How did you personally go about choosing your language's syntax and other related attributes?

[Eti]

I'm brand new to this scene, so while I am (relatively) clueless when it comes to certain concepts in how this whole construction process works, I will say now that I'm aware that this process differs from person to person. Being fresh out of the gates I'd rather have at least something as my point of reference. Apologies ahead of time if anything I ask is considered taboo!

Now to get to the cut. The LCK has certainly shown me a lot of possibilities for how to handle my language, but more often than not I find myself getting quite overwhelmed with the options. I know the general idea of how to choose them (e.g. what matches the culture, context of the world, etc.) but there's a lot of options that really don't work with this linear thought process as simply as I thought they might.

When you made your conlang(s), how did you decide which models to use for how the grammar works? I feel like just flipping a coin or rolling some dice to select isn't that good of an idea, even if I constrain the options to what would most likely make sense in the context of the world and culture down to a couple options, but I'm afraid I simply do not know what the alternatives to this really are. Is there any other process aside from answering the question of "What would work in the context of your culture?" that can be applied to try to select some of these options? On top of this, some of these don't really work as exclusive options (e.g. p. 58 goes over fusional, agglutinative, and isolating languages, and from what I gathered, it relays that languages tend to use a certain balance of these with a huge preference to one specifically). How do I determine what balance makes the most sense, assuming this is even something that can be answered without the abstract experience?

Part of me says that I'm making this way too hard for myself. I'm really good with logically oriented things, and had assumed that the construction of a language would follow a sort-of-simple model, but it seems there's a lot more holes in the logic that I don't know how to fill up than I was ready for. My aim with this thread here is to try to get some assistance in filling in these blanks, and to get a better understanding what a good workflow might look like. I'll do everything in my power to make the most out of what I'm given, at the very least, since I can only assume that this concept isn't something that can be explained over text very easily. It seems like one of those things that just comes from trying and crashing a few times, like riding a bike.

[Pabappa]

My vision for Pabappa* has been the same since I was young .... I was trying to write a kids' novel about a league of child superheros, and in that novel the narrator meets up with some younger kids from another planet who speak a language that sounds like baby talk. So people, particularly adults, assume that these kids are very simple-minded, and that their language will be easy to learn. They soon learn, however, that Pabappa is an extremely difficult language, with far more inflections and irregular verbs than their own language, and decide to let the kids have their language to themselves.

I gave up on writing the story long ago, but the same idea still works well ... if I say that Pabappa is literally baby talk, I can say that it has a syntax so complicated that we adults can never hope to learn it, and must merely sit back and wonder whenever we hear it.

Unlike many others who post here, I dont have a great deal of knowledge about linguistics, so I have been brute-forcing everything, trying one idea after another and discarding the ones that I decide I dont like. Possibly, one major motivator for me is that there are a lot of ideas I just dont like .... Pabappa is what it is because if I were to change even minor details, I wouldnt enjoy it anymore. That is also why after more than twenty years of conlanging I still have only two fully functional conlangs .... my world map is full, but the dozens of other languages on it are so underdeveloped that they dont even have names and I cant write any sentences in any of them.

Basically, my advice for now is, go with what you like, and try writing sentences in it, maybe even paragraphs if you get to that stage. The writing tests will help you ensure that your ideas fit together and aren't contradictory. Writing tests have helped me discover a lot of issues with Pabappa that I wouldnt have noticed if I had always just plowed ahead blindly.

Welcome to the board. I wish you the best.

*Poswa is my main conlang now, which carries Pabappa to an even greater extreme, but that's a technicality. I like the name Pabappa better so I picked that as my username.

[bradrn]

I'm brand new to this scene, so while I am (relatively) clueless when it comes to certain concepts in how this whole construction process works, I will say now that I'm aware that this process differs from person to person. Being fresh out of the gates I'd rather have at least something as my point of reference. Apologies ahead of time if anything I ask is considered taboo!

Welcome to the board Eti! And if you’re worried, I can’t see anything you’re asking which could possibly be considered taboo.

