Preface Note: This page is a work in progress and is subject to revision.
- The phones [ŋ], [z], [w], [ɫ], [ɰ] and [ʔ] are not phonemic. [ʔ] is only found in the phonemic /ʔ͡n̩/ co-articulation, but is never used as a stand-alone phoneme. The others are allophonic realizations of /n/, /s/, /u/, /l/, and /h/, respectively.
- The phonemes /m/, /n/, /p/, /t/, /k/, /s/, /ʃ/, /l/, and /t͡ʃ/ have geminated (long) counterparts as additional phonemic consonants.
- There can only ever be up to one gemination (long consonant) in an individual morpheme (root or affix). Morphemes which do contain a gemination will not contain a long vowel (ie: you could have katta or kaata, but not kaatta).
- Long /sː/ can lose its length and become realized as [s] in most cases. This is because the short /s/ phoneme is allophonically realized as [z] in most cases. When a long /sː/ phoneme occurs in a position where a short /s/ phoneme would be read allophonically as [z], the long /sː/ can instead be realized as a short [s]. The gemination is still always written , regardless of whether or not it has a shortened articulation. This shortening is optional. When utilized, it does not affect the stress patterning. The syllable containing the shortened long /sː/ phoneme still carries the weight of a gemination.
- The phones [b] and [f] are not found in any native Duojjin words, but they can be reliably distinguished by Duojjin speakers (who like, y’know, totally exist). These completely not fictional speakers will also be familiar with the letters ⟨b⟩ and ⟨f⟩ as being representative of those sounds, which are included in the orthography for the sake of writing onomatopoeia, foreign words, and foreign names.
- The voiceless bilabial stop /p/ is always aspirated.
- The voiceless stops /t/ and /k/ are always aspirated before /u/, but are otherwise unaspirated.
- The glottal fricative /h/ is realized as [ɰ] intervocalically between mid-vowels (/ɛ/, /e/, /ə/, /o/), or intervocalically between any vowels when the previous syllable also contains an /h/.
- Vowel phones in the chart above are stamped in colours which represent phonemic ‘families’. Every phone which shares the same colour of stamp are all allophones of the same phoneme and are all written the same way orthographically.
- The schwa /ə/ is fully phonemic and forms minimal pairs with other vowels.
- The phones [ɪ], [ʊ], and [ɐ] are not phonemic. They are reduced allophonic variations of /i/, /u/, and /a/ respectively, and are only found in unstressed syllables.
- The reduced allophone [ɪ] is only found in unstressed closed syllables with a voiced onset, but the other reduced allophones [ʊ], and [ɐ] are only found in unstressed open syllables, regardless of onset.
- The allophones of the /y&ɯ/ phoneme are in free variation. Some speakers might choose to use [ɯ] as the reduced form of /y/, but there are no rules either way. A speaker could use exclusively [y], or exclusively [ɯ], or they might go back and forth between the two haphazardly. In IPA transcriptions, this phoneme is always transcribed as /y/.
- The phoneme /e/ only exists in the diphthong /eɪ̯/, wherein the /ɛ/ phoneme is allophonically realized as [e] (being pulled by the off-glide vowel). However, this otherwise phonemic diphthong is usually realized simply as the monophthong [e] in actual speech, especially in fast speech, but also at casual speeds. The diphthong is always fully realized in deliberately slow speech. This reduction causes [e] to behave as though it were a separate phoneme from /ɛ/.
- The close (high) vowel phonemes /i/, /y/, and /u/, as well as the open (low) vowel phonemes /a/ and /æ/ can occur doubled phonemically as a phonetic increase in length.
- Syllables containing a long vowel are always stressed.
The table below lists all of the diphthongs that can be found in Duojjin. In speech, a diphthong does not sound like a sequence of two different vowels; instead, the sound of the first vowel gradually glides into the sound of the second vowel with full vocalization lasting through the whole sound. That is to say, the two portions of the diphthong are not broken by a pause or stress pattern. In Duojjin, diphthongs are considered phonemic units, contrasting with both long vowels and with short vowels. Phonologically, however, most Duojjin diphthongs usually are analyzed as sequences (this in contrast to languages like English, where the diphthongs are best analyzed as independent phonemes). However, the diphthongs /aɪ̯/, /oɪ̯/, and /uɪ̯/ are usually analyzed as independent phonemes, which is reflected by their representation in the orthography wherein they are assigned to the letters ⟨å⟩, ⟨ø⟩, and ⟨ů⟩, respectively.
