Project Mammoth Hunter - Help me make my first conlang

[Këkkytir]

(Also, I’m fairly sure that German is fusional, not agglutinative.)

You're right, my mistake. What I meant is that I can easily imagine an agglutinative version of german.

What exactly do you mean by this? I’m struggling to understand what exactly it would mean for a language to have an ‘agglutinative version’.

Sorry, I wasn't very clear here. By 'agglutinative version' I mean that you can replace all the fusional suffixes with agglutinative ones (marking gender, number, case, etc. separately) without radically changing anything else in the language.

[bradrn]

You're right, my mistake. What I meant is that I can easily imagine an agglutinative version of german.

What exactly do you mean by this? I’m struggling to understand what exactly it would mean for a language to have an ‘agglutinative version’.

Sorry, I wasn't very clear here. By 'agglutinative version' I mean that you can replace all the fusional suffixes with agglutinative ones (marking gender, number, case, etc. separately) without radically changing anything else in the language.

That makes sense.

[Xwtek]

I read (in WALS article?) the reason for that is because language with SVO order can parse the sentence easier. In language with neither head marking or dependent marking, it's pretty difficult to parse whether "man stone break" means "A man breaks stone" or "A man stone breaks" (This can be mitigated with obligatory article, or case marking, or to the lesser extent, head marking (person stone 3SG-3SG-break can only ever mean "A man breaks stone"))

That's an obvious motivation for case in verb-peripheral languages, sure---but does it play out the same way in SOV and verb-initial languages? In particular, are case-markers as likely to be aggluted affixes in both sorts of languages?

When TAM markers come after the verb, there's a very strong tendency for them to be affixes (aggluted or fused or whatever). You don't get nearly as strong a tendency when they occur before the verb, from what I understand. It's hard to see what that could have to do with distinguishing the verb's arguments. (There's an idea that the general preference for suffixes over prefixes has something to do with preserving the salience of word-beginnings, that could be part of the story.)

I don't know about the rest, except that case+number exponence is itself restricted to indo-european languages (and semitic?) https://wals.info/feature/21A#2/25.5/152.1. (also, by case marking, I don't exclude adpositions, as long as it distinguishes actor and undergoer).

[KathTheDragon]

case+number exponence is itself restricted to indo-european languages (and semitic?)

You don't seem to have read your own reference properly. WALS lists Finnish, Nenets, Chukchi, West Greenlandic, and Yaqui as all having case+number exponence.

[Salmoneus]

case+number exponence is itself restricted to indo-european languages (and semitic?)

You don't seem to have read your own reference properly. WALS lists Finnish, Nenets, Chukchi, West Greenlandic, and Yaqui as all having case+number exponence.

And Lugbara, according to the notes - I think they don't show that on the map because Lugbara's one of only two case+TAM languages, and they didn't want to make a new case+TAM+number category for just one language... (either that or the notes are wrong).

However, WALS also notes that it only 'properly' occurs in Indo-European and Uralic (in their survey), and is more marginal or accidental in the other languages where it occurs - only for certain nouns in Chukchi, and in Yaqui only because case is unmarked in the plural. So in an extensive form, it is more Indo-European that it might first look.

That said, the fact that IE, Uralic, partially Chukchi and in some unspecified way partially Greenlandic - i.e. two siberian language families, a language family from the steppe and a language family recently emigrated from siberia - is a little bit suggestive of perhaps a former sprachbund in north asia. So it would certainly be appropriate for the OP's palaeosiberian language (although again, given the time depth, there's nothing that would be INappropriate really).

[Qwynegold]

What do you think: is it a good idea to go for an agglutinative language as a beginner, or should I stick with my original plan of an analytic language?

I'm equally open to both possibilities, but I'm not sure I can do a complex agglutinating grammar properly. Hungarian can help, yes, but it gets super complicated after a while. Maybe a simpler agglutinating system like in german (without gender and inflection) would be the golden middle way. If the setting is the only obstacle, I wouldn't mind changing it at all.

From what I've heard, it's typical for beginning conlangers to make agglutinative languages. I think it's because it's easy to think of what grammatical categories you want and to make up affixes for those. You don't have to complicate things the way Hungarian does just because your doing agglutination. On the other hand, I think it would be interesting to see what you can come up with if you go analytic.