Now to get to the cut. The LCK has certainly shown me a lot of possibilities for how to handle my language, but more often than not I find myself getting quite overwhelmed with the options. I know the general idea of how to choose them (e.g. what matches the culture, context of the world, etc.) but there's a lot of options that really don't work with this linear thought process as simply as I thought they might.

You can actually get quite far with a language without figuring out its context — for my most well-developed language, I still know absolutely nothing about the world it’s spoken in. The context only really affects the lexicon, semantics and pragmatics of a language — the rest is more or less orthogonal to context.

When you made your conlang(s), how did you decide which models to use for how the grammar works?

Personally, I would say: go with what you like. I really like agglutinative languages, verging on polysynthetic but not quite there, so most of my languages have been of that type. But more recently I wanted to make a more isolating language, and that’s what I’m doing now. In general, try looking at a bunch of languages, see which aesthetic you like most, and then make a language like that.

EDIT: In case you want a list of languages you can look at for inspiration, here’s some languages and families that have been helpful for me. Arranged very roughly from most agglutinative to most isolating: NW Caucasian, Chukchi, Salishan, Ojibwe, San Miguel Chimalapa Zoque, Quechua, Mayan, Basque, Dyirbal, Bororo, Chalcatongo Mixtec, Shilha Berber, Yapese, Marshallese, Northeast Ambae, Tibetan, Burmese, Yoruba, Mandarin.

Is there any other process aside from answering the question of "What would work in the context of your culture?" that can be applied to try to select some of these options?

I’m not entirely sure why you would ask this… as I said, most of the basics of language is more or less orthogonal to culture. There isn’t really any commonality between the cultures which speak agglutinative languages, for instance.

On top of this, some of these don't really work as exclusive options (e.g. p. 58 goes over fusional, agglutinative, and isolating languages, and from what I gathered, it relays that languages tend to use a certain balance of these with a huge preference to one specifically).

I dunno… I’d say those options are pretty mutually exclusive, even if it is possible to combine them to a small degree. But I do get what you’re saying; there do exist other options which can only be combined and balanced in many ways.

How do I determine what balance makes the most sense, assuming this is even something that can be answered without the abstract experience?

Unfortunately, the answer to this is mostly knowledge: I found myself improving dramatically at figuring out this balance when I actually started to read more about linguistics. But one easy ‘shortcut’ is to ask people here: for instance, elsewhere on the board, I’m currently discussing the best ways to balance pronominal clitics and zero-anaphora.

Part of me says that I'm making this way too hard for myself. I'm really good with logically oriented things, and had assumed that the construction of a language would follow a sort-of-simple model, but it seems there's a lot more holes in the logic that I don't know how to fill up than I was ready for. My aim with this thread here is to try to get some assistance in filling in these blanks, and to get a better understanding what a good workflow might look like. I'll do everything in my power to make the most out of what I'm given, at the very least, since I can only assume that this concept isn't something that can be explained over text very easily. It seems like one of those things that just comes from trying and crashing a few times, like riding a bike.

I can empathise, as someone whose brain works in much the same way. Luckily, not everything in language is quite as subtle as you seem to be imagining, and there are a bunch of basics about a language which it is possible to answer in a simplistic way:

• What is the sound inventory?
• Is it going to be more agglutinative or more isolating?
• What is the basic word order?
• If it’s agglutinative, then which categories get marked on the noun and the verb?
• Does it have noun case? Does it have verbal agreement? (Few languages have both.)

Most of these are fairly independent of each other — the answer to one doesn’t really depend on the answer to any of the others. (One notable exception: verbal agreement is more common in agglutinative languages.) Once you’ve answered those, it should be easier to figure out the rest.

[Eti]

Awesome responses, and thanks for the welcomes! Now - I can say that I'm quite relieved to see these various examples and how things work. I was definitely overcomplicating things, it seems.

I think most of my main concerns stemmed from the fact that I was worried that coming back to something earlier in the pipeline was a bad idea, that it was a strictly constructive process where it would require completely scrapping work and rebuilding it from where the change was made no matter what. On consideration of the points given, it's assuring to know that it's perfectly safe to jump around. I'm a lot more used to planning ahead of time. I suppose when I look back at when I first learned programming, I did a lot more improvisation and jumping around. Food for thought.

