How would one form roots?
-
- rupestrian
- Posts: 10
- Joined: 03 Dec 2022 09:05
How would one form roots?
I've been considering a language inspired by Turkish and other Oghuz languages, yet I'm unsure of how to properly form roots off of which to base some of the words I have in mind. Does it come down to concocting something that sounds nice, and going from there, or is there/can there be more rhyme or reason to it?
- LinguistCat
- sinic
- Posts: 325
- Joined: 06 May 2017 07:48
Re: How would one form roots?
Different people have different ways of going about it. Since you're basing your language on existing languages, you could look up how their roots are shaped.
As an example that I know more about, Japanese on its face is a mostly (C)V language, but can have long vowels, a syllabic nasal or a lengthening of certain consonants. On top of this, while surface realizations of verbs (and adjectives) have to follow the morae and syllable rules of the language, root forms of many verbs can end in consonants besides the moraic ones above. (Verbs and certain adjectives are the main word classes to change from their roots in Japanese. Most adverbs that do are actually the adverbial forms of those adjectives.) Vowels are added both to show what form the verb is conjugated into as well as making a valid mora.
I could take all this into consideration when making roots for a language based on Japanese. I could also just keep all that in my head and make roots or words that "sound right" without following this exact set up. So if you know how roots work in the languages you're basing your language on, then you could take either course to form roots and derive words.
You could also just make a bunch of words without regard to any of this and then tweak them to sound and look the way you'd like.
It's really a matter of what works best for you.
As an example that I know more about, Japanese on its face is a mostly (C)V language, but can have long vowels, a syllabic nasal or a lengthening of certain consonants. On top of this, while surface realizations of verbs (and adjectives) have to follow the morae and syllable rules of the language, root forms of many verbs can end in consonants besides the moraic ones above. (Verbs and certain adjectives are the main word classes to change from their roots in Japanese. Most adverbs that do are actually the adverbial forms of those adjectives.) Vowels are added both to show what form the verb is conjugated into as well as making a valid mora.
I could take all this into consideration when making roots for a language based on Japanese. I could also just keep all that in my head and make roots or words that "sound right" without following this exact set up. So if you know how roots work in the languages you're basing your language on, then you could take either course to form roots and derive words.
You could also just make a bunch of words without regard to any of this and then tweak them to sound and look the way you'd like.
It's really a matter of what works best for you.
- Creyeditor
- MVP
- Posts: 5123
- Joined: 14 Aug 2012 19:32
Re: How would one form roots?
The other possibility is the use of regular sound changes. This is often called 'a posteriori' conlanging. It can be used to diachronically derive a daughter language but it does not have to be that way.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
"Thoughts are free."
Produce, Analyze, Manipulate
1 2 3 4 4
Ook & Omlűt & Nautli languages & Sperenjas
Papuan languages, Morphophonology, Lexical Semantics
- eldin raigmore
- korean
- Posts: 6356
- Joined: 14 Aug 2010 19:38
- Location: SouthEast Michigan
Re: How would one form roots?
Whatever you find out, let us know! We’ll be interested!Hugh_Capet wrote: ↑19 Sep 2023 09:56 I've been considering a language inspired by Turkish and other Oghuz languages, yet I'm unsure of how to properly form roots off of which to base some of the words I have in mind. Does it come down to concocting something that sounds nice, and going from there, or is there/can there be more rhyme or reason to it?
My minicity is http://gonabebig1day.myminicity.com/xml
Re: How would one form roots?
(Much of this advice is intended with synthetic and agglutinating languages in mind.)
One thing that's important is that you should be prepared to make roots a little general, since you're going to be deriving new words from them via the morphology. In English, we might say "throw" is a root, but you can mod it using affixes - say "throw up" "throw out" "throw down" etc. There's no need, therefore, to make more words for these. (Mind you, English does that anyways, i.e. vomit. And you can - and probably will regardless - do that in your language. Plus, you're probably looking for advice for making languages much more synthetic than English.) Another example would be Semitic verb constructs, and the way Semitic languages can flip a noun into a verb and then into a noun again. Another, whatever the heck Navajo did to make its version of the word "tank" (chidí naaʼnaʼí beeʼeldǫǫh bikááʼ dah naaznilígíí). Another might be Hungarian's mad long words, and how a suffix doesn't need to have just one very specific and confined use. These are tools that make having fewer roots a real possibility.
There are no rules, but making it proportional makes it easier. That is, number of roots to complexity of morphology. Semitic languages, for example, have far less developed derivational morphologies than Apache languages - and so, as one might expect, Semitic languages tend to have more roots than Apache ones. (I have no idea how this applies to "Altaic" languages.)
Another thing to consider with roots is evolution in phonology. A language may start off, let's say, with two-consonant roots, that can take some modding consonants. Over time, these may become three-consonant roots (in other words, hooray more roots!!!), and out from that stage, four-consonant ones. (This is something of a controversy in Hebrew etymology.) Or the other way around - lop off phonological info (evolve the sound inventories to be smaller) and work around it. (Not sure of an irl example. Algic and Algonquian maybe?? I didn't look into it.)
There're other ways to play with evolution but this is all I can think of at the moment.
Also, consider noun morphology vis á vis verb morphology.
- Think basics
- Think of how you mod the basics
- Think of how many basic roots vs. how many ways to mod
- Think of the different ways your basics (and mods) evolve
- Think of how changes in other features affects your basic roots
- Think of different kinds of basic roots vis á vis each other
- Ask people who know more than me. I beg of you. I could be wrong about literally everything.
One thing that's important is that you should be prepared to make roots a little general, since you're going to be deriving new words from them via the morphology. In English, we might say "throw" is a root, but you can mod it using affixes - say "throw up" "throw out" "throw down" etc. There's no need, therefore, to make more words for these. (Mind you, English does that anyways, i.e. vomit. And you can - and probably will regardless - do that in your language. Plus, you're probably looking for advice for making languages much more synthetic than English.) Another example would be Semitic verb constructs, and the way Semitic languages can flip a noun into a verb and then into a noun again. Another, whatever the heck Navajo did to make its version of the word "tank" (chidí naaʼnaʼí beeʼeldǫǫh bikááʼ dah naaznilígíí). Another might be Hungarian's mad long words, and how a suffix doesn't need to have just one very specific and confined use. These are tools that make having fewer roots a real possibility.
There are no rules, but making it proportional makes it easier. That is, number of roots to complexity of morphology. Semitic languages, for example, have far less developed derivational morphologies than Apache languages - and so, as one might expect, Semitic languages tend to have more roots than Apache ones. (I have no idea how this applies to "Altaic" languages.)
Another thing to consider with roots is evolution in phonology. A language may start off, let's say, with two-consonant roots, that can take some modding consonants. Over time, these may become three-consonant roots (in other words, hooray more roots!!!), and out from that stage, four-consonant ones. (This is something of a controversy in Hebrew etymology.) Or the other way around - lop off phonological info (evolve the sound inventories to be smaller) and work around it. (Not sure of an irl example. Algic and Algonquian maybe?? I didn't look into it.)
There're other ways to play with evolution but this is all I can think of at the moment.
Also, consider noun morphology vis á vis verb morphology.
- Think basics
- Think of how you mod the basics
- Think of how many basic roots vs. how many ways to mod
- Think of the different ways your basics (and mods) evolve
- Think of how changes in other features affects your basic roots
- Think of different kinds of basic roots vis á vis each other
- Ask people who know more than me. I beg of you. I could be wrong about literally everything.