[Ares Land]

So, my first question to you is:

What do you think: is it a good idea to go for an agglutinative language as a beginner, or should I stick with my original plan of an analytic language?

I'm equally open to both possibilities, but I'm not sure I can do a complex agglutinating grammar properly. Hungarian can help, yes, but it gets super complicated after a while. Maybe a simpler agglutinating system like in german (without gender and inflection) would be the golden middle way. If the setting is the only obstacle, I wouldn't mind changing it at all.

If that was your original idea, why not go for an analytic language? They're particularly interesting, and you can try deriving an agglutinating system from it later on!

[Këkkytir]

According to the the feedback you guys have provided, and seeing how excited you were about the
paleosiberian concept (although some of you would be equally interested in an analytic language), I've decided on an agglutinative language inspired by today's paleosiberian languages . This way, I can also take advantage of my understanding of hungarian.
Plus, I can have fun learning about paleosiberian languages. This, however, will probably cause delays in the project, so please forgive me if I respond slowly.

As a start, I'd like to follow the steps described in this series of videos by Biblaridion:
https://www.youtube.com/watch?v=FHK1gO2 ... 2O3YO0b2nN
In short, the steps are: phonology (+syllable structure), syntax, grammar and lexicon (then phonological evolution, grammatical evolution and writing systems, but I think these are optional). On the other hand, the LCK suggests phonology, lexicon (+syllable structure), grammar (+extending the lexicon) and syntax (semantics & pragmatics) instead.

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

Next time, I'm going to write about the phonological system of the language (which doesn't have a name yet because I'd like to give it a native name later).

[Këkkytir]

What do you think: is it a good idea to go for an agglutinative language as a beginner, or should I stick with my original plan of an analytic language?

I'm equally open to both possibilities, but I'm not sure I can do a complex agglutinating grammar properly. Hungarian can help, yes, but it gets super complicated after a while. Maybe a simpler agglutinating system like in german (without gender and inflection) would be the golden middle way. If the setting is the only obstacle, I wouldn't mind changing it at all.

From what I've heard, it's typical for beginning conlangers to make agglutinative languages. I think it's because it's easy to think of what grammatical categories you want and to make up affixes for those. You don't have to complicate things the way Hungarian does just because your doing agglutination. On the other hand, I think it would be interesting to see what you can come up with if you go analytic.

So, my first question to you is:

What do you think: is it a good idea to go for an agglutinative language as a beginner, or should I stick with my original plan of an analytic language?

I'm equally open to both possibilities, but I'm not sure I can do a complex agglutinating grammar properly. Hungarian can help, yes, but it gets super complicated after a while. Maybe a simpler agglutinating system like in german (without gender and inflection) would be the golden middle way. If the setting is the only obstacle, I wouldn't mind changing it at all.

If that was your original idea, why not go for an analytic language? They're particularly interesting, and you can try deriving an agglutinating system from it later on!

Thank you for the support! The others convinced me that an agglutinative language is better suited for me. But I won't throw away my original ideas and maybe create an analytic proto-language later on. That'd be fun to do.

[bradrn]

According to the the feedback you guys have provided, and seeing how excited you were about the
paleosiberian concept (although some of you would be equally interested in an analytic language), I've decided on an agglutinative language inspired by today's paleosiberian languages . This way, I can also take advantage of my understanding of hungarian.
Plus, I can have fun learning about paleosiberian languages. This, however, will probably cause delays in the project, so please forgive me if I respond slowly.

That sounds like a great plan! And I for one don’t mind if you respond slowly — from what I know of it, conlanging isn’t supposed to go at a breakneck pace.

As a start, I'd like to follow the steps described in this series of videos by Biblaridion:
https://www.youtube.com/watch?v=FHK1gO2 ... 2O3YO0b2nN
In short, the steps are: phonology (+syllable structure), syntax, grammar and lexicon (then phonological evolution, grammatical evolution and writing systems, but I think these are optional). On the other hand, the LCK suggests phonology, lexicon (+syllable structure), grammar (+extending the lexicon) and syntax (semantics & pragmatics) instead.

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

Next time, I'm going to write about the phonological system of the language (which doesn't have a name yet because I'd like to give it a native name later).