I'll see what I can't make of the points given. So far I think I've already cleaned up a couple of the roadblocks in my workflow thanks to some of the points given. Oh, and a huge thanks for that list of languages, it'll be instrumental in deciding how I want my language to behave. I'll have to do some tinkering.

[bradrn]

Awesome responses, and thanks for the welcomes! Now - I can say that I'm quite relieved to see these various examples and how things work. I was definitely overcomplicating things, it seems.

You’re not alone — I do the same thing.

I think most of my main concerns stemmed from the fact that I was worried that coming back to something earlier in the pipeline was a bad idea, that it was a strictly constructive process where it would require completely scrapping work and rebuilding it from where the change was made no matter what. On consideration of the points given, it's assuring to know that it's perfectly safe to jump around. I'm a lot more used to planning ahead of time. I suppose when I look back at when I first learned programming, I did a lot more improvisation and jumping around. Food for thought.

Yep, it’s perfectly fine to jump around. For instance, often I’ll just do something really simple for an area where I’m not quite sure, then I’ll come back to it later when I’ve decided what I want. And I’ll often redo a section I’ve decided I don’t like. (Actually, I probably do this far more than I should!)

I'll see what I can't make of the points given. So far I think I've already cleaned up a couple of the roadblocks in my workflow thanks to some of the points given. Oh, and a huge thanks for that list of languages, it'll be instrumental in deciding how I want my language to behave. I'll have to do some tinkering.

You’re welcome! The languages are just some which I found interesting; you should do some research on your own to discover which ones you personally like best. In case you do want to do some more investigation into natlangs, one resource which for me has been incredibly useful is the Grammar Pile: a huge collection of reference grammars for hundreds of languages around the world. And while I’m giving recommendations, another resource I like is the Stack, a similarly huge collection of linguistics books and articles. (Although, if you haven’t yet, you should probably read zompist’s Advanced Language Construction and Lexipedia before reading any of those.)

[Raphael]

Welcome to the Board from me, too, Eti!

I'm "only" doing naming languages so far, so I should theoretically have a lot less work than a "full-fledged" conlanger, but I definitely know the feeling of not knowing what to do because there are so many options. As for your main question, in my main proto-naming-language, very innovatively named "Proto-1", I decided that I would try to make it agglutinative for the fairly unoriginal reason that I thought isolating would be too complicated and fusional would be too SAE. I also decided that I would try to do as much as possible with prefixes rather than suffixes because I thought prefixes would be a greater deviation from SAE. I basically tried to make up for the fact that the phonology would be pretty SAE (which, in turn, was probably because I wanted to make sure that I'd be able to pronounce the names myself). Those are arguably not really "good" reasons for any conlanging decisions, though.

[bradrn]

… my main proto-naming-language, very innovatively named "Proto-1", …

Ah, I see you have the same naming conventions as I do!

[Frislander]

Welcome! The gherkins and tea will be arriving shortly...

I tend to find I create languages (really most of the time language sketchs) based on some particular thing or other that I have had the idea to work with after having read about it. My currently most developed conlang Asta, for example, sprung from a desire to do stuff with noun classes and a particular style of polysynthesis and has grown also to incorporate a few other bits and pieces that I've come across and liked, such as having the schwa be altered by following consonants, pervasive sandhi processes and a particular set of consonant contrasts with some interesting gaps partly driven by a desire to create a romanisation that doesn't use letters with ascenders (so no voiced stops and no voiceless velar stop either). I tend to prefer this style as it leads to a more interesting variety in the conlangs you create that better reflects the diversity of natural languages, rather than having to feel like you're filling out a checklist of typological features that seems a tad parochial in its particular focuses. Like for example saying to myself "I'm gonna craft a system of aspectual contrasts and see what I can do with them" is a much more interesting proposition that simply asking "does my conlang have aspect?".

Now to get to the cut. The LCK has certainly shown me a lot of possibilities for how to handle my language, but more often than not I find myself getting quite overwhelmed with the options. I know the general idea of how to choose them (e.g. what matches the culture, context of the world, etc.) but there's a lot of options that really don't work with this linear thought process as simply as I thought they might.

