Hello.
I had an idea to make English, but each phoneme is swapped for an opposite. This is easy to do for vowels - for example, /i/ and /u/ are opposites, because one is front and the other is back. I tried to figure out a similar system for consonants and have been stuck ever since. So, how would you calculate an opposite for /m/, or /h/, or /r/? I’m sure there must be a way to do it, but for the life of me I can’t figure it out.
If /i/ is the opposite of /u/ due to front/backness, what happens to central vowels? If you also flip height, what happens to schwa?
You could “flip” consonants in a similar kind of way; place and manner of articulation are essentially two linear dimensions (with place being how far forward or back in the mouth the consonant is articulated, and manner being how much the airstream is interrupted - but this is approximate). In this scheme, /b/ would become /ʔ/. There would be some difficulty because not all consonants can be made or distinguished in practice.
Is this cipher supposed to be spoken or written? You will have difficulties either way: If spoken you will end up with sounds not in the English sound inventory, so they will be hard for speakers of English (or probably: of any single particular language) to say. Some pairs of sounds that are distinct in English will be mapped to allophones. If written you will end up with sounds where there is no standard way to denote them except with IPA, in which case you’ve really made an alphabetic substitution cipher where you:
In this scheme it doesn’t actually matter that you’re “flipping” the sounds; the transformation could be anything and it would be encoded, decoded and cryptanalysed in the exact same way.