UniLang

rubs wrote:not sure what i'm meant to argue.

I'm still trying to figure out why you call Navajo "incredibly regular" when, as far as I can see, it has roughly the same amount of irregularity as most other languages I've studied.

rubs wrote:I never intended you to read that I thought it was particular to amerindians.
Other than finner nuances, I wanted to give somewhat varried examples.

Just naming a language isn't an example of anything. If you think Yup'ik is "expressing concepts previously entirely alien to the culture without requiring any degree of lexicalisation", then I'd like to see some instances of that.

rubs wrote:But if you mean In the sense that categories exist at some level and that their must be some cognitive coding or translation, and that language plays some role in this, and I understood what you said about lojban being an exception in some sense to this.
But I can assure you that lojban certainly has sets and categories.
even though I’m in no way fluent.

But they're not structured the same. Please, if you're not familiar with prototype theory, read up on it before you try dispute which languages use graded categories (i.e. all natural human languages) as opposed to monothetic Aristotelian ones (e.g. Lojban). That's why I provided a link.

I'm revamping my conlang's grammar.

I don't expect the phonology or vocabulary to change much, but I want my conlang to be topic-prominent.

Can a Verb initial word order be compatible with a topic prominent language. Japanese and Korean are topic-prominent with an SOV word order while Chinese is SVO.

What about gender? All of the examples I know of are genderless languages.

Most topic prominent languages seem to be either agglutinating or isolating. Are there any fusional natlangs with this structure?

Atluk wrote:Can a Verb initial word order be compatible with a topic prominent language. Japanese and Korean are topic-prominent with an SOV word order while Chinese is SVO.

Topic-prominence is compatible with any default word order (the keyword being "default"). If you want an example of a VSO language with syntactic topicalisation, look no further than Welsh.

Atluk wrote:What about gender? All of the examples I know of are genderless languages.

Why would gender make a difference?

Atluk wrote:Most topic prominent languages seem to be either agglutinating or isolating. Are there any fusional natlangs with this structure?

Brazilian Portuguese.

linguoboy wrote:

Losam wrote:Phonology comes before Grammar, right?

Not necessarily. You could easily decide, "I want to make a default VSO fluid-S polysynthetic language" and start mapping out which categories you plan to inflect for before you make a single choice about what combinations of sounds will be present in your language.

Losam wrote:But how about Grammar? Could be an alphabet and after pronouns? What you can suggest?

Alphabets have nothing to do with grammar. They should probably be the last thing you create, since they require a pretty solid knowledge of the language's phonology and external history to get right.

Pronouns don't play a prominent role in all languages. Look at pro-drop languages for examples of how easily they can be dispensed with much of the time.

Thank you so much linguoboy for the answer. I'm too noob with conlang yet, and when is possible, I study and search to improve my skills in this field. But, talking about which properties of a language come first to build a language, I saw in a comment (in a youtube video) that first come phonology, then, morphology and after syntax. What do you recommend or are used to do when you start to create a language?
I'm asking that because for my first conlang, I started with a phonemic inventory, a basic of orthography (which letter correspond to which sound)then some spelling rules, and personal pronouns.
For a person like me (the biggest beginner in linguistics, conlang and similar), again, what do you recommend or which steps come first (besides zoompist kit, swadesh list and so on)?

Thank you again for your attention and help.

Losam wrote:Thank you so much linguoboy for the answer. I'm too noob with conlang yet, and when is possible, I study and search to improve my skills in this field. But, talking about which properties of a language come first to build a language, I saw in a comment (in a youtube video) that first come phonology, then, morphology and after syntax. What do you recommend or are used to do when you start to create a language?

I don't have a recommendation. Conlangs are extremely varied, as are conlangers. I suspect that this order is the most common for the simple reason that this is how most published grammars (of both natural and constructed languages) are ordered. There's also an argument to be made that phonology is easier to learn about than morphosyntax and that makes it a better place to start, but I don't find this entirely convincing.

I've seen conlangers adopt a range of successful approaches. As I said before, some decide from the start what kind of syntax they want to have and only settle on a phonology much further down the line. Others know exactly what they want their language to sound or look like long before they know what the grammar should be. So without knowing what your goals as a conlanger are, I can't really tell you what's the best place to start.

Losam wrote:I'm asking that because for my first conlang, I started with a phonemic inventory, a basic of orthography (which letter correspond to which sound)then some spelling rules, and personal pronouns.
For a person like me (the biggest beginner in linguistics, conlang and similar), again, what do you recommend or which steps come first (besides zoompist kit, swadesh list and so on)?