You can actually get quite far with a language without figuring out its context — for my most well-developed language, I still know absolutely nothing about the world it’s spoken in. The context only really affects the lexicon, semantics and pragmatics of a language — the rest is more or less orthogonal to context.

Is there any other process aside from answering the question of "What would work in the context of your culture?" that can be applied to try to select some of these options?

I’m not entirely sure why you would ask this… as I said, most of the basics of language is more or less orthogonal to culture. There isn’t really any commonality between the cultures which speak agglutinative languages, for instance.

Small quibble, but this may not necessarily be true for naturalistic languages if the view held by Trudgill and Mcwhorter and co that smaller more intimate social networks tend to accumulate complexity/irregularity more easily is correct (and I tend to think that this is probably true to some degree), in which case more complex morphology with more irregularity (of whatever kind) should probably restricted to smaller groups of people, i.e. not large empires. This of course is not the same as saying that smaller tribes speak universally fusional languages - even highly morpholgically transparent agglutinative languages can be highly complex/irregular in other ways.

On top of this, some of these don't really work as exclusive options (e.g. p. 58 goes over fusional, agglutinative, and isolating languages, and from what I gathered, it relays that languages tend to use a certain balance of these with a huge preference to one specifically).

I dunno… I’d say those options are pretty mutually exclusive, even if it is possible to combine them to a small degree. But I do get what you’re saying; there do exist other options which can only be combined and balanced in many ways.

I also quibble this, but more because I think the fusional-agglutinative-isolating trichotomy is bad pop-typology that actually hinders understanding of any language whether natural or constructed.

[bradrn]

Welcome! The gherkins and tea will be arriving shortly...

I was wondering whether we still did that! I’m glad we still are.

Now to get to the cut. The LCK has certainly shown me a lot of possibilities for how to handle my language, but more often than not I find myself getting quite overwhelmed with the options. I know the general idea of how to choose them (e.g. what matches the culture, context of the world, etc.) but there's a lot of options that really don't work with this linear thought process as simply as I thought they might.

You can actually get quite far with a language without figuring out its context — for my most well-developed language, I still know absolutely nothing about the world it’s spoken in. The context only really affects the lexicon, semantics and pragmatics of a language — the rest is more or less orthogonal to context.

Is there any other process aside from answering the question of "What would work in the context of your culture?" that can be applied to try to select some of these options?

I’m not entirely sure why you would ask this… as I said, most of the basics of language is more or less orthogonal to culture. There isn’t really any commonality between the cultures which speak agglutinative languages, for instance.

Small quibble, but this may not necessarily be true for naturalistic languages if the view held by Trudgill and Mcwhorter and co that smaller more intimate social networks tend to accumulate complexity/irregularity more easily is correct (and I tend to think that this is probably true to some degree), in which case more complex morphology with more irregularity (of whatever kind) should probably restricted to smaller groups of people, i.e. not large empires. This of course is not the same as saying that smaller tribes speak universally fusional languages - even highly morpholgically transparent agglutinative languages can be highly complex/irregular in other ways.

I personally think this idea is highly dubious — there’s been plenty of complex languages spoken by large communities (e.g. Latin, Mayan, Quechua, Turkish, Fillipino, Swahili…). And anyway, how do you define complexity? To my understanding, most languages are about as complex as each other.

On top of this, some of these don't really work as exclusive options (e.g. p. 58 goes over fusional, agglutinative, and isolating languages, and from what I gathered, it relays that languages tend to use a certain balance of these with a huge preference to one specifically).

I dunno… I’d say those options are pretty mutually exclusive, even if it is possible to combine them to a small degree. But I do get what you’re saying; there do exist other options which can only be combined and balanced in many ways.

I also quibble this, but more because I think the fusional-agglutinative-isolating trichotomy is bad pop-typology that actually hinders understanding of any language whether natural or constructed.

I don’t think it’s particularly good either (I’m particularly suspicious of ‘fusional’, which seems to apply to IE and nothing else), but it’s certainly useful for getting a good idea for what a language might be like.

