[drieddust]

> It never really worked in those instances. What you see in the books and what you hear in real life are often two quite different things.

You are basing your assumption on languages using Roman writing system which actually isn't capable of expressing all sounds used within the language. All language systems doesn't suffer this. For example most Indian languages have accurate sound to letter mapping even to the level of defining short and long sounds of same alphabet differently. In order to speak correctly you just have to say the individual sounds together. There is no weirdness involved.

[dmitriid]

> You are basing your assumption on languages using Roman writing system which actually isn't capable of expressing all sounds used within the language.

Romanian and Turkish disagree. And the spoken language still differs (probably less so in Romanian, more so in Turkish).

[drieddust]

In my experience letters A-Z doesn't capture the whole gamut of utterances used in the language. Most languages use more sounds than 26 letters can capture. I don't know about Turkish or Romanian so can't say. As an example my mother tongue Hindi and Sanskrit maps the most of utterances of hard, soft, and nasal varieties originating from throat, palate, cerebral, dental, and lips into specific symbols. This reduces the friction between writing and speaking.

One can still speak wrong if they learn to speak the letters wrong but chances are reduced. Are Romanian and Turkish similar?

[dmitriid]

> In my experience letters A-Z doesn't capture the whole gamut of utterances used in the language. Most languages use more sounds than 26 letters can capture.

It's more than just sounds. Turkish and Romanian alphabets capture the sounds of Turkish and Romanian alphabets quite well.

But then...

In Turkish "I will do something" is written like this: "Bir şey yapıcaǧım". And if you're going for proper enunciation, that's what you will say. In most situations most people will say "Bi şey yapıcam" (note the omission of "r" in "Bir" and "ǧı" in "yapıcaǧım").

Do the letters correspond to the sounds? Yes. Does it help? Nope, people will not speak the way it's written, because written rules describe a very specific rigid set of rules. And, for example, elisions and contractions [1] are very common in nearly every language and are often frowned upon in written texts (except for a small number of unavoidable ones).

And that's before we go into the plethora of sounds between dialects and regional variations. For example, in Swedish, there's a combination of letters, `sj`[3] that has at least four different pronunciations across Sweden. So, the word "seven", "sju" will be pronounced with [ɕ] in one part of Sweden, with [ɧ], in a different part of Sweden and so on. It's the same word, should it be spelled differently for each group of people?

And note, we're just touching just 1% of 1% of the complexities of pronunciation :) And they rarely if ever can be captured in written text and rules for the written text.

[1] https://en.wikipedia.org/wiki/Elision

[2] https://en.wikipedia.org/wiki/Contraction_(grammar)

[3] https://en.wikipedia.org/wiki/Swedish_phonology#Fricatives Wikipedia has a full paragraph dedicated to this alone