Toki Pona Syllabics

Here’s how to write Toki Pona using Canadian Aboriginal syllabics! One of the world’s coolest writing systems!1 (The simplest way to refer to this writing system in Toki Pona is to simply call it ᓯᑌᓐ ᕐᑲᓇᑕ sitelen Kanata.)

When you’re done reading this, you might wanna practice your Toki Pona syllabics reading skills with jan Pijes lessons for Toki Pona (with all the Toki Pona written in syllabics), try out my own little Toki Pona dictionary, or learn the writing with my Toki Pona Syllabics course on Memrise!

ea
oi
u
pepa
popi
pu
teta
toti
tu
keki
koka
ku
memi
moma
mu
neni
nona
nu
le2li
lola
lu
sesi
sosa
su
je3ji
joja
ju
we4wa
wowi
wu
Use ◌ᓐ for n at end of syllable (e.g. ᑐᓐᓯ tonsi).5
Syllabics Sort Order6
-e-i-o-u-afinal
vowel
only
p-
t-
k-
m-
n-◌ᓐ
l-
s-
j-
w-
ᕐ◌ (naming mark) is ignored

Syllabics characters are categorized as either symmetric, or asymmetric, depending on the way they are modified when the vowel change. The symmetric shapes (ᐁ e, ᐯ pe, ᑌ te, and ᕓ we) point in the four cardinal directions of the compass (south/north/east/west) and are rotated 90° between the vowels (these are also called cardinal). The asymmetric shapes (ᑫ ke, ᒣ me, ᓀ ne, ᓓ le, ᓭ se, and ᔦ je) are flipped, not rotated, to get the different vowels (these are also called ordinal, since they point in the ordinal directions of the compass: NW, NE, SW, SE).—When talking about the orientation of a character, the word rotation is used, regardless of whether the specific character is actually rotated or flipped.

The word series is often used to group characters with, either the same consonant (a line in the sort table), or vowel (a column). The t-series thus includes all rotations of the character ᑌ te, while the e-series include the characters with a rotation that corresponds to the vowel e: ᐁ e, ᐯ pe, ᑌ te, ᑫ ke, ᒣ me, etc.

When sorting Toki Pona syllabics (see table), the Eastern Cree/Inuktitut order is used, except for the naming mark (ᕐ◌), which is ignored (and does not affect the sort order).6

Punctuation

Many texts use only space between words and a small cross (᙮) at the end of sentences as punctuation—other texts instead use a full set of Latin punctuation marks (including the Latin period). Syllabics uses a double horizontal line (᐀) as a hyphen, this is to avoid confusion with the syllable final c (◌ᐨ) used in Western Cree.

Syllabics does not have upper and lower case, and so names (that is, unofficial words) can’t be marked with capitalization. Instead, to mark these words, put a naming mark ᕐ in front of it: ᑐᑭ ᕐᓯᓇᓐ toki Sinan (the Klingon language), ᔭᓐ ᕐᓱᓐᔭ jan Sonja, and ᒪ ᕐᑲᓇᑕ ma Kanata (Canada).8

Sample Text

ᑕᐸᕙᐊᐳᑐ








A Toki Pona syllabics star chart.

ᒥ ᔭᓐ ᕐᓴᔦᓐ. ᒥ ᓗᓐ ᒪ ᑐᒧ ᕐᐆᐸᓚ. ᒪ ᑐᒧ ᕐᐆᐸᓚ ᓕ ᓗᓐ ᒪ ᕓᓐᓴ.

ᑌᓐᐳ ᒨᑌ ᐱᓂ ᓚ ᔭᓐ ᕐᕕᑭᓐ ᓗᓐ, ᓗᓐ ᒪ ᒥ. ᐅᓇ ᓕ ᔭᓐ ᐆᑕᓚ ᓕ ᑕᕙ, ᑕᕙ ᒪ ᓲᓕ ᐁᓗᐸ ᐊᓓ. ᐅᓇ ᓕ ᐸᑲᓚ ᐁ ᒪ ᑐᒧ ᒨᑌ, ᓕ ᒧᓕ ᐁ ᔭᓐ ᒨᑌ. ᐅᓇ ᓕ ᔭᓐ ᐃᑫ. ᑕᓱ ᐅᓇ ᓕ ᐸᓕ ᑭᓐ ᐁ ᑭᕓᓐ ᓲᓕ. ᐅᓇ ᓕ ᓯᑌᓓᓐ ᐁ ᓯᑌᓓᓐ ᐱ ᔭᓐ ᕐᕕᑭᓐ, ᓗᓐ ᑭᕓᓐ ᓂ. ᑭᕓᓐ ᓂ ᒨᑌ ᓗᓐ, ᓗᓐ ᒪ ᑐᒧ ᒥ. ᑭᕓᓐ ᓂ ᓕᓕ ᓗᓐ ᑕᓱ, ᓗᓐ ᒪ ᐊᓐᑌ. ᓯᑌᓓᓐ ᓂ ᓕ ᐳᓇ, ᑕᕙ ᒥ !

