F.W. Goudy has a nice bit in one of his books where it is shown how the lowercase _g_ developed from the uppercase --- I also have a photocopy of page from a calligraphy text where it advocates for using the older script forms for headings, more recent for subheads, and then setting body text in current/recent styles, using age as a guide to hierarchy, which I've done for a couple of projects and it can have a nice effect.
behnamoh · 7h ago
Why do Ç, å, é, etc. exist? Why didn't people come up with new characters? Why use "sh" and "ch" instead of making a new character for those sounds? (Maybe other languages do this but English doesn't).
tokai · 1m ago
Å is a new character. It is not an a with an accent. It was created to replace the non letter aa.
aarroyoc · 1h ago
In Spanish, we consider Ñ to be a different letter, with its key in keyboards and under a different chapter in dictionaries. Of course you could say that it's just an N with a ~. But that tilde is not defined for any other letter. Why did they do it? In medieval Spanish it was common to write two Ns in many words, with a different sound. For example: "anno", which is now: "año" (year). Two Ns was very similar to M so the tilde was added to remark that it was two Ns, not one M. Later, they just wrote one N with the tilde.
bloak · 3h ago
It seems a bit of an arbitrary choice. For example, Polish has digraphs, but Czech has diacritics, and Icelandic has a couple of additional letters that aren't modified Latin letters.
Old English had four letters that are not in today's US-ASCII, two of which are borrowed from a runic alphabet rather than created by modifying Latin letters.
It's also a bit arbitrary whether a modified Latin letter is regarded as a new character or an existing character modified by a diacritic: take Ø, for example. And there are characters like Æ. And never forget what Turkish did: add dotless i so that ordinary i could then be regarded as dotless i plus a diacritic (though of course it isn't usually regarded that way).
guappa · 2h ago
Consider that runic letters come from latin letters normally.
DuMOHsmol · 54m ago
Slavic languages originally used the very unique Glagolitic script, developed by saints Cyril and Methodius. However, their disciples later created an alternative Greek-based script, which eventually prevailed.
My guess is that the educated people of the time were very familiar with Greek, so it was easier for them to work with Greek letters rather than the newly invented ones. It probably was the same for Latin-based scripts as well.
jhanschoo · 2h ago
- Historically in writing, many accents come from letters written above or below, or there was an existing use of the same letter for distinct sounds, so it was natural to mark existing letters rather than create new ones out of whole cloth.
- The use of digraphs is frequently due to historical sound changes; first, certain sound combinations changed valence, but others do not. Then one rewrites with the digraphs with the other instances by analogy. Sometimes it comes from borrowing; for example the historical use of "ch" to denote the historic greek "chi", or aspirated chi sound, as in "chronograph"
- Finally, it is easier to modify a existing type, or arrange movable type, to include an accent rather than to create one from scratch.
HelloNurse · 2h ago
Altering letters with combining marks (often derived from other letters and not invented from scratch) is not only easier to learn than complete novel letters, but also consistent enough to allow learning by analogy: dieresis, grave accent, acute accent, tilde, macron and the like affect different phonemes in similar ways.
nlitened · 3h ago
Written characters and sounds don't usually correspond because pronunciation drifts very quickly in time (even two consecutive generations might have noticeably different pronunciation), and in different directions in different locations, while rules of "proper writing" change rarely, usually as a result of people agreeing to change them in order to catch up to an apparent accumulated drift of pronunciation.
As I understand (in a simplified way), every once in a while (once in a few centuries) there's a major language reform to make written form reflect current pronunciation as much as possible, but already in 100—200 years people would start asking "why do we write not like we pronounce", until some political movement decides reforming orthography would promote their agenda. Then the cycle continues.
kergonath · 1h ago
> Written characters and sounds don't usually correspond because pronunciation drifts very quickly in time
In space as well. Local accents can be quite variable. This alone dooms any “we should write words as we pronounce them” to fail unless every region has its own orthographic rules, which would be honestly terrible.
create-username · 3h ago
People must come up with new characters because they lack their own script. You can either have a script adapted to your language, or make the best of what you’ve got.
