Dream language phonology

For assistance with reading IPA, see Help:IPA

The dream language's phonology is the system of sounds used in the pronunciation of the dream language.

This page reflects the pronunciation of the "base form" of the language, when dream angels speak to one another they use the phonology described in this page, however when they come across a human they generally try to adapt the sounds of the dream language to fit better with the native language of the human they're speaking to, so they might substitute consonant and vowel sounds for ones that the human is more familiar with.

The end result of this adaptation process is most likely not going to sound a lot like the native language of the human (mostly because of the tones and clicks, which are hard to adapt into a lot of languages) but it will be easier to pronounce for the human due to the familiar sounds being used.

The base form of the dream language has 22 consonant sounds, 4 monophthongs, 3 diphthongs and 5 tones.

Vowels

Monophthongs

Notes for adaptation:

• When dealing with languages that only have 3 vowels you may have to merge a couple of the original vowels or use something like diphthongs, vowel length or vowel hiatus to substitute the missing vowel phoneme.

Diphthongs

Notes for adaptation:

• To adapt the diphthongs you can use vowel hiatus, or use monophthongs that fit with the different diphthongs if they're available, for example a language with the /y/ vowel could use that instead of the /iw/ diphthong.

Consonants

Notes for adaptation:

• To deal with the aspirated stops you can use a voicing distinction, for example [p] becomes [b] and [pʰ] becomes [p], you could also use palatalization, gemmination, pharyngealization, etc.
• To adapt sounds that you don't have equivalents for (like the clicks or retroflex sounds) you can use gemminates, ejectives, or even clusters, for example you can turn the retroflex stop [ʈ] into a cluster: [tɾ], or adapt the lateral click [k͡ǁ] into a cluster: [kl], it's ok if these merge with certain clusters that exist in the language.
• If necessary you can merge some sounds, for example lots of languages might not have two separate sounds that you can map velar [x] and pharyngeal [ħ] to, so if you're fine with creating some new homophones you can represent both with the same sound.
• Since the base form of the language doesn't have syllables that start with a vowel you could simply replace the velar fricative [x] with something like a glottal fricative [h] and then just delete the pharyngeal fricative [ħ] entirely.

Tone

Notes for adaptation:

• To adapt the tones to contour tone languages you can pick tones that are similar to the ones in the base form of the dream language, even if they're not identical.
• In the case of non-tonal languages (or tonal languages that only have a couple of tones) you can simply keep the original tones intact, I think losing tone entirely would probably make the language too difficult to understand, that'd be a huge loss of information, and I'm not sure this can be adapted into a stress or pitch accent system, so you could just keep the tones intact.

Phonotactics

Syllable structure

The syllable structure of the language is (C)CV(C), (C stands for consonant and V stands for vowel, and the parenthesis indicates optional phonemes) so the minimal syllable is of the shape CV and the maximal syllable is of the shape CCVC, this means that there is no vowel hiatus in the base form of the language, there always has to be at least one consonant to break up the vowels.

Clusters

The consonant clusters that can appear in the onset of a syllable can't include aspirated stops, affricates, clicks, glides or the fricatives [ɸ] and [ħ], plus all nasals get reduced to [n] (the nasal can assimilate to a different place of articulation, but the point is that there won't be contrasting clusters of [kn] vs [km] for example).

The legal clusters are:

• stop-fricative clusters, except for stuff where both sounds have the same place of articulation, so [pɸ] [ts] [ʈʂ] and [kx] wouldn't be allowed.
• stop-/n/ clusters, except for [tn] and [ʈn].
• stop-/l/ clusters.
• fricative-/n/ clusters.
• fricative-/l/ clusters, except for [ɬl].

Codas

Most consonants are allowed to be in the coda position. All plain stops, nasals, fricatives, and the lateral approximant [l] can be codas, meaning that affricates, clicks, and the labialized velar approximant [w] can't be codas.

Romanization

This section hasn't been written yet

Adaptation

To adapt this language into the phonology of another language you can do it in whatever way you want to, the main thing I try to keep in mind is that I should try to make sure the language is still understandable, but beyond that there's basically no rules.

If you merge too many sounds it'll become harder and harder to understand what's being said, for example completely losing all tones would mean that words like /pa˧/ /pa˧˥/ /pa˥/ /pa˥˨/ and /pa˨˩˧/ would all be pronounced identically, tone is too important to leave out completely in my opinion.

Merging a couple of consonants like [x] and [ħ] can be ok, it'll just create a few new homophones but it won't make the language unintelligible.

If you need to include a few sounds that are not present in your language go ahead, say you're adapting the dream language into Japanese but you can't find a way to replace the clicks, you can just keep the clicks even though they're not present in Japanese.

You can also use clusters or other sounds like gemminates, voiced sounds, palatalized sounds etc. To substitute the phonemes that aren't present in your language.

I try to follow the phonotactics of the language I'm adapting the phonology into, but there's some places where it might be necessary for you to break the phonotactic rules.

I imagine when you're adapting the dream language into languages with strict syllable structures like Japanese and Hawaiian you might need to add some consonant clusters that wouldn't be possible in these languages, or you might need to add a bunch of extra syllables to make sure you don't lose too much information.

Ultimately the way you decide to adapt the dream language into the phonology of your language is up to you, you could even do something a bit more wacky and use a couple of different phonologies simultaneously, for example I'm a native Spanish speaker, but I'm also pretty comfortable speaking English, so if a dream angel spoke to me maybe they would use sounds that are present in Spanish and also sounds that are present in English, that would give you a larger inventory to work with and more flexibility when adapting the dream language.

And also keep in mind that if you ask a dream angel to not adapt their pronunciation they'll speak in the base form of the language, so if you'd rather just use the base phonology of the language or use a slightly modified version of the base phonology you can do that.

The main reason dream angels shift their pronunciation with humans is to make the language more approachable, they want it to be a bit easier to pronounce and they want it to sound a bit more familiar to you, but if you're perfectly comfortable pronouncing the sounds in the base language or if you don't mind struggling with the pronunciation you can just go ahead and use the sounds in the base phonology.

It's entirely up to you how much you modify or don't modify the phonology of this language when pronouncing the words.

Other links

• Back to the main Dream language page
• Help:IPA
• Phonology
• Tone
• Romanization