Tokiponidos

are languages based on or influenced by Toki Pona, sometimes I think they are nice

Introduction to tokiponidos

Tokidopido?

A tokiponido (formed with Esperanto -ido "offspring", cf. esperantido, rhymes with knee dough) is a constructed language that is in some way based on Toki Pona, itself a conlang that was first shared online in 2001 by lili Sonja. As a language both of creativity and of the digital age, Toki Pona inspires many people across the web to make many sorts of things. Beside creating things in TP, some take inspiration from it when they construct languages of their own. How exactly they do that varies from conlang to conlang.

Because of that, it’s hard to say what exactly should count as a tokiponido and what shouldn’t. Is it a changed form of standard Toki Pona, with only a few nonstandard words and constructions? Is it a language that was evolved from Toki Pona, with changes made to the sounds, words, rules…? Suppose it didn’t resemble Toki Pona at all, but its creator had some knowledge of it and was inspired to add a few similar features, consciously or not? There’s a lot of places where one could draw the line, and you can probably find someone using the word in each of these ways. Still, even in Toki Pona circles the term “tokiponido” is uncommon and not universally understood (understandably so, since it's not a Toki Pona word).

Why is this a thing?

People invent languages for a variety of reasons. But why tokiponidos in particular? I'll admit it isn't exactly the pinnacle of originality, it can feel pretty derivative sometimes. An explanation many people arrive at is that some folks didn't like or get Toki Pona, and thought it needed fixing. This does play a role sometimes (see Mini (2020), Tipunsin (2020)). Most of the time, the attitude is more positive: when people make something they like, it can resemble Toki Pona simply because Toki Pona is also a thing they like.

Personally, I got into tokiponidoing because I was having trouble conlanging - or rather, trouble starting from scratch and not giving up halfway through. By building off a preëxisting base like TP's, I found it easier to regularly occupy myself with conlanging in general and with Toki Pona in particular, at a time when I struggled with both. To me, tokiponidos are an accessible entry point to the bigger world of conlanging, and fun in their own right. Here's a few examples:

A few examples:

Sometimes, the name of a language tells you a lot about its resemblance to Toki Pona. Languages with Toki Pona words in the name, like luka pona and toki ma, are often very similar:

A conlang’s name is one way the creator can tell people they were inspired by TP. Another way to tell them is, well, to actually just tell them:

A few more, slightly less typical, examples:

There are times when a conlang’s name is deceptive. For example: there’s a language out there called Sitelen Pona, a written-only language that’s meant to be convertable to any spoken language in the world. As such, it’s based more on a Western understanding of Chinese writing than on Toki Pona. It also bears little to no resemblance to Toki Pona's writing system (also called sitelen pona). For lack of a single native pronounceable name, the author picked one out from another conlang with its own logography… I guess he liked the sound? The implied relation to TP doesn’t amount to anything when I look closer; I don’t consider this a tpido.

Conversely, someone can publish a suspiciously Toki Pona shaped language without mentioning Toki Pona at all. Such is the case of Jackson Moore’s musical language Moss/Maas. His 2009 description sounds a lot like (early) descriptions of Toki Pona: for one, Moss “has just 120 words and a few grammatical rules of thumb”. Moore also called it a “pidgin language”, while Sonja called Toki Pona a “pidgin-like language” in 2001 (this disappeared quite quickly from the website and is no longer said by people today). It would be a stunning coincidence if this didn’t have to do with Toki Pona. Either way, it’s not at all a rip-off but a valuable work of art in its own right! You can listen to some Moss performances, or conversations, on Jackson Moore's FreeMusicArchive page. Milamber made a great comparison of Maas & TP in her video about Toki Kalama (which, like Sitelen Pona, picked a name from Toki Pona because it has no spoken name in itself).

So that's one group of edge cases: the middle ground between "tokiponido" and "regular conlang". I see these two categories as overlapping circles in a Venn diagram. Ditto for edge cases between "tokiponido" and straightup Toki Pona:

My thing

I became interested in tokiponidos in 2020, the same year I started Tuki Tiki. In the summer, a TP community member called Robbie (jan Wapi) made a YouTube video rating different tokiponidos, following similar content about esperantidos. I remember being surprised that something I had “made” got any amount of attention online – I mean, all I’d done was share a Google Spreadsheets link in a couple Discord servers. Turns out Tuki Tiki would become one of the most well-known tokiponidos among Toki Pona Folk, along with Luka Pona and Kokanu (Toki Ma).

Although I made it, Tuki Tiki never kept my attention for very long; and I have no doubt that there are people who use it better than me. On occasion, it has been featured in Ma Toki Pona VR, and also as an April Fools joke in lipu tenpo (IItiki, #kokosila) and ma pona pi toki pona (2025). What has kept me interested over the years is Toki Pona, and stuff related to it. When I say I like tokiponidos, it's not that I like all or most of them; most of them are not very interesting. Just as I don't especially like most conlangs, I'm more interested in the fact that this phenomenon exists, and why.