English doesn’t. Bernard Shaw tried inventing new letters. I guess that changing the English alphabet os a slippery slope. If you make it as phonetic as the Latin script is meant to be, and with special characters, people would have to relearn how to read from scratch
ryao · 3h ago
If people followed the alphabetic principle for English, the written language would become unintelligible due to all of the regional variations in pronunciation causing there to be many variations of the same words.
guappa · 2h ago
you mean like there is color and colour?
pessimizer · 1h ago
English has between 16 and 22 vowels (iirc) depending on the variety. English speakers that use a different set of vowels than each other often cannot understand each other at all at speed.
Color and colour are different pictures of the same word.
yorwba · 5h ago
The timeline starts with 22 characters for Capitalis Monumentalis and ends with 26 characters in upper and lower case each for Sans-Serif. People did come up with new characters (JKUW), you're simply used to them.
adrian_b · 2h ago
There has never been a 22-character Latin alphabet.
The original Latin alphabet had 21 letters. Then G has replaced Z, without changing the number of letters. Then, by the time of the Empire, Y and Z have been added at the end (the same Z that had been removed earlier, while Y had the same origin as V, but by this time it had acquired a distinct pronunciation in Greek).
Then the Latin alphabet had 23 letters for more than 1000 years.
It has grown to 26 letters during the Middle Ages, with the addition of J, U and W ("u" was originally the small letter form of "V"; distinct letters U and V have been created by making a capital "U" and a small "v").
gwd · 3h ago
In the case of K, U, and W, those were borrowed from Greek, weren't they?
Not sure where J came from.
One question is why we borrowed K when we already had C. In modern English, C is more or less a completely superfluous letter, adding unnecessary complexity to pronunciation. Seeing my son try to pronounce "cycle" is one of many examples.
ryao · 3h ago
The Latin alphabet was based on the Etruscan alphabet, which was based on the Greek alphabet, which was based on the Phoenician alphabet. In early Latin, C, G and K were all pronounced the same way. They later dropped G, only to bring it back when they realized that they used C to refer to two different consonants, and wanted to disambiguate them. They assigned the less common constant to G.
J was introduced because I had a similar problem, except I could be either a consonant or a vowel, rather than any of two consonants. The same applied to the introduction of U for V. Well, almost the same. In the case of I, they added J to be the consonant, while in the case of V, they added U to be the vowel.
Finally, people changed the pronunciation of V over time to be something new, and W was introduced to be the original sound in English. W does not exist in Latin since V fulfills its role, unless you are using a modern pronunciation that changes how V is pronounced and then you still do not need W. There were other sound changes (see Italian), but this is sufficient to explain all of the letters. Well, there is also Z, which they removed from early Latin and later reintroduced to represent a Greek sound.
adrian_b · 2h ago
All the Latin letters, except for G, which is a modified Greek letter invented by a Roman, come from the Greek alphabet, only at different times.
The original Latin alphabet had 21 letters:
ABCDE FZHIK LMNOP QRSTV X
The letter V (Greek u-psilon) was used to write the sound U, both as vowel and as consonant (i.e. like English W).
The letter F (Greek di-gamma) was used in Greek for consonant U (English W), but in Latin it was used to write the sound F, also a labial consonant, which did not exist in Greek.
In Latin, in the beginning 3 different letters were used to write the sound K, C before E or I, K before A and Q before O or U. So K belonged to the Latin alphabet since its very beginning.
Later, the rules for writing K have been simplified, so it was always written as C, except before a V (i.e. U sound) that was consonant, not vowel, i.e. like English W. Writing K has been retained only in a few traditional expressions, e.g. in "KALENDAE", which was used when writing dates.
Initially, the letter C (Greek gamma) was used both for the sound K and for the sound G. Later, the letter G has been created, by modifying the letter C. Since then, G was used for the sound G, except in a few traditional expressions, e.g. the name "Gaius" has continued to be written as "Caius", together with a few other traditional names.
The new letter G has substituted in the Latin alphabet the letter Z (Greek zeta), which was not used in Latin. Therefore the 21-letter Latin alphabet has become:
ABCDE FGHIK LMNOP QRSTV X
Several centuries later, during the Roman Empire, 2 additional Greek letters have been added at the end of the alphabet, so the 23-letter Latin alphabet was:
ABCDE FGHIK LMNOP QRSTV XYZ
The letter Z was reintroduced, but not in its original place, because it was contained in some borrowed Greek words. The same with Y. While both V and Y come from Greek u-psilon, by the time when Y has been added to the Latin alphabet it was pronounced as a vowel different from both Latin V and I, i.e. as a front rounded closed vowel, like the Scandinavian Y, German Ü (U with Umlaut) or French U.
