says that the relationship is that they are both distant descendants of Phoenician, and that Orkhon appears to be a descendant of a non-cursive Sogdian.

Futhark is definitely an alphabet, like the writing systems neighbouring Scandinavia. Orkhon is sort of like an abjad, like many of the writing systems in the vicinity of Mongolia.  This to me is the most persuasive argument.

The earliest Orkhon artifacts that are known date from the early 700s, and the earliest Yenisei from the 800s, while the earliest Futhark artifacts date from 160 AD.  This is not difinitive proof that Orkhon did not influence Futhark: earlier Orkhon artifacts might have been written on perishable objects (which perished).  The strong dialectical differences between Yenisei and Orkhon (Mongolian) suggest to me that Yenisei and Orkhon diverged at an earlier point than the early 700s.  However, this would also argue that Futhark and Orkhon would have needed to diverge at an even earlier point than 160 AD, which I have a harder time believing.  If the Scandinavians traveled widely enough in 100 BC to meet the Mongloians and take the script, why didn’t they meet up with the Latins first and take the Latin script?

Orkhon is unusual in two ways:

  • Orkhon is the only writing system I know of (yet) which is written right-to-left in rows, but writing the rows from bottom-to-top.  (The only other script I know of that goes bottom-to-top at all is Tifinagh, which sometimes was written vertically in columns from bottom-to-top.)
  • Turkic languages have vowel harmony, which means that words only have either front vowels or back vowels.  While vowels were frequently not written (like abjads), the consonants were written differently depending upon whether the next vowel was a front vowel or back vowel.  Thus it has more vocalic information than a strict abjad, but less information than a strict alphabet.



Links: Wikipedia, Ancient Scripts, Omniglot

Posted in Abjad, Rating: 4 "Huh, interesting!" | 2 Comments

Younger Futhark — 800 AD, Scandinavia

Younger Futhark character for Golden Year 18.

Elder Futhark evolved into Younger Futhark, with the transition happening between 650 AD and 800 AD.  Younger Futhark was most different from Elder Futhark in the number of characters: Younger Futhark had only two-thirds as many letters as Elder Futhark.  You might think that this was because the language simplified, and there weren’t as many sounds in Norse during the Younger Futhark period.  You might be wrong, however.  Linguists believe that the number of sounds in the spoken language actually increased over the transition period.

Archeologists also believe that literacy significantly increased during this period.  Maybe the new learners were lazy; maybe as time went on, people discovered that you could still read things even without as many characters as the language needed.  I don’t know.

One common use of this writing system was to keep track of the Christian calendar, which uses the 19-year Metonic cycle of “golden years”.  They used one Younger Futhark character for each of the years, which meant that they were three characters short.  To handle that, they invented three characters which were only used to represent those three years’ numbers.

Futhark was written in all kinds of ways.  In addition to left-to-right, right-to-left, and boustrophedon, the writing also frequently followed the outlines of the object that it was inscribed upon.

Younger Futhark had two styles: the long-branch (or Danish) and short-twig (or Swedish/Norwegian) versions.  Like the names suggest, the short-twig version had shorter and/or more abbreviated strokes.

A variant, used in a area of Sweden between the 10th and 12th century, called Hälsinge Runes took this to extremes.  In these runes, the vertical bar was completely removed from most glyphs, leaving very short lines to represent the runes.  It might be that Hälsinge Runes were the very first shorthand.

Links: Wikipedia (Younger Futhark), Ancient Scripts, Omniglot, Wikipedia (Hälsinge Runes)

Posted in Alphabet, Evolved slowly from parent, Rating: 4 "Huh, interesting!" | Leave a comment

Elder Futhark — 160 AD, Denmark

Futhark "d"

Elder Futhark, an early runic script, was definitely used in 160AD in Denmark.  Based on linguistic clues, some people think that it is much older.

One of the clues cited is that it is written both left-to-right and right-to-left, like Greek or Etruscan before 200 BC.   Another clue is that there are characters for sounds which linguists believe had faded from the spoken language by 160 AD.

Futhark is an unusual alphabet in that it does not use either of the alphabetization orders that come from Ugaritic, unlike almost all other alphabets. It takes its name from the first six letters: f, u, th, a, r, and k. I think its unusual ordering argues for a single person developing the script –perhaps a trader who wandered down to Greece and brought the idea of writing back  — instead of it evolving gradually among a group of people.  (Perhaps Bob Futhark brought the script to Scandinavia, and he thought it would be fun to spell out his name in the first few letters.)  It is also possible that my imagined trader simply never learned the standard ordering, so came up with his own.

You might think that the ordering just wasn’t important to them, and the f-u-th-a-r-k is just based on perhaps one person who happened to write them down in that order.  However, the ordering was perhaps more important in Futhark than most writing systems: in an alternate glyph scheme, characters were written based on their place in the ordering.  Characters were grouped into three sets of eight and could  be represented as two sets of horizontal lines on either side of a vertical line.  The number of lines on the left represented which set of three the letter was in; the number of lines showed which number in the set of eight it was. (You could imagine this as an early octal representation.)

Futhark had a very angular look to it, which probably came from the common writing technology of that place and time: scratching or carving into hard objects like wood, bone, and stone.  Straight lines are easier to make than curved when carving.