So, in 2024, I started keeping a list of tokiponidos in a spreadsheet, including bits of information on each of them: year of origin, amount of words/meaningful units, amount of sounds, stated goal, etc. As of early 2026 I have not updated it in a long while, having lost steam sometime during 2025. Tpidos made after that period are either underrepresented or totally absent. This is a real shame, as the number of tokiponidos is exploding right now - to the point that it overwhelmed me! All the same, I'd like to share what I got.

What I got

For each tokiponido, I kept track of some basic identifying info: name, creator, year of origin; as well as some more specific details about the language, which for lack of a better idea I will just call classifying info: modality (spoken, signed, written, played, eaten, etc.), goal, amount of phonemes, amount of lexemes or morphemes (yes I know they're different), degree of similarity to TP, other languages it is similar to, and progress of its development. The last two columns in the spreadsheet are for extra info: other remarks from the author or observations I made, plus links to documentation. I'll go over these in more detail.

First three (identifying) columns are pretty self-explanatory, other than the notation of years: since tokiponidos have only existed for the last few decades of history, they don't care about the first three digits and prefer "1" to mean 2001 (the year of Toki Pona's online publication). Tokiponidos are necessarily younger than Toki Pona.

For columns D and E there is a limited set of values: D ("nasin sijelo", modality) has us(l), us, l, a. The letters stand for Toki Pona words: uta (mouth), sitelen (writing), luka (hand) and ante (other). Almost all tpidos are "us" (spoken and written), some also have signed versions ("us(l)"), a few are signed ("l"), just one is "a" (Maas, discussed above). Column E ("wile", goal) has a few more:

Naturally this categorisation is arbitrary, most conlangs are not so cut and dry and neither are most tokiponidos. Understand that I am trying to describe many tokiponidos, not each in their own terms, but all in the same terms. This should aid in comparison with TP and one another; this relatively fine-grained system took a few weeks to form based on 100-odd entries. One common overlap is between "kulupu pi lon ala" and "tenpo ante": if a tpido is both, I give it the value "tenpo ante" (since the fictional speakers are kind of implied by the fictional language). It happens often that a language could sit well in multiple groups: the wile is not based directly on the language, but on the goal that is stated (or, failing that, implied) in the description. This is what I put in Column F ("toki lon wile"). If I found no explicit statement about the goal, I used another source, my own mind, or a question mark.

Next four columns are for sounds/kalama (G-H) and morphemes or lexemes/nimi (I-J). The first column in each pair compares the amount relative to Toki Pona: about the same (=), significantly larger (V) or smaller (ˇ). The symbols used stand for Sitelen Pona symbols (the writing system, not the language): anything over 150 is V, 100 or less is ˇ. The two other columns are for absolute amounts. I should also justify the conflation of morpheme and lexeme: most tpidos inherit TP's lexical minimalism. Some diverge by embracing compounding or inflexion or somesuch, in which case morphemes can be meaningfully distinguished from words. Even then, a tpido often has baked into its design that it's gotta be small about some things (by merit of being a TPido), and either morphemes or lexemes will be more salient as the primary building blocks of the language. My judgment on which is definitely subjective in some cases, and this column is probably less interesting for more diverging tokiponidos.

Column K is for how much the tpido is like Toki Pona ("sama toki pona"). As with "wile", the categories I made are broad. Here they work like four gradations: r for relex (TP but all the word forms are replaced), m for modified (TP but with superficial changes), d for derived (started by applying changes/derivations to TP), i for inspired (conlang partly inspired by the general concept of TP; or a derivation of an earlier tokiponido). Of course, every tpido is inspired by Toki Pona, just as every square is a trapezium and every pigeon a bird. Conversely, not every tpido is a Toki Pona reskin.

Columns L ("toki sama ante") is where I put any other languages the conlang may have been taking inspiration from. Column M ("pini") is for the completion/progress of the tpido. A lot of tpidists would probably object to my calling their conlang "pini" or finished; I just use the word to mean "more or less finished or functional". Many conlangs are never truly "finished" to their makers, but they are far along enough that you could take a snapshot now and see a full-fledged (minimalist) conlang. I reserve "pini ala" for unfinished but active tpidos, and "lape" for unfinished inactive ones. The value "weka" means the documentation is lost to time. "?" means ?.

Columns N & O ("ante" & "lipu"): extra remarks and links.

Why I did that, and how

In difficult times I get joy out of fixating on insignificant things. Also, I found existing lists pretty short and thought the tpidos that are (still) viewable online should be archived better (many old ones have missing links, or are completely lost to time). Things on the internet do just go away after a while, so archiving stuff you care about is important.

I collected these in a pretty informal way: I found them through Google, conlanging and neography subreddits, conlanging sites like Conworkshop, old forums and Toki Pona adjacent Discord groups I’m in. That last one contributed a lot: Discord is probably the primary communication channel of the Toki Pona community these days. Since Discord servers are relatively private (you need an invite link to see or search for messages), they’re not as easy to search through, and people have sometimes expressed discomfort about their messages being screenshotted or linked elsewhere. Combine that with link rot and server deletions, and it gets confusing (for my brain) to figure out what to do. Generally I preferred to link to (archived) image links than Discord messages.

