For a great many of us language learners, knowing when a particular case is used, and its associated endings are probably frustrating enough. Nouns and adjectives are probably the word categories where one would find these endings the most, while other languages might do things a bit differently and add case particles or markers instead. But what if I told you that there is a language that does the same thing, but for verbs as well?

In the South Wellesley Islands of Queensland, Australia, there is an indigenous language spoken by the Kaiadilt and Yangkaal people groups, who are among the last of the Aboriginal people groups who made full contact with Europeans, with some as late as 1948. But this language is extremely close to extinction. Among the several hundred Kaiadilt and Yangkaal peoples, only 7 or 8 native speakers remain. The last speaker of the classical form of the language passed away in 2015, although there is one other speaker reported as late as 2017. In any case, this is a language on the brink of extinction, much like many of the 300 or so indigenous languages spoken in Australia. This is Kayardild.

The Kayardild language is part of the Tangkic language family, although some linguists hypothesise that this could be a divergent branch from the proposed Greater-Pama-Nyungan language family. This puts it related to languages like Yukulta, remarked to be mutually intelligible with Kayardild, and the relatively divergent Lardil (and its special register called Damin).

Kayardild’s phonological inventory looks… fairly typical of the sounds of Aboriginal Australian languages, missing a voied-voiceless consonant distinction, lacking sounds like /s/ and /z/, and having a very minimal set of vowels, with Kayardild having three vowels /a/, /i/, and /u/, distinguished by vowel length. And for each of the plosive consonants like /k/, /t/, and /p/, there are nasal parallels /ŋ/, /n/, and /m/ respectively. This makes for a total of 17 consonants and 3 vowels (or 6 when factoring in vowel length), which makes for a fairly average set of sounds in an Aboriginal Australian language.

But that is not what you are here for, right? You clicked here to look at the grammar of this language. And that is where the more interesting bits are.

Kayardild is one of the few languages in the world that features case stacking. This phenomenon, also curiously known as “Suffixaufnahme” (or literally “taking up of suffixes” in German), features a noun phrase with a possessor and a possessed noun. All the words of that noun phrase can share the same case that is designated to the entire phrase. As the possessor is usually the one denoted in the genitive case in most scenarios, the additional information that is conjugated to it (i.e., noun cases) become stacked, hence we get a case of Suffixaufnahme.

In Kayardild, the possessor can be denoted by the genitive or the ablative case, and additional case functions like those pertaining to time, space, and other relations to nouns in the clause can be applied to the entire noun phrase. Take, for example, this gloss provided by Evans (1995):

• dan-kinaba-nguni dangka-naba-nguni mirra-nguni walbu-nguni
• this-ABL-INS man-ABL-INS good-INS raft-INS
• ‘With this man’s good raft’

In this noun phrase, ‘this man’ is shown here as the possessor of the ‘good raft’, taking up an ablative case. When the speaker wants to express the use of ‘this man’s good raft’ as a tool for something, the instrumental case is used, and is applied to the entire noun phrase, including that of the possessor. This makes it such that the instrumental case is attached to the conjugation of the words ‘this’ and ‘man’ in the ablative case, forming Suffixaufnahme. This also works with the modal ablative case, which is used when the verb is conjugated with a suffix marking the past tense. In Kayardild, this case stacking phenomenon works for up to four levels of stacked cases, giving speakers additional layers to perform linguistic functions like pointing out interclausal relations.

This leads us to this conjugation of verbs by case. Interestingly, case stacking does indeed apply to verbs as well. For example, the oblique case marker can be used to convey a non-indicative mood, such as a deduction like “Person A must have done something”. Such a case could be applied for the entire clause, including the verb as well.

For a verb, it could have a structure consisting of a stem, final inflection, and a complementising case or a nominal suffix. This is largely limited to the locative or the oblique cases, but serves a grammatical purpose in signaling subordinate clauses. Additionally, some final inflections have corresponding cognate cases, such that the noun that functions as an ‘object’ in the clause has to agree in case. One example is the “directed” final inflection in the noun, which always require the noun to take the allative modal case. One special final inflection unique to Kayardild is the “resultative nominalisation”, which denotes a current state that implies a preceding event. Although technically a nominalisation rather than purely a “case ending”, it still is a rather unique feature among its Tangkic counterparts. Other nominalisations can be inflected for case as well, with similar case-stacking rules applying to them.

Lastly, we also have to talk about the verbal case. No, those are not really an extension of conjugating verbs by case, but more rather, conjugating a noun with verbal endings that agree with verbal tenses and moods. With verbal cases, it is indeed possible to do away with the main verb in favour of something that only uses verbal cases. Take, for example, the Kayardild word for “to go”, warra. This can be omitted in the sentence “I will go to that cave”, using instead the verbal allative case to denote a direction of motion towards “that cave”. That glosses as:

• ngada dathin-kiiwa-thu ngilirr-iiwa-thu
• 1sgNOM that-VALL-POT cave-VALL-POT
• ‘I will go to that cave.’

There are a total of seven verbal cases in Kayardild, and their presence has sparked some debate over how these verbal cases arise.

While the most significant advances in understanding of the Kayardild language were made by Norman Tindale, more recent publications covering this language were written by Nicholas Evans. The 1995 publication of Evans’ A Grammar of Kayardild is the primary source for this introduction to Kayardild. I encourage you to check it out, although there are many linguistics jargon you might have to get used to. Nevertheless, there are some pages dedicated to the culture and geography of the people who speak Kayardild, which undoubtedly have been interesting reads.

Even as new deductions and discoveries about Kayardild and its interesting grammatical phenomena are continually made, there are some concerns over the language’s survival over the coming years and decades. With barely a handful of native speakers remaining, it is heartening to see some community projects aimed at preserving the recordings of Kayardild, garnering interest in this Aboriginal Australian language. However, from what I could find, there does not seem to be any plans to teach Kayardild to a wider audience like the Kaiadilt and Yanggal communities, or perhaps I was not looking in the right places. If you are in Australia and have come across more Kayardild revitalisation projects or movements, please share it with us.