# Kuku Yalanji — Sample Production Pages *What the three living books look like when the raw working files are run through an editor into finished form. Every example below is a real, verified sentence from the corpus (or a minimal clause drawn from one). Three products grow out of the corpus: a **Teaching Grammar**, a **Learner's Dictionary**, and an **Interlinear Reader**. One sample page of each follows.* *Current build note: refreshed 2026-07-10 after Batch 399 / Rule 357 at 20,047 verified sentences; process tag `batch-399`. The full 50-row exact complement, not a three-row threshold patch, completes the selected non-singular 100-cell grid and all 200 fish-set-F cells. Every candidate was absent and every opposite was the corresponding Batch 398 row. Patz clusivity, the dictionary-pronoun defect, grammatical deferred `ngana`, two-card fish inventories, linter exclusions, whole-head analysis, and all sense/property/source/use/event boundaries remain explicit. The separate milestone completion audit has now passed.* *Glossing follows the Leipzig conventions. Small capitals (rendered here as UPPERCASE) mark grammatical categories: `ERG` ergative, `ABS` absolutive (unmarked), `DAT` dative, `LOC` locative, `ABL` ablative, `COMIT` comitative, `POSS` possessive, `INST` instrumental, `PURP` purposive, `SUB` subordinate, `REDUP` reduplication, `KPL` kinship plural, `NPST` non-past, `PST` past, `1/2/3` person, `sg/pl` number. Kuku Yalanji is written in the practical (community) orthography; `ng` and `ny` are single nasal sounds, `rr` a trilled *r*, plain `r` a glide.* --- --- # BOOK ONE · The Teaching Grammar ### *Yalanji Kuku* — A Grammar for Learners ## Chapter 4. The Case System: How Kuku Yalanji Marks Who Does What ### 4.1 An ergative language English tells you who did what mainly by **word order**: *the dog bit the man* and *the man bit the dog* use the same words and mean opposite things only because of the order. Kuku Yalanji works differently. Word order is free — you can put the words in almost any order you like — because the **endings on the nouns** carry the information instead. The central pattern is called **ergativity**, and once you see it, a great deal of the language falls into place. The rule is this: > The **doer of an action that affects something else** (the subject of a transitive verb) takes a special ending, the **ergative** (`-ngka`, `-bu`, `-ju`, and other forms). Everything else — the **thing affected** (the object), and the **subject of an action that affects nothing** (the subject of an intransitive verb) — takes **no ending at all**. That plain, unmarked form is called the **absolutive**. So the doer is marked and stands out; the plain form does double duty for the object and for the simple subject. This is the opposite of English (and Latin, and most European languages), and it is the single most important thing to learn. **(1)** A simple intransitive subject — no ending: | | *Karrkay* | *dunga-y.* | |---|---|---| | | child.ABS | go-NPST | | ‘The child is going.’ | **(2)** A transitive doer — the ergative ending `-ngka`: | | *Karrkay-angka* | *mayi* | *nuka-ny.* | |---|---|---|---| | | child-ERG | food.ABS | eat-PST | | ‘The child ate the food.’ | Notice that in (2) *karrkay* ‘child’, now the **doer**, takes `-angka`, while *mayi* ‘food’, the **thing affected**, stays bare. Compare the bare *karrkay* in (1), where the child is simply moving and affecting nothing. Same noun, two roles, two forms. ### 4.2 The "potent" ergative and instruments The ergative ending also marks the **instrument** — the thing an action is done *with*. Kuku Yalanji treats "the man speared it **with a spear**" and "**the spear-wielding man** speared it" as closely related, and the corpus glosses the transitive-subject ergative as `ERG:pt` ("potent") to keep it distinct from the pure instrumental where that helps the learner. **(3)** Instrument in `-bu`, alongside an ergative subject: | | *Bubu* | *marri* | *juku-bu* | *balka* | *nyubun-bu* | *jilba-nyjiku.* | |---|---|---|---|---|---|---| | | ground.ABS | always | stick-INST | hit.NPST | one-INST(=first) | walk-before | | ‘Always beat the ground with a stick before you walk (through).’ | | | | | | | ### 4.3 The one place there IS an object ending: pronouns There is a single, important exception to "the object takes no ending." **Pronouns** — *I, you, he* — *do* have a distinct object form, called the **accusative**. This is a genuine feature of the language, not a slip: while a common-noun object is bare, a pronoun object is marked. | person | subject (‘…’) | object (‘…me/you/him’) | |---|---|---| | 1sg | *ngayu* | **nganya** | | 2sg | *yundu* | **yununy** | | 3sg | *nyulu* | **nyunguny** | **(4)** A pronoun object in the accusative: | | *Mukul-dingkar-angka* | *nyunguny* | *balka* | *kuli-bu.* | |---|---|---|---|---| | | old-man-ERG | 3sg.ACC | speak.NPST | anger-INST | | ‘The old man speaks to him sternly (lit. with anger).’ | | | | | > **⚠ Common learner error.** Do **not** invent an object ending for ordinary nouns on the English model. There is no `*-nha` "accusative" on a noun like *mayi* ‘food’ or *kadar* ‘wallaby’ — the object is simply bare. The ending in (4) is there only because *nyunguny* is a **pronoun**. (For the addressee of *balka* ‘tell, speak to’, use the **dative** `-nka`, not any accusative.) ### 4.4 Case summary | case | typical ending | marks | |---|---|---| | absolutive | — (bare) | intransitive subject; object | | ergative / instrumental | `-ngka` ~ `-bu` ~ `-ju` | transitive subject; instrument | | dative | `-nka` | goal, purpose, addressee, beneficiary | | locative | `-ngu`~`-nga`~`-a`~`-ba`~`-mba` (by final sound) | location, time | | ablative | `-muny` | source, ‘from’, ‘since’ | | comitative | `-jV(rr)` / `-ji` | ‘with, having’ | The locative's five shapes are governed by the last sound of the word (Chapter 3.4); e.g. *jukar-a* ‘on the sand’ (after `-r`), *kalbay-mba* ‘far off’ (after `-y`), *wumbul-ba* ‘in the heat’ (after `-l`). --- ## Chapter 12. Asking Questions Kuku Yalanji has two ways of asking, and they are kept strictly apart. ### 12.1 Content questions (who, what, where…) To ask *who? what? where?* you use a question word. Unlike English, you do **not** have to move it to the front — it is stressed, and it can stand where the answering word would stand. The verb keeps its ordinary tense. The two most basic question words behave differently from each other, in a way that reflects §4: - **‘who’ = *wanya*** inflects like a **noun**, on the ergative/absolutive pattern: bare *wanya* for a subject-or-object, but *wanya-ngka* (ergative) when ‘who’ is the doer of a transitive verb. - **‘what’ = *wanyu*** takes the neutral case set, and — beautifully — its case-forms *are* the other question words: the dative *wanyurrimun* ‘for what (purpose)’ **is the word for ‘why’**, and the comitative *wanyurri-nji* ‘with what’ is the fixed way to ask after an illness. **(5)** ‘what’ as the plain subject of a description: | | *Wanyu* | *yinya* | *yalbay* | *ngumbu* | *janka* | *wuna-nya* | *jukar-a?* | |---|---|---|---|---|---|---|---| | | what | that | big | dark | still | lie-SUB | sand-LOC | | ‘What is that big dark thing lying still there on the sand?’ | | | | | | | | **(6)** ‘where’, with the verb in its ordinary non-past — this is also the everyday greeting: | | *Yurra* | *wanjabu* | *dunga-y?* | |---|---|---|---| | | 2pl | where | go-NPST | | ‘Where are you (all) going?’ | | | | **(7)** ‘what… with?’ — the comitative of ‘what’, used for sickness: | | *Yundu* | *wanyurri-nji* | *bambay?* | |---|---|---|---| | | 2sg | what-COMIT | sick | | ‘What are you ill with?’ | | | | The full inventory: *wanya* who · *wanyu* what · *wanjabu* where · *wanjamuny* where-from · *wanja-wanja* when · *wanjarr* how · *wanjarrinya* how many · *wanyurrimun* why. For ‘something’ (never *wanyu*, which is only ever a question) use the indefinite *junjuy*. ### 12.2 Yes/no questions — nothing but the voice A yes/no question has **no question word and no little question particle**. It is spoken in exactly the same order as a plain statement; only the **rising tone of voice** marks it as a question. In writing, only the question mark tells them apart. **(8)** | | *Yundu* | *ngananji* | *kadar* | *balbi-nka?