My motivation went up and down over time: I did most of it in 2024, a little in 2025, plus some cleaning up while I’ve been writing on this. There’s plenty of things that I could have found but didn’t: I mostly searched the English-language internet, and translation engines only helped me so far in finding keywords in Russian and Chinese and such. Also, there are plenty of tpidos here with their documentation partly or wholly missing, and doubtless there were plenty out there on now-defunct websites, deleted YouTube videos etc. The fact that there are more tokiponidos from the year I made the list is at least partly due to recency: I'm not some neutral observer, I mostly found these in spaces that I was already in and from a time when they were still easy to find.

I tried to finish the incomplete entries in the list before making little charts about them, but I also had to draw a line to stop my own perfectionism. In the end, I was left with 370 tpidos. I did not think there would be this many. At the same time, it isn't that big a number, so it's hard to say how all the different variables interact with each other.

Charts

This section isn't done: I am struggling both to decode my own year-old thoughts and to find motivation for writing about a niche within a niche within a niche. All the same there's some stuff here that I hope will be found interesting by some. My intention with the rest of this page is to go over every one of the variables I was trying to look at one by one and, maybe after that, to treat a few of them together. Not sure how valuable that last part will be; but the charts give me something to hang my thoughts on.

start time / tenpo open

Most tpidos in the lipu are from 2020 or later, with a peak in '23-'24 (the drop in 2025 is only cause I lost interest in keeping track). Hard to say how much this reflects an actual increase in tpido creation rather than an increase in my own interest for them, which happens around the same time. Keep in mind I got many tokiponidos from Discord servers I'm in or could join, and occasionally from old Google Drive docs that were once shared with me in servers I'd left (since 2019). Yet I think this period was important for many other TOK people, too: our growing adoption of Discord catalysed the language's growth, right when a lot of people were looking for ways to spend the lockdown. The more people know about Toki Pona, the more tokiponidos happen, I would think.

There is a noticeable peak in 2010: this is mostly one guy who kept making similar languages from 2007-'11.

Amount of tokiponidos (y-axis) from each year (x-axis; 1 = 2001)

goal / wile

I made a lot of categories here, so I will try to elaborate on those. 81 tpidos wanted to go smaller than Toki Pona in some way(s) ("mute lili"). Some well-known examples are Toki IO (6 phonemes, 110 words), Puna (9 phonemes, 18 "words") and Tuki Tiki (8 phonemes, 39 words). 71 are in the very general "musi" category, made for art and/or fun, e.g.: Kisuwi ("made to sound cute"), Minecraftian aka Toki Lojbanto (a mix with Lojban and Esperanto, for use on an old Minecraft server) and Kyrete (an "engineered personal art language"). 43 are nasa, by which I mean: jokelangs (mu, a subset of TP), shitposts (Koka Kola) and "cursed" tokiponidos (kornlang).

Minimalist or "mute lili" tokiponidos are very well represented, probably because they take less time to "finish" – at least, in a form that you'd be willing to share online. Many of the lower-effort tpidos fall into this group, as well as in "nasa". I see reasons to think there are many conlangs people spend more time on, but never properly publish online (having never reached that stage of "good enough to show to people", which is harder to reach the more seriously you take the project). One tpido example is J. Rhiemeier's Quetch/Quetig (2006): despite very little (documentation of) development, the author still expressed interest in it in 2023. Based on my own experience in making (and abandoning) conlangs, this could be a very common fate. Non-minimalist tokiponidos may or may not be less common objectively; they are at least less likely to be shared.

Also, I am personally more acquainted with small tokiponidos, and have spent more time in spaces dedicated to microlangs than, say, auxlangs.

Moving further down the list, I'm starting to see flaws in the granular system past-me came up with, but also in his identifications of some languages. That makes it a good deal harder to figure out what to say about it. I'm not going to go through it all over again, but I do wanna point out some weird examples before I move on. (Also, I will try to treat "tawa jan ale"/IALs before "nasin ante", since they're easier to explain). Sorry if this reads like a note to future me, it is.

The next two biggest groupings are "nasin ante" (a different take on pona) and "tawa jan ale" (auxiliary languages). Most nasin ante's are personal languages of their creator(s), diverging from Toki Pona's concept or design in a few places as a matter of taste, philosophy etc. Some are self-declared dialects or styles, like tok' apona (2017) and tok' pi mi (2019); others are less obviously related to Toki Pona, like Bleep (2020), which was kvk's "personal ideal simplified not-quite-a-language"; and nemune (2023), which was initially an "offshoot" before becoming "a piece of art with an approach of its own".

Amount of tokiponidos per wile/goal category Amount of tokiponidos by degree of similarity to TP Amount of tokiponidos by degree of similarity to TP Amount of tokiponidos by degree of similarity to TP

gotta put link somewhere more noticeable too this is WIP so will finish some other time link to my tokiponido spreadsheet