The types of words that might get one’s speech stereotyped as “lazy” or “disfluent” or “uneducated” or whatever else if used excessively or in too formal a setting, but which in truth are vital for fluency and listening comprehension.
I dunno, this is just an impression because I don’t interact much with the broader conlang community, but I feel like these words often end up being sort of overlooked by many conlangers. I certainly overlooked them for a long time myself. But to me these words make a language feel that much more alive, you know, that different people talk in different ways with different registers.
Do any of you have any interesting thoughts or experiences with these types of words? How are they handled in your own conlangs?
I guess that those aren’t seen often because they require the phonology and grammar to be already close to finished - or at least enough to know which constructions are used so often that get contracted.
That said I full agree with you, they’re awesome when done right. They’re when the conlang stops being a bunch of sketches in a book to become something living, at least in the mouths (or gestures) of imaginary speakers.
Since the phonology of my main conlang (Tarune) is finished, but the grammar is still heavily WIP, my only progress in this regard was creating a formal register vs. local pronunciations. Not quite what you’re asking about, but close enough, so I’ll share two examples here:
Hiatuses between words
In the formal register you’re supposed to dissolve them with [h]. However, people in Central/Northern cities don’t do this bother in quick speech. Example:
- Romanised: ⟨Sobeca ep Lorā⟩
- Phonemic: /su.bi.ca ip lu.ɾa:/
- Phonetic (formal): [sʊ˥.bɪ.cɐ hɪp lʊ˥.ɾä:]
- Phonetic (Central/Northern urban, quick speech): [so˥.be.cɐɪ̯p lo˥.ɾä:]
- Translation: “Sun and Moon”
And it’s hard to represent in IPA, but Central speakers have a tendency to shorten the long vowels. They’re still distinct from the short vowels, but in quick speech you’re telling who’s who by the quality, not by the quantity.
Rendering of voiced consonants
In the formal register, when a voiced consonant or consonant cluster is near a nasal vowel, you’re “supposed” to nasalise it midway: a single consonant gets pre-/post-nasalised, and in a cluster only one consonant gets nasalised. In practice… well, only people in the coast do this in a natural way. The others either don’t nasalise the consonant at all, or do it fully, like this:
- Romanised: ⟨Duamde⟩, ⟨ṭelsemd⟩
- Phonemic: /dwã.di/, /ʈil.sĩd/
- Phonetic (formal): [dw̃ɐ̃˥.n͜dɪ], [ʈɪl˥.z̃ɪ̃n͜d]
- Phonetic (C/N, informal but increasingly common): [dwɐ̃˥.ne], [ʈel˥.zẽn]
- Phonetic (Southern, #1): [nw̃ã˥.ni], [ʈil̃˥.z̃ĩn]
- Phonetic (Southern, #2): [nw̃ã˥.di], [ʈil̃˥.z̃ĩd]
- Translation: “Southern Wind”, “45 days month/season”
So it’s a lot like the Central/Northern speakers shifted the nasalisation to the right, while Southern speakers either spread it further or shift it left.
EDIT: ah, Southern backchannel ['u:˥˩ʔu]; typically spelled ⟨ōho⟩. This… interjection? has a weird story - it was initially used by cattle herders to direct their cattle. Eventually the usage spread towards humans too, to convey “are you following?”; and then as backchannel, to convey “I’m following it, go on”.
All introduction to conlanging contain some degree of prescriptivism focusing on grammar and vocabulary building, so I’m not surprised when most famous examples also emphasize on them to also ease their introduction. Regular lexicon, consistent spelling and pronunciation are a huge help when learning a new language.
However, I have to agree that irregularity and going off the beaten path is part of human nature. I used to imagine what a future version of current day languages would split into based on regional dialects and features considered “un-educated” but also popular.
