sitelen pona

lipu pona: This is a good article.
lipu pona: This is a good article.
From sona pona, the Toki Pona wiki
(Redirected from Learn sitelen pona)

sitelen pona (sitelen-pona) is a logographic writing system for Toki Pona designed by Sonja Lang, the creator of the language. The system is described in the book Toki Pona: The Language of Good,[1] alongside sitelen sitelen. As a logography, each word is written as a symbol. Glyphs are typically written from left to right, in horizontal lines from top to bottom.

sitelen pona is the most used original writing system in Toki Pona, and the second-most after sitelen Lasina (the Latin script).[2] Learners might find sitelen pona useful for memorizing words' meanings.

The Toki Pona logo (toki-pona) functions in sitelen pona as the language's name, toki pona, written as a combined glyph.

History[edit | edit source]

Since first appearing in Toki Pona: The Language of Good in 2014, many Tokiponists and communities have expanded sitelen pona, integrated it into general Toki Pona learning resources, and created fonts and text encoding schemes.

Glyphs[edit | edit source]

Original pu glyphs[edit | edit source]

The original edition of the book Toki Pona: The Language of Good (lipu pu) includes the following glyphs for each of its main 120 words.

Esperanto pu glyphs[edit | edit source]

Following the recommendations of the Toki Pona Dictionary, the Esperanto edition of lipu pu includes several additional glyphs that had already been established among sitelen pona users. Alternative glyphs for three nimi pu are presented as footnotes in the main section on sitelen pona, "Hieroglifoj" (Hieroglyphs).

Glyphs for nimi ku suli are presented in the appendix "Aldonaĵo: nimi ku suli pi pu ala".

Other glyphs[edit | edit source]

There are alternative glyphs for more of the aforementioned words than those presented in Tokipono: La lingvo de bono. Some of the words have multiple notable alternatives. Tokiponists have also created glyphs for many nimi sin. See Category:Words with sitelen pona glyphs.

Radicals[edit | edit source]

Many sitelen pona glyphs have shapes in common, which have been analyzed as "radicals".

Punctuation[edit | edit source]

Use of punctuation in sitelen pona is not officially defined. In general, the main use of punctuation is to mark the boundary between sentences. Most people use only a sentence boundary mark, and some will use a form of quotation marking.

Using Latin-script punctuation, as in English, can cause confusion. The question mark (?) and seme (seme) are homoglyphs, and the exclamation point (!) resembles the glyphs for a (a), kin (kin), and o (o). In fact, these words largely make these punctuation marks redundant.

Sentence boundaries[edit | edit source]

The only full sitelen pona sentence in pu, seen below, is written with no ending punctuation. This style can extend to multiple sentences. The end of a sentence can be indicated by simply adding a space, or starting the next sentence on a separate line (as used in su).

ma [kasi alasa nasin awen telo a] li suli

ma Kanata li suli.

Canada is huge.

Some speakers use explicit sentence-ending punctuation. The most common choices are a Latin-style period (.), the middle dot (·), and a CJK-style circular period (。). A common argument for the middle dot is that sitelen pona is typically written centered around a midline, rather than a baseline. nasin sitelen kalama also uses the middle dot inside cartouches.

Question marks[edit | edit source]

The glyph for seme (seme) already looks just like a question mark.

A question mark is largely unnecessary. Most standard questions are already marked with seme (seme) or the pattern X ala X (ijoalaijo). Due to the glyph for seme, using a Latin-style question mark in sitelen pona causes confusion rather than solving any, and is considered bad style.

A dedicated question mark would only be needed to clarify whether the phrase X anu X (ijoanuijo) marks a statement or question. However, nowadays, the majority of proficient speakers do not use this construction, so that all questions are explicitly marked in all modalities of speech.

Because some fonts stretch the glyph for seme to fill a fullwidth character space, there is a misconception that it looks wider or more stylized than a question mark. This can result in juxtaposing the glyph in 2 widths (so that seme? looks similar to ??). In reality, seme looks just like a normal question mark in Sonja Lang's handwriting, and many other fonts follow suit.

Colons[edit | edit source]

Similar to Toki Pona in the Latin script, colons are frequently used in sitelen pona. A colon can be written centered between the words on either side (ijo:: ijo), or closer to the word on its left (ijo: ijo).

Some people use a right-facing ni (ni2) to avoid the need for colons. For example, below are three different ways to write the sentence mi wile e ni: sina pilin pona.

mi wile e ni:: sina pilin-pona
mi wile e ni2:: sina pilin-pona
mi wile e ni2   sina pilin-pona

nasin sitelen kalama also uses the colon inside cartouches.

