Talk:Toki Pona
This is not a forum for general discussion of the subject of the article.
- Add new text under old text.
- New to Wikipedia? Welcome! Learn to edit; get help.
- Assume good faith
- Be polite and avoid personal attacks
- Be welcoming to newcomers
- Seek dispute resolution if needed
| Former good article | Toki Pona was one of the Language and literature good articles , but it has been removed from the list. There are suggestions below for improving the article to meet the good article criteria. Once these issues have been addressed, the article can be renominated. Editors may also seek a reassessment of the decision if they believe there was a mistake. | |||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||
It is of interest to multiple WikiProjects.
Archives
The toki pona translation of "The Lord's Prayer" is gone for some reason. Where did it go? Ayslays (talk) 06:13, 3 January 2023 (UTC) Reply
- Nevermind, it's back. 118.137.29.117 (talk) 12:28, 6 January 2023 (UTC) Reply
- Sorry, i wasn't logged in. Ayslays (talk) 12:30, 6 January 2023 (UTC) Reply
- Where is it moved to? I can't find any of the examples texts, neither the prayers nor the poems. Ike a! Oogalook (talk) 17:13, 28 January 2024 (UTC) Reply
- @Oogalook the example texts were removed in this revision due to concerns over their copyright status discussed at Wikipedia:Copyright problems/2023 December 9. Although it appears they missed something in that discussion: the Lord's Prayer had already been replaced with a wholly different, probably non-copyright violating translation supplied by @Casperdewith, in this revision from January 2023. I remember this happening because it caused these weird formatting problems that made it invisible for some users for some reason; the comments you're replying to here were about this!
- Because this isn't a copyright issue (anymore), the question then is whether it's fine to have an example text in a language article that's written by a Wikipedian and not published anywhere else. I actually have no idea. Alternatively we could copy (with attribution) something short out of lipu tenpo, as it has the same license as Wikipedia. twotwos (talk) 00:04, 29 January 2024 (UTC) Reply
- What an excellent answer, thank you! And that seems like a wise solution. Oogalook (talk) 14:06, 29 January 2024 (UTC) Reply
the question then is whether it's fine to have an example text in a language article that's written by a Wikipedian and not published anywhere else
- I'd be happy to publish it on my site if this becomes and issue and @Casperdewith is okay with that --u9000 (they/them • talk) 00:33, 17 July 2024 (UTC) Reply
- Hi all – sorry for the late reply. I don’t need my translation to stay on the Wikipedia page. I just wanted to rid the earlier, terrible one, so I made my own as a temporary replacement. It’s licenced under a CC BY-SA 4.0 licence, though. So @U 9 000, feel free to use it on your website.
- The random Toki Pona texts seem to have been removed in favour of texts with a clear source and licence that are sure to be of good quality: a poem that got first place in the annual literature competition (licenced under CC BY-SA 4.0), and an excerpt from jan Sitata, a book that’s linked on the official Toki Pona website (licenced under CC0 1.0). I’m satisfied with this new ‘policy’.
- Using lipu tenpo articles, as @Twotwos mentioned, is also fine. They are all licenced under a CC BY-SA 4.0 licence. And as for the measure of quality, they’re proofread by at least two (but usually more) proficient tokiponists.
- Greetings from jan Kasape, lipu tenpo designer and proofreader. Casperdewith (talk) 14:12, 12 November 2024 (UTC) Reply
- Where is it moved to? I can't find any of the examples texts, neither the prayers nor the poems. Ike a! Oogalook (talk) 17:13, 28 January 2024 (UTC) Reply
- Sorry, i wasn't logged in. Ayslays (talk) 12:30, 6 January 2023 (UTC) Reply
I understand that the file mentioned is the most reliable source for a speaker count, but the toki pona Subreddit (is 'subreddit' a proper noun?) alone has 20 thousand members, and the Discord server has 14 thousand more. I'm certain not all of those people are speakers, but these are indicators that the number is far higher than "500–5000". ThaNook (talk) 00:16, 19 May 2024 (UTC) Reply
- We're to use more reliable sources, even if they're not that new. Subreddit population is too suspicious to be used as a reference to speaker count. CuSO4, who non-natively speaks English 14:52, 20 May 2024 (UTC) Reply
- The Toki Pona Census should probably be considered as reliable, is there some kind of guideline for how proficient speakers should be? 2A00:801:715:B810:0:0:4707:EE00 (talk) 06:48, 4 November 2025 (UTC) Reply
- I do not think it meets WP:RS. Kovcszaln6 (talk) 16:49, 4 November 2025 (UTC) Reply
- You know, the current source is the ISO request, which uses Toki Pona census data as its lower bound. Using that, the lower bound should be 1000-1100. The upper bound uses (outdated) membership counts of online communities, which have since then grown fourfold (except Facebook, with ~50% growth. Approximately applying their method for newer statistics, we would get 7000-20000 as the upper bound. ~2025-43336-78 (talk) 00:21, 27 December 2025 (UTC) Reply
- Or at least something of that order ~2025-43336-78 (talk) 00:41, 27 December 2025 (UTC) Reply
- I've thought about this, and using the Toki Pona Census should be fine as long as we use in-text attribution, something along the lines of "According to the community-run 2024 Toki Pona Census, ..." --Kovcszaln6 (talk) 09:10, 27 December 2025 (UTC) Reply
