Romanization

[Teck]

I'm working on my first real conlang and was wondering whether my romanization makes any sense (especially for ʂ ʐ x ɣ) and if there could be improvements.
/m n ɳ ŋ/ <m n gn ng>
/p b t d k g q ʔ/ <p b t d k g q ‘>
/f v s z ʂ ʐ x ɣ h/ <f v s z sh j ch gh h>
/ʙ r/ <w r>
/l/ <l>

/ɪ ɛ/ <i e>
/ʉ ə ɐ/ <uu u a>
/ɒ/ <o>

[zyma]

I'm working on my first real conlang and was wondering whether my romanization makes any sense (especially for ʂ ʐ x ɣ) and if there could be improvements.
/m n ɳ ŋ/ <m n gn ng>
/p b t d k g q ʔ/ <p b t d k g q ‘>
/f v s z ʂ ʐ x ɣ h/ <f v s z sh j ch gh h>
/ʙ r/ <w r>
/l/ <l>

/ɪ ɛ/ <i e>
/ʉ ə ɐ/ <uu u a>
/ɒ/ <o>

Welcome to the board!

This looks good and generally makes sense to me.

I'm guessing you asked about /ʂ ʐ x ɣ/ in particular because you have <s z> for /s z/, but <sh j> for /ʂ ʐ/, and <k g> for /k g/, but <ch gh> for /x ɣ/. I don't think these are problems, personally. <k ch> for /k x/ reminds me of German.

I do have a few more specific comments, but please know that I am not asking you to change any of the things I'm pointing out. They're not bad, just noteworthy to me in some way.

One thing that stands out is <gn> for the retroflex nasal. I'd be interested in hearing the reasoning behind this choice, since I tend to associate <gn> with the palatal nasal /ɲ/, based on how that digraph is used in French and Italian.

In my experience, there's a tendency for anglophone conlangers to avoid using <u> for /ə/ because it feels "Englishy". However, like I said above, just because this isn't something I've seen a lot, that doesn't mean you shouldn't do it. If you like /ə/ <u>, don't let other people's preferences pressure you into changing it. Again, I don't think there's anything wrong with /ə/ <u>, but, as you move forward with this language, be warned that you might see people saying it's too "Englishy".

[Teck]

I'm guessing you asked about /ʂ ʐ x ɣ/ in particular because you have <s z> for /s z/, but <sh j> for /ʂ ʐ/, and <k g> for /k g/, but <ch gh> for /x ɣ/. I don't think these are problems, personally. <k ch> for /k x/ reminds me of German.

Yes, I asked because I'm not really familiar with the IPA (saw it for the first time about a week ago) and don't know how these would normally be romanized and wanted to be sure I wasn't doing really weird things. And I get why it reminds you of German. My first language is Dutch which is very closely related to German, and in Dutch /x/ is <ch> according to what I could find and /k/ is also <k>, so that could be why it reminds you of German.

I do have a few more specific comments, but please know that I am not asking you to change any of the things I'm pointing out. They're not bad, just noteworthy to me in some way.

Please tell me everything, that's why I made the post. I love to learn. (And thank you for the reply).

One thing that stands out is <gn> for the retroflex nasal. I'd be interested in hearing the reasoning behind this choice, since I tend to associate <gn> with the palatal nasal /ɲ/, based on how that digraph is used in French and Italian.

There's not really that much reasoning behind it. I chose it because I have /ŋ/ as <ng> and I thought /ɳ/ sounded quite similar when I listened to it so I switched the n and g of /ŋ/ to make /ɳ/ <gn>. How would /ɳ/ be romanized most of the time if <ng> is mostly used for /ɲ/?

In my experience, there's a tendency for anglophone conlangers to avoid using <u> for /ə/ because it feels "Englishy". However, like I said above, just because this isn't something I've seen a lot, that doesn't mean you shouldn't do it. If you like /ə/ <u>, don't let other people's preferences pressure you into changing it. Again, I don't think there's anything wrong with /ə/ <u>, but, as you move forward with this language, be warned that you might see people saying it's too "Englishy".

As I said above, my native language is Dutch, and because /ə/ in Dutch is written as <e>, I chose to use <u> instead to make it less like Dutch, but if that makes it to Englishy that's also not really what I'm going for. How else is that vowel written when people don't use <u>?

[zyma]

Yes, I asked because I'm not really familiar with the IPA (saw it for the first time about a week ago) and don't know how these would normally be romanized and wanted to be sure I wasn't doing really weird things.

Something I've found helpful, although I'm sure you've already done this, is looking at the orthographies of natural languages written in the Latin alphabet, or the romanization systems used for natural languages written in other scripts, to see how certain speech sounds tend to be represented.

And I get why it reminds you of German. My first language is Dutch which is very closely related to German, and in Dutch /x/ is <ch> according to what I could find and /k/ is also <k>, so that could be why it reminds you of German.

Right. I said it reminds me of German because that was the first language that uses <k ch> for the phonemes /k x/ that came to mind at the time. I didn't mean to imply that it was the only language that does that. I'm not surprised to hear that Dutch does the same thing. I think a number of Balto-Slavic languages do as well, including Polish, Czech, Slovak, and Lithuanian, if I remember correctly.

Please tell me everything, that's why I made the post.

The other comments I had were the ones I made in the paragraphs following that statement.

There's not really that much reasoning behind it. I chose it because I have /ŋ/ as <ng> and I thought /ɳ/ sounded quite similar when I listened to it so I switched the n and g of /ŋ/ to make /ɳ/ <gn>. How would /ɳ/ be romanized most of the time if <ng> is mostly used for /ɲ/?