I don’t think you have to go in any particular order, although I would highly recommend doing the phonology first. Then maybe make a couple of words, just so you can write some minimalist sample sentences. After this, I often end up gradually filling out bits and pieces of the morphology and syntax sections in no particular order other than doing the simplest bits first and working up to the more complex bits.

(On the other hand, I’ve never actually gotten any further than this with any conlang, so it may be worth doing something different to what I say!)

___________

Anyway, all of these guidelines seem to agree that you should do the phonology first, so from the PDF that was linked earlier, here’s some things to consider:

• Some kind of vowel harmony
• Presence of /ɯ/ or /ɨ/
• Contrast between /m n ŋ/, often including /ɲ/ as well. /ŋ/ is allowed in syllable onsets.

Apart from those, it may be worth reading the PDF itself (although it doesn’t say much of use other than the above points), or looking through the phonologies of individual Paleosiberian languages and seeing if you can find any other tendencies.

[Qwynegold]

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

You don't have to follow any particular order, though naturally lexicon will have to be started some time after the phonology is finished. The lexicon is something that you will work on forever though. My advice is to work on it little by little while you do other things as well. If you don't have the words you need when you're working with the grammar, you can always leave placeholders in their stead, and fill those in later.

As for the evolution of the language, you can create one or several daughterlangs from the current conlang. But that is by no means necessary. Natural languages always have some irregularities and oddities in them due to the way they have evolved. This is hard to emulate in a conlang that doesn't have any earlier stage, but you can always try to put some things in it that hint at it having evolved from some earlier language.

[Ares Land]

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

I don't know if there's any specific order to follow, except that you should decide on a phoneme inventory and constraints first, and have some idea of typology as well...

Maybe I can tell you how I work instead?

I start with a very basic sketch: phonemic inventory, a few basics on syntax (what kind of inflections are there? what's the basic word order?). I don't spend much time on this, I'm going to revise it anyway.

Then I figure out the proto language. You don't need to make up a full-fledged language at this point -- indeed it would be extremely counter-productive! A rough sketch fitting on a page or two is sufficient. I decide on a phoneme inventory, some constraints, some very basic grammar stuff (a few ideas of syntax, some inflections). It's best to keep things as simple as possible.
Also, I figure out a lexicon for the proto-language. One possible approach is to start with the swadesh list and use a vocabulary generator.

Then I figure out sound changes from the proto-language to the final language and run then through the inflections and word list until I get something I like.

Once you have that, you can start work on your language, beginning on pretty much anything you like.

Why I do it that way?

• I get more naturalistic results that way. Interesting irregularities, constraints I wouldn't have thought of, and so on.
• If I ever want to create a sister language, I can!
• If I need to create a new word, I can either make up a new root, or reuse an existing root with a compound, or another derivation

I should point out that it's most easier to work forwards than backwards (Once I found an old grammar sketch I had made up fifteen years earlier, with no info on the proto-language... Figuring out why the language worked as it did was a nightmare). And it's not that long. You can spend a few hours on the proto-language and get more than enough for your needs. It's actually realistic (how much do we really now for sure about PIE or proto-Semitic anyway?)
and you can flesh it out later when you work on a sister language (if you've kept things vague on the proto-language, if you find out you need it to have a particular feature, or more roots, you can add these retroactively).

And of coure, that's just the ways I do things. It's not mandatory or anything, just something I find convenient.

[Ares Land]

Just popping here to ask how that project's coming along

[Këkkytir]

Just popping here to ask how that project's coming along

Thanks for asking!
I didn't have much freetime last week and I'm still working on the phonology part. I've been looking at phonetic charts of many different siberian languages (maybe I overdid it a little bit ) I also noticed that I can't pronounce many of the sounds, so maybe it's best to avoid those ones for now (is it?). Things that I want in some form: vowel-harmony, a series of palatalized consonants, geminated voiceless stops, very few fricatives/affricatives, CV(C) syllables, stress on first syllable. I'll do a phonetic chart proposal on the weekend.

[bradrn]

Just popping here to ask how that project's coming along

Thanks for asking!
I didn't have much freetime last week and I'm still working on the phonology part. I've been looking at phonetic charts of many different siberian languages (maybe I overdid it a little bit ) I also noticed that I can't pronounce many of the sounds, so maybe it's best to avoid those ones for now (is it?).