ᔭᓐ ᕐᓴᔦᓐjan SajenZrajm (me)
ᒪ ᑐᒧ ᕐᐆᐸᓚma tomo UpalaUppsala (a city)
ᔭᓐ ᕐᕕᑭᓐjan Wikinthe Vikings

Pointing

Syllabics can be written either pointed, or unpointed, referring to the dots (or points) above the characters. Text written in syllabics is very often unpointed (fluent speakers have no problem recognizing the words, even without vowel length markings). In the few cases where ambiguity may rear its ugly head, one may either rely on context to clear it up, or add dots to clarify only as needed (this is known as strategic pointing).9

Unpointed syllabics are very common, even on road signs, inscriptions, and other quite formal contexts.

Toki Pona Unpointed!

The word ᑰᓘᐴ kulupu, if written unpointed becomes ᑯᓗᐳ kolopo. Since the word ᑯᓗᐳ kolopo does not exist in the Toki Pona lexicon, any reader fluent enough in the language would thus quickly realize that ᑯᓗᐳ kolopo could not be the intended word and that ᑰᓘᐴ kulupu is the only possible substitute!

A closer look at pu10 reveals that, even if you replace all occurrences of u in the official dictionary with o, this doesn’t lead to any ambiguities! Sure, words like ᑯᓗᐳ kolopo and ᒧᑯ moko may sound funny, but they don’t cause any confusion, since none of these words overlap with any existing words containing o!11

There are a couple of cases where strategic pointing might be especially useful:

First, with names (capitalized words in the Latin orthography), whether you need to point them depends on your reader. Are they likely to know the name? With a famous, and easily recognized name like ᒪ ᑐᒧ ᕐᓅᔪᑲ ma tomo Nujoka (New York City) you can probably get away with dropping the dot from the start, but if I were to talk about my friend ᔭᓐ ᑕᑐ, how could you possibly tell whether I’m talking about someone named jan Tato, or jan Tatu?—I’m probably better off referring to them as ᔭᓐ ᑕᑑ jan Tatu, at least the first time I mention their name.

The second case is the official word ᑑ tu, which can be confused with the (post-pu) word ᑐ to (used to delimit quotes with ᑌ … ᑐ te … to).12 So, if you find yourself quoting passages in which ᑑ tu appear, a strategically placed dot might be in order.

Compared to Other Toki Pona Syllabics

As I started to researching how to write Toki Pona with Canadian Aboriginal syllabics, I figured I couldn’t be the first person to try this, so I did some googles. I did a lot of googles, actually, but all my efforts came up empty. Naturally enough, after having written this article, I stumbled across some earlier Toki Pona syllabics (one hidden in a video, the others inside walled gardens). So I thought I’d take a moment to talk about them here.

By jan Misali (2017)

The first version I stumbled upon was the following table of all possible Toki Pona syllables, written in Canadian Aboriginal syllabics, and displayed for a few seconds in one of jan Misali’s videos on YouTube:

Screenshot of table from video Conlang Critic Episode Twelve: Toki Pona (03:49) by jan Misali (23 February 2017).

The video is a review of Toki Pona and the above table is shown for only a few seconds, in the midst of a stream of similar tables, each showing the syllables of Toki Pona written with different writing systems.—Looking at the table more carefully, it seems as if it was quite quickly thrown together (to be expected, given the amount of screen time it gets) but most of it makes perfect sense. The first row contains a couple accidentally doubled characters, but since the columns below are consistent in their use of the overdot (◌̇) the intent is easy enough to see. (Also, a long vowel is written with an overdot when using syllabics, but in Latin transcriptions it is written by doubling the vowel.) Below I will simply assume ᐃᐃ was intended as ᐄ, and ᐅᐅ as ᐆ.