What inspired you to make this language? Is it a priori (invented from scratch) or a posteriori (based on an existing language or languages)? Are you creating it for a particular project (literary, scientific, or otherwise)? Each of these factors might give me some clue what is the best way to proceed. But ultimately it depends on what works for you, not what I or anyone else finds useful.

Thanks again. Well, it's a little bit of derived from other languages and from sketch. I decided to create a language to understand more from linguistics and languages in general. My goal with my first conlang is just to create a language for personal use (for example: write down something that no one knows what means), to teach someone who wants to learn and to use it (for a book for example). Also, to improve my languages skills.

Losam wrote:Thanks again. Well, it's a little bit of derived from other languages and from sketch. I decided to create a language to understand more from linguistics and languages in general. My goal with my first conlang is just to create a language for personal use (for example: write down something that no one knows what means), to teach someone who wants to learn and to use it (for a book for example). Also, to improve my languages skills.

Then the logical thing would be to figure out what language skills you want to work on and tackle those parts first.

I've seen this approach before. A lot of people are curious about how ergative languages work, so they create an ergative language to find out. The project forces them to do research, and then whatever they learn gets incorporated into the conlang.

linguoboy wrote:

Losam wrote:Thanks again. Well, it's a little bit of derived from other languages and from sketch. I decided to create a language to understand more from linguistics and languages in general. My goal with my first conlang is just to create a language for personal use (for example: write down something that no one knows what means), to teach someone who wants to learn and to use it (for a book for example). Also, to improve my languages skills.

Then the logical thing would be to figure out what language skills you want to work on and tackle those parts first.

I've seen this approach before. A lot of people are curious about how ergative languages work, so they create an ergative language to find out. The project forces them to do research, and then whatever they learn gets incorporated into the conlang.

That's exactly what I did with my conlang. For awhile, all of my conlangs were nominative-accusative because I was afraid to experiment with ergativity.

Does anybody know whether the language spoken by Watto (and Anakin) in Star Wars is an actual conlang or just gibberish?

https://www.youtube.com/watch?v=OLnhvoOKNvA

Is there any link or information that explains which sounds cost less effort to produce? I'm thinking about creating combinations of sounds to make my language flows, I mean, with an easy articulation. For example, the schwa vowel costs less effort to produce than any other vowel in the English language, isn't?
Above all, thanks for the help.

IpseDixit wrote:Does anybody know whether the language spoken by Watto (and Anakin) in Star Wars is an actual conlang or just gibberish?

I don't know for sure, but I'd put my money on gibberish.

Suomalainen Varis wrote:Is there any link or information that explains which sounds cost less effort to produce? I'm thinking about creating combinations of sounds to make my language flows, I mean, with an easy articulation. For example, the schwa vowel costs less effort to produce than any other vowel in the English language, isn't?
Above all, thanks for the help.

I don't know about vowels specifically. The closest thing I can think of for language sounds in general is a sonority hierarchy (Wikipedia). Generally, languages like to put sounds with the highest sonority (the loudest ones, like vowels) toward the middle of a syllable, while less sonorous sounds (like voiceless stops) tend to go to the edges of a syllable. Additionally, less sonorous sounds are often changed to be more sonorous over time. Wikipedia gives the example of the Finnish potential mood, where -tne- becomes -nne-. Both of these tendencies tend to make words easier to pronounce. But there's probably a limit to how useful this is; there's a loose preference for higher sonority in languages, and vowels are the most sonorous sounds we can make, but no natural language is composed of just vowels.

As a side note, while ease of pronunciation is definitely a thing in languages, also very important is ease of understanding. Perhaps schwa is the easiest vowel to pronounce, but the most common vowels across languages are [i a u]. That's hypothesized to be because they are at the corners of the vowel space; they sound as different from each other as any three vowels possibly can, so there's no mistaking one for another.

IpseDixit wrote:Does anybody know whether the language spoken by Watto (and Anakin) in Star Wars is an actual conlang or just gibberish?

It's a half-baked conlang. It has a distinct lexicon, but the grammar is pretty much identical to English.

Thank you so much Dormouse559. Nice argument. If I'm not wrong, Arabic only have /a/, /i/ and /o/ as well, a long form of these three vowels. Probably, there's no conflating about vowels, I think.

