Keeping track of sound changes
-
linguistcat
- Posts: 427
- Joined: Sun Jul 08, 2018 12:17 pm
- Location: Utah, USA
Keeping track of sound changes
Aside from a general list of sound changes occurring in your conlang, how do you keep track of what sound changes happen, when leveling happens, etc?
A cat and a linguist.
Re: Keeping track of sound changes
Write down every word in which an irregular change happens and beneath each one list what sound change and between what other changes did it happen. Save it in the same document/section as you keep the list of regular sound changes.
ìtsanso, God In The Mountain, may our names inspire the deepest feelings of fear in urkos and all his ilk, for we have saved another man from his lies! I welcome back to the feast hall kal, who will never gamble again! May the eleven gods bless him!
kårroť
kårroť
Re: Keeping track of sound changes
The lists of sound changes work just fine for me. I have the lists for Poswa, Pabappa, Bābākiam, and Gold all memorized ... so I can do the diachronics in my head. Ive always preferred to write the sound changes out with descriptive words, as in http://www.frathwiki.com/Languages_of_T ... .284700.29 (which appears in color for me).
Analogical levelling of various grammatical paradigms takes place whenever a sound change opens up a new opportunity. For convenience's sake I assume that the analogy happens immediately, and I haven't yet run into a situation where I'd get a different outcome if the analogy happened a few hundred years later on, so I'm fine with that for now.
For irregulars, I try to define them away as much as possible .... irregular sound changes can be fit into one of these categories:
1) rare but reliable sound changes that affect only a few words. For example, in Khulls, /ʕəs/ > /ż/ unconditionally. Such words are so rare that the /ż/ seems out of place in the language, but the shift is regular and there are no examples where it fails to complete.
2) sound changes across morpheme boundaries that can resolve in more than one way. In Poswa, the sequence /sb/ can shift to /sp/, /žb/, or /b/, depending on when the compound containing the sequence was formed. I consider all three outcomes to be regular and i list the etymology with the two words separated by a space to remind myself that it was originally a compound. This category can include exceptions to category 1), where an already rare sound change becomes even rarer by the introduction of later-coined counterexamples, but these will be held distinct because of the space in the phrase given in the etymology.
3) Placenames, given names, and frequently used words that disobey all the rules. I avoid these in my conlangs whenever possible, because the phonologies of my languages are simple enough that there isn't really a great need to further reduce commonly used words and names. These are the only type of words that I mark as irregular; when I do, I put a note after the etymology saying "as if from /pi pa/" or something similar, to indicate that the speakers could reanalyze the irregular word as a regular one whose origin was a different word or phrase. This matters because my languages are highly fusional and even placenames need to change their stems to take on various inflections.
Analogical levelling of various grammatical paradigms takes place whenever a sound change opens up a new opportunity. For convenience's sake I assume that the analogy happens immediately, and I haven't yet run into a situation where I'd get a different outcome if the analogy happened a few hundred years later on, so I'm fine with that for now.
For irregulars, I try to define them away as much as possible .... irregular sound changes can be fit into one of these categories:
1) rare but reliable sound changes that affect only a few words. For example, in Khulls, /ʕəs/ > /ż/ unconditionally. Such words are so rare that the /ż/ seems out of place in the language, but the shift is regular and there are no examples where it fails to complete.
2) sound changes across morpheme boundaries that can resolve in more than one way. In Poswa, the sequence /sb/ can shift to /sp/, /žb/, or /b/, depending on when the compound containing the sequence was formed. I consider all three outcomes to be regular and i list the etymology with the two words separated by a space to remind myself that it was originally a compound. This category can include exceptions to category 1), where an already rare sound change becomes even rarer by the introduction of later-coined counterexamples, but these will be held distinct because of the space in the phrase given in the etymology.
3) Placenames, given names, and frequently used words that disobey all the rules. I avoid these in my conlangs whenever possible, because the phonologies of my languages are simple enough that there isn't really a great need to further reduce commonly used words and names. These are the only type of words that I mark as irregular; when I do, I put a note after the etymology saying "as if from /pi pa/" or something similar, to indicate that the speakers could reanalyze the irregular word as a regular one whose origin was a different word or phrase. This matters because my languages are highly fusional and even placenames need to change their stems to take on various inflections.