In the table below, the column on the left represents the onset vowel of a diphthong, and the row along the top represents the off-glide of a diphthong. Empty entries do not exist as diphthongs in any native Duojjin words. Vowel combinations created at morpheme boundaries are read as two distinct syllables when they result in a combination which is blank on this table. However, vowel combinations created at morpheme boundaries which are not blank on this table will be diphthongized, except for ⟨ae⟩ and ⟨äe⟩, which are each realized with two distinct syllables at morpheme boundaries. Within roots, however, ⟨ae⟩ and ⟨äe⟩ are always diphthongized.
Note: A few of these diphthongs are rarely (if ever) used, and so I’m thinking about maybe just axing them entirely.
|⟨a⟩ /ɐ̯/||⟨e⟩ /ɛ̯/||⟨i⟩ /ɪ̯/||⟨o⟩ /o̯/||⟨u⟩ /ʊ̯/||⟨y⟩ /y̯/||⟨ä⟩ /æ̯/||⟨õ⟩ /ə̯/|
|⟨a⟩ /a/||⟨ae⟩ /aɛ̯/||⟨å⟩ /aɪ̯/||⟨au⟩ /aʊ̯/|
|⟨e⟩ /ɛ/||⟨ea⟩ /ɛɐ̯/||⟨ei⟩ /eɪ̯/||⟨eu⟩ /ɛʊ̯/||⟨eõ⟩ /ɛə̯/|
|⟨i⟩ /i/||⟨iõ⟩ /iə̯/|
|⟨o⟩ /ɔ/||⟨ø⟩ /oɪ̯/|
|⟨u⟩ /u/||⟨ů⟩ /uɪ̯/||⟨uõ⟩ /uə̯/|
|⟨y⟩ /y/||⟨yõ⟩ /yə̯/|
|⟨ä⟩ /æ/||⟨äe⟩ /æɛ̯/||⟨äu⟩ /æʊ̯/|
|⟨õ⟩ /ə/||⟨ø⟩ /əɪ̯/|
Note: This section needs to be revisited.
The phonemic template for the first syllable of a Duojjin morpheme is (X(L)/S(P))V(C), where X is representative of a non-liquid consonant; L is representative of a consonant classified as a liquid; S is representative of any consonant classified as a sibilant; P is representative of any consonant classified as a plosive; V is any vowel, long vowel, or diphthong, and C is representative of any consonant (see the Full Phonetic Inventory section below for classification types).
From a more basic perspective, the basic phonemic template for the first syllable of a Duojjin morpheme is (C(C))V(C). The basic phonemic template for a root is typically C(C)VC(C)V(V).
The phonemic template for any further syllables of a Duojjin morpheme is (C)V(C).
The above structure is subject to the following stipulations:
- In onset position, no consonant may follow a liquid, including other liquids.
- Stops and non-sibilant fricatives can only be followed by liquids.
- In clusters within morphemes, the liquid /l/ can only follow voiced consonants.
- In clusters within morphemes, the liquid /ɾ/ can only follow voiceless consonants.
- In clusters within morphemes, the nasal /n/ can only be followed by /j/ or [w].
- In clusters within morphemes, the non-sibilant fricative /v/ can only follow sibilant fricatives.
- In clusters within morphemes, the voiceless stops can only follow voiceless sibilant fricatives.
- In clusters within morphemes, the voiced stops cannot follow any consonants. However, at morpheme boundaries, they can.
- In clusters at morpheme boundaries, any nasal can be followed by any stop, except for /np/.
- In clusters at morpheme boundaries, any nasal can be followed by any fricative, except for /Nh/
- In clusters at morpheme boundaries, /j/ and [w] can follow nasals (except for /ʔ͡n̩/), but other liquids cannot.
- Voiceless sibilant fricatives can only be followed by voiceless stops, /j/, [w], and /v/.
- Voiced sibilant fricatives can only be followed by /j/, [w], and /v/.
- Clusters created at morpheme boundaries which break the above rules are split into two distinct syllables.
- The liquid /ɾ/ cannot directly precede any other alveolar or post-alveolar consonant, regardless of whether or not that consonant crosses the syllable boundary. In such circumstances, the /ɾ/ changes to /l/ both phonetically and orthographically.
