Robert Dixon went to the University of Edinburgh in 1961 to study linguistics. M.A.K. Halliday^§MS0.3a was teaching there at the time. “After about four months of learning about phonemes and morphemes, I went to Halliday and said that I wanted a language of my own – some interesting and complicated tongue that was in need of study.” Dixon had been reading the extensive literature on Native American languages, and the “linguistic relativity” thesis^§AS1.3.1c which suggests that very different language forms reflect very different ways of thinking and being human.

“No,” said Halliday, “leave those to the Americans. What you want is an Australian language.”

That language ended up being Dyirbal. Dixon’s journey took him from the depths of linguistic theory – attending a debate at MIT in 1962 between two of the great linguistic paradigm-setters, Chomsky and Halliday – to the heights of the Australian rainforest in 1963.

Not at MIT, but in the rainforest, he found something perplexing. “When I returned … and explained the structure of Dyirbal to M.A.K. Halliday, he told me that these unusual-looking grammatical patterns I had uncovered were ‘ergative’.” Dixon went on to document Dyirbal in a landmark book,¹⁰² then to write the definitive text on ergativity across a number of languages,¹⁰³ then to develop his own comprehensive theory of speech, developing a frame of reference different again from Halliday’s and Chomsky’s.¹⁰⁴

Ergativity in Dyirbal, to speak in Dixon’s technical terms, is the existence not only of the nominative and accusative cases that are familiar aspects of English transitivity, but ergative and absolutive cases where “the subject of an intransitive clause is treated in the same way as the object of a transitive clause, and different from a transitive subject.” In such situations, the ergative marks the transitive subject, contrasting with an absolutive which marks both the intransitive subject and transitive object¹⁰⁵ – the bush hen in the poem/song performed by Jimmy Murray.^§MS2.1.2c In such constructions, the parallel to passive voice is called, again obscurely, the “anti-passive” – here, the bush hen clapping its wings.

Meanings of agency are never so carefully and clearly distinguished in English. But Charles Fillmore^§MS2.1.2a nevertheless attempts to find them hiding in the opening of doors. In the clauses “the door will open” and “the janitor will open the door,” there is a relation between door and open that is the same in both sentences, the first sentence intransitive where the door would seem to be nominative case, and the second intransitive where the door would seem to be in the accusative case. If this parsing is to be revised to capture the shared reality of the door in both sentences, door would need to be another case (absolutive) and janitor also another case (ergative).

These, Fillmore argues, are things that Noam Chomsky^§MS3.1.2f neglects because he starts with the grammar of English in order to extrapolate an ostensible universal grammar applicable to all languages. Chomsky’s grammar is limited by its starting point. Some of these nuances in agency may be captured by elaboration in preposition phrases (“The janitor will open the door with this key.” “This key will open the door.”)¹⁰⁶ But prepositions like “with” are notoriously polyvalent, ambiguous, and for these reasons demand a great deal of shared contextual meaning^§AS1 to make sense. Just because English speech does not recognize these nuances of agency in an explicit way does not mean that such agency cannot exist in our experiences and meanings. It’s only that English speech doesn’t do them very well. We have to rely on multimodal transpositions to make sense, and to mitigate the risk of error in communication and interpretation.

So what do people mean, so carefully and clearly in Dyirbal and other such languages? University of Chicago linguist Michael Silverstein suggests a seven-step hierarchy across case combinations in languages that are both nominative/accusative and ergative/absolutive, ranging from first and second person actors to inanimate things.¹⁰⁷ As for the meaning of the distinction between accusative and ergative, Anna Wierzbicka concludes that differentiating nominative/accusative from ergative/absolutive indicates that the speaker is more acutely aware of his own agency or subjecthood in relation to others. “The ergative encodes the role of someone who is causing things to happen to other people or objects.”¹⁰⁸ It distinguishes the incidental agency of immanent effect.

William McGregor comes to parallel conclusions following a study of the phenomenally complex North West Australian language, Gooniyandi.¹⁰⁹ In this language, a distinction is made between accusative patterns in which direct agents and indirectly causing actors behave alike, and differently from undergoers; compared to ergative patterns in which indirect agents and undergoers are classed similarly.¹¹⁰ This leads him to develop a theory of agency ranging through the most direct forms of agency (such as body moves, vocalizations) to the most indirect or immanent (attribution, existence).¹¹¹

Fillmore suggests six or more cases.^§MS2.1.2a Halliday proposes a taxonomy of transitivity spanning material, behavioral, mental, verbal, relational, and existential processes.¹¹² Dixon builds out a taxonomy of verbs based on qualitative differences in kinds of agency.¹¹³

Explicit ergativity, Dixon estimates, occurs in about a quarter of the world’s languages: in Europe only in Basque, a rare survivor of the colonization of Eurasia by Indo-European speakers in Neolithic times; in some languages of the Caucasus; but mostly in the languages of first peoples of Australia, Melanesia, and the Americas.¹¹⁴ Traces of explicit ergativity can also sometimes be found in the archaic progenitors of today’s European languages.¹¹⁵

Such delicate construals of agency must surely be connected to the forms of life of first peoples. In English, by failing to make these distinctions clearly, we can elide different kinds of power and responsibility, perhaps lazily, perhaps deceptively. Methodologically, this much is clear: when our grammars use words like “subject” and “object,” related in a clause via a verb through a highly restricted set of options for voice (active/passive) and transitivity (transitive/intransitive), and when we have so few case and number markers, our tools for making sense are decidedly inadequate. Languages like Dyirbal can show us what we are missing.

