In my introductory posting to this website I wrote, “Another new thing I’ve included is a blog category on ‘Grammar musings.’ […] [They] will hopefully give you some insight into the process that is behind conlanging.” Here is one such insight, and a very basic one at that.

The other day I remembered an episode from when my dad learned – fascinatedly – that I’m doing this language inventing stuff. He asked me first, what e.g. the word for ‘house’ is (nanga) and then, why it’s not something different. The same question also comes up in the forums every now and then.

Case 1: Make up a word from scratch

As I found out in an analysis of my dictionary that I did last year, Ayeri has a number of phonotactic constraints that restrict words considerably. Some of them arose purposefully, others accidentally. Based on this analysis, I programmed a script to make up words for me. However, this does not mean I simply take those generated words and fully randomly assign a meaning to them. Technology and statistics aside, I have an idea in my mind about how the language should sound like and all analysis was done in retrospect. In fact, I often only take my list as a way to help me find a suitable word if I can’t think of one offhand. For example, right now I still need a word for ‘poison’ and don’t feel like duplicating German Gift, lit. ‘something given’ (with some semantic drift …). Using my list for inspiration, I find that I somehow like mikam. There is no word that begins with mik- in the dictionary yet, so I don’t need to tweak it further. But how and why? Why not kotas or desay? I don’t know! Incidentally, I think kotas has something piercing that fits ‘thorn’ (k-t-s sounds hard and pricky) and desay sounds like it could best be an adjective, by analogy with other adjectives in -ay (atay, dakay, gibay, kebay, …).

Case 2a: Extend an existing word’s meaning

This is something I find myself doing a lot, because it’s boring to have 1:1 equivalents to German-English-French words. As an example, take sihiru- ‘to translate’. I wanted to translate ‘to adopt’ the other day and was thinking about whether to coin a new word, or to reuse an old one. I decided for the latter strategy and after a little brainstorming, I thought that translating is also a way to ‘adopt’ a text into one’s own language, thus another possible meaning of sihiru- could be ‘adopt’.

Another example is pray ‘smooth’. When I made that word for Conlang Relay 18, I had to make it up from scratch. I also used it when I needed a word for ‘even number’. ‘Smooth’ and ‘even’ seem to be very English-y by being synonyms, but ‘odd number’ does not re-use the Ayeri word for ‘odd’. Instead, I chose baras ‘rough’ for consistency.

Case 2b: Extend an existing word’s meaning by changing its noun class

Ayeri distinguishes two noun classes, animate and inanimate. Sometimes it’s neat to add a meaning to a word not simply by extending it, but by also by changing its noun class. One such example is the word for ‘navel’, terpeng. This word existed previously as the inanimate terpeng ‘middle’. However, body parts are animate neuter in Ayeri, since they are things that are associated with living entities, thus asking for a category switch. A change in animacy can thus be used to derive a new meaning, whether motivated by grammatical constraints or freely.

Case 3: Derivation from existing words

Take the word minjisān ‘candidate, electee’ for example. I used it in a previous posting and commented on how unwieldy I found it. Nonetheless, let’s have a look at how I made it. First of all, I needed a word that means ‘candidate’. A candidate in this case was someone who is set up as an electable person. Someone to choose, one could say. Searching my dictionary for possible words to derive this from, I found mindoy- ‘to choose’ and mindoyam ‘choice, option’. Since the choosing is applied to someone, I added the causative suffix -isa to the verb, which is a valid way in Ayeri to derive a non-noun with a resumptive meaning – English would make that ‘chosen’ as an adjective. This results in mindoyisa, which then got nominalized to mindoyisān. Since that’s a mouthful at four syllables, I applied reduction and got minjisān.

Case 4: Nick etymologies, but reasonably so

So you have a word, say, ‘bunch’. A ‘bunch’ in English can refer to a number of things, but let us focus on this meaning:

A collection or cluster of things of the same kind, either growing together (as a bunch of grapes), or fastened closely together in any way (as a bunch of flowers, a bunch of keys) (OED, “Bunch.”)

I suspected that ‘bunch’ is maybe somehow related to ‘bind’, as it’s the case in German:

Bund […] ist eine Bildung zu dem unter binden behandelten Verb und bedeutet eigentlich “Bindendes, Gebundenes”.[2. “Bund ‘bundle’ (…) is a formation belonging to the verb discussed under binden ‘to bind’ and in fact means ‘binding thing, bound thing.'”] (Duden Herkunftswörterbuch, “Bund.”)

Upon further investigation, though, I found out that it is of unknown origin and possibly onomatopoeic (cf. OED, “Bunch.”). But anyway, the German etymology doesn’t seem unreasonable for other languages to come up with independently, so let’s simply look whether there’s a word for ‘to bind’ already. And indeed, there is: disy-. Thus, applying regular nominalization, the word for ‘bunch’ in the meaning above is disyan. And it is probably best categorized as inanimate, since it does neither refer to a living thing, nor would I associate it with one off the top of my head (food is inanimate). There should also be a possibility to pluralize it.