The 23-letter Latin alphabet has become a 26-letter alphabet more than a millennium later, when the letters J, U and W have been added (J and U were required because in the Romance languages the old I consonant and U consonant had become fricative sounds, completely distinct from the I and U vowels, while W has been initially added for English, which is one of the few Indo-European languages that has retained the original pronunciation of consonant U).
Interesting! I'm curious, in the Rustic Capitals, would different forms of what looks like the same letter imply different pronunciations, or did those variations have other functions?
bryanrasmussen · 3h ago
anyone have any knowledge of what libraries, visualization tools were used for this?
rkagerer · 6h ago
Interesting how clear the original, pre-200BC ones are to read, and how they get less legible in the interim until modern times. Guess what's old is new again.
jhanschoo · 2h ago
The Roman cursive is illegible to modern eyes, as is the archaic script. You may think that the archaic script would be legible once you learn the correspondence, but don't be fooled by their geometric simplicity; they possess a lot of variation in forms (still geometric due to their inscriptional nature) across time and space. Likewise, the modern lowercase letters are likely illegible to ancient Roman eyes. The inscriptional Roman styles have remained legible, because Italic typesetters and printers (see Renaissance Antiqua) have consciously revived the monumental forms for the capitals. (Contrast the ornate capitals of Fraktur, or of English Roundhand).
raldi · 6h ago
They went from chiseling stone to writing on parchment.
t-3 · 4h ago
Probably more that the artifacts with the most ancient handwriting have been destroyed by time and so all we have left are the chiseled and carved samples. Monuments designed for display are going to have very clear and legible text which is hard to misread so that everyone knows that the person who commissioned it was really awesome, while handwritten cursive often only needs to mostly legible to one reader.
paganel · 3h ago
Really, really cool visualisation!
Even though I know almost nothing at all about the subject of medieval palaeography in Western Europe, I did read recently the write-downs of a Italian conference focused on the subject of palaeography in the Italian Middle Ages, and that is how I learned about stuff like La minuscola cancelleresca and La mercantesca. A link (in Italian, google translate can help) [1] about these two styles of writing and how they were formed:
> la MINUSCOLA CANCELLERESCA, usata dai notai e dalla classe colta non universitaria, e la MERCANTESCA, scrittura professionale dei mercanti, usata anche per testi letterari (ma solo in lingua volgare).
via google translate
> the MINUSCULE CANCELLERESCA, used by notaries and the non-university educated class, and the MERCANTESCA, the professional writing of merchants, also used for literary texts (but only in the vernacular).
Old English had four letters that are not in today's US-ASCII, two of which are borrowed from a runic alphabet rather than created by modifying Latin letters.
It's also a bit arbitrary whether a modified Latin letter is regarded as a new character or an existing character modified by a diacritic: take Ø, for example. And there are characters like Æ. And never forget what Turkish did: add dotless i so that ordinary i could then be regarded as dotless i plus a diacritic (though of course it isn't usually regarded that way).
My guess is that the educated people of the time were very familiar with Greek, so it was easier for them to work with Greek letters rather than the newly invented ones. It probably was the same for Latin-based scripts as well.
- The use of digraphs is frequently due to historical sound changes; first, certain sound combinations changed valence, but others do not. Then one rewrites with the digraphs with the other instances by analogy. Sometimes it comes from borrowing; for example the historical use of "ch" to denote the historic greek "chi", or aspirated chi sound, as in "chronograph"
- Finally, it is easier to modify a existing type, or arrange movable type, to include an accent rather than to create one from scratch.
As I understand (in a simplified way), every once in a while (once in a few centuries) there's a major language reform to make written form reflect current pronunciation as much as possible, but already in 100—200 years people would start asking "why do we write not like we pronounce", until some political movement decides reforming orthography would promote their agenda. Then the cycle continues.
In space as well. Local accents can be quite variable. This alone dooms any “we should write words as we pronounce them” to fail unless every region has its own orthographic rules, which would be honestly terrible.
English doesn’t. Bernard Shaw tried inventing new letters. I guess that changing the English alphabet os a slippery slope. If you make it as phonetic as the Latin script is meant to be, and with special characters, people would have to relearn how to read from scratch
Color and colour are different pictures of the same word.