Which sounds exactly can’t you pronounce? I may be able to help. Also, in case you haven’t found this yet, Wikipedia has a separate page for each phoneme in the IPA; each page includes a breakdown of the pronunciation, plus an audio sample. Other sites with audio include http://www.ipachart.com/ and https://web.uvic.ca/ling/resources/ipa/ ... IPAlab.htm. The UCLA Phonetics Lab Archive is quite useful as well: it contains real audio samples from many natural languages. I’ve found all these resources to be incredibly useful when trying to learn how to say new sounds.

Things that I want in some form: vowel-harmony, a series of palatalized consonants, geminated voiceless stops, very few fricatives/affricatives, CV(C) syllables, stress on first syllable. I'll do a phonetic chart proposal on the weekend.

That sounds great! I can easily see this language as fitting in with the various other Siberian languages.

[Këkkytir]

Just popping here to ask how that project's coming along

Thanks for asking!
I didn't have much freetime last week and I'm still working on the phonology part. I've been looking at phonetic charts of many different siberian languages (maybe I overdid it a little bit ) I also noticed that I can't pronounce many of the sounds, so maybe it's best to avoid those ones for now (is it?).

Which sounds exactly can’t you pronounce? I may be able to help. Also, in case you haven’t found this yet, Wikipedia has a separate page for each phoneme in the IPA; each page includes a breakdown of the pronunciation, plus an audio sample. Other sites with audio include http://www.ipachart.com/ and https://web.uvic.ca/ling/resources/ipa/ ... IPAlab.htm. The UCLA Phonetics Lab Archive is quite useful as well: it contains real audio samples from many natural languages. I’ve found all these resources to be incredibly useful when trying to learn how to say new sounds.

Well, there's a couple of them, but I think the the sites you linked could help me. Especially the UCLA Phonetics Lab Archive. I use Wikipedia's IPA chart a lot but sometimes I don't feel confident enough about my pronunciation. Even if I follow the instructions mechanically, it feels as if I'm missing something.

The two things I struggled the most with recently:
-velar vs. uvular plosives: [k]-[q], [g]-[ɢ], [x]-[X] and [ɣ]-[ʁ]. I can pronounce the velars more back in my throat but they doesn't sound very different. Also, I wouldn't be able to tell them apart if I hear them in speech. Maybe I'm just not used to this distinction yet, idk.
-the vowel [ɯ]: I want to include this badly in my conlang but every time I try to pronounce it, it sounds either like [y] or [ø]. But I won't give up yet and countinue practising

[bradrn]

The two things I struggled the most with recently:
-velar vs. uvular plosives: [k]-[q], [g]-[ɢ], [x]-[X] and [ɣ]-[ʁ]. I can pronounce the velars more back in my throat but they doesn't sound very different. Also, I wouldn't be able to tell them apart if I hear them in speech. Maybe I'm just not used to this distinction yet, idk.

Velars and uvulars don’t sound all that different anyway — only /q/ and /χ/ are at all common, and most languages don’t even have a contrast between /x/ and /χ/. To make a uvular, just put your tongue as far back in your mouth as you can without it becoming pharyngeal.

-the vowel [ɯ]: I want to include this badly in my conlang but every time I try to pronounce it, it sounds either like [y] or [ø]. But I won't give up yet and countinue practising

Start with a fully back, highly rounded /u/, then without moving your tongue, slowly change your lips from being rounded to being spread out. It’s a bit tricky at first, but if you practise then you should be able to pronounce it easily.

(If you still can’t pronounce /ɯ/, you could use /ɨ/ instead: there are very, very few languages which contrast the two, but /ɨ/ is slightly easier to say.)

[Këkkytir]

As a start, I'd like to follow the steps described in this series of videos by Biblaridion:
https://www.youtube.com/watch?v=FHK1gO2 ... 2O3YO0b2nN
In short, the steps are: phonology (+syllable structure), syntax, grammar and lexicon (then phonological evolution, grammatical evolution and writing systems, but I think these are optional). On the other hand, the LCK suggests phonology, lexicon (+syllable structure), grammar (+extending the lexicon) and syntax (semantics & pragmatics) instead.

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

Next time, I'm going to write about the phonological system of the language (which doesn't have a name yet because I'd like to give it a native name later).

