4 ms·

I wish they also invented new alphabets. The need for diacritics (and letter groups) indicates a non-ideal choice of the alphabet. Vietnamese is a good example

I wish they also invented new alphabets. The need for diacritics (and letter groups) indicates a non-ideal choice of the alphabet. Vietnamese is a good example of that, closely followed by Polish and Finnish. (English, on the other hand, has no diacritics, and therefore the Latin alphabet seems like a perfect fit; Russian is another example of a good match.)

gnulinux 4y ago
Very misleading, English doesn't have diacritics but it has copious amounts of digraphs like ch, sh, tch, eu, oe, oo etc... All of which may represent a different sound (being a Germanic language, English has a huge variety of vowels). Some Latin-based alphabets such as Turkish and Eastern European specifically attempted to solve this problem (e.g. use [Ç] instead of [ch] in Turkish or [Š] instead of [sh] in Czech et al). You may not consider digraph usage a problem, but historically, for some reason, many alphabet designers went this route. So, from this perspective it doesn't hold true English seems the most suitable for Latin alphabet. I think quite the opposite, Latin alphabet completely lacks the resources English needs to present its vowels consistently.
avgcorrection 4y ago
English arguably has 14 distinct vowel sounds while the Latin script has six vowel. So it seems like a bad match. Spanish has a few more letters and more accents. But the correspondence between spelling and pronounciation is amazing. And you always know which syllable to stress thanks to the acute accents on the vowels. EDIT: Sibling comment mentioned Turkish. Turkish also has very regular spelling apparently (my impression and what someone has told me). And they apparently needed those extra letters in order to achieve that.
- gnulinux 4y ago
  Depends on how you define regular really but Turkish alphabet also doesn't map 1-to-1 with IPA sounds we can transcribe Turkish language with. Of course, this very much depends on regional accent, but overall Turkish accents are pretty consistent that they present both /c/ and /k/ sounds yet they're both transcribed as [k] since it's obvious which sound should be produced for a native speaker (depends on the vowel). Similar anomaly exists for [g] where there are palatal and velar varities. I think the concept of language fitting an alphabet is flawed. As I mentioned in my sibling comment, although GP considers digraph usage more natural than diacritics, many alphabet designers preferred the opposite. I.e. they turned English digraphs into their own diacritics. Does this make the modified alphabet more suitable for the language? I don't know if there is an objective way to answer that. If there is, let me know. Realistically, any of the worlds' languages would be usable with any of the alphabets.
projektfu 4y ago
New alphabets are a huge barrier to overcome in learning a language. I studied Japanese at the college level for two years and never could keep the kana straight. (Kanji are a lost cause for me as well). I forget Cyrillic pronunciations all the time and my head gets jammed by things like P meaning R. My lips have to move while reading. But in every Latin-script language I have worked on it has been a non-issue for this native English speaker. It would take a while to learn to read in rot13. Now, add in that I am learning the language, and learning Spanish in rot13 would be hard for me.
chmod775 4y ago
> English, on the other hand, has no diacritics, and therefore the Latin alphabet seems like a perfect fit. English as commonly written using the Latin alphabet is objectively an awful fit, because one cannot tell how a word is pronounced from how it is written. One has to memorize the proper pronunciation of hundreds (if not thousands) of words beforehand in order to correctly read a text aloud. Almost all languages have a few words like that (usually at least some loanwords), but English is quite extreme compared to languages like German, where pronunciation can be trivially inferred from spelling in almost all cases. So even if written English doesn't have diacritics, it certainly could use some to give readers more information on how something is supposed to be pronounced.

gnulinux 4y ago

Very misleading, English doesn't have diacritics but it has copious amounts of digraphs like ch, sh, tch, eu, oe, oo etc... All of which may represent a different sound (being a Germanic language, English has a huge variety of vowels). Some Latin-based alphabets such as Turkish and Eastern European specifically attempted to solve this problem (e.g. use [Ç] instead of [ch] in Turkish or [Š] instead of [sh] in Czech et al). You may not consider digraph usage a problem, but historically, for some reason, many alphabet designers went this route. So, from this perspective it doesn't hold true English seems the most suitable for Latin alphabet. I think quite the opposite, Latin alphabet completely lacks the resources English needs to present its vowels consistently.

avgcorrection 4y ago

English arguably has 14 distinct vowel sounds while the Latin script has six vowel. So it seems like a bad match. Spanish has a few more letters and more accents. But the correspondence between spelling and pronounciation is amazing. And you always know which syllable to stress thanks to the acute accents on the vowels. EDIT: Sibling comment mentioned Turkish. Turkish also has very regular spelling apparently (my impression and what someone has told me). And they apparently needed those extra letters in order to achieve that.

gnulinux 4y ago
Depends on how you define regular really but Turkish alphabet also doesn't map 1-to-1 with IPA sounds we can transcribe Turkish language with. Of course, this very much depends on regional accent, but overall Turkish accents are pretty consistent that they present both /c/ and /k/ sounds yet they're both transcribed as [k] since it's obvious which sound should be produced for a native speaker (depends on the vowel). Similar anomaly exists for [g] where there are palatal and velar varities. I think the concept of language fitting an alphabet is flawed. As I mentioned in my sibling comment, although GP considers digraph usage more natural than diacritics, many alphabet designers preferred the opposite. I.e. they turned English digraphs into their own diacritics. Does this make the modified alphabet more suitable for the language? I don't know if there is an objective way to answer that. If there is, let me know. Realistically, any of the worlds' languages would be usable with any of the alphabets.

projektfu 4y ago

New alphabets are a huge barrier to overcome in learning a language. I studied Japanese at the college level for two years and never could keep the kana straight. (Kanji are a lost cause for me as well). I forget Cyrillic pronunciations all the time and my head gets jammed by things like P meaning R. My lips have to move while reading. But in every Latin-script language I have worked on it has been a non-issue for this native English speaker. It would take a while to learn to read in rot13. Now, add in that I am learning the language, and learning Spanish in rot13 would be hard for me.

chmod775 4y ago

> English, on the other hand, has no diacritics, and therefore the Latin alphabet seems like a perfect fit. English as commonly written using the Latin alphabet is objectively an awful fit, because one cannot tell how a word is pronounced from how it is written. One has to memorize the proper pronunciation of hundreds (if not thousands) of words beforehand in order to correctly read a text aloud. Almost all languages have a few words like that (usually at least some loanwords), but English is quite extreme compared to languages like German, where pronunciation can be trivially inferred from spelling in almost all cases. So even if written English doesn't have diacritics, it certainly could use some to give readers more information on how something is supposed to be pronounced.