In the previous post, I talked about how easy Gen is. In this post, I’ll show what a generator in Gen looks like and what the limitations are. I’m working on a language called Maanxmuʃt, which I had decided I wanted to be Germanic-like so that there would be a bunch of sound changes when words where borrowed, since Nyji words are basically CVC. (I actually stole the phonology and phonotactics from some Germanic language, I don’t remember which) So I wrote up some guidelines for the phonotactics.
Words take the form of (C)(C)CV(C)(C)(C), where the onset can be:
any consonant
ʃ(any non-voiced nasal, stop, or fricative)1(r or l)
a stop or fricative + r or l, but not dl or vl
1 = m, n, p, t, k, θ, x
Codas consist of:
any non-voiced consonants, but r and ʀ can’t follow diphthongs and m can’t follow long vowels or diphthongs
(r or l)x
x(ʃ or t)
n(coronal and dorsal stops and fricatives)2
2 = t d k ʃ θ x
Long vowels are rare in unstressed syllables and the stress is always on the first syllable. ɦ and ʀ only appear outside of consonant clusters. After a nasal x changes to d. And after a vowel h changes to x.
Then I had to convert that to Gen, which resulted in the below (with category types noted):
Vowels = V Unvoiced Stops and Fricatives = F Consonants = C Consonants minus d & v = X Stops and Fricatives except glottal = S V=iueoaɐɑɒɓɔIUEOA R=rl T=ʃt F=tdkʃθx C=rʀθmnʃptdkvxɦljw X=mnptkʃθxɦljwrʀ N=mn S=ptdkʃvθx Rewrite rules: ɐr|ir ɐʀ|iʀ ɑr|er ɑʀ|eʀ ɒr|ar ɒʀ|aʀ ɓr|ur ɓʀ|uʀ ɔr|or ɔʀ|oʀ ɐm|im Im|im ɑm|em Em|em ɒm|am Am|am ɓm|um Um|um ɔm|om Om|om I|ii U|uu E|ee O|oo A|aa ɐ|iu ɑ|ei ɒ|ie ɓ|uo ɔ|ou dl|dr vl|vr mx|md nx|nd aɦ|ax eɦ|ex uɦ|ux oɦ|ox iɦ|ix Syllable types: CV CVX CVxT CVnF CVRx ʃNRV ʃNV SRV ʃNRVX ʃNVX SRVX ʃNRVxT ʃNVxT SRVxT ʃNRVnF ʃNVnF SRVnF ʃNRVRx ʃNVRx SRVRx
Definitions have to be one letter, so long vowels I turned into capitals (if you do this, make sure you don’t use those capitals as categories!) and diphthongs into more or less random unicode. The rewrite rules fixed that, as well as the two phonotactics rules. Then I had to figure out all the possible syllables. You’ll notice I can’t do anything with stress, at least not easily.
Running it on the default settings (except output type changed to list) gives results like this:
ʃnlimʀanʃmu
ʃneθaxʃʃmiʃriruxʃ
vrinθʃnlu
monkʃniʃnrixtʃraxʃ
tret
ʃmimʀond
driiθupʃmoθ
muninʃ
ʃnlummink
ʃmulxʃu
θuxʃʃnrom
I’m not fond of the long words, so switching the monosyllable frequency to “mostly” changes it to this:
ʃnrei
plent
θuxʃ
ʃmlo
θla
mexʃ
ʀe
ʃexʃ
vrorx
ʃmiʃ
ʃnli
These results are fine, but repetitive. I could fiddle with the order of the definitions or the dropoff, but it’s tedious. Usually what I do is copy the list to a text file and start deleting things, followed by picking favorites or using random.org to pick randomly (which helps somewhat with having too many similar words if certain sounds appeal to you more).
So that’s alright if you want something quick and dirty. Next time we’ll look at Inspiration Pad Pro, which gives you more control.
Enjoy this post? Support me on Patreon or Ko-fi
