Writing Klingon in UCAS (Unified Canadian Aboriginal Syllabics), part 1: Inuktitut-style

[pne]

So I’ve been wondering whether it would be possible to write Klingon in UCAS.

One problem is that the syllabics were developed for languages with fewer consonants and fewer vowels.

My first attempt was to start from Inuktitut writing conventions: both because the Inuktitut variant of syllabics was the one I’m most familiar with and because it has writing conventions for syllables starting in /q/ and /ŋ/, both of which Klingon needs.

For vowels, I had the idea that I could use the Inuktitut three-vowel system as a base and then differentiate e from I and o from u by using the dot above that usually marks vowel length, since Klingon has no phonemic vowel length. (Later, when reading about Carrier syllabics—more on that in part 2—, I also saw a version that uses the dot above only for o/u, distinguishing e/i by orientation. Also a possibility, though slightly less symmetric.)

I also got the idea of using the side-dot (either left-dot à la eastern Cree or right-dot à la western Cree) that is traditionally used for labialised consonants or medial -w- to mark voiceless/voiced distinctions. This could work well for the p/b and ch/j distinctions, and possibly also for the H/gh and t/D ones.

I’m not sure quite what the H/gh distinction is; the Klingon Dictionary claims that gh is “the same as Klingon {H} (see below), but with the vocal cords vibrating at the same time”, which implies that they are at the same point of articulation, i.e. voiced and voiceless velar fricatives (since gh has “the tongue in the same position it would be in to say English <g> as in <gobble,>”); but it also says that H is “just like <ch> in the name of the German compose <Bach>”, which I consider a uvular fricative. On the other hand, uvular q has “the main body of the tongue touch[] the roof of the mouth at a point farther back than it does for {gh} or {H.}”, so I suppose H is velar after all and H/gh are a pair distinguished only by voicing.

So the only not-quite-pair is t/D, since t is “as in English”, i.e. alveolar, while D is retroflex. But since there is no voiced alveolar plosive and no unvoiced retroflex stop, the two can probably be paired together.

Some of the assignments were pretty straightforward. For example:

• glottal stop went to the vowel-initial syllabics, since Klingon doesn’t have any V or VC syllables: 'a would be ᐊ.
• w went to the vowel-initial sign with dot beside: ᐗ wa. (I’m using left-dot in this and subsequent examples, but I haven’t decided whether I like left-dot or right-dot better. It should be possible to substitute all left-dot syllables with right-dot ones since I think they’re always encoded in Unicode in pairs.)
• p/b went to p-: ᐸ ᑄ pa ba. (Or vice versa; I haven’t decided whether the voiced or the voiceless one of a pair gets the side-dot.)
• t/D went to t-: ᑕ ᑡ ta Da.
• m went to m-: ᒪ ma.
• n went to n-: ᓇ na.
• l went to (Inuktitut) l-: ᓚ la.
• y went to y-: ᔭ ya.
• r went to (Inuktitut) r-: ᕋ ra.
• v went to (Inuktitut) v-: ᕙ va.

Another one that presented itself was using ᕦ for tlha: a modified form of /t/ for an affricate starting with [t].

Ones where I was less sure were:

• whether to assign S (a retroflex fricative) to ᓴ (the more common sign, usually for s /s/) or to ᔕ (less common; usually for š /ʃ/). I’ll probably go with ᓴ, though.

• Inuktitut has no /rq/ clusters, to ᖃ is unambiguously qa and can’t be read as ᕐᑲ rka (ᕐ = r; ᑲ = ka). But Klingon has no phonotactic constraints on consonant clusters that I’m aware of: any coda can be followed by any initial. So there, ᕐᑲ would probably be confused with ᖃ easily, and so the ᖃ series probably won’t be used after all.

• which sounds to assign to ᑲ and ᒐ. The former is usually k /k/, and so q seemed appropriate, while the latter is g /ɣ/ in Inuktitut, so seems appropriate for gh.

  However, ᒐ was originally c /ts/ (in the Athabaskan languages that syllabics were first designed for), so ᒐ ᒜ could also be cha ja.

  Also, if ᒐ ᒜ isn’t cha ja, there’s no obvious (to me) pair that could be used for those sounds. On the other hand, there’s no obvious pair for gh H except for ᒐ ᒜ, either.

  Also, if I use ᖓ for ng-, then having ᒐ be ch- seems odd. But then, /ŋ/ isn’t particularly close to /ɣ/, either, yet Inuktitut probably manages fine. (Though they do have the same place of articulation, at least.)

  An additional consideration for the assignment of gh is the fact that Klingon allows either CV or CVC syllables, plus the three additional syllable codas rgh w' y'. And it seems tempting to assign a possible final to the rgh ending: specifically, ᖅ (Inuktitut /-q/) seems really tempting. This would then confirm that ᕋ should be r, and would seem to imply that ᑲ should be gh.

  This leaves open how to represent q and Q… my current lame idea is to put q on ᕹ and Q on ᖬ, but neither is particularly satisfying.

So that’s syllable onsets and nucleuses taken care of; now how to deal with syllable codas.

Here, there are traditionally two systems: a western one which uses geometric shapes to indicate codas (or, often, freestanding consonants in clusters, if the language allows for clusters in syllable onsets), and an eastern one which uses reduced-size copies of the syllable (typically the -a shape/orientation) to indicate a coda: so a hypothetical tat would be either ᑕᐟ (western style) or ᑕᑦ (eastern style).

I decided to make use of these two methods to have distinct letters for the voiced pairs: so for p/b, for example, the western final can stand for the voiced consonant and the eastern final for the voiceless one (or vice versa).