Indeed, for these seemingly so-very-ordinary things – opening doors in one cosmology, or the clapping of the bush hens’ wings in another – we have in English grammar surprisingly limited tools for analysis.

When the linguists try to get to the bottom of the nuances, Halliday, Chomsky, Dixon, Silverstein, Wierzbicka, and McGregor all come up with systems that are, for what one would think to be such obvious everyday things, surprisingly different. The more elaborated their arguments, the more their theories diverge. Not only this, the more they work over these things, the more they seem to collapse in on themselves with qualifications and complexities. The 387 pages of the first edition of Halliday’s grammar became the 786 pages of his fourth.¹¹⁶ John Searle argues that Chomsky's universal grammar eventually becomes so complex that it implodes upon itself.¹¹⁷ This speaks to the ultimate failure of structuralist projects that attempt to distinguish components of meaning and order them into essentially static systems. The reality of meaning is the constant movement that we call transposition, where the patterning occurs across vectors of transposability and the potentially endless range of possibilities these offer.

For all their technical finery, these minutely elaborated structuralist theories fail to engage anyone other than their own technical kind: neither ordinary first peoples nor their everyday colonizers; neither ordinarily living first peoples nor moderns without carefully considered pasts. Worse, they rarely get to the meanings of the differences, and these surely, we all should know.

---

¹⁰² R.M.W. Dixon, 1972, The Dyirbal Language of North Queensland, Cambridge, UK: Cambridge University Press.

¹⁰³ R.M.W. Dixon, 1994, Ergativity, Cambridge, UK: Cambridge University Press.

¹⁰⁴ R.M.W. Dixon, 1991, A New Approach to English Grammar, on Semantic Principles, Oxford, UK: Oxford University Press; R.M.W. Dixon, 2010a, Basic Linguistic Theory, Volume 1: Methodology, Oxford, UK: Oxford University Press; R.M.W. Dixon, 2010b, Basic Linguistic Theory, Volume 2: Grammatical Topics, Oxford, UK: Oxford University Press; R.M.W. Dixon, 2012, Basic Linguistic Theory, Volume 3: Further Grammatical Topics, Oxford, UK: Oxford University Press.

¹⁰⁵ R.M.W. Dixon, 1994, Ergativity, Cambridge, UK: Cambridge University Press, p. 1.

¹⁰⁶ Charles J. Fillmore, 1966, “Toward a Modern Theory of Case.” Columbus, OH: Ohio State University, pp. 1–7.

¹⁰⁷ Michael Silverstein, 1976, “Hierarchy and Features of Ergativity,” pp. 112–71 in Grammatical Categories in Australian Languages, edited by R.M.W. Dixon, Canberra, Australia: Australian Institute of Aboriginal Studies.

¹⁰⁸ Anna Wierzbicka, 1981, “Case Marking and Human Nature,” Australian Journal of Linguistics 1(1):43–80, pp. 50, 76.

¹⁰⁹ William McGregor, 1990, A Functional Grammar of Gooniyandi, Amsterdam, NL: John Benjamins.

¹¹⁰ William B. McGregor, 2009, “Typology of Ergativity,” Language and Linguistics Compass 3(1):480–508, p. 481. doi: 10.1111/j.1749-818x.2008.00118.x.

¹¹¹ William B. McGregor, 1997, Semiotic Grammar, Oxford, UK: Oxford University Press, pp. 110–11.

¹¹² M.A.K. Halliday and Christian Matthiessen, 2004, An Introduction to Functional Grammar, London, UK: Routledge, p. 65.

¹¹³ R.M.W. Dixon, 1991, A New Approach to English Grammar, on Semantic Principles, Oxford, UK: Oxford University Press, pp. 94–204.

¹¹⁴ R.M.W. Dixon, 1994, Ergativity, Cambridge, UK: Cambridge University Press, pp. 2–5.

¹¹⁵ Eleni Karantzola and Nikolaos Lavidas, 2014, “On the Relation between Labilizations and Neuter Gender: Evidence from the Greek Diachrony,” Linguistics 52(4):1025–59.

¹¹⁶ M.A.K. Halliday and Christian M.I.M. Matthiessen, 2014, Halliday’s Introduction to Functional Grammar (Edn 4), Milton Park, UK: Routledge.

¹¹⁷ John R. Searle, 2002, “End of the Revolution,” New York Review of Books 49(3).

---

Cope, Bill, and Mary Kalantzis. 2020. Making Sense: Reference, Agency and Structure in a Grammar of Multimodal Meaning. Cambridge UK: Cambridge University Press, pp. 207-211. https://doi.org/10.1017/9781316459645

R.M.W. Dixon’s "Ergativity" §MS2.1.2d