- I removed a citation to the Toki Pona Census a few years ago. Despite its general acceptance as semi-authoritative in the Toki Pona community, I can't point to anything that would make it a valid use of a self-published source. In-text attribution doesn't give carte blanche to cite unreliable sources, and the census is just that: There's no institutional oversight of who ran it, no public audit (AFAIK) of whether there were polling irregularities or anything, and even assuming no issues on either of those fronts it's still based entirely on self-reporting. If it gets cited somewhere else, then it would be appropriate for an in-text mention, ideally accompanied by a paraphrase of however that secondary source characterizes its reliability. I'm not even sure if the current citation to the ISO request is appropriate. That's still a self-published primary source, and SIL granting the request doesn't mean they endorsed every claim made in it. At a minimum we should probably change "500–5000 (2021)" to "500–5000 (2021 community estimate)" or something else that makes clear this was some ballpark math done by some community representatives, not a scientific enumeration. -- Tamzin [cetacean needed] (they|xe|🤷) 09:48, 27 December 2025 (UTC) Reply
- I agree. On the same note, we should probably remove the other statement that is cited to the census, as that is basically just original research. Kovcszaln6 (talk) 09:57, 27 December 2025 (UTC) Reply
- I, for one, disagree. Numbers of speakers (personally, I prefer the term "users", BTW) are a notoriously difficult issue when it comes to constructed languages, but that doesn't make them irrelevant. That's why it would be good to at least mention the figures that we do have. And for the record, let's not discard primary sources just because they are primary sources, because in many cases they are all we have, and at least in some cases (such as membership of Facebook groups and the like) they are even more reliable than any second-party source. Of course, membership figures of internet groups are problematic in so far that we all know that only a small percentage are active users. Based on my own experience, I'd say that perhaps 1% are hardcore users, about 5% have some limited command of the language, another 10% are interested bystanders and the rest are sympathizers at best, or might not even be aware of the fact that they ever subscribed at all. Nobody can tell how much overlap there is between groups or how many people know the language without being active on social media, and there is hardly any way of establishing one's real proficiency. Still, 1% of 20,000 is not the same thing as 1% of 200. And as for those censuses: even if we assume that every conlang community has a vested interest in making it look more than it actually is, I think they provide valuable info and absolutely deserve to be mentioned with a disclaimer. —IJzeren Jan Uszkiełtu? 12:09, 27 December 2025 (UTC) Reply
- The lack of available reliable sources doesn't make an unreliable source reliable. Yes, a primary source is useful for a narrow claim like "the Facebook group has $amount users". The corresponding claim for the census would be something like "$amount people completed the community-run Toki Pona Census in $year", which is worded narrowly enough to satisfy WP:PRIMARY, but runs into a due weight issue: Are there reliable sources that say the Toki Pona Census is a thing of any significance? Do we have any reason to consider it such other than some of us having personally favorable opinions of it? -- Tamzin [cetacean needed] (they|xe|🤷) 12:23, 27 December 2025 (UTC) Reply
- My point is that the difference between primary and secondary is not the same as the difference between reliable and unreliable. People on Wikipedia often mix up these things, but the difference is crucial. Primary sources don't make a subject notable, but they are not unreliable if placed in the right perspective. If reliable sources that say tokipona.org is indeed the official website of the language, that makes it inherently reliable as a source of information about it. And if that website says that the language has X speakers, there's nothing wrong with writing something like "According to its own website, Toki Pona had X speakers in year Y." —IJzeren Jan Uszkiełtu? 14:38, 27 December 2025 (UTC) Reply
- The lack of available reliable sources doesn't make an unreliable source reliable. Yes, a primary source is useful for a narrow claim like "the Facebook group has $amount users". The corresponding claim for the census would be something like "$amount people completed the community-run Toki Pona Census in $year", which is worded narrowly enough to satisfy WP:PRIMARY, but runs into a due weight issue: Are there reliable sources that say the Toki Pona Census is a thing of any significance? Do we have any reason to consider it such other than some of us having personally favorable opinions of it? -- Tamzin [cetacean needed] (they|xe|🤷) 12:23, 27 December 2025 (UTC) Reply
- I, for one, disagree. Numbers of speakers (personally, I prefer the term "users", BTW) are a notoriously difficult issue when it comes to constructed languages, but that doesn't make them irrelevant. That's why it would be good to at least mention the figures that we do have. And for the record, let's not discard primary sources just because they are primary sources, because in many cases they are all we have, and at least in some cases (such as membership of Facebook groups and the like) they are even more reliable than any second-party source. Of course, membership figures of internet groups are problematic in so far that we all know that only a small percentage are active users. Based on my own experience, I'd say that perhaps 1% are hardcore users, about 5% have some limited command of the language, another 10% are interested bystanders and the rest are sympathizers at best, or might not even be aware of the fact that they ever subscribed at all. Nobody can tell how much overlap there is between groups or how many people know the language without being active on social media, and there is hardly any way of establishing one's real proficiency. Still, 1% of 20,000 is not the same thing as 1% of 200. And as for those censuses: even if we assume that every conlang community has a vested interest in making it look more than it actually is, I think they provide valuable info and absolutely deserve to be mentioned with a disclaimer. —IJzeren Jan Uszkiełtu? 12:09, 27 December 2025 (UTC) Reply