Quotation marks[edit | edit source]

For marking quoted text, Latin-style quotation marks are commonly used, but corner brackets (te to) are widespread, being symmetrical around the midline. These glyphs are also used for the quotative nimi sin, te and to. It may be unclear whether to read them as words or punctuation. However, when used as punctuation, a reader may interpret them as te and to without misunderstanding the author's intent.

Word spaces[edit | edit source]

Many fonts with ASCII transcription expose a ligature using the string zz to produce a glyph-sized space character, for cases where the author wants the text to be visually indented or separated. This is best rendered using the ideographic space   instead. This is especially important for people who use screen readers, which will read "zz" out loud rather than silently skipping over it.

An alternative may be typing a double space, which is even more common in fonts to get a word-sized space. However, this can cause problems on websites, because HTML condenses whitespace.

Combined glyphs[edit | edit source]

Examples of combined glyphs in pu: pilin*ike pilin ike (scalar), telo*lete telo lete (scalar), kala^lili kala lili (stacked), toki*pona toki pona (scalar)

The glyph of a head word may be combined with the glyph of one modifier. These are called combined glyphs[3] or compound glyphs. There are 2 main ways of doing this:

  • Stacked: The modifier glyph goes above the head glyph. kala lili becomes kala^lili.
  • Scaled: The modifier glyph goes inside of the head glyph. kala lili becomes kala*lili. To allow for scalar combination, the head glyph generally must contain a single, sufficiently large, main negative space.

Extended glyphs[edit | edit source]

An extended form of the word pi (pi) is very common. The low horizontal line continues under all glyphs in the pi phrase (pi(ijo ijo)). Other characters such as prepositions are sometimes also extended in this way.

Names[edit | edit source]

Cartouches[edit | edit source]

ma Kanata li suli. "Kanata" is spelled out inside a cartouche, as [kasi alasa nasin awen telo a].
jan Sonja li pali e toki ni. ona li sitelen e lipu pu. Typeset in linja sike. "Sonja" is spelled out inside a cartouche, as [sike olin noka jelo akesi].

Proper adjectives are written like acronyms inside a cartouche [ ], inspired by Egyptian hieroglyphs. Within a cartouche, only the first sound of each word is read, each then strung together to form the name.

For example, ma Kanata can be written as ma [kasi alasa nasin awen telo a].

Any word that starts with the right letter can be used. Specific glyphs may be chosen to reflect the meaning or associations of the name, and people might have preferred ways to write their own name in sitelen pona. The glyphs in cartouches can even encode extra sentences, like spelling toki Inli ("English") as toki [ijo ni li ike] to also convey "ijo ni li ike".

There is also a nonstandard but fairly common extended system for writing cartouches using syllables or morae, called nasin sitelen kalama.

Name glyphs[edit | edit source]

Some people design custom sitelen pona glyphs for their names. Like the glyphs for common words, most of these do not indicate pronunciation.

Flexible glyphs[edit | edit source]

Many sitelen pona characters are not static, and aren't supposed to be drawn the same way every time.

akesi, pipi, soweli (akesi pipi soweli)
The amount of legs is sometimes reduced, especially in combined glyphs to reduce glyph density.
jaki (jaki)
Any scribble, usually in a single stroke dense with overlaps and (mostly) curvy lines. For example, scribbles submitted by community members to be added to linja lipamanka:
jaki1 jaki2 jaki3 jaki4 jaki5 jaki6 jaki7 jaki8 jaki9 jaki10 jaki11
ko (ko)
Any blobby closed shape. For example, blobby closed shapes submitted by community members to be added to linja lipamanka:
ko1 ko2 ko3 ko4 ko5 ko6 koDitto
ku (ku)
The symbol representing Toki Pona on the cover of lipu ku can range anywhere from a simple dot to a full Toki Pona symbol.
lete, kin (lete kin)
The rotation of the asterisk shape is not important. In some cases it may be written with four intersecting lines instead of three, and in kin some people reduce the asterisk to an x shape.
linja, kon, telo, pakala, mun (linja kon telo pakala mun)
May be mirrored.
ni (ni)
May point to the part of the text that ni is referencing; left, up, right, down (󱥁‍← 󱥁‍↑ 󱥁‍→ 󱥁‍↓), or other directions, instead of the standard downwards arrow. In handwriting, it may be drawn in an extended manner, pointing toward the relevant reference. A right-facing arrow (ni>) is often used in linking ni statements. For example, mi pilin e ni> zz toki-pona li musi (mi pilin e ni: toki pona li musi). Sometimes it is extended horizontally (ni>(--) ni<(--)).

