April 28, 2011 § 1 Comment
Here’s a phonology I cooked up in a few minutes that’s pretty much a copy of Japanese, with some of my own touches.
The voiceless labial stop /p/ only occurs doubled; historically, it has merged with /h/ when single. The glottal fricative is [ɸ] before /u/. The siblants /ts dz s z/ are palatalized to [tɕ dʑ ɕ ʑ]. The voiced affricate usually loses it’s stoppage between vowels, becoming either [z] or [ʑ]. The sonorant /r/ is a flap unspecified for lateralness.
I’m thinking of adding /x/ for some flavor/
The vowels are pretty simple. The mid vowels are laxed before a syllable final consonant to [ɛ ɔ] and are always lax in diphthongs and triphthongs. The vowels /y u/ do not occur in the same words, participating in vowel harmony.
The chart above shows the cooccurance of the glides with vowels and falling diphthongs. The glides do not occur before homoorganic high vowels.
The maximum syllable is CGVC, where C is a consonant, G a glide and V a vowel or falling diphthong. The glide /j/ occurs with all initials, while the glide /w/ occurs only after velar initials or /m/. Word initial, there may be an open onset. Word medially, there must be an onset, a glide, or both.
Syllable final consonants are severely restricted. Word internally, there may only be a copy of a following voiceless obstruent or nasal /ŋ/ . Long /ss/ > [tts]. These stops obstruent sequences are often realized as a [ʔC] sequence. The nasal is realized as slightly postvelar in this position. Before labials, the nasal becomes labio-velar: [ŋ͡m]. Word finally, the choices are slightly larger: /t k ŋ/ all occur.
April 26, 2011 § 2 Comments
This is a phonology I thought up last night, intended for a close-to-monosyllabic isolating language I’m thinking about making. The consonants are shown below:
There are five simple vowels: /i e a u o/ which may occur short or long. The high vowels /i u/ may occur as glides, yielding both rising and falling diphthongs, as well as triphthongs. These are shown below:
The glides /i u/ are spelled <j> and <w> after a vowel and word initially, respectively.
All morphemes are mono- or bisyllabic. Monosyllables have the form V, VC, CV, or CVC where C is a consonant and V is a vowel or diphthong or triphthong. The initial C slot may be filled by any consonant, though there are some restrictions based on the following V. The labial fricative /f/ may only precede a rounded vowel or glide [u̯]. Before glide [i̯], the sibilant fricatives have all palatalized historically, leaving only /tɕ dʑ ɕ ʑ/. This does not occur before vowel [i].
Codas are much more restricted: only /h t k n ŋ/ occur. The voiceless finals /h t k/ only occur after the short vowels and rising diphthongs. The nasal finals may occur after the short, long, and rising diphthongs. The falling diphthongs and triphthongs only occur in open syllables. The glottal fricative does not surface as an actual coda, but instead lowers the tone of the preceding syllable nucleus. This is written as a grave over the vowel: kiè ‘room’ is /kieh/ [ki̯e˨].
Bisyllabic words consist of a main syllable of the form CV or CVC (limited by the above form rules) and a presyllable of the form (C)V(h, N), where C and V are both extremely limited. Only three underlying vowels occur in presyllables: /e u a/. These are realized [ɛ ʉ a], respectively, except before main syllables containing [i] or on-glide [i̯], in which case /u/ > [ɨ]. Presyllables may be open or begin in a consonant from the following set: /p t k s h m n l/. Presyllables may be open or closed by /h/ or a placeless nasal /N/. After open presyllables, all initials occur. After /h/, an obstruent must be voiceless. After a nasal presyllable obstruents must be voiced or /h/. The nasal assimilates to the following place of articulation.
April 20, 2011 § 4 Comments
As happens to me a lot, a phonology goes through several stages in the beginning phases of a language. As can be seen from my post last night on some adjustments I’d just made, the phonology is still in a high state of flux. I’ve come to a point where I might be changing a key aspect of the phonology I first posted about, and I’d like some feedback. There are two options:
1) Leave the phonology as it is, plus the changes from yesterday. This would leave coronal harmony in place.
2) Scrap coronal harmony. Move /t̪/ to /d/ and add a voiced labial stop /b/. Also, to compensate further for the lack of coronal harmony, add /ɬ/, perhaps with some kind of remnant coronal relationship with /ts/.
Anyway, if anyone has ideas/feedback about this, I’d really like to here it.
April 19, 2011 § Leave a comment
P1 syllables may begin and end in exactly one consonant. Word initial syllables may begin in a vowel, a glide, or a consonant (or CG combo), while word internal syllables must begin with either a consonant or a glide. There are no syllable-internal consonant clusters, and those consonant clusters that do appear are the result of a closed syllable followed by the next onset. There are some general principles as to how this clustering works:
1) The consonants /h ð̞/ never occur in clusters, and /ɾ l/ do not occur as the second member of a cluster
2) There are no clusters with two identical consonants
3) Nasals assimilate to the following obstruent when they occur in a cluster
4) Stops/the affricate may only be followed by obstruents
Based on these principles, the valid consonant clusters that occur in stems can be summarized in the following table. Final consonants are represented along the vertical access, initial consonants along the horizontal. “N” indicates a nasal that assimilates to the following obstruent An X indicates that the cluster occurs, while lack of one indicates that it does not:
So, as can be seen from the chart above, most consonants can stand in the coda of a stem internal syllables. However, the consonants that can occur in word and stem final syllables are much more limited. They are /t̪ t ts m n k l/.