- I agree. On the same note, we should probably remove the other statement that is cited to the census, as that is basically just original research. Kovcszaln6 (talk) 09:57, 27 December 2025 (UTC) Reply
- I removed a citation to the Toki Pona Census a few years ago. Despite its general acceptance as semi-authoritative in the Toki Pona community, I can't point to anything that would make it a valid use of a self-published source. In-text attribution doesn't give carte blanche to cite unreliable sources, and the census is just that: There's no institutional oversight of who ran it, no public audit (AFAIK) of whether there were polling irregularities or anything, and even assuming no issues on either of those fronts it's still based entirely on self-reporting. If it gets cited somewhere else, then it would be appropriate for an in-text mention, ideally accompanied by a paraphrase of however that secondary source characterizes its reliability. I'm not even sure if the current citation to the ISO request is appropriate. That's still a self-published primary source, and SIL granting the request doesn't mean they endorsed every claim made in it. At a minimum we should probably change "500–5000 (2021)" to "500–5000 (2021 community estimate)" or something else that makes clear this was some ballpark math done by some community representatives, not a scientific enumeration. -- Tamzin [cetacean needed] (they|xe|🤷) 09:48, 27 December 2025 (UTC) Reply
- I've thought about this, and using the Toki Pona Census should be fine as long as we use in-text attribution, something along the lines of "According to the community-run 2024 Toki Pona Census, ..." --Kovcszaln6 (talk) 09:10, 27 December 2025 (UTC) Reply
- Or at least something of that order ~2025-43336-78 (talk) 00:41, 27 December 2025 (UTC) Reply
- You know, the current source is the ISO request, which uses Toki Pona census data as its lower bound. Using that, the lower bound should be 1000-1100. The upper bound uses (outdated) membership counts of online communities, which have since then grown fourfold (except Facebook, with ~50% growth. Approximately applying their method for newer statistics, we would get 7000-20000 as the upper bound. ~2025-43336-78 (talk) 00:21, 27 December 2025 (UTC) Reply
- I do not think it meets WP:RS. Kovcszaln6 (talk) 16:49, 4 November 2025 (UTC) Reply
- The Toki Pona Census should probably be considered as reliable, is there some kind of guideline for how proficient speakers should be? 2A00:801:715:B810:0:0:4707:EE00 (talk) 06:48, 4 November 2025 (UTC) Reply
Since this is a constructed language, it would be nice to have a criticism section. Consider, e.g., https://www.reddit.com/r/tokipona/comments/sp52ap/some_critique_from_an_outsider_trying_to_learn/, https://www.reddit.com/r/tokipona/comments/k8n2f4/sorry_but_toki_ponas_ideology_is_scarily_similar/ , but there may be better sources. Kdammers (talk) 18:30, 23 June 2024 (UTC) Reply
- Wikipedia policy is to avoid Criticism sections. But a section detailing 3rd party analysis of its scope, strengths and weaknesses would be useful. Ashmoo (talk) 10:07, 18 October 2024 (UTC) Reply
- so in the intro it says toki pona translates to the language of good, although in the book made by sonja lang it says and i "toki pona itself means 'good language' or 'simple language'" 108.83.210.204 (talk) 03:22, 25 September 2025 (UTC) Reply
- When a content word follows another content word, the second content word needn't always act as a modifier. Such a construction is also used to mark possession. Take, for example, the phrase "tomo jan," which means something along the lines of "the person's building," but not necessarily.
- Thus, "toki pona" can mean both "the good/simple language" or "the language of good/simplicity."
- -Sincerely and aufrichtig, Babelball | talk 22:50, 25 September 2025 (UTC) Reply
- Most words (70%) are disyllabic; about 20% are monosyllables and 10% trisyllables.
Given that there are fewer than 200 words in all, why not use exact numbers here? —Tamfang (talk) 23:56, 16 July 2024 (UTC) Reply
- Percentages give a better idea of the proportions of those slices, which I think is more important than exact numbers. Plus, seeming-rounded percentatges allow for slightly different total word counts.
- I'm open to other reasoning though! --u9000 (they/them • talk) 00:18, 17 July 2024 (UTC) Reply
- There are some words that some people count and others don't. Brain Boy 53 (talk) 20:20, 8 June 2026 (UTC) Reply
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
- Article (edit | visual edit | history) · Article talk (edit | history) · Watch • Watch article reassessment page • Most recent review
- Result: Ijzeren Jan has presented evidence that the article is not broad in coverage enough for GA status. ~~ AirshipJungleman29 (talk) 11:40, 8 December 2024 (UTC) Reply
Article has 4 citation needed tags, two non-primary source needed tags, and two failed verification tags. I think a cleanup might be needed to maintain its GA status. ❤History Theorist❤ 03:38, 25 November 2024 (UTC) Reply
- Three citation needed tags now. The claim of sitelen suwi being inspired by Woodring and Scharf is verifiable in [1] . Bowler the Carmine | talk 16:56, 25 November 2024 (UTC) Reply
References
- ↑ Gabel, Jonathan (2021). "sitelen sitelen acknowledgements and etymology". Jonathan Gabel. Archived from the original on 2022年01月25日. Retrieved 2021年10月22日.