Alternative glyphs[edit | edit source]

For various reasons, people have designed new glyphs for words that already have them. The Esperanto edition of Toki Pona: The Language of Good includes drawings of some of these.

In many sitelen pona fonts with ASCII transcription, alternative glyphs can be typed by appending a number to the end of a word, such as kala1. The default variants and list of supported alternative glyphs are not consistent between fonts, so this is not a consistent method to identify glyph variants. The added numbers may also interfere with screen readers.

Word sitelen pona Notes Included in pu Epelanto?
Main Alt.
a a a Diagonal punctuation stem
kin kin kin
n n n
o o o
akesi akesi akesi Only 2 strokes for legs.
epiku epiku epiku

Main: pona with emitters.
Obsolete:[i] Upwards arrow, now more often used for directional ni.

jaki jaki May be any scribble. Four different variations are included in the Esperanto translation of lipu pu.
kala kala kala Two dots for eyes to match the other animal glyphs.
kokosila kokosila kokosila

Main: toki with diagonal slash. Obsolete:[i] Star, representing the Verda Stelo.

lanpan lanpan lanpan Obsolete:[i] Turned pana.

Main: Top end has a circle representing a head, similarly jo.
Obsolete:[i] Top end is a hooklike curve.

majuna majuna majuna

Main: Turned sin.
Alternative: Five rays up similar to a lotus flower.

meli meli meli Derived from the planetary gender symbols, possibly by analogy with tonsi (from the transgender symbol). These alternatives avoid physical stereotypes.
Uncommon due to gendern't philosophy, which avoids meli and mije regardless of glyph choice.
mije mije mije
meso meso meso

Main: 2 vertical lines with dot between.
Alternative: ante with dot between strokes.

mi mi mi Curved finger stroke.
ona ona ona
sina sina sina
moli moli moli Low line for mouth instead of circle for head. Often used in pixelated fonts due to space constraints.
namako namako namako

Main: sin with an extra ray below.
Alternative: Hot pepper with emitters above.

olin olin olin pilin with emitters above
sewi sewi sewi Uncommon: Turned anpa. Commonly called "secular sewi".
It has been suggested elsewhere to use other religious symbols, with the possible downside of being less legible.[4]
soko soko soko2 The stem varies from a closed round rectangular shape to a vertical line, increasingly with a horizontal stroke for an annulus. The line variants are meant to avoid confusion with mama. includes only
uta uta uta Dotless. Can create ambiguity with moku (otherwise moku; here confusable with luka uta). However, this ambiguity is not more than that which already exists with pali and kepeken (normally read as ilo and kepeken respectively, but technically possible to read as luka ijo and luka ilo).
wile wile wile Closed to become a turned heart or pilin. While the separation of the line endings varies in handwriting, this alternative is most likely influenced by sitelen pona pona (a font and derivative script). Most differences in sitelen pona pona are rejected by sitelen pona users, but some occasionally slip in when writers want to introduce oddities.


  1. 1.0 1.1 1.2 1.3 Obsolete glyphs are still likely to be encountered because the main glyph is relatively recent. However, the main glyph is still most widely accepted.

Fonts such as sitelen seli kiwen also support OpenType character variants and stylistic sets, which can be enabled:

  • In Microsoft Word, in the Font dialog box → Advanced → Stylistic sets.
    • In many versions of Word, the Font dialog box is directly accessed from the Ribbon → Home tab → Font group, in the lower right corner.
    • In other recent versions, you can use Tell me → search for "Font Settings".
  • In CSS, with the font-feature-settings property.[5]

Mirrored glyphs[edit | edit source]

Caution: The subject of this section is nonstandard and will not be understood by most speakers.
It is recommended to familiarize yourself with the standard style, and to be informed and selective about which nonstandard styles you adopt.

sitelen pona is uncommonly written from right to left. There is no set practice for doing so. Some speakers recommend that all glyphs should be mirrored vertically, so that, for example, the glyph for soweli (soweli) faces left (soweli).

linja sike includes experimental mirrored glyphs for e, li, pi, tan (visually elipitan) at codepoints U+EC9B–U+EC9E.[6]

Text encoding[edit | edit source]

As of February 2024, there is no official Unicode support for sitelen pona. The Under-ConScript Unicode Registry (UCSUR), an unofficial registry for constructed languages, is used instead.