Regarding that last sentence, it's <gn> that's used for /ɲ/ in French and Italian, not <ng>, although I suspect this may have been simply a typo.

For the retroflex nasal, /ɳ/, there are two ways of writing it in the Latin alphabet that come to mind right away for me, personally. The first is <ṇ>, which I've seen in the romanization systems used for various languages spoken in India. The second is <rn>, based on the orthographies of some of the indigenous languages of Australia, as well as some North Germanic languages, in which some underlying sequences of a rhotic followed by an alveolar consonant can surface as retroflex.

But, as I said before, if you like <gn> for /ɳ/, you don't absolutely need to change it.

As I said above, my native language is Dutch, and because /ə/ in Dutch is written as <e>, I chose to use <u> instead to make it less like Dutch, but if that makes it to Englishy that's also not really what I'm going for. How else is that vowel written when people don't use <u>?

Based on what letters you're already using, some possibilities that come to mind are <ë> (used in, for example, Albanian), <ă> (used in Romanian), <y> (which represents [ə] in Welsh in at least some contexts, if I understand correctly), <ơ> (used for the longer of Vietnamese's two central mid vowels), <â> (used for the shorter of Vietnamese's two central mid vowels), or perhaps simply <ə>. I don't know if any of these fit the aesthetic you're going for.

[Teck]

Something I've found helpful, although I'm sure you've already done this, is looking at the orthographies of natural languages written in the Latin alphabet, or the romanization systems used for natural languages written in other scripts, to see how certain speech sounds tend to be represented.

That's a very good idea. Thanks, I'll try that next time.

Right. I said it reminds me of German because that was the first language that uses <k ch> for the phonemes /k x/ that came to mind at the time. I didn't mean to imply that it was the only language that does that. I'm not surprised to hear that Dutch does the same thing. I think a number of Balto-Slavic languages do as well, including Polish, Czech, Slovak, and Lithuanian, if I remember correctly.

And I didn't mean to imply that you implied that. I just thought it's kind of cool that you know it enough that <k ch> for /k x/ reminds you of certain languages.

For the retroflex nasal, /ɳ/, there are two ways of writing it in the Latin alphabet that come to mind right away for me, personally. The first is <ṇ>, which I've seen in the romanization systems used for various languages spoken in India. The second is <rn>, based on the orthographies of some of the indigenous languages of Australia, as well as some North Germanic languages, in which some underlying sequences of a rhotic followed by an alveolar consonant can surface as retroflex.

But, as I said before, if you like <gn> for /ɳ/, you don't absolutely need to change it.

I think I'm going to keep <gn> or else use <ṇ>. I'll think about it. I do like the way ṇ looks, but I'm going to have to look into how to type it.

Based on what letters you're already using, some possibilities that come to mind are <ë> (used in, for example, Albanian), <ă> (used in Romanian), <y> (which represents [ə] in Welsh in at least some contexts, if I understand correctly), <ơ> (used for the longer of Vietnamese's two central mid vowels), <â> (used for the shorter of Vietnamese's two central mid vowels), or perhaps simply <ə>. I don't know if any of these fit the aesthetic you're going for.

I like <ë> and <â>, so I'll use one of those two. But would /ʉ/ as <uu> still make sense if /ə/ is no longer <u> but <ë>, wouldn't it make more sense for /ʉ/ to be <u> then?

[sangi39]

One thing to think about (I'm not sure if Shimobaatar mentioned it, I skimmed the rest of the posts) is the history of the romanisation. For example, in areas of the world previously under English control, a lot of romanisations, especially the earlier ones, tend to use English orthography as a base (for example, Bloodywood, a band that tends to sing in Punjabi, uses a romanisation that has <oo> and <ee> for /u:/ and /i:/ respectively, although, IIRC, the most commonly used romanisation uses <ū> and <ī> instead). Similarly, languages spoken in areas formerly under Spanish control tend to have features like <x> for shibilants or <j> and <y> for /x/ and /j/ respectively.

[zyma]

And I didn't mean to imply that you implied that. I just thought it's kind of cool that you know it enough that <k ch> for /k x/ reminds you of certain languages.

Ah, sorry about that. It can be quite easy to misunderstand exactly what someone else meant over the internet. And thank you.

I think I'm going to keep <gn> or else use <ṇ>. I'll think about it. I do like the way ṇ looks, but I'm going to have to look into how to type it.

For symbols I'm unable to type, I usually either copy and paste them, or "insert" them using a special menu that the web browser I use (Google Chrome) gives me access to.

I like <ë> and <â>, so I'll use one of those two. But would /ʉ/ as <uu> still make sense if /ə/ is no longer <u> but <ë>, wouldn't it make more sense for /ʉ/ to be <u> then?

I would definitely make /ʉ/ <u>, personally.

One thing to think about (I'm not sure if Shimobaatar mentioned it, I skimmed the rest of the posts) is the history of the romanisation. For example, in areas of the world previously under English control, a lot of romanisations, especially the earlier ones, tend to use English orthography as a base (for example, Bloodywood, a band that tends to sing in Punjabi, uses a romanisation that has <oo> and <ee> for /u:/ and /i:/ respectively, although, IIRC, the most commonly used romanisation uses <ū> and <ī> instead). Similarly, languages spoken in areas formerly under Spanish control tend to have features like <x> for shibilants or <j> and <y> for /x/ and /j/ respectively.

A very good point. If your conlang is meant to exist within a certain context, or is meant to have a history of being spoken by a fictional group of people, that context/history could absolutely influence its romanization.

[Teck]

Thank you both for helping me with this. I think I have all the information I need for now