- Comments by User:IJzeren Jan
Let me start with two disclaimers. First of all, I am not too fond of the circus around featured and good articles. And secondly, I know little about Toki Pona, and although I am impressed by its achievements, it's not really my thing. That said, since I was asked for an opinion, I'll be happy to share some thoughts.
Although there's no doubt that this is a really nice, decent article, I can see some space for improvement. To sum it up:
- The article is extremely heavy on references. Not that there's anything wrong with references, au contraire, but there's no reason to provide three or four different sources for just one simple, unquestioned fact. This comes across as an effort to justify the article's existence rather than to prove a fact, by making the references section look more impressive than it actually is. Which is understandable in so far that constructed languages are frequently under attack for being not notable enough, self-promotional etc., but Toki Pona is way beyond that point by now.
- Speaking about sources, it can't go unnoticed that the vast majority of them are primary sources. I would recommend cutting their number somewhat, at least in cases were a reference is not really necessary.
- On the other hand, the sentence "Toki Pona was the subject of some scientific works" is followed by only one footnote, referring to an article of five pages, which has been used for no less than 29 references and whose author is an active member of the Toki Pona community himself. I'm sure there must be some truly independent academic research on Toki Pona by now.
- In the same sentence, it is also written that "it has also been used for artificial intelligence and software tools". However, after briefly skimming through the source I get the impression that it is about potential uses rather than actual usage. The same goes for the third part of the sentence ("as well as a therapeutic method"): the source writes that the professor is "studying the language's usefulness", not that he actually uses it.
- The "purpose" section discusses the design principles of Toki Pona, which is fine. But what I would expect in this section is rather something about its intended goals. Was it a private thing of its creator that unwittingly happened to achieve some popularity, or was it actually intended to be spoken by others? And if so, by whom and why?
- What I would also like to know, and perhaps there are sources indicating it, is this: what exactly explains the popularity of Toki Pona? It's not uncommon for a constructed language to have a few fans, but rarely more than a handful. Why do so many people learn it? Is it because they want to be part of some experiment? Or because of the community? Is Toki Pona to them what Esperanto is to Esperantists, or is it rather what Klingon is to Klingonists?
- The article should at least explain why there are three different writing systems, and also if they are used, and if so, by whom and for what?
- Apparently, Toki Pona has no less than two different signed versions, and the phrase "more widely used in the Toki Pona community" seems to suggest that both are actually in use, which is surprising for such a small language. If that is true, I really would like to know more about that.
- Toki Pona uses capitals only for proper names. Why?
- And speaking about names, it seems like members of the community use a Toki Pona pen name, consisting of "jan" and a bastardised form of their first names, right? Which I guess would make me "jan Jan".:) This would be worth mentioning in the article, too. And what happens if two people have the same name?
- Nevertheless, I would suggest using real names instead of "in-universe" names. Same goes, BTW, for phrases like "Toki Pona: The Language of Good (known as pu)". Things like that make the article look like a from-fans-for-fans kind of thing.
- In-person meetups: it would be interesting to mention the number of attendees. And if it's just two or three people, better leave it out.
- Not that I am such a fan of "criticism" sections, but given the extremely low number of words, the article should present viewpoints on questions like: in how far can Toki Pona be considered a language at all? How do non-users see it? Etc. Several of such questions can be found on Meta.
- Including this one: how does Toki Pona handle complex (f.ex. technical) texts, especially since circumlocutions are subjective? Even if it is possible to translate complicated texts to Toki Pona, can the result be translated back in a reasonably reliable way?
- And at last, lipu tenpo looks great, but adding every individual issue to the list of publications seems a bit of an exaggeration. Instead, I would suggest adding https://liputenpo.org/ to the external links section.
Hope this helps! —IJzeren Jan Uszkiełtu? 21:58, 1 December 2024 (UTC) Reply
The History section claims that pu ‹features 120 main words, plus 4 words presented as synonyms of these›, citing an arXiv article. However, this claim is wrong. There are only three synonyms in The Language of Good: ‹kin›, ‹namako›, and ‹oko›, and they are synonyms of the words ‹a›, ‹sin›, and ‹lukin›, respectively. It would be more appropriate to cite pu here instead of the arXiv article. – Casper (talk) 10:43, 6 December 2024 (UTC) Reply
- There's also ale and ali. Kovcszaln6 (talk) 12:23, 7 December 2024 (UTC) Reply
- ali is a pronunciation variant, which doesn't count as a synonym. ~2026-42359-49 (talk) 20:10, 30 July 2026 (UTC) Reply
- Nevermind. I read a previous thread and it seems to be fixed. ~2026-42359-49 (talk) 20:13, 30 July 2026 (UTC) Reply
- ali is a pronunciation variant, which doesn't count as a synonym. ~2026-42359-49 (talk) 20:10, 30 July 2026 (UTC) Reply
In the History section, a study is mentioned in which ‹it was revealed that the modified vocabulary of Toki Pona significantly outperformed English› in terms of speech recognition. I consider this claim misleading.