Based on the fact that -rgh “wants” to be represented by ᖅ, I suppose that the voiced consonants should get the eastern finals. This may mean that undotted shapes should be voiced consonants, so that you have pairs like ᐸ:ᑉ and ᑄ:ᑊ; ᑕ:ᑦ and ᑡ:ᐟ.

The full list, at this point, is ᐊᐞ 'a'; ᐗᐤ waw; ᐸᑉ bab; ᑄᑊ pap; ᑕᑦ DaD; ᑡᐟ tat; ᑲᒃ ghagh; ᑾᐠ HaH; ᒐᒡ jaj; ᒜᐨ chach; ᒪᒻ mam; ᓇᓐ nan; ᓚᓪ lal; ᓴᔅ SaS; ᔭᔾ yay; ᕋᕐ rar; ᕙᕝ vav; ᕦᕪ tlhatlh; ᕹᕻ qaq; ᖓᖕ ngang; ᖬᖮ QaQ. For the coda consonant clusters, next to ᖅ -rgh, I propose ᕀ -y' and ᕽ -w'.

Some notes:

• I’m not completely happy with ᐞ U+141E CANADIAN SYLLABICS GLOTTAL STOP due to the orientation; I would have preferred a left-pointing triangle, but that doesn’t seem to exist. ᣟ U+18DF CANADIAN SYLLABICS FINAL RAISED DOT might work better (and that sign is apparently used for glottal stop in Carrier), but since it’s a comparatively new addition to Unicode (added in 5.2, I think), font support is likely to be poor. So I’ll go with ᐞ for now.

• ᐤ seems to be the standard for final -w.

• ᕀ is western Cree for -y, so it’d be the final for a hypothetical *ᔹ syllable, but since I don’t need that one, I’ll use the final for -y'.

• ᕽ is originally -hk. Its cross shape seems to go well with ᕀ for -y', which is the main reason I chose to use it for -w'.

  An alternative would be ᣜ/ᣝ U+18DC/D CANADIAN SYLLABICS EASTERN/WESTERN W, but again, font support is likely to be poor due the placement of those characters in the new-to-Unicode-5.2 UCAS Extended block. But besides, dotted initials correspond to geometric, not iconic, finals in this system.

So the Klingon pangram qajunpaQHeylIjmo' batlh DuSuvqang charghwI' 'It would be rendered as ᕹᒎᓐᑄᖮᑶᔾᓕᒡᒧᐞ ᐸᕪ ᑐᓱᕝᕹᖕ ᒜᖅᐐᐞ ᐄᐟ.

And the complete table would be:

Consonant \ Vowel	a	e	I	o	u	Final
'	ᐊ	ᐃ	ᐄ	ᐅ	ᐆ	ᐞ
w	ᐗ	ᐎ	ᐐ	ᐒ	ᐔ	ᐤ
b	ᐸ	ᐱ	ᐲ	ᐳ	ᐴ	ᑉ
p	ᑄ	ᐼ	ᐾ	ᑀ	ᑂ	ᑊ
D	ᑕ	ᑎ	ᑏ	ᑐ	ᑑ	ᑦ
t	ᑡ	ᑙ	ᑛ	ᑝ	ᑟ	ᐟ
gh	ᑲ	ᑭ	ᑮ	ᑯ	ᑰ	ᒃ
H	ᑾ	ᑶ	ᑸ	ᑺ	ᑼ	ᐠ
j	ᒐ	ᒋ	ᒌ	ᒍ	ᒎ	ᒡ
ch	ᒜ	ᒔ	ᒖ	ᒘ	ᒚ	ᐨ
m	ᒪ	ᒥ	ᒦ	ᒧ	ᒨ	ᒻ
n	ᓇ	ᓂ	ᓃ	ᓄ	ᓅ	ᓐ
l	ᓚ	ᓕ	ᓖ	ᓗ	ᓘ	ᓪ
S	ᓴ	ᓯ	ᓰ	ᓱ	ᓲ	ᔅ
y	ᔭ	ᔨ	ᔩ	ᔪ	ᔫ	ᔾ
r	ᕋ	ᕆ	ᕇ	ᕈ	ᕉ	ᕐ
v	ᕙ	ᕕ	ᕖ	ᕗ	ᕘ	ᕝ
tlh	ᕦ	ᕠ	ᕢ	ᕤ	ᕥ	ᕪ
q	ᕹ	ᕵ	ᕶ	ᕷ	ᕸ	ᕻ
ng	ᖓ	ᖏ	ᖐ	ᖑ	ᖒ	ᖕ
Q	ᖬ	ᖨ	ᖩ	ᖪ	ᖫ	ᖮ

with the additional finals (mentioned above) ᖅ ᕀ ᕽ -rgh -y' -w'.

And if I go the Kirkby 1881 Northeast Athabaskan route, then 'e we be pe De te ghe He je che me ne le Se ye re ve tlhe qe nge Qe would be ᐁ ᐌ ᐯ ᐺ ᑌ ᑗ ᑫ ᑴ ᒉ ᒒ ᒣ ᒬ ᓀ ᓓ ᓭ ᔦ ᕂ/ᕃ* ᕓ ᕞ ᕴ ᙰ ᖧ, and 'I wI bI pI DI tI ghI HI jI chI mI nI lI SI yI rI vI tlhI qI ngI QI would lose their dot, turning into merely ᐃ ᐎ ᐱ ᐼ ᑎ ᑙ ᑭ ᑶ ᒋ ᒔ ᒥ ᒮ ᓂ ᓕ ᓯ ᔨ ᕆ ᕕ ᕠ ᕵ ᖏ ᖨ. (*I’m not sure which of those two shapes to use for re: whether it should be a reflection or a rotation of the rI shape. If using Inuktitut syllabics as a base, it’d be ᕂ.)