[akam chinjir]

Like for example saying to myself "I'm gonna craft a system of aspectual contrasts and see what I can do with them" is a much more interesting proposition that simply asking "does my conlang have aspect?."

For myself, I'm usually pretty unsatisfied if I set out to design a paradigm---an aspectual paradigm, and agreement paradigm, whatever. With some things (pronouns, agreement, numbers), it sort of goes with the territory, but with TAM stuff I usually end up trying to do things more piecemeal. Which is to say, I'm more likely to start out thinking, say, "I want an 'already' perfect," and let whatever system there's going to be emerge from the interaction of different ideas.

...Which is to say I agree with this other bit:

I tend to find I create languages (really most of the time language sketchs) based on some particular thing or other that I have had the idea to work with after having read about it.

I also want to take up the Trudgill/McWhorter stuff, but I'll move over to the Linguistic Miscellany Thread so as not to derail things here.

(Edit. I got distracted and forgot what I was going to say about that.)

[Vardelm]

I'm brand new to this scene

Welcome, BUT....

The gherkins and tea will be arriving shortly...

... may I offer some bread & butter pickles instead of gherkins if you prefer?

When you made your conlang(s), how did you decide which models to use for how the grammar works?

There's good input so far, but here's my 2¢.

I've been at this for a while, but what has started to work for me is picking a few core features that I want to base the language on. Mostly this is morphsyntactic stuff, since you can plop just about any phonology on top of that. It may be alignment, tense but no morphological aspect, few or lots of cases, evidentials, etc. Think of 1-3 features that are core, non-negotiable, but make sure they aren't mutually exclusive. Get a rough idea of how you might want to handle those via morphology (suffixing, particles, or whatever) and a vague list of what cases, evidentials, etc. you might like to include. Then, get a rough idea of other features you might like to consider adding, but aren't crucial. At this point, I start thinking of overall typology: fusional, isolating, suffixes, infixes, etc.

Once I have this general idea, I start making sample "sentences" with glosses, but it's really JUST the glosses. I don't do any phonology or word forms yet. Just something like:

man-ERG PAST-kick-EVID.VIS-3P.SNG dog-ABS
The man kicked the dog.

I make sure my core features will work, and I think about overall syntax like word order. Then, I start adding more elements that are the other, non-core features. I also start thinking about how to handle adjectives, adjunct/oblique phrases, etc. Also, if the language has particles or is fusional, I think about what grammatical features will be combined into a single affix or particle. I keep with "gloss sentences".

a.IND.SNG man-ERG kick-EVID.VIS.PAST-3P.SNG dog-ABS morning-ABS in.POSTP
A man kicked the dog in the morning.


Once I have a whole bunch of sentence types worked out (equational, intransitive, transitive, adjunct phrases, questions, orders, etc.), I start thinking in more detail about phonology & morphology. I decide what I want the phonology to be like (maybe combine a couple different natlangs) and figure out if its prefixing, suffixing, etc. This is when I finally start making paradigms and actual sentence samples.

The whole approach helps me think about how the language is going to function without worrying about stylistic issues yet. It's like making an outline before ever writing a paper. You focus on function first and form later.

Welcome again & good luck!

[vegfarandi]

I usually have 1-3 natlangs in mind that I've been reading about recently and I pick a couple of core features I like the best from each of them. Then I triangulate the remaining features based on these core features. I find that, once I have a core idea of a language, a large portion of the grammar feels like it "writes itself". Personally, I typically get stuck later in the process once the basic phonology and morphosyntax are fairly fleshed out and it comes to describing things like conditionals and the nuance of tense-aspect-mood usage and I have a tendency to go back and restructure my grammars, overcomplating the description and being paralyzed with inaction at 85% done

Anyway, how the core features come to me has changed over time. My first conlang, Lomanin was inspired by French and Arabic which I was reading about at the time, as well as another conlang, Golic Vulcan from the original Star Trek movies. And probably some Quenya. Stuff I was exposed to at the time in my early days of conlanging. The main ideas were heavily skewed towards the phonology side of things, so there's a lot going on with fricatives in there.