* | |---|---|---|---|---| | | 2sg | 1pl.COMIT | wallaby.ABS | hunt-PURP | | ‘Are you coming (with us) to hunt wallaby?’ | | | | | Said with a falling tone, the very same words would be a statement, ‘You are coming to hunt wallaby with us.’ The question lives entirely in the melody. --- --- # BOOK TWO · The Learner's Dictionary ### *Kuku Yalanji – English* · Sample page: **K – ka·lnga to ku·ru** *Each entry gives the headword in the community orthography, the part of speech, the meaning(s), and — where the corpus attests it — a real example sentence with translation. ▸ marks a usage or cultural note; → is a cross-reference. Kinship abbreviations: M mother, F father, B brother, Z sister, so MB = mother's brother.* --- **kalnga** *n.* — **mother's brother** (MB); maternal uncle. : *Karrkay-mu kalnga dingkar junkurr.* ‘A child's mother's brother is a man of special weight.’ : ▸ A figure of his own authority, not a "lesser father." The *kalnga* gives a boy his first spear and teaches him the hard law, speaking sternly where the father is gentle. → **nganjan** (F), **bimakay** (FZ). **kami** *n.* — **grandfather**, specifically the father's father (FF); paternal grandfather. : ▸ Distinguished from **ngaji**, the mother's father. Grandparents are the keepers of deep knowledge and the ones who have time to pass it down. → **ngaji**, **babi**. **karrkay** *n.* — **child**; (reduplicated *karrkay-karrkay*) children. : *Karrkay-angka mayi nuka-ny.* ‘The child ate the food.’ : ▸ Pluralised by ordinary reduplication, **not** by the kinship plural *-karra* (contrast **yaba** below). → **buban** ‘small’. **kadar** *n.* — **wallaby**. : *Yundu ngananji kadar balbi-nka?* ‘Are you coming with us to hunt wallaby?’ → **balbi-l** ‘to hunt’, **minya** ‘game meat’. **kaday** *v.* — **to come**; *(past)* **kada-n** came. : ▸ The bare stem *kada* never stands alone: use *kaday* (non-past) or *kada-n* (past). A yes/no question *Yundu kaday?* ‘Are you coming?’ is marked by tone alone (→ Grammar §12.2). **kalbay** *adj.* — **long; far; tall; deep**; also *(of time)* long ago. : *jikan kalbay-mba* ‘in the long grass’; *juku kalbay* ‘a tall tree’. : ▸ Takes the *-mba* form of the locative because it ends in *-y*: *kalbay-mba*. **kalnga → see above.** **kaminjarr** *n.* — **grandchild** (also **jija**). : *…ngayku kaminjarr.* ‘…my grandchild.’ → **jija**, **kangkal** ‘one's own child’. **kima** *n./adj.* — **soft; fine (sand, ground)**. : *jukar kima* ‘soft sand’. **kuku** *n.* — **word; language; story; speech**. : ▸ The name of the language itself, *Kuku Yalanji*, is literally ‘the *yalanji* speech’ (*yalanji* ‘this-way, thus’). → **balka** ‘to speak’. **kuli** *n.* — **anger**; *(instrumental* **kuli-bu***)* angrily, sternly. : *…nyunguny balka kuli-bu.* ‘…speaks to him sternly (lit. with anger).’ → **kuli-ma-l** ‘to get angry’. **kulbul** *adj.* — **ripe** (of fruit). : *junda kulbul* ‘ripe figs’. → **junda** ‘fig’, **mayi** ‘plant food’. **kuru** *n.* — **affection, love, fondness**. : ▸ The warm regard that binds kin; distinct from *wawu-mini* ‘happy’ (lit. ‘good breath/spirit’). --- *Continued: **kuyu** ‘fish’ · **kubu** ‘smoke’ · **kija** ‘moon’ …* --- ### Cross-reference box: the KINSHIP PLURAL *-karra* Kinship terms do **not** reduplicate to form a plural (the way *karrkay* → *karrkay-karrkay* does). They take the special ending **-karra** (KPL): | singular | plural | | |---|---|---| | **yaba** eB | **yaba-karra** | ‘brothers’ | | **babarr** eZ | **babarr-karra** | ‘sisters’ | | **manyarr** wife | **manyarr-karra** | ‘wives’ | | **jija** grandchild | **jija-karra** | ‘grandchildren’ | *The one exception is **kangkal** ‘one's own child’, which reduplicates like an ordinary noun.* --- --- # BOOK THREE · The Interlinear Reader ### Text 7 · *Bayan-ba* — “At the Camp” *A short reader drawn from the corpus, arranged as a day at an old people's camp. Each sentence is given in the elder surface, a word-by-word gloss, and a free translation, with notes on points of grammar and culture. The Reader is where a learner meets the language as continuous, natural speech rather than isolated drills.* --- **7.1** *Wanya-ngka junda kulbul yaluy-mba wubulku nuka-n?