The study is not about Toki Pona, but about ROILA. And yes, according to section 2 (p. 251), the initial vocabulary of ROILA is the 115 Toki Pona words*. However, a genetic algorithm is then run on them, in order to construct a vocabulary where the words are as distinct as possible. The eight allowed word types are cvcv, vcvc, vccv, cvcvc, vcvcv, cvcvcv, vccvcv, vcvccv, of which only three follow Toki Pona phonology.
* The study refers to tokipona.org from 2008. I can’t verify which words were in use then, because tokipona.org is excluded from Wayback.
So what does the final vocabulary of ROILA look like? The study regrettably didn’t include it, but from the few examples in table 1 (p. 254), we get to learn that a ROILA translation of ‹I walked to the house› is ‹pito fosit jifi bubas› (I walk past-tense house), probably pronounced as /pɪtɑ fɑsɪt d͡ʒɪfɪ bʌbæs/. This ridiculously far from Toki Pona – I almost feel insulted by how far. For reference, a Toki Pona translation would be ‹mi tawa tomo› (I toward house).
All in all, although mentioning an appearance of Toki Pona in scientific literature is nice, the relation of ROILA to Toki Pona is so weak that I’m very tempted to delete the whole ROILA bit. – Casper (talk) 15:14, 6 December 2024 (UTC) Reply
- I agree. Feel free to remove that bit. Thanks. Kovcszaln6 (talk) 12:30, 7 December 2024 (UTC) Reply
- how on earth is ‹pito fosit jifi bubas› so different (semantically) from ‹mi tawa tomo›? the only difference is the fact that there is a past tense marker. that's all
- i do agree that the phonology is much different NS1729 (talk) 08:57, 17 December 2024 (UTC) Reply
- oh and also i forgot to mention, the current vocab of roila can be found at https://roila.org/language-guide/vocabulary/ which is extremely far from toki pona NS1729 (talk) 09:00, 17 December 2024 (UTC) Reply
ok so first the toki pona dictionary did add 16 community selected words, but also autocratically the word "ku" (referring the the book itself) was added which is now widely accepted making it 17. this itself is not a large problem and i see the benefit of including it this way. however,
"120 words and 4 synonyms" is just wrong. i have no idea where the source just made a 4th word appear from but it isn't there in the book itslef (Which i have a copy of). I do not have however a better source that mentions the true count so i haven't been bold and changed it yet. NS1729 (talk) 08:56, 17 December 2024 (UTC) Reply
- There's also ale and ali. Kovcszaln6 (talk) 15:42, 17 December 2024 (UTC) Reply
- but ale and ali are variants of the same word, not synonyms like namako and sin.
- Kugelfisch002 (talk) 19:35, 1 February 2025 (UTC) Reply
- @Casperdewith, NS1729, and Kugelfisch002: What do you think about this: we change it to "120 words and 3 synonyms", cite pu, and add a footnote about ale/ali? Kovcszaln6 (talk) 19:44, 1 February 2025 (UTC) Reply
- I am in favour of this option. Then we can say that pu defines 3 synonyms, with a correct citation. And we mention ‹ali› separately, which is better, because it doesn’t occur in pu. – Casper (talk) 22:32, 1 February 2025 (UTC) Reply
- @Casperdewith, NS1729, and Kugelfisch002: What do you think about this: we change it to "120 words and 3 synonyms", cite pu, and add a footnote about ale/ali? Kovcszaln6 (talk) 19:44, 1 February 2025 (UTC) Reply
@Kovcszaln6 hi, thanks for clarifying your revert. I want to ask whether it is appropriate to use a primary source to describe a person versus secondary sources before that. while the label of "linguist" is not wrong, it may be misleading to some. would it be better to add a note explaining that? cheers! Juwan (talk) 12:53, 15 March 2025 (UTC) Reply
- @JnpoJuwan: A note explaining this is fine by me; feel free to write one. Thanks. Kovcszaln6 (talk) 12:57, 15 March 2025 (UTC) Reply
- @Kovcszaln6 should I keep it as such: linguist[note] Sonja Lang with the note explaining the context? Juwan (talk) 13:02, 15 March 2025 (UTC) Reply
I've long been confused by this article's use of the term "root". Toki Pona words don't inflect or compound. Very few phrases can even be considered distinct lexical units with meanings greater than the sum of their parts; wikt:Category:Toki Pona multiword terms is mostly proper names and phrasebook listings (and I'm not sure many of the other dozen-or-so entries should be there). "Root" seems misleading, implying that one of these details is not the case. It seems like most sources just speak of "words" anyway. Am I missing any reason for the article not to use "word" instead? AgentMuffin4 (talk) 23:55, 2 May 2025 (UTC) Reply
- @AgentMuffin4: I agree. Feel free to change it. Kovcszaln6 (talk) 08:06, 3 May 2025 (UTC) Reply
- I think "content words" might be a better description. 2A00:801:715:B810:0:0:4707:EE00 (talk) 06:56, 4 November 2025 (UTC) Reply
You are invited to join the discussion at Talk:List of Toki Pona words § Words to include. Kovcszaln6 (talk) 08:10, 3 May 2025 (UTC) Reply
I noticed that in the "Sentence structures" part of the article, it's stated that there are two ways to ask yes/no questions: using "X ala X", and adding "anu seme" to the end of a sentence. The way that it's worded suggests that they are equivalent andhave the same meaning. However, in my experience, "tomo li lete ala lete?" means "Is the house cold?"; whereas "tomo li lete anu seme?" is more like "The house is cold, isn't it?". "X ala X" is like asking a question, and "anu seme" is more like asking for confirmation. I didn't want to change it because I wasn't sure if I had a conflict of interest as a toki pona speaker/learner. Xgav2o (talk) 21:49, 3 June 2025 (UTC) Reply
- Merely speaking a language or being a fan of something isn't enough be a conflict of interest - editors having knowledge and interest in an article's subject matter is usually quite helpful! Though, you'd need a reliable source for this beyond just your own experience.