I don’t think you have to go in any particular order, although I would highly recommend doing the phonology first. Then maybe make a couple of words, just so you can write some minimalist sample sentences. After this, I often end up gradually filling out bits and pieces of the morphology and syntax sections in no particular order other than doing the simplest bits first and working up to the more complex bits.

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

You don't have to follow any particular order, though naturally lexicon will have to be started some time after the phonology is finished. The lexicon is something that you will work on forever though. My advice is to work on it little by little while you do other things as well. If you don't have the words you need when you're working with the grammar, you can always leave placeholders in their stead, and fill those in later.

You're right. I can see the advantages of such an iterative approach, so I'll do that. In this way, I can extend the language bit by bit and modify it whenever I want.

My question:

What do you think about Biblaridion's process? Is it good? Would you change the order of
the steps (like in LCK)? Would you add something else to the list? Is phonological and grammatical evolution a must-have, or is it an optional step that can be added later when I have more experience with it?

I don't know if there's any specific order to follow, except that you should decide on a phoneme inventory and constraints first, and have some idea of typology as well...

Maybe I can tell you how I work instead?

I start with a very basic sketch: phonemic inventory, a few basics on syntax (what kind of inflections are there? what's the basic word order?). I don't spend much time on this, I'm going to revise it anyway.

Then I figure out the proto language. You don't need to make up a full-fledged language at this point -- indeed it would be extremely counter-productive! A rough sketch fitting on a page or two is sufficient. I decide on a phoneme inventory, some constraints, some very basic grammar stuff (a few ideas of syntax, some inflections). It's best to keep things as simple as possible.
Also, I figure out a lexicon for the proto-language. One possible approach is to start with the swadesh list and use a vocabulary generator.

Then I figure out sound changes from the proto-language to the final language and run then through the inflections and word list until I get something I like.

Once you have that, you can start work on your language, beginning on pretty much anything you like.

Why I do it that way?

• I get more naturalistic results that way. Interesting irregularities, constraints I wouldn't have thought of, and so on.
• If I ever want to create a sister language, I can!
• If I need to create a new word, I can either make up a new root, or reuse an existing root with a compound, or another derivation

I should point out that it's most easier to work forwards than backwards (Once I found an old grammar sketch I had made up fifteen years earlier, with no info on the proto-language... Figuring out why the language worked as it did was a nightmare). And it's not that long. You can spend a few hours on the proto-language and get more than enough for your needs. It's actually realistic (how much do we really now for sure about PIE or proto-Semitic anyway?)
and you can flesh it out later when you work on a sister language (if you've kept things vague on the proto-language, if you find out you need it to have a particular feature, or more roots, you can add these retroactively).

And of coure, that's just the ways I do things. It's not mandatory or anything, just something I find convenient.

Thank you for sharing your method, I like it! I think it'd work well with the iterative approach mentioned above. I can start with a rough sketch of a proto-language and iteratively modify it until I like the result. It can also explain the irregularities in the language.

I had the same idea of using the swadesh list for basic vocab. Actually, I've been working on an extended swadesh list for some time (mostly based on the LCK's suggestions).

[Këkkytir]

The two things I struggled the most with recently:
-velar vs. uvular plosives: [k]-[q], [g]-[ɢ], [x]-[X] and [ɣ]-[ʁ]. I can pronounce the velars more back in my throat but they doesn't sound very different. Also, I wouldn't be able to tell them apart if I hear them in speech. Maybe I'm just not used to this distinction yet, idk.

Velars and uvulars don’t sound all that different anyway — only /q/ and /χ/ are at all common, and most languages don’t even have a contrast between /x/ and /χ/. To make a uvular, just put your tongue as far back in your mouth as you can without it becoming pharyngeal.

Wow this already helped a lot, thanks!

[bradrn]

The two things I struggled the most with recently:
-velar vs. uvular plosives: [k]-[q], [g]-[ɢ], [x]-[X] and [ɣ]-[ʁ]. I can pronounce the velars more back in my throat but they doesn't sound very different. Also, I wouldn't be able to tell them apart if I hear them in speech. Maybe I'm just not used to this distinction yet, idk.

Velars and uvulars don’t sound all that different anyway — only /q/ and /χ/ are at all common, and most languages don’t even have a contrast between /x/ and /χ/. To make a uvular, just put your tongue as far back in your mouth as you can without it becoming pharyngeal.

Wow this already helped a lot, thanks!

You’re welcome!