- A free root morpheme may never end in a consonant (with a small handful of (probably fossilized) exceptions that end in /n/), but words and bound morphemes can.
- Affricates and co-articulated consonants count as single consonants for purposes of syllable structure.
- A single morpheme will only ever contain up to one gemination or long vowel, but never both, or more than one of either.
- A morpheme will never begin with a gemination.
- A free root morpheme will never begin with a vowel. However, bound morphemes can.
- Only /j/, and [w] cannot appear word-finally. /h/ can appear word-finally as a phoneme, but it is allophonically realized as [ç] in this position, so it will not appear word-finally phonetically.
- Only [ŋ] cannot appear word-initially.
Duojjin stress-patterning is weight sensitive. The heaviest syllable in a word will take primary stress. Every second syllable from the heaviest syllable in both directions will take secondary stress. Every second stressedsyllable from the heaviest syllable in both directions will take primary stress.
By default, the heaviest syllable will be the penultimate (second-to-last) syllable (ie: míri). However, if the penultimate syllable is composed of nothing but an epenthetic sandhi vowel, then the extra syllable weight will move to the antepenultimate (third-to-last) or ultimate (last) syllable, whichever one is heavier. If they are both of equal weight, then it will move to the antepenultimate.
Long vowels and geminated consonants add significant weight to a syllable. Because of this weight, they pull main stress, and the stress sequence will pattern around them as their starting point (ie: víkkipàl). In words which contain more than one such syllable, the final one will be considered a little heavier to break the tie. If the stress pattern created would result in a syllable being stressed which is composed of nothing but an epenthetic sandhi vowel, then that syllable will pass that stress along to the next syllable, rendering the epenthetic syllable unstressed, despite the pattern.
Compound words (words which are composed of two or more roots) sometimes have the stress pattern of the final root morpheme flipped. For example, the word hjeivamiri (good cat) can be realized as hjèivamirí, as opposed to hjèivamíri. As of the time of this writing, this flip is optional. However, I may one day think of some sort of distinction which could/should be drawn from this flip, so it’s subject to some potential future asterisk for now.
Sequences of two consecutive unstressed syllables are possible, but sequences of two consecutive stressed syllables are not.
Diphthongs are heavier than short vowels, but lighter than long vowels.
Full Phonetic Inventory & Orthography:
The following charts illustrate all of Duojjin’s phonemes, including allophonic variants, and their orthographic representations. The left column denotes manner of articulation in consonants and height in vowels. The second column features the phonemes and allophonic variants in Duojjin of the appropriate type. The phonemes are surrounded by /slashes/, and the allophonic variants are surrounded by [brackets]. The allophones are separated from the phonemes in the second column by a |pipe|. The third column features the orthographic representation of each corresponding phoneme and allophone from the second column. These orthographic forms are surrounded by ⟨chevrons⟩. In other words, the second column shows you the sounds of the language in IPA, and the third column shows you how those sounds are spelled/written in the Duojjin alphabet.
Short Consonant Phonemes and Allophones:
|SHORT CONSONANTS||Phonemes and Allophones:||Orthographic Representations:|
|Nasal:||/m/ /n/ /ʔ͡n̩/ | [ŋ]||⟨m⟩ ⟨n⟩ ⟨ŋ⟩ | ⟨ng⟩|
|Plosive:||/pʰ/ /t/ /d/ /k/ /g/ | [tʰ] [kʰ]||⟨p⟩ ⟨t⟩ ⟨d⟩ ⟨k⟩ ⟨g⟩ | ⟨t⟩ ⟨k⟩|
|Sibilant Fricative:||/s/ /ʃ/ /ʒ/ | [z]||⟨s⟩ ⟨š⟩ ⟨jj⟩ | ⟨s⟩|
|Non-Sibilant Fricative:||/v/ /ç/ /h/||⟨v⟩ ⟨q⟩ ⟨h⟩|
|Liquid:||/ɾ/ /l/ /j/ | [w] [ɫ] [ɰ]||⟨r⟩ ⟨l⟩ ⟨j⟩ | ⟨u⟩ ⟨l⟩ ⟨h⟩|
|Affricate:||/t͡ʃ/ /d͡ʒ/||⟨c⟩ ⟨djj⟩|
Long Consonant Phonemes and Allophones:
|LONG CONSONANTS||Phonemes and Allophones:||Orthographic Representations:|
|Nasal:||/mː/ /nː/||⟨mm⟩ ⟨nn⟩|
|Plosive:||/pː/ /tː/ /kː/||⟨pp⟩ ⟨tt⟩ ⟨kk⟩|
|Sibilant Fricative:||/sː/ | [s]||⟨ss⟩ | ⟨ss⟩|
Short Vowel Phonemes and Allophones:
|SHORT VOWELS||Phonemes and Allophones:||Orthographic Representations:|
|High/Close:||/i/ /u/ /y*/ | [ɪ] [ʊ] [ɯ*]||⟨i⟩ ⟨u⟩ ⟨y⟩ | ⟨i⟩ ⟨u⟩ ⟨y⟩|
|Mid:||/ɛ/ /ə/ /o/ | [e]||⟨e⟩ ⟨õ⟩ ⟨o⟩ | ⟨e⟩|
|Low/Open:||/a/ /æ/ | [ɐ]||⟨a⟩ ⟨ä⟩ | ⟨a⟩|
*: The realization of the /y/ phoneme is in free variation between [y] and [ɯ].