- On that, I was able to find that the Wasona course describes anu seme as having a different "vibe": "If you use anu seme, it’s more likely that you expect the other person to agree with you", but the nasin toki pona reference says instead that "it suggests that a less simple answer than just "yes" or "no" may be expected". Both of these are linked from tokipona.org which gives some authority to them, but they're definitely still both self-published, so probably not usable here anyway... I think it's honestly fine to just leave the article as-is on this? It is true that both constructions form yes/no questions; there might be some vibes-based difference between them in common usage, but it seems hard to pin down what exactly. twotwos (talk) 07:56, 4 June 2025 (UTC) Reply
- Anu seme means literally "or what" and, I believe, is used the same way. 2A00:801:715:B810:0:0:4707:EE00 (talk) 06:59, 4 November 2025 (UTC) Reply
- We need reliable sources; a belief is not reliable. Kovcszaln6 (talk) 16:51, 4 November 2025 (UTC) Reply
- sona pona is probably not a reliable source in itself, especially as the link to the toki pona forums doesn't work anymore. However, a few courses mention it as expecting the answer to be "yes", most notably wasona and jan Kekan San's Learn Toki Pona!. ~2025-31157-62 (talk) 09:44, 6 November 2025 (UTC) Reply
- Those are still not reliable. Kovcszaln6 (talk) 16:54, 6 November 2025 (UTC) Reply
- sona pona is probably not a reliable source in itself, especially as the link to the toki pona forums doesn't work anymore. However, a few courses mention it as expecting the answer to be "yes", most notably wasona and jan Kekan San's Learn Toki Pona!. ~2025-31157-62 (talk) 09:44, 6 November 2025 (UTC) Reply
- We need reliable sources; a belief is not reliable. Kovcszaln6 (talk) 16:51, 4 November 2025 (UTC) Reply
- Anu seme means literally "or what" and, I believe, is used the same way. 2A00:801:715:B810:0:0:4707:EE00 (talk) 06:59, 4 November 2025 (UTC) Reply
The following edit was reverted without explanation:
- Furthermore, most words can be used in multiple grammatical categories, with related meanings. For example tawa can mean for or towards as a preposition, moving as an adjective, or to go as verb.
According to 'Toki Pona: The Language of Good', this is an essential feature of the language, and the reason she was able to keep the vocabular so low. Why was this reverted? Perhaps I am just missing this explanation elsewhere in the article? seberle (talk) 11:36, 12 August 2025 (UTC) Reply
- @Seberle:
The following edit was reverted without explanation
Incorrect. I reverted your edit with the explanationUnsourced + already in the Grammar section (as moku)
. Please note that Toki Pona#Grammar already states this but with moku:the word moku means "to eat" in the verb position, but means "food" (that which is eaten) in the noun position, and might mean "edible" (of or relating to eating) as an adjective.
So I find it unnecessary to basically repeat this. Also, you did not cite a (secondary) source. Kovcszaln6 (talk) 11:56, 12 August 2025 (UTC) Reply- Thanks for the explanation. I'm glad to know this is covered in the article after all. When I saw the undo, it said something about reverting a good faith effort, but no other explanation. But now I see it shows the explanation you stated. I guess I must have misclicked or misread something! Sorry. seberle (talk) 22:28, 15 August 2025 (UTC) Reply
The description of colors mentions combining specific color words, but makes no mention of the practice of using other words to further describe color. A good example (Given the previous text about excluding colors common in western languages) might be laso kasi to describe green distinctly from blue.
I appreciate that combining words this way is a core part of the language described elsewhere, but I think this small example could be helpful to unfamiliar readers. Does anyone have any objections to adding this in? Damp Cuttlefish (talk) 14:18, 12 August 2025 (UTC) Reply
- @Damp Cuttlefish: Sure, but only if you can cite a secondary reliable source that explicitly states this example, otherwise it would be original research. Kovcszaln6 (talk) 15:03, 12 August 2025 (UTC) Reply
- A trickier task than I had imagined! I would suggest given the fairly uncontroversial nature of the statement that a primary source might suffice, but even then I can't find anything that meets the bar for reliability. Damp Cuttlefish (talk) 16:07, 12 August 2025 (UTC) Reply
- https://nimi.li/ilo-ku/laso-kasi uses lipu ku as a source, so it's reliable.