April 19, 2011 § 2 Comments
I’ve had a thing for tiny, minimalist phoneme inventories lately. Here are a couple of them. One of these may end up being recruited to serve as the mother for an extremely isolating artlang idea I’m playing around with.
The first one has 10 consonants and 4 vowels, for a total of 14 phonemes
/t k ʔ m n ŋ s h ʋ l e a u o/
Valid consonant clusters number four: /ts ks hl hʋ/. They probably only occur intervocalically. The second inventory is inspired by Lilipu, sort of, and is generally kind of polynesian. It has 11 phonemes, 8 consonants and 3 vowels. I may substitute /e/ for /i/ if i end up using this:
/p k ʔ f s m n l a i u/
This one would have a CV syllable structure. No consonant clusters.
April 18, 2011 § Leave a comment
As mentioned in my previous post, P1 syllables conform the shape (C)(G)V(G)(C), where C is a consonant, G a glide (/i u/), V a vowel and parenthesis indicate optionality. So, a vowel may be flanked on both the left and right by a glide, derived from the high vowels /i/ and /u/. Glides may not precede an identical high vowel: */ii/ and */uu/ are therefore disallowed. If such a sequence would occur, one of the high vowels is lowered: /ii/ > [ie] or [ei] and /uu/ > [ou] or [uo].
The sequences /eu au/ have undergone monothongization, becoming [ɯ] and [ɔ], respectively. Thus, the word keu ‘tree’ is phonetically [kɯ]. The fact that are vowel + glide sequences and not simple underlying vowels is confirmed by the fact that when a vowel initial suffix is attached to a word ending in such a sequence, it breaks and syllabifies as either [e.wV] or [a.wV]. For example, the plural of keu is keue [kewə].
Sequences of non-high vowels are not permitted in P1. If such a sequence occurs because of affixation, there are two processes of simplification: 1) two identical non high vowels (/ee aa oo/) simplify into one or 2) in a sequence of dissimilar non-high vowels, one of the mid-vowels will raise to become a corresponding high vowel.
April 18, 2011 § 15 Comments
Over the past week or so, I’ve had an idea (or group of ideas) rumbling around my brain for a conlang heavily inspired by Matt Pearson’s brilliant language Tokana (seriously, follow the link, it’s an awesome conlang). The project is as of yet without a name, so I’ll be referring to it as “project 1” (or P1, maybe?) until I settle down on a name.
Anyway, I’d like to post the first sketch of what I’m thinking for the phoneme inventory and the general shape of the phonology.
P1 has a rather cute little consonant inventory, with just 12 phonemes (orthographic representations in bold):
/p t̪ t k ts s h m n ɾ l ð/ p th t k ts s h m n r l j
The sound /ð/ is a somewhere between a voiced interdental approximant (close transcription /ð̞/) and interdental fricative, though it is generally produced with very little frication. The dental stop /t̪/ is also produced interdentally. The remaining coronal consonants differ in their aritculation: /t ts/ are always alveolar, while the phonemes /s n ɾ l/ are underspecified for the dental-alveolar distinction. This distinction as important, as the two places participate in a harmonic process, detailed below.
In addition to the 12 consonantal phonemes above, there are two glides, [j] and [w] which are allophones of the high vowels /i/ and /u/, respectively.
The language has a simple six vowel system that’s actually rather bland: /i u e o a ə/, represented orthographically as i u e/è o a e. There is no length contrast. The schwa is a marginal phoneme, occurring only in final syllables, where it is spelled e; The phoneme /e/ in final syllables is spelled è. As mentioned above, the high vowels /i u/ occur as glides, becoming [j w] when appearing adjacent to another vowel.
A central aspect of P1 phonology is the presence of harmonic system in the coronal consonants. There are three classes of coronal consonants:
Dental- /t̪ ð/
Alveolar- /t ts/
Neutral- /s n ɾ l/
In any given word, dental segments and alveolar segments may not coexist, while neutral coronal segments may cooccur with either type. The neutral type of segment is underspecified for the distinction; when occurring in a word alone or with an alveolar consonant, their values are alveolar. When occurring with a dental segment, they harmonize to the dental place, becoming: [θ n̪ ɾ̪ l̪].
This feature plays an important role in morphology, as there are bound morphemes contain a non-neutral (‘harmonizing’) segment. When attached to a base with a coronal segment differing in POA from the, the bound morpheme harmonizes with the base. This sumarized in the table below:
So, the causative suffix -tsa will remain as such when suffixed with the verb stem tukia ‘plant’, yielding [tukjatsa] ‘make X plant’, but will become -sa [θ] after the verb stem thu ‘say’, yielding thusa [t̪uθa] ‘make X say’.
P1 has a syllable structure of (C)(G)V(G)(C). Where V is a vowel, G a glide /i u/ [j w], and C a consonant. So, every syllable consists minimally of a vowel, and this vowel may be optionally preceded and followed by a glide and one consonant. Word initially, any consonant except /ɾ/ may occur in initial position. Word medially after a vowel or glide, any consonant may open a syllable. All word medial syllables must begin in either a consonant or glide (no V.V sequences). Final consonants are more constricted, and I haven’t figured out this part very well. Word medial coda consonants are less restricted than word final codas.
The language has a preference towards open syllables wherever possible. Syllabification is pretty straight forward, and guided by the following rules:
VGV > V.GV
VCV > V.CV
VCGV > V.CGV
VGCV > VG.CV
VCCV > VC.CV