e: The writing of e is just odd!—I figure jan Misali can only ever have been looking at the Inuktitut vowels (ᐁ ai, ᐃ i, ᐅ u, ᐊ a, but lacking e) and worked his way from there without ever seeing (the, to Toki Pona, much better fitting) the Cree vowels (ᐁ e, ᐃ i, ᐅ o, ᐊ a). This lead to the counter-intuitive use of ᐃ for e (it means i in all other languages) while not making use of the rotation ᐁ (e in Cree). Not using all the possible rotation also means that, in any given text, there are more characters with overdots, possibly impacting legibility.

w: I fell in love with how jan Misali write w!—In the Canadian syllabics languages that use w, the sound is added to the beginning of a syllable by putting a dot, either before (ᐧ◌), or after (◌ᐧ), the syllabic character (the dot position depends on the language). However, rather than using a dot for w, jan Misali opted to spell ‘w’ as ‘v’ using the v-series of characters (ᕓ we, ᕕ wi, ᕗ wo, ᕙ wa). While this is not technically correct it leads to no conflict since Toki Pona does not have a v sound. Also the v-series shapes are much prettier than adding dots! (Though, here, too, I suspect that jan Misali only ever looked at Inuktitut, since that is one of the few languages which use Canadian Aboriginal syllabics, but which do not have a w sound.)

Conclusion: It is quite clear that this is just a throwaway suggestion, created, not to be the final word in Toki Pona syllabics, but as window-dressing for a couple of frames in a video. It is also quite obvious that jan Misali’s syllabics only draw from the Inuktitut usage of the script, ignoring Cree (which is a shame, since Cree’s vowel mapping is a much better fit for Toki Pona). Despite this is jan Misali’s syllabics pretty good (the only major mistake being the vowels). The (however accidental) idea to use the v-series for Toki Pona w (ᕓ we, ᕕ wi, ᕗ wo, ᕙ wa) is a good one, as it cuts down on dot clutter, and increases cuteness. (I just love those loopy ᕓᕕᕗᕙ characters!)

Update, 13 March 2021: I originally wrote w in the Eastern Cree style (ᐎ wi, ᐌ we, ᐗ wa, ᐒ wo), but I have since changed my mind, and now use the Inuktitut v-series (ᕓ we, ᕕ wi, ᕗ wo, ᕙ wa) instead. This is partly on jan Misali’s (and jan Katelin’s) suggestions, and partly because I, in practice, have found it easier to read.4

By David King (2018)

I found King’s take on syllabics for Toki Pona in a posting to the toki pona Facebook group. It is based on a single Canadian Aboriginal syllabics star chart, which, at the time, had been doing the rounds on Twitter for a while.

David King
29 August 2018 ·

I recently found this chart, by Matt Baker (https://twitter.com/usefulcharts/status/866363656568619008) concerning Canadian Aboriginal Syllabics (https://en.wikipedia.org/wiki/Canadian_Aboriginal_syllabics).

Except for some awkwardness about u / oo, it seems like an easily learned system that would adapt well for toki pona.

I’m thinking a umlaut over syllables with u? Maybe an appended character if you don’t have an accented syllable in your typeset. Pu would be ᐵ…

Facebook posting by David King in the toki pona group (29 August 2018).

I have not been able to determine with absolute certainty which language this set of characters is used for, but the r-series (re, ri, ro, ra in the image above) are unique to the Western Oblate missionary style of syllabics,13 which I presume is used for Western Cree. (This r-series is not even included in Unicode!)

w: King writes w as a dot after the syllabic character (ᐍ we, ᐏ wi, ᐓ wo, ᐘ wa) in the Western Cree style.

l: Curiously, the l-series of characters found in King’s chart differs from the Western Oblate style. Between the two styles, all characters, except ᕊ lo, are mirrored along center line of the U-shape (so that little serif-like line, that juts out from one of the tines in the U-shape, sticks out in the opposite direction). Compare: ᕂᕃ le, ᕄᕆ li, ᕊᕊ lo, and ᕋᕍ la (in each pair, the first character is in King’s style, and the second is in the Western Oblate style).14

j: The “j-series” of characters chosen by King (which are more commonly written with c in the Latin script; ᒉ ce, ᒋ ci, ᒍ co, ᒐ ca) are, in Western Cree, pronounced with [t͡s] or [t͡ʃ].—The Toki Pona j is softer, and its pronunciation is much closer to the Cree y sound, so a better fit would have been the y-series of characters (ᔦ ye, ᔨ yi, ᔪ yo, ᔭ ya) which are also present in the star chart.