My next one was Uscaniv which I essentially based on Etruscan (phonology and basic morphosyntax) but there's a lot of Latin in there as well because I was studying it at high school at the same time.

Since then I've been mostly interested in exploring ever more exotic ideas and less about phonology and more about morphosyntax. So Kin-Sang was tones, tones as inflection and English-level syntheticism (so mostly analytic but some synthetic categories).

Imutan was complex morphophonology, VSO word order, minimal conjunctions and lots of reliance on non-finite verb forms.

Proto-Ronquian (which was a team project) was serial verb constructions, exceptional case marking, monosyllables, minimal inflection and the transparent use of content words as grammatical words (in other words, there's almost no words that exclusively function as grammar words).

Duriac, which I'm working on now, is based on word classes – half-way between Bantu- and Caucasian-style indexing, asymmetric case marking based on animacy (animates don't mark ergative; inanimates don't mark accusative), polysynthesis on verbs and verb chaining à la Amazonian languages, with a lot of quasi-non-finite verb forms.

I used to get my basic ideas reading the high-level language descriptions in the Encyclopedia Britannica, now I tend to read linguistics papers on various phenomena on Academia.edu.

[Eti]

man-ERG PAST-kick-EVID.VIS-3P.SNG dog-ABS
The man kicked the dog.

I make sure my core features will work, and I think about overall syntax like word order. Then, I start adding more elements that are the other, non-core features. I also start thinking about how to handle adjectives, adjunct/oblique phrases, etc. Also, if the language has particles or is fusional, I think about what grammatical features will be combined into a single affix or particle. I keep with "gloss sentences".

a.IND.SNG man-ERG kick-EVID.VIS.PAST-3P.SNG dog-ABS morning-ABS in.POSTP
A man kicked the dog in the morning.


Once I have a whole bunch of sentence types worked out (equational, intransitive, transitive, adjunct phrases, questions, orders, etc.), I start thinking in more detail about phonology & morphology. I decide what I want the phonology to be like (maybe combine a couple different natlangs) and figure out if its prefixing, suffixing, etc. This is when I finally start making paradigms and actual sentence samples.

The whole approach helps me think about how the language is going to function without worrying about stylistic issues yet. It's like making an outline before ever writing a paper. You focus on function first and form later.

Welcome again & good luck!

I really like this. Use sort of placeholder things for structure before trying to find out the words I need. I never thought of that. Given that I think in a fairly strict ordered manner, this should be perfect. Huge +1 for proposing this.

Welcome! The gherkins and tea will be arriving shortly...

I was wondering whether we still did that! I’m glad we still are.

First thing in mind was Vinny playing AI Dungeon. I look forward to this event if it's gonna have me laughing as hard as this stream was.

You can actually get quite far with a language without figuring out its context — for my most well-developed language, I still know absolutely nothing about the world it’s spoken in. The context only really affects the lexicon, semantics and pragmatics of a language — the rest is more or less orthogonal to context.

I’m not entirely sure why you would ask this… as I said, most of the basics of language is more or less orthogonal to culture. There isn’t really any commonality between the cultures which speak agglutinative languages, for instance.

Small quibble, but this may not necessarily be true for naturalistic languages if the view held by Trudgill and Mcwhorter and co that smaller more intimate social networks tend to accumulate complexity/irregularity more easily is correct (and I tend to think that this is probably true to some degree), in which case more complex morphology with more irregularity (of whatever kind) should probably restricted to smaller groups of people, i.e. not large empires. This of course is not the same as saying that smaller tribes speak universally fusional languages - even highly morpholgically transparent agglutinative languages can be highly complex/irregular in other ways.

I personally think this idea is highly dubious — there’s been plenty of complex languages spoken by large communities (e.g. Latin, Mayan, Quechua, Turkish, Fillipino, Swahili…). And anyway, how do you define complexity? To my understanding, most languages are about as complex as each other.

I dunno… I’d say those options are pretty mutually exclusive, even if it is possible to combine them to a small degree. But I do get what you’re saying; there do exist other options which can only be combined and balanced in many ways.

I also quibble this, but more because I think the fusional-agglutinative-isolating trichotomy is bad pop-typology that actually hinders understanding of any language whether natural or constructed.

