Languages of Taiwan — Seediq (Kari Sediq, Kari Seediq, Kari Seejiq)

Moving down the list of Formosan languages in Taiwan, we have the other member of the Atayalic branch, Seediq. Spoken in the mountains of Central and Eastern Taiwan by the Seediq and Taroko, this language is predominantly found in the counties of Hualien and Nantou. However, the number of native speakers prove concerning, although we do not really know how many speakers of Seediq remain. One of the few studies was in 2012, when it was recorded that there were less than 10 000 native speakers of Seediq. If this number were true, it would place the language on the border of being endangered, or vulnerable to endangerment.

Also known by Kari Seediq, Kari Sediq or Kari Seejiq to native speakers or 賽德克語 (Hanyu Pinyin: sài dé kè yǔ) by Mandarin speakers, Seediq is largely split into three main dialects, and members of each dialect group often refer to themselves by the name of the dialect, rather than the language. The Amis, on the other hand, refer to Seediq speakers as “Taroko”, which sounds a bit like one of the dialect groups of Seediq. In no order, the three main dialects are:

  • Truku (Truku, 德路固方言 dé lù gù fāng yán), similar to Taroko
  • Toda (Tuuda, 都達方言 dōu dá fāngyán
  • Tgdaya (Tkdaya, Paran, 德固達雅方言 dé gù dá fāng yán)

While speakers of these three dialects often refer to themselves by their dialect group, these dialects still share similar grammars, words, and are still mutually intelligible with one another. Interestingly, the Truku dialect has similar origins of the Taroko, and some argue that these are basically the same language. In the field of language classification, this forms a really tricky grey area to study. In school systems, the Seediq language certification tests and assessments in Taiwan is divided into the three respective dialect groups, rather than being assessed as one “standard” language altogether.

The phonology of the Seediq language features a total of 19 consonants and 4 vowels. There are not really any special consonant sounds to take note of, but among the vowels, only the /a/, /i/, /u/ and the schwa are used. Linguists argue that Seediq syllable structures consist of consonant-only, consonant-vowel, and consonant-vowel-consonant order. Some interjections, or colloquial exclamations, can accommodate two consonants in a cluster, such as “sawp”, the sound of an object blown by the wind. However, in dialects like Taroko/Truku, consonant onsets of up to six are allowed. But…there is a catch.

See, Seediq, like Atayal, may use the Latin alphabet to write the language, but there are a few quirks some would like to take note. One might encounter words with seemingly impossible-to-pronounce consonant clusters, something that one might relate to Kartvelian languages like Georgian. Words like “qmpah dxral” (farmer, Toda dialect), “qnpahan” (farmland, Toda dialect) and “lmlung” (think, Toda dialect) would pop up, raising questions on how these words are pronounced. One must notice, that the schwa vowel is often not written in some dialects, hence reflecting as crazy consonant clusters in text.

At this point, it should go without saying that Seediq bears many characteristics of Austronesian languages, along with similar grammatical bits shared among the Atayalic and Formosan languages. Seediq uses the verb-object-subject word order, pretty much like in Malagasy, but there are some elements of the use of the predicate model as discussed in Atayal and Amis. Similarly, there are modifying particles added to spice up the grammatical features of a sentence or clause, as well as predicate extenders to indicate further information such as some adverbial concepts and tense. One important feature to take note of is the use of reduplication of some syllables to form the plural, as this is pretty much observed in many Austronesian languages like Malay. There are two patterns of reduplication though, one which involves only the first syllable of the stem, with structure Cə-CV(C), and one which involves the last pair of syllables of the stem excluding the final consonant of the syllable, having structure CəCə-CV(C)CV(C). So for the first pattern, one example is “qhuni”– “qqhuni” (tree — trees), and for the second pattern, an example is “se’diq” — “sedese’diq” (person — people). This is just one way how nouns are modified in the language’s grammar.

Interestingly, the numerical system of Seediq seems to bring back more cognates as compared to Atayal, giving us slightly more familiar words for certain numbers. The cardinal numbers from 1 to 10 are as follows:

  1. kingal
  2. deha
  3. teru
  4. sepat
  5. rima
  6. mataru
  7. mpitu
  8. maspat
  9. mengari
  10. maxal

Among these, the Seediq words for “five” and “four” ring closely to the words for the same numbers in other Austronesian languages, although other numbers may still sound quite unusual when compared to the other Austronesian languages.

Just like the Atayal language, if you want to learn the Seediq language, I have to tell you that many resources are in traditional Chinese. However, there is one central site that documents the words, dialogues, audio and pronunciation of Seediq and its respective dialects, along with most of the other Formosan languages. With multimedia materials, translations of books and stories into respective indigenous languages, this serves as a platform for people to learn about the cultures and languages of Taiwan, as well as preserving the Formosan languages in the digital world. Access it here at: This is so detailed, that even known individual dialects are compiled, and users are prompted to choose a dialect to learn that falls under the target language.


The first time I heard about the Seediq language was on some Taiwanese reality television programme, though I could not remember the details and stuff beyond that. There were several terminology and classification issues I encountered when writing this post, such as the consideration of the Taroko dialect group, and how Seediq and Atayal were related in one way or another. Should you, readers, have a more complete picture about these classifications, please let me know about them through the contact form.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s