n: King uses a Western Cree form of the syllable-final n: ◌ᐣ.

u: King suggests adding an umlaut or double overdot (◌̈) to the o-series of characters in order to write a syllable with u (ᐇ u, ᐵ pu, ᑒ tu, ᑱ ku, ᒩ mu, ᓆ nu, ᓙ lu, ᓳ su, ᔬ ju, ᕗ̈ wu).—I (having independently come up with the same idea) can do nothing but approve of the fact that he writes syllables with u by modifying the o-series. Though I personally think a single overdot (◌̇) is a better fit, since the single overdot is used (to mark that a vowel is long) in all styles of syllabics, while the doubled overdot is only rarely used. (Presumably King did not know of the prevalence of the single overdot in Canadian Aboriginal syllabics, since this information is not included in the star chart he used.)

Conclusion: It is clear that King based his Toki Pona syllabics off of that single star chart of Western Oblate syllabics, meaning that his style also inadvertently inherited all the idiosyncrasies of that particular chart (like the odd j/c mix-up, and the mirrored characters of the l-series). With that in mind, I like his take on it, and the fact that both King and I independently decided to write u in a similar way strengthens my belief in that idea.

Epilogue: As it turns out, King’s star chart did not originate with Matt Baker (who never claimed to have created it himself—he only tweeted about it, but neglected to give a source). A reverse image search instead revealed the source to be the web page of University Blue Quills, of Blue Quills First Nation in Alberta, Canada. (Unfortunately I could find no information about what language the chart is intended for.) In addition to the chart used by King, their web page also feature a different star chart, with more common characters (with characters that are all found in Unicode). For completeness, I’ll include it here.

A different Canadian Aboriginal syllabics star chart published by University Blue Quills (13 July 2015).

By jan Katelin (2021)

I found jan Katelin’s syllabics when searching the Toki Pona Discord server for all things related to Canadian Aboriginal syllabics. She describes her syllabics style in the following messages to the #toki-pona channel.

ᔭᓐ ᕐᑲᑏᓕᓐ (jan Katelin) 09/01/2021
aiueo
p
t-
k
l
w--
j-
s
m
n(edited)
the biggest change is that I made overdot mean a more open vowel (ie i -> e, u -> o) instead of a long vowel
oh and no consonant is just ᐊ ᐃ ᐅ ᐄ ᐆ
trailing ᓐ is “n”, leading ᕐ serves the same purpose as capitals in latinized tp
Discord message by jan Katelin to #toki-pona on the ma pona pi toki pona server (9 January 2021).

In contrast to the others, jan Katelin seemed to have spent quite some time with her syllabics. She mentioned in #toki-pona that she’d learned the use of it while she was learning the Toki Pona language.—She later verified this in private conversation, also mentioning that her starting point had been the table of syllabics in jan Misali’s video (above) but that she’d regularized, and expanded upon it where she’d found a need for it.

FrontBack
Closeiu
Mideo
Opena

Vowels: This syllabics style pays homage to its Inuktitut roots by keeping the vowels that are used in the Inuktitut language (a, i, and u) unpointed, while marking the vowels that do not occur in Inuktitut (e and o) with an overdot. This also means that the overdot can be thought of as a diacritic to indicate a “more open” vowel (turning ᐃ i into ᐄ e, and ᐅ u into ᐆ o).—The drawback of this vowel system is that it adds (a little bit of) needless graphical complexity by preferring points over using all available glyph rotations. This, in turn, causes some words to become indistinguishable homographs when writing in an unpointed style (though this actually only affects ᑮᓐ ken and ᑭᓐ kin).

Names: Where the Latin orthography of Toki Pona uses capitalization to indicate names/unofficial words, jan Katelin instead marks names by putting the Inuktitut uvularization character (ᕐ◌)15 in front of the word.—This is neat! Originally my own syllabics did not include a naming mark like this, but in part inspired by jan Katelin, I later added one.

Conclusion: The syllabics style of jan Katelin is an improvement on jan Misali’s, and both her additions are good ones. Her addition of a naming mark (ᕐ◌) is an especially good idea and is original to her system.

Update, 21 March 2021: I originally used the Western Cree final ᑊ p as a naming mark, but, after seeing the symbol that jan Katelin chose, and thinking about it for a bit, I adopted jan Katelin’s symbol.16

Footnotes