User:Elector Dark/Conlang Advice

After that initial hurdle of getting past the entry barrier to decent conlanging, the conlangers that are no longer completely new or prospective, but at the same time not yet firmly established in the art and trade of language design, face unexpected and unusual challenges they most likely aren't used to dealing with. Getting past that initial barrier may seem challenging to a beginner, but it basically just means having to learn some elementary techniques and tools of conlanging and memorising the basic theory and guidelines behind languages and language design. What happens when you master the principles behind the IPA and figure out how case marking works in general? There hasn't been any guide that deals with progressing while at this stage, there aren't any obvious issues waiting to be resolved, there probably aren't any blatantly gaping knowledge holes, so what does an establishing conlanger do?

Well, he reads the Elector's 'Conlang Advice' article, of course!

When you get to the point where the basics are behind you, but what you produce just doesn't cut it, you'll know you've reached the emerging zone, so to say. Let me break it to you, though: you'll be in it forever! There will always be things you won't know or just won't be sure about, but the objective is to try and fill the gaps as thoroughly as possible.

Where do you start, though? I've assembled some notes and guides here, divided into topics, that are supposed to assist anyone with knowledge of the rudimentary basics in acquiring a firmer grasp on the things. Some of the areas have some decent bibliographical coverage as well, but others are just my retelling of the things I've learnt.

Work Method and Ethics
How to do the work? What to do and what not to do!

Inspiration
When it comes to work method involving inspiration, I consider some people might have it a bit wrong. Usually, I've observed that, when most beginners start work on an a priori language, they begin by trying to incorporate some feature from some language intentionally, and then pass it off as inspiration. These inspirations are usually actually taken as models or reference material; the conlanger then starts copying it while changing some details and ends up with their own thing that looks like the model they started off with.

This is not inspiration. It is perfectly okay to incorporate features of a natural language into your own constructed one, but that's what you're meant to be doing anyway. What you aren't meant to be doing is incorporating language chunks into conlangs. A complex case marking system in nouns is perfectly okay, but something that rips everything but the foundations off Sanskrit's noun case system is obviously not as fine.

I can't even say how many times I've seen languages that look moderately okay from a linguistics point of view, but look so thoroughly uninspired and unoriginal that it's not even funny. A present indicative conjugation of the shape <-m, -s, -t, -mo, -te, -n> just spells laziness. Someone might say "But how is that uninspired? I can see the Latin inspiration!" and, well, they'd be pretty right. There's Latin inspiration in there. About as much of it as there's inspiration in a copy of an original.

What is inspiration then? Well, in my opinion, it's taking a liking to an idea, concept or feature, then adapting it. Taking chunks out of languages isn't inspiration, inspiration is seeing a pattern and making it work elsewhere. Old Irish has this insane system of preverbs and object infixes and mad irregularity; it's perfectly okay if a conlang has something similarly mad, but it becomes questionable when all those "ní"s and "no"s and "indech"s and oddly Gaelic orthographies keep appearing for no reason at all.

Now, don't get me wrong, it is okay to do a heavily-inspired language, but trying to recreate existing languages is what's not conlanging is really about. Inspiration can be anything from just a single syntactical detail up to full paradigm shapes: the amount of it doesn't change its nature.

As an example of an inspired conlang, I'd put forward my Xwarṣa, an a priori language isolate that draws significantly from Avestan and Sogdian but doesn't actually copy them. Poor examples of 'inspiration' can be found pretty much anywhere, and there is no need to name them.