Links: WikipediaAncient Scripts, Ominglot,

Posted in Alphabet, first in its area, Rating: 4 "Huh, interesting!", technology influenced | 3 Comments

Gyaru-moji — 2000 AD?, Japan

a Gyaru-moji "ki"

Gyaru-moji is sort of like a Japanese Leet: a variant orthography for Japanese.  Unlike Leet, which was developed in the predominantly male hacker culture, Gyaru-moju (which means “girl characters”) appears to have been developed by schoolgirls.  In both cases, by excluding those not “in the know”, use of the variant orthography serves to strengthen intimacy among the in-group.

Unlike the relatively small number (around one hundred) of printable symbols that computers usually allow in Latin text, Japanese textual communication has tens of thousands of glyphs to draw upon.  For a long time, Japanese users have been able to type Katakana, Hiragana, Kanji (“Chinese characters”), Latin, Greek, Cyrillic, and a number of symbols — even on cell phones.

It is difficult to describe rules for Gyaru-moji; it seems that one of the rules is “there are no rules”.  I will make an attempt, but understand that reality is more complicated than what I describe here.

Glyphs in one writing system are sometimes substituted for similar-looking glyphs in a different writing system, e.g. Katakana チ becoming Kanji 干, or Hiragana す becoming the section symbol §.  Characters with two disconnected sub-elements sometimes turn into two distinct characters, e.g. 私 becoming 禾+ム. (Note that Japanese writing is almost always monospaced — each character is centered in a square with the same amount of horizontal and vertical room.  私 takes one square; 禾ム takes two squares.)

Like Leet, Gyaru-moji uses unusual grammatical structures, but unlike Leet, those structures are not usually incorrect (by the standards of the dominant dialect), merely uncommon.  Like Leet, there is not a one-to-one relation between a standard character and one Gyaru-moji character: there are many possible variations of each character.

Unlike Leet, Gyaru-moji does not appear to use the rebus principle (as Leet uses in “gr8” and “&hill”).  However, they do sometimes transcribe characters phonetically in Latin, then alter the Latin characters.  Thus も, the character representing the syllable “mo”, can become “мσ”.   (I do not know if they ever would use the Leetish “AA()”.)

Posted in Logograms, private or secret, Rating: 5 "Whoa!!", significant female influence, Syllabaries | Leave a comment

Leet — ~1980 AD, USA

a Leet "B"

Leet, also known as “1337” is a writing system developed for the English language which gains some of its value in being difficult to understand — but not too difficult.  It is in some respects a code-substitution cipher, where glyphs — especially numbers — are used in place of letters which look similar.  For example, “3” is frequently used instead of “E”, “7” in place of “T”, etc.  (This makes Thaana‘s use of numerals from other writing systems look somewhat less odd.)

Combinations of glyphs are also substituted for single Latin characters sometimes, e.g. “|<” for “K” or “/\/\” for “M”.  Unusually among writing systems, there is a great abundance of “legal” ways of writing a character.  Just a few of the ways to write”M” include “|\/|". "44", "IYI", "AA".

Leet is also somewhat logographic, using rebus principles.  For example, “banned” can be written “b&”; “great” can be written “gr8”.

There are number of other “misspellings” and non-standard grammatical forms used to transcribing standard English to Leet in the service of making the communications less understandable.

Note: I put “USA” down for the country of origin in the title, but that is undoubtedly an over-generalization.  Leet developed in the BBS culture, a majority of which probably originated in the US, but Leet is an international ad-hoc collaborative evolutionary endeavour.

Links: Wikipedia

Posted in Alphabet, private or secret, Rating: 3 "I did not know that" | 1 Comment

Gond — 2010 AD, India

Gond "kha"

It is not very common for someone to create a new script.  Cherokee, Ol Chiki, Pin Cin Hau logograms, Gurmukhi, Hangul are just a few of the scripts which we know were created or invented more-or-less from scratch.

However, in all the cases that I have written of up until now, the scripts were invented by men*.  This makes it all the more surprising that a female professor of English in India, Dr. Prasanna Sree, has developed not one, but eighteen writing systems for minority languages in India, including another script for the Gondi language.  I do not know why Dr. Sree felt that the 1928 Gondi script was inadequate.

I can’t tell how widely adopted Dr. Sree’s writing systems have been (if at all).  This raises a question that I have touched on briefly before: how well accepted does a writing system need to be before it is a legitimate writing system and not a piece of art?  Omniglot’s page on constructed scripts has 175 entries, many of which were designed more-or-less for fun.    (For example, see Thoorsha.)  Should these be considered artworks more than writing systems?

*Yes, Empress Wu Zetian did introduce some characters into Chinese.  However, it appears clear that someone else suggested the characters and she merely approved them.

Links: Omniglot, Unicode proposal

Posted in Abugida, inventor known, Rating: 2 "Not all that interesting", significant female influence | 1 Comment

Tolong Siki — 1999 AD, India

Tolong Siki "cha"

Tolong Siki was developed rather recently for the Kurukh spoken language.  Previously, Devanagari was used (and is still used in large part a decade later).

Tolong Siki is one of the few languages that was created from scratch collaboratively that we know of.  For most writing systems, their origins are lost in antiquity.  Many clearly evolved rather than “were created”.  For recently-created writing systems, when we know who the creator was, there was almost always single creator.  Tolong Siki is an exception: while Narayan Oraon (a medical doctor) gets most of the credit, he collaborated with Francis Ekka (the former director of the Central Institute of Indian Languages), Ramdayal Munda (former Vice Chancellor of Ranchi University), and Nirmal Minz.

Links: Unicode proposal, Kurukh World

Posted in Abugida, National pride, now ceremonial, Rating: 3 "I did not know that" | Leave a comment