Long Vowel Phonemes and Allophones:
|LONG VOWELS||Phonemes and Allophones:||Orthographic Representations:|
|High/Close:||/iː/ /uː/ /yː*/ | [ɯː*]||⟨ii⟩ ⟨uu⟩ ⟨yy⟩ | ⟨yy⟩|
|Low/Open:||/aː/ /æː/||⟨aa⟩ ⟨ää⟩|
*: The realization of the /yː/ phoneme is in free variation between [yː] and [ɯː].
Diphthong Phonemes and Allophones:
|DIPHTHONGS||Phonemes and Allophones:||Orthographic Representations:|
|/i/ Offset:||/aɪ̯/ /oɪ̯/ /əɪ̯/ /uɪ̯/ /eɪ̯/||⟨å⟩ ⟨ø⟩ ⟨ø⟩ ⟨ů⟩ ⟨ei⟩|
|/u/ Offset:||/aʊ̯/ /æʊ̯/ /ɛʊ̯/||⟨au⟩ ⟨äu⟩ ⟨eu⟩|
|/ə/ Offset:||/ɛə̯/ /iə̯/ /uə̯/ /yə̯*/ | [ɯə̯*]||⟨eõ⟩ ⟨iõ⟩ ⟨uõ⟩ ⟨yõ⟩ | ⟨yõ⟩|
|/ɛ/ Offset:||/aɛ̯/ /æɛ̯/||⟨ae⟩ ⟨äe⟩|
*: The realization of the /yə̯/ phoneme is in free variation between [yə̯] and [ɯə̯].
Not much to say here, I guess. This is the native Duojjin alphabet. The letters themselves do not have names yet at this time. I mean, they do have names, but I don’t like them; I need to revisit letter names. So I’m not going to add the names of the letters until after I’ve redone them. But, for now at least, here’s the native Duojjin alphabet:
A C D E G H I J K L M N O P Q R S T U V Y Ä Å Õ Ø Ů Š Ŋ
a c d e g h i j k l m n o p q r s t u v y ä å õ ø ů š ŋ
The letters ⟨b⟩, ⟨f⟩, ⟨w⟩, ⟨x⟩, and ⟨z⟩ are absent from the native Duojjin alphabet. The [z] sound exists in Duojjin as a very common allophonic variant of the /s/ phoneme, and the [w] sound exists in Duojjin as an allophonic variant of the /u/ phoneme before other vowels. However, the sounds [b] and [f] do not exist in any native Duojjin words. Despite this, however, Duojjin speakers ought to have no trouble producing, recognizing, or distinguishing these sounds or letters, and they could theoretically show up in loanwords. I haven’t yet imported any such loanwords, however, and I remain unsure as to whether or not I will. The letter ⟨x⟩ means too many different things in too many languages, but whatever your language uses for ⟨x⟩, it probably doesn’t exist in Duojjin.
I have been wondering for a time if I should add the letter ⟨z⟩ into the native Duojjin alphabet to be used to represent the voiced alveolar affricate /d͡ʒ/, which is presently spelled with the trigraph ⟨djj⟩. It seems to me like the letter ⟨z⟩ would probably be much more practical, much more balanced, and much more natural. But I find myself hesitant, as I’ve been writing it this way for ten years, and I do like my ⟨j⟩s. It might happen. It might not. I’m still unsure.