- -Sincerely and aufrichtig, Babelball | talk 00:20, 12 September 2025 (UTC) Reply
- Thank you! I have since dug my copy of ku out of storage and can confirm that particular example is there with frequency index 41-60%. I've added it to the section with the qualifier 'common amongst english speakers' referencing the relavant pages of ku directly. Damp Cuttlefish (talk) 17:04, 27 April 2026 (UTC) Reply
- The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
- The result of this discussion was not merged mostly due to page length, although editors have noted its use by other languages. Kovcszaln6 (talk) 13:11, 29 December 2025 (UTC) Reply
Proposed merge from Sitelen Pona to Toki Pona : Article is very niche and is a topic about an uncommonly used, niche writing system in an uncommonly used, niche constructed language; it does not meet WP:N. It has its own section on the Toki Pona article, thus can be merged thereinto. from Piperium (chit-chat, i did that) at 13:26, 10 December 2025 (UTC) Reply
- Sitelen Pona is not exclusive to Toki Pona. Other conlanguages can and use this script. Also they seem to be 2 different, yet related, entities. In that case, sanskrit should be merged with India or the region, German Fraktur with all the germanic regions that used/use it, Hiragana, katakana and so on. I think it is fair to have them seperated. Toki Pona also being the second largest, in community size, con-lang after Esperanto and recently having gotten tok.wikipedia.org, does suggest that is not more niche than many other articles on wikipedia.
- A script may be related in origin to a language or region, but it does not make it exclusive to it. Peter Alfons Loch (talk) 15:16, 11 December 2025 (UTC) Reply
- The article does not mention any use beyond Toki Pona; I know there's some but not, I don't think, in any language we could make reliably-sourced statements about. To me the main question is one of page length. The Sitelen Pona article has two reasonably long prose sections and two large graphical elements. That may be reason enough to keep it separate. -- Tamzin [cetacean needed] (they|xe|🤷) 21:05, 13 December 2025 (UTC) Reply
- No opinion on the merger itself but your comparison to the other language script seems inaccurate. These scripts weren't invented in 2014, and even if they were, in the cases of some languages not written down untill recently, if they have enough coverage (there are usually linguists who study these) then they can have their own articles. User:Easternsahara review this 02:18, 14 December 2025 (UTC) Reply
- I believe the articles should stay separate, just like there is a difference between the articles for Chinese and Hanzi. --ValtteriLahti12 (talk) 14:32, 15 December 2025 (UTC) Reply
- Do we have any evidence for this being used in other languages, or just a baseless claim? ~2025-42832-93 (talk) 23:37, 24 December 2025 (UTC) Reply
- I am unsure of what degree they mean, but I do know quite a few languages based off toki pona do use either sitelen pona outright, else a form of it;
tuki tiki is one that uses a sitelen pona based script, for instance.
for the most part, several other constructed scripts made solely for constructed languages, do in fact have their own pages, given they (like this one) have sufficient length and reason for separation.
DigiSpaceProductions (talk) 01:03, 25 December 2025 (UTC) Reply- I'll reïterate, contra a few comments below, that it doesn't matter if other conlangs use Sitelen Pona, or similar scripts, if that isn't verifiable in reliable sources. That said, I still lean opposed to a merge due to page length. -- Tamzin [cetacean needed] (they|xe|🤷) 03:51, 29 December 2025 (UTC) Reply
- I am unsure of what degree they mean, but I do know quite a few languages based off toki pona do use either sitelen pona outright, else a form of it;
- Weak oppose Given what Peter and Tamzin said, I think there is a decent case to keep them separate. I was going to support, and say we could always split later, but if thats a significant possibility, best to keep them separate. ←Metallurgist (talk) 00:07, 16 December 2025 (UTC) Reply
- Weak support I personally think that Sitelen Pona and Toki Pona are in the same sort of category, and SP is a writing system specifically tailored to Toki Pona. There really isn't much to say about Sitelen Pona aside from its use in Toki Pona, and it doesn't have much history besides being icons associated with Toki Pona words. There's not a big chance it might be important enough to warrant a separation in my opinion, but I really do think that the two are very connected. The reason I don't fully commit to yes is because it's still very big so it might be hard to include as a chapter. AtTheTownHouse (talk) 02:32, 21 December 2025 (UTC) Reply
- Oppose
- Though Sitelen Pona is made for (and used by) toki pona, it isn't the sole writing system for the language (though it is the 2nd most used, 2nd to Latin), and if the articles were merged
- In their current states, it might be a bit large for a chapter in the toki pona article; and due to that, I think having it a separate article works well.
- Additionally, there are other conlangs (normally based off toki pona) that use scripts inspired by (or outright just using) the Sitelen Pona script, one case of one using a modified Sitelen Pona is tiki tuki. (Which is one of the more popular toki pona-based languages)
Also, other constructed scripts have their own articles, such as Klingon's different scripts and Tolkien's Sarati and Tengwar scripts, as a couple examples.
If anything, a article for Toki Pona's various scripts could be done instead, discussing Sitelen Pona and any other scripts that may be discussed, similar to the Klingon Script article.