I don’t think it’s particularly good either (I’m particularly suspicious of ‘fusional’, which seems to apply to IE and nothing else), but it’s certainly useful for getting a good idea for what a language might be like.

In my specific case, given the context I'm designing this in, the language would be relatively solid and not see much derivation. It's a bit of an interesting concept, but the simple version here is that the language is spoken universally across an entire species across their entire world since its primary means of propagation is, when not verbally spoken, a network of information not unlike what could be best considered a global phone line. The only cultural differences and differences in dialect would stem from the context of where people live (e.g. in a cold environment, there may be animals akin to bears that don't live in other warmer environments, so people in the warm env don't necessarily talk about them much and by extension, the given word doesn't see usage and some may not even know what it means under specific scenarios). In this case, it's more a thing of word frequency in favor of changing the dialect.

Still, lots of stuff still stands here so the points here will be something I consider as I work, especially the objections to different inflections and their usage.

[Curlyjimsam]

A common motivation for me is to try stuff I haven't done before. Admittedly that doesn't narrow things down very much for your very first conlang!

[Vardelm]

A common motivation for me is to try stuff I haven't done before. Admittedly that doesn't narrow things down very much for your very first conlang!

I'd think motivation for new (most?) conlangers is stuff that isn't in your L1.

[WeepingElf]

When you made your conlang(s), how did you decide which models to use for how the grammar works? I feel like just flipping a coin or rolling some dice to select isn't that good of an idea, even if I constrain the options to what would most likely make sense in the context of the world and culture down to a couple options, but I'm afraid I simply do not know what the alternatives to this really are. Is there any other process aside from answering the question of "What would work in the context of your culture?" that can be applied to try to select some of these options? On top of this, some of these don't really work as exclusive options (e.g. p. 58 goes over fusional, agglutinative, and isolating languages, and from what I gathered, it relays that languages tend to use a certain balance of these with a huge preference to one specifically). How do I determine what balance makes the most sense, assuming this is even something that can be answered without the abstract experience?

I don't do any random decisions. Indeed, what I do is to make the conlang such that it "fits" the culture it is meant for - which is of course a very subjective notion. Most of my conlangs are placed in a setting which is essentially the real world with some languages and cultures added; though in some cases the latter boils down to "there are a few people who still speak a language of an otherwise extinct lineage in place XY".

The natlang which actually had the greatest impact on the formation of my conlinguistic taste was Latin. The sound of the language and especially those inflectional paradigms - not very different from the paradigms of my native German, what concerns the categories expressed and the means used, but in what vivid colours compared to the phonetically bleached version of German - that spoke to me, and my early conlangs were much like that, and in some ways, my current conlangs are still influenced by it.

When I started the project which is now Old Albic, I had something very different in mind from what it is now. It was this Tolkien fan fiction story which got the ball rolling, as I asked myself, "What kind of language would those Elves speak?", and set forth to build a descendant of Sindarin. However, I soon ran into a problem, namely that the documentation of Sindarin was incomplete, and I had to invent a lot of words and even inflections on my own. Also, of course, I had no good idea how language change works, so I did a lot of research on that, checking out textbooks from the local university library, and reading a lot of stuff on the Net, including the CONLANG mailing list.

But I got carried away. Quite early in the process, I came up with a new, entirely non-Tolkienian idea about the Elves who spoke Nur-ellen (as I called my conlang): they were no immortal, magical race at all, but humans with an "Elf-like" culture. What they are now. And there was the idea that this "Elvish" language would be the substratum responsible for the "weirdness" of the Insular Celtic languages (VSO word order, initial mutations and all that). Also, I got some new ideas about the language that had nothing to do with Tolkien's Elven languages. I hit upon ergativity, and as I did research on that, I found that some languages had an even funkier morphosyntactic alignment: active-stative. I felt that this was just right for my Elves, so I decided to use it! My research on this topic first threw up some indigenous American languages, but also Georgian and even a reconstruction of an early stage of PIE as an active-stative language (Indo-European and the Indo-Europeans by T. Gamkrelidze and V. Ivanov, the same book in which they laid out their famous "glottalic theory"). This gave me the idea that my language could be a descendant of such an early stage of PIE, diverging even earlier than Anatolian; and in the following process, it was more and more divorced from Tolkien's languages, instead being built on PIE and on words in western IE languages such as Celtic and Germanic which lack good PIE etymologies and are suspected by many scholars to have been borrowed from a substratum language. This, essentially, was the road I travelled, and am still travelling.