Layman’s Pronunciation Guide:
The written versions of all the sounds in the language have been outlined above in IPA. However, if you don’t know IPA, that might be less than helpful. If so, then this section is for you. I very much dislike this sort of pronunciation guide, because it is very imprecise by nature. As such, please understand that several items here will be only approximate pronunciations. Not to mention that there’s so many different dialects of English that a pronunciation guide like this will doubtless be rendered useless for speakers of many dialects. Such approximations will be marked with an asterisk (*).
⟨Letter⟩ : Pronunciation
⟨a⟩ : * Like the ⟨a⟩ in father, car, tar, bar, mall, Las Vegas
⟨c⟩ : Like the ⟨ch⟩ in choose, chain, patch, chill, cherry
⟨d⟩ : Like the ⟨d⟩ in dive, dare, dip, pads, dads
⟨e⟩ : Like the ⟨e⟩ in pen, hen, elevator, end, ten, red
⟨ei⟩ : Like the ⟨ay⟩ in day, tray, may, bay, hay, ray, nay, slay
⟨g⟩ : Like the ⟨g⟩ in go, gear, pegs, gap, dogs, leggings
⟨h⟩ : Usually like the ⟨h⟩ in he, home, hill; but sometimes like a ⟨g⟩, except much ‘softer’
⟨i⟩ : Usually like the ⟨ea⟩ in mean, bean, glean; but sometimes like the ⟨i⟩ in pin, tin, fit
⟨j⟩ : Like the ⟨y⟩ in yes, year, yellow, yin-yang, you, yet
⟨jj⟩ : Like the ⟨s⟩ in treasure, measure, pleasure, leisure
⟨djj⟩ : Like the ⟨j⟩ in jump, jack, jell, jolly, jaunt
⟨k⟩ : Usually like the ⟨c⟩ in scar, scope, scary, but sometimes like the ⟨c⟩ in car, coal, cream, cut
⟨l⟩ : Like the ⟨l⟩ in land, lot, live, lever, love, let
⟨m⟩ : Like the ⟨m⟩ in mind, mellow, map, tremor, mouse
⟨n⟩ : Like the ⟨n⟩ in narrow, nine, not, pen, cane, brown
⟨o⟩ : * Like the ⟨o⟩ in hold, told, bold, bone, cone, lone, home
⟨p⟩ : Like the ⟨p⟩ in pine, power, peel, pepper, pale, trapper
⟨q⟩ : * This cannot even be approximated in English. It’s like a hissing cat sound. It’s spelled ⟨ch⟩ in German: ich, mich, dich
⟨r⟩ : Like the ⟨tt⟩ in better, fatter, matter, otter, hotter in some North American dialects.
⟨s⟩ : Usually like the ⟨z⟩ in zebra, breeze, sneeze, brazen; but sometimes like the ⟨s⟩ in snake, sour, stream, sing
⟨t⟩ : Usually like the ⟨t⟩ in stand, sterile, pastor; but sometimes like the ⟨t⟩ in tower, ten, tarrif
⟨u⟩ : Usually like the ⟨oo⟩ in spoon, cartoon, boon; but sometimes like the ⟨oo⟩ in took, rook, hook; and sometimes like the ⟨w⟩ in water, wary, weary, weather
⟨v⟩ : Like the ⟨v⟩ in very, vale, liver, volume, vector
⟨y⟩ : * Like the the ⟨oo⟩ in the way a Scottish person might say food. To produce this sound, say the ⟨ee⟩ sound in teen, but round your lips while you’re doing it. If you don’t feel silly doing it, you’re probably not doing it right.
⟨ä⟩ : Like the ⟨a⟩ in cat, hat, rat, mat, map, tack, jack, trap, smack, back, track, slap
⟨å⟩ : Like the ⟨y⟩ in why, try, bye, fly, rye, shy, sly, my
⟨õ⟩ : Schwa. Like the bold vowels in photograph, recognize, neutral, formal, similar
⟨ø⟩ : Like the ⟨oy⟩ in boy, toy, Roy, or alternatively like <õi>
⟨ů⟩ : * Like ⟨uoy⟩ in buoy or the ⟨ooey⟩ in phooey
⟨š⟩ : Like the ⟨sh⟩ in she, shell, shone, cash, slash, sheep, shot
⟨ŋ⟩ : * Like the ⟨tton⟩ of button, mutton, glutton or the ⟨tten⟩ in smitten, kitten, bitten in some North American dialects.