Sitelen Pona
Under-ConScript Unicode Registry

  0 1 2 3 4 5 6 7 8 9 A B C D E F
U+F190x 󱤀 󱤁 󱤂 󱤃 󱤄 󱤅 󱤆 󱤇 󱤈 󱤉 󱤊 󱤋 󱤌 󱤍 󱤎 󱤏
U+F191x 󱤐 󱤑 󱤒 󱤓 󱤔 󱤕 󱤖 󱤗 󱤘 󱤙 󱤚 󱤛 󱤜 󱤝 󱤞 󱤟
U+F192x 󱤠 󱤡 󱤢 󱤣 󱤤 󱤥 󱤦 󱤧 󱤨 󱤩 󱤪 󱤫 󱤬 󱤭 󱤮 󱤯
U+F193x 󱤰 󱤱 󱤲 󱤳 󱤴 󱤵 󱤶 󱤷 󱤸 󱤹 󱤺 󱤻 󱤼 󱤽 󱤾 󱤿
U+F194x 󱥀 󱥁 󱥂 󱥃 󱥄 󱥅 󱥆 󱥇 󱥈 󱥉 󱥊 󱥋 󱥌 󱥍 󱥎 󱥏
U+F195x 󱥐 󱥑 󱥒 󱥓 󱥔 󱥕 󱥖 󱥗 󱥘 󱥙 󱥚 󱥛 󱥜 󱥝 󱥞 󱥟
U+F196x 󱥠 󱥡 󱥢 󱥣 󱥤 󱥥 󱥦 󱥧 󱥨 󱥩 󱥪 󱥫 󱥬 󱥭 󱥮 󱥯
U+F197x 󱥰 󱥱 󱥲 󱥳 󱥴 󱥵 󱥶 󱥷 󱥸 󱥹 󱥺 󱥻 󱥼 󱥽 󱥾 󱥿
U+F198x 󱦀 󱦁 󱦂 󱦃 󱦄 󱦅 󱦆 󱦇 󱦈
U+F199x 󱦐 󱦑 󱦒 󱦓 󱦔 󱦕 󱦖 󱦗 󱦘 󱦙 󱦚 󱦛 󱦜 󱦝
U+F19Ax 󱦠 󱦡 󱦢 󱦣

Resources[edit | edit source]

A pattern of neon sitelen pona glyphs, inspired by 1990s galaxy-print carpeting

Learning resources[edit | edit source]

Dictionaries[edit | edit source]

Fonts[edit | edit source]

ijo Linku maintains a comprehensive list of all fonts.[7]

Fonts that support ASCII transcription ligatures can be used as a learning tool. If you type a supported word correctly, the font will turn it into a sitelen pona glyph. Building this connection between pronounceable sitelen Lasina spellings and visually meaningful sitelen pona ideographs can also help you memorize the meanings of words.

See also[edit | edit source]

Notes[edit | edit source]

  1. 4-legged (2-stroke) form of akesi
  2. Secular form of sewi
  3. Glyph version used since 2016.
  4. Alternate version designed by jan Sonja for personal use and published in 2022.

References[edit | edit source]

  1. Lang, Sonja. (25 May 2014). Toki Pona: The Language of Good. Tawhid. ISBN 978-0978292300. OCLC 921253340. pp. 104–111.
  2. jan Tamalu. (Updated November 3, 2022). Results of the 2022 Toki Pona census. Toki Pona census, Github. Retrieved 6 November 2023.
    The data regarding writing systems is quite similar to the last [2021] census. The only significant difference is that now 71% of people reported to at least know sitelen pona, while this figure was 61% before. This might be due to new better systems to write and interact with sitelen pona in different platforms and the ensuing new content created in sitelen pona. It is nice that so many people are learning sitelen pona. To put this number in perspective, more people report to know sitelen pona than people that report that they know 8 or more of the 17 Ku suli words.
  3. Lang, Sonja. (25 May 2014). Toki Pona: The Language of Good. Tawhid. ISBN 978-0978292300. OCLC 921253340. p. 110.
  4. soko weka [@vanapiton]. (12 June 2022). [Informal poll posted in the #sona-kulupu channel in the ma pona pi toki pona Discord server]. Discord. Retrieved 6 November 2023.
    Concerning the sitelen pona glyph for sewi, which of these (choose all that apply)...
    (Multiple allowed)
    (n = 75)
    Do you like the usage of: Do you use in your nasin:
    Arabic sewi 73 72
    Secular sewi 48 22
    Other religious symbols 8 2
    Other 2 3
  5. font-feature-settings. font-feature-settings. MDN Web Docs. Retrieved 7 November 2023.
  6. linja sike documentation. Google Docs. Retrieved 24 October 2023.
  7. fonts. Google Spreadsheets. Retrieved 21 October 2023.