Guys, I'm creating the basics about phonology for my first conlang, and I'd like to receive your feedback, if you don't mind of course. The sound system are:

• Vowels:
  
  • Front: /i/, /y/, /e/, /ɛ/ and /a/;
  • Central: /ɐ/;
  • Back: /u/ and /o/

• Consonants:
  
  • Plosives: /p/, /b/, /t/, /d/, /k/ and /G/;
  • Nasals: /m/ and /n/;
  • Trill: /r/ (alveolar);
  • Flap/Tap: /r/ (alveolar);
  • Fricatives: /f/, /v/, /s/, /z/ and /h/;
  • Approximant: /j/;
  • Lateral Approximant: /l/;

I also, have another question: how do you guys, handle stress in your conlang?
I was thinking about put stress only on the first syllable, always (from left to right).
However, the exception will be: for words with only one syllable and for the second syllable to the last one, that contains /y/, /e/, /ɐ/, /u/ and /o/, as pure vowels, will not be stressed.

Suomalainen Varis wrote:

• Vowels:
  
  • Front: /i/, /y/, /e/, /ɛ/ and /a/;
  • Central: /ɐ/;
  • Back: /u/ and /o/

• Consonants:
  
  • Plosives: /p/, /b/, /t/, /d/, /k/ and /G/;
  • Nasals: /m/ and /n/;
  • Trill: /r/ (alveolar);
  • Flap/Tap: /r/ (alveolar);
  • Fricatives: /f/, /v/, /s/, /z/ and /h/;
  • Approximant: /j/;
  • Lateral Approximant: /l/;

This looks just fine to me. I like the slight asymmetry of having /ɣ/ instead of /g/.

Suomalainen Varis wrote:I also, have another question: how do you guys, handle stress in your conlang?

In my main conlang (Silvish; Romance language) stress goes either on the last or second-to-last syllable.

Suomalainen Varis wrote:I was thinking about put stress only on the first syllable, always (from left to right).
However, the exception will be: for words with only one syllable and for the second syllable to the last one, that contains /y/, /e/, /ɐ/, /u/ and /o/, as pure vowels, will not be stressed.

I'm afraid I don't understand what you want the exception to be. Could you rephrase or give examples?

Initial stress is of course a fine idea, though.

No worries . The stress works on the following way:

• Stress is applied only to vowels;
• The first syllable is always stressed (from left to right);
• The exception is: If the first syllable contains one of the following vowels at the beginning: /y/, /e/, /ɐ/, /u/ or /o/, the word won't be stressed;

For example: /mɐ/, /ne/, /moi/, /jai/, /ki.ra/, /rok.sa/, /rak.sa/;

The sounds in bold are the ones that receive the stress. I hope that now it's clear. If not, tell me.
Thank you again for the comment.

Okay, I understand now. I have doubts about how naturalistic this restriction is. When a natural language has a different number of vowel distinctions between stressed and unstressed syllables, it normally distinguishes more vowels in stressed syllables. The effect of your restriction is that only /i ɛ a/ can occur in certain stressed syllables, compared to eight vowels in unstressed syllables, the reverse of that trend. But you've applied it only to one- and two-syllable words, so that makes it less of an issue for me.

Note: Up to now, I've been assuming you are going for a naturalistic language, one where the goal is to make it seem like it might be a natural language. There are other possible goals for conlangs, and my advice may or may not apply depending on what yours are.

Suomalainen Varis wrote:No worries . The stress works on the following way:

• Stress is applied only to vowels;
• The first syllable is always stressed (from left to right);
• The exception is: If the first syllable contains one of the following vowels at the beginning: /y/, /e/, /ɐ/, /u/ or /o/, the word won't be stressed;

For example: /mɐ/, /ne/, /moi/, /jai/, /ki.ra/, /rok.sa/, /rak.sa/;

The sounds in bold are the ones that receive the stress. I hope that now it's clear. If not, tell me.
Thank you again for the comment.

What happens to words which aren't stressed? Are they cliticised?

Thank you linguoboy for the comment. I never heard about cliticised before. It means that "non-stressed words" are function words?

Dormouse559, thanks. Yes, you're right: I'm trying to creating a conlang more natural. But since I'm a beginner, I have a few ideas about how to achieve such goal.
My basic goals with my first conlang, are not to create a full language, or a language that I can express complexity thoughts. Just for my everyday use, basic things and to allow to write things down with nobody having any idea of what's written. As well, to learn more about linguistics and languages.
Maybe I'll create this conlang as a sort of proto language.

Suomalainen Varis wrote:Thank you linguoboy for the comment. I never heard about cliticised before. It means that "non-stressed words" are function words?

Not exactly. It means that stress is part of the definition of a (phonological) word, so any morphemes without stress are attached to an adjoining word.

I don't have a clear idea what the role is of stress in your conlang.

I see. I'm not sure if I'll use it at all, because I'm not sure if stress can really makes any difference. I know that for English, certain words have the same written form, but the only difference is on the stress.