*     who-ERG fig ripe here-LOC all eat-PST     ‘**Who** ate all the ripe figs that were sitting right here?’ > The old woman's accusation opens with *wanya-ngka* — ‘who’ in the **ergative**, because whoever ate the figs was the doer of a transitive act (§4.1, §12.1). Had she asked merely *who is coming?*, ‘who’ would stand bare as *wanya*. **7.2** *Yurra wanjabu dunga-y wunkuwingku ngiki niji?*     2pl where go-NPST morning cold now     ‘**Where** are you all going this cold morning?’ > A content question with the verb in its plain non-past; *wanjabu* ‘where’ sits in the middle of the clause, not fronted. This same sentence, *Yundu wanjabu dunga-y?*, is the ordinary Kuku Yalanji greeting — one asks *where are you going?* the way English asks *how are you?* **7.3** *Yundu ngananji kadar balbi-nka madja-ka niji?*     2sg 1pl.COMIT wallaby hunt-PURP scrub-DAT now     ‘Are you coming **with us** to hunt wallaby in the scrub?’ > A **yes/no** question — no question word at all, only the rising voice (§12.2). Note *balbi-nka* ‘in order to hunt’, the **purposive** (§—): the whole reason for the going. *ngananji* is ‘with us’, the comitative of the inclusive ‘we’. **7.4** *Yundu wanyurri-nji bambay, ngayku kaminjarr?*     2sg what-COMIT sick my grandchild     ‘**What are you ill with**, my grandchild?’ > The fixed idiom for asking after sickness: literally ‘you are sick **with what?**’, using *wanyurri-nji*, the comitative of ‘what’. The old woman lays her cool hand (*mara ngiki*) on the child's hot forehead as she asks. **7.5** *Bindimu jarba janka wuna jikan kalbay-mba bana ngundu-ba.*     black.snake snake still lie.NPST grass long-LOC water near-LOC     ‘The **black snake** lies still in the long grass near the water.’ > *bindimu jarba* ‘the black snake’ — the generic *jarba* ‘snake’ narrowed by the colour-species word *bindimu*. Locatives stack naturally: *kalbay-mba* ‘in the long (grass)’ (after *-y*) and *ngundu-ba* ‘near (the water)’. **7.6** *…bubu marri juku-bu balka nyubun-bu jilba-nyjiku.*     ground always stick-INST hit.NPST one-INST(=first) walk-before     ‘…so always beat the ground **with a stick** before you walk through.’ > The old man's caution completes 7.5. *juku-bu* ‘with a stick’ is the **instrumental** — the same `-bu` ending that marks a transitive doer (§4.2). *jilba-nyjiku* ‘before walking’ carries the ‘lest / before’ suffix that frames the whole warning. **7.7** *Yaba-karra muru bundanday wula-nya narmba.*     brother-KPL together stay.NPST die-SUB whole (=all their lives)     ‘The **brothers** were together all their lives, right to the end.’ > *yaba-karra* ‘brothers’ shows the **kinship plural** *-karra*, never reduplication (Dictionary, cross-reference box). The clause *wula-nya narmba* ‘for the whole (of their lives), unto dying’ uses the **subordinate** *-nya* to fold one event inside another — the reach of Kuku Yalanji clause-chaining, and the natural close of the day's talk at the camp. --- *Notes to Text 7. The camp (*bayan*) is the ground of daily life; the sentences here move from the morning's small accusation over the figs, through the greetings and the setting-out to hunt, to the tending of a sick child, a warning about snakes on the path to water, and an old man's memory of two brothers. Every form in this Reader has been checked against Patz's grammar and the community dictionary; nothing has been invented. A learner who can read these seven sentences aloud, hear the questions in the melody, and feel why ‘who’ changes shape when it does the eating — that learner has the living bones of the language.* --- --- *These three pages are cut from one cloth: the same verified sentences serve as grammar illustrations, dictionary citations, and reader texts. At 20,047 verified sentences, the corpus has crossed its production target and passed the 2026-07-10 milestone completion audit. The audit proves source/process/review/export integrity; it does not turn synthetic research rows into elder-certified speech. The next editorial task is to expand these samples into the finished Grammar, Dictionary, and Reader while keeping that distinction explicit.*