The original Latin alphabet had 21 letters. Then G has replaced Z, without changing the number of letters. Then, by the time of the Empire, Y and Z have been added at the end (the same Z that had been removed earlier, while Y had the same origin as V, but by this time it had acquired a distinct pronunciation in Greek).
Then the Latin alphabet had 23 letters for more than 1000 years.
It has grown to 26 letters during the Middle Ages, with the addition of J, U and W ("u" was originally the small letter form of "V"; distinct letters U and V have been created by making a capital "U" and a small "v").
Not sure where J came from.
One question is why we borrowed K when we already had C. In modern English, C is more or less a completely superfluous letter, adding unnecessary complexity to pronunciation. Seeing my son try to pronounce "cycle" is one of many examples.
J was introduced because I had a similar problem, except I could be either a consonant or a vowel, rather than any of two consonants. The same applied to the introduction of U for V. Well, almost the same. In the case of I, they added J to be the consonant, while in the case of V, they added U to be the vowel.
Finally, people changed the pronunciation of V over time to be something new, and W was introduced to be the original sound in English. W does not exist in Latin since V fulfills its role, unless you are using a modern pronunciation that changes how V is pronounced and then you still do not need W. There were other sound changes (see Italian), but this is sufficient to explain all of the letters. Well, there is also Z, which they removed from early Latin and later reintroduced to represent a Greek sound.
The original Latin alphabet had 21 letters:
ABCDE FZHIK LMNOP QRSTV X
The letter V (Greek u-psilon) was used to write the sound U, both as vowel and as consonant (i.e. like English W).
The letter F (Greek di-gamma) was used in Greek for consonant U (English W), but in Latin it was used to write the sound F, also a labial consonant, which did not exist in Greek.
In Latin, in the beginning 3 different letters were used to write the sound K, C before E or I, K before A and Q before O or U. So K belonged to the Latin alphabet since its very beginning.
Later, the rules for writing K have been simplified, so it was always written as C, except before a V (i.e. U sound) that was consonant, not vowel, i.e. like English W. Writing K has been retained only in a few traditional expressions, e.g. in "KALENDAE", which was used when writing dates.
Initially, the letter C (Greek gamma) was used both for the sound K and for the sound G. Later, the letter G has been created, by modifying the letter C. Since then, G was used for the sound G, except in a few traditional expressions, e.g. the name "Gaius" has continued to be written as "Caius", together with a few other traditional names.
The new letter G has substituted in the Latin alphabet the letter Z (Greek zeta), which was not used in Latin. Therefore the 21-letter Latin alphabet has become:
ABCDE FGHIK LMNOP QRSTV X
Several centuries later, during the Roman Empire, 2 additional Greek letters have been added at the end of the alphabet, so the 23-letter Latin alphabet was:
ABCDE FGHIK LMNOP QRSTV XYZ
The letter Z was reintroduced, but not in its original place, because it was contained in some borrowed Greek words. The same with Y. While both V and Y come from Greek u-psilon, by the time when Y has been added to the Latin alphabet it was pronounced as a vowel different from both Latin V and I, i.e. as a front rounded closed vowel, like the Scandinavian Y, German Ü (U with Umlaut) or French U.
The 23-letter Latin alphabet has become a 26-letter alphabet more than a millennium later, when the letters J, U and W have been added (J and U were required because in the Romance languages the old I consonant and U consonant had become fricative sounds, completely distinct from the I and U vowels, while W has been initially added for English, which is one of the few Indo-European languages that has retained the original pronunciation of consonant U).
Even though I know almost nothing at all about the subject of medieval palaeography in Western Europe, I did read recently the write-downs of a Italian conference focused on the subject of palaeography in the Italian Middle Ages, and that is how I learned about stuff like La minuscola cancelleresca and La mercantesca. A link (in Italian, google translate can help) [1] about these two styles of writing and how they were formed:
> la MINUSCOLA CANCELLERESCA, usata dai notai e dalla classe colta non universitaria, e la MERCANTESCA, scrittura professionale dei mercanti, usata anche per testi letterari (ma solo in lingua volgare).
via google translate
> the MINUSCULE CANCELLERESCA, used by notaries and the non-university educated class, and the MERCANTESCA, the professional writing of merchants, also used for literary texts (but only in the vernacular).
[1] https://spotlight.vatlib.it/it/latin-paleography/feature/17-...