DigiSpaceProductions (talk) 21:27, 21 December 2025 (UTC) Reply - Neutral I am not really convinced that Sitelen Pona is independently notable from Toki Pona. Indeed, finding adequate sourcing for the various aspects of Toki Pona itself is difficult, and Sitelen Pona is no exception to that (which is a shame because SP is fairly interesting as it is). I am not personally aware of any serious usage of Sitelen Pona outside of Toki Pona, and the derived scripts do not seem to be significant enough for me to support retaining the article. However, I am concerned about the page length of including everything that's in the Sitelen Pona article into the Toki Pona article. Windfarmer — talk 23:23, 26 December 2025 (UTC) Reply
- Oppose the merger.
- This merge would simply increase the confusion around this language in the same way that merging the latin alphabet and the Latin language would. Sitelen pona is not the only way to write toki pona, and toki pona is not the only language that uses a sitelen-pona-family script, in a manner similar to the latin-Latin situation. To equate sitelen pona with toki pona, insted of simply relating them, is to show a fundamental misunderstanding of this writing system, which is one possible option for written sitelen pona.
- The sitelen pona article is multiple paragraphs already and is still growing. There exists the possibility of adding entire sections about nasin sitelen kalama pi linja lili and other common techniques. This article merger would likely lose information by crammimg what should honestly be more than two articles into one. It would also provide editors with the difficult task of managing one central article for an entire community instead of having specialty articles about the distinct options available to users. CodingPhys (talk) 20:44, 28 December 2025 (UTC) Reply
- Oppose You saying that Toki Pona is an "uncommonly used, niche constructed language" suggests you have little to no, if not zero, knowledge about the conlang and its community. Toki Pona has thousands of speakers, its own ISO 639-3 language code, and even its own Wikipedia version. There is no shortage of sources. Even this page already has 21 sources and sona pona has a media coverage page with a bunch of times the conlang has been mentioned in media. Also, Sitelen pona is by far not the only writing system for Toki Pona anyways. Anpang (talk) 03:28, 29 December 2025 (UTC) Reply
nasin nanpa pona is a more advanced number system that makes it somewhat reasonable to express large numbers. It's used in Sonja Lang's more recent offical titles, specifically to transcribe the copyright year. For numbers less than 100, it matches the method detailed in The Language of Good, though beyond this, it functions as a base-100 system with digits separated by the word ale.
For example, 2025 becomes mute ale mute luka ({\displaystyle 20\times 100+20+5}).
I'd like to add this to the section on numbers, though unless there are good secondary sources it's probably not going on there. SuperminerJG (talk) 05:24, 23 December 2025 (UTC) Reply
- I've removed this addition because there is no secondary coverage. Wikipedia follows reliable sources, even if that sometimes means leaving out things we know to be true and personally find interesting. -- Tamzin [cetacean needed] (they|xe|🤷) 07:22, 23 December 2025 (UTC) Reply
The creator of Toki Pona has been using the name "Amatullah" for herself quite consistently over the past few months as opposed to the name "Sonja". Hence, I propose changing all mentions (other than where her birth name is listed) of the name "Sonja" to "Amatullah" to address her in the way that she wishes. ItsRadeZ 01:12, 28 April 2026 (UTC) Reply
- Wikipedia writes what reliable sources state, and they use "Sonja Lang". Kovcszaln6 (talk) 15:26, 28 April 2026 (UTC) Reply
- Would she not be a reliable source given that it's her own name? ItsRadeZ 01:31, 29 April 2026 (UTC) Reply
- That's not the problem. Wikipedia articles reflect on and summarize what reliable sources write. If they starting using a different name, then we'll switch to that. But until then, we'll be using what is already widespread in sources. Kovcszaln6 (talk) 13:54, 29 April 2026 (UTC) Reply
- But see also e.g. MOS:DEADNAME, where we are to use the subject's latest preferred name regardless of sourcing. However, based on the apparent circumstances regarding this name change (now removed and revdel'd from the article), we can afford to wait a little bit. Children Will Listen (🐄 talk, 🫘 contribs) 15:53, 31 May 2026 (UTC) Reply
- That's not the problem. Wikipedia articles reflect on and summarize what reliable sources write. If they starting using a different name, then we'll switch to that. But until then, we'll be using what is already widespread in sources. Kovcszaln6 (talk) 13:54, 29 April 2026 (UTC) Reply
- Would she not be a reliable source given that it's her own name? ItsRadeZ 01:31, 29 April 2026 (UTC) Reply
- the name change is complicated due to the nature of its circumstances. we should not make any decisions as of now, until that is first resolved. to further note, under a strict reading, MOS:DEADNAME does not apply, though the recent RFC regarding other types of name changes might. Juwan 🕊️ 🌈 10:43, 2 June 2026 (UTC) Reply
- Regarding Wikipedia's deadname policy, that may be subject to change soon as discussion has been sparked on how it handles some cases differently for lack of an apparent reason (ie. using slave names instead of chosen names when referring to people's past) so we should look out for that as well. ItsRadeZ 10:30, 3 June 2026 (UTC) Reply
- this is what I was referring to in the last sentence. Juwan 🕊️ 🌈 10:35, 3 June 2026 (UTC) Reply
- Regarding Wikipedia's deadname policy, that may be subject to change soon as discussion has been sparked on how it handles some cases differently for lack of an apparent reason (ie. using slave names instead of chosen names when referring to people's past) so we should look out for that as well. ItsRadeZ 10:30, 3 June 2026 (UTC) Reply