[bradrn]

man-ERG PAST-kick-EVID.VIS-3P.SNG dog-ABS
The man kicked the dog.

I make sure my core features will work, and I think about overall syntax like word order. Then, I start adding more elements that are the other, non-core features. I also start thinking about how to handle adjectives, adjunct/oblique phrases, etc. Also, if the language has particles or is fusional, I think about what grammatical features will be combined into a single affix or particle. I keep with "gloss sentences".

a.IND.SNG man-ERG kick-EVID.VIS.PAST-3P.SNG dog-ABS morning-ABS in.POSTP
A man kicked the dog in the morning.


Once I have a whole bunch of sentence types worked out (equational, intransitive, transitive, adjunct phrases, questions, orders, etc.), I start thinking in more detail about phonology & morphology. I decide what I want the phonology to be like (maybe combine a couple different natlangs) and figure out if its prefixing, suffixing, etc. This is when I finally start making paradigms and actual sentence samples.

The whole approach helps me think about how the language is going to function without worrying about stylistic issues yet. It's like making an outline before ever writing a paper. You focus on function first and form later.

Welcome again & good luck!

I really like this. Use sort of placeholder things for structure before trying to find out the words I need. I never thought of that. Given that I think in a fairly strict ordered manner, this should be perfect. Huge +1 for proposing this.

I also found doing this to be really helpful when I was less experienced at conlanging — maybe not for the whole grammar as Vardelm recommends, but just for a few key structures. These days I feel confident enough not to need it, but I still do it on occasion.

Welcome! The gherkins and tea will be arriving shortly...

I was wondering whether we still did that! I’m glad we still are.

First thing in mind was Vinny playing AI Dungeon. I look forward to this event if it's gonna have me laughing as hard as this stream was.

Nope, it’s a board tradition. All newcomers get given pickles and tea. Let me see if I can still find some… ah, here they are…



(Hold on… if newcomers are given pickles and tea, and I’m fairly new around here, why did no-one give me pickles and tea? Ah well, no great loss; I don’t like either particularly well.)

In my specific case, given the context I'm designing this in, the language would be relatively solid and not see much derivation. It's a bit of an interesting concept, but the simple version here is that the language is spoken universally across an entire species across their entire world since its primary means of propagation is, when not verbally spoken, a network of information not unlike what could be best considered a global phone line. The only cultural differences and differences in dialect would stem from the context of where people live (e.g. in a cold environment, there may be animals akin to bears that don't live in other warmer environments, so people in the warm env don't necessarily talk about them much and by extension, the given word doesn't see usage and some may not even know what it means under specific scenarios). In this case, it's more a thing of word frequency in favor of changing the dialect.

Quick question here: what do you mean by not having much ‘derivation’? Do you mean that there isn’t any derivational morphology (which sounds implausible)? Or do you mean that there isn’t any language evolution? (Which sounds equally implausible.)

[Vardelm]

(Hold on… if newcomers are given pickles and tea, and I’m fairly new around here, why did no-one give me pickles and tea? Ah well, no great loss; I don’t like either particularly well.)

Oh, fine. Here's your pickles:


...and here's your tea:


Welcome to the ZBB. Happy?

[bradrn]

(Hold on… if newcomers are given pickles and tea, and I’m fairly new around here, why did no-one give me pickles and tea? Ah well, no great loss; I don’t like either particularly well.)

Oh, fine. Here's your pickles:


...and here's your tea:


Welcome to the ZBB. Happy?

Thank you so much! The sentiment is greatly appreciated. But I did tell you, I don’t like pickles or tea! Maybe I’ll give them to my father; he loves both.

[KathTheDragon]

Did you ever announce you were new? If not, that's why.