Early Descriptions of Gender in Pama-Nyungan Languages

Clara Stockigt
University of Adelaide

There is little correlation between the existence of a system of gender in Pama-Nyungan languages and the inclusion of a discussion of these systems under the heading “Gender” in early grammatical sources.

Of the small minority of Pama-Nyungan languages which have a system of gender, a handful exhibit systems of noun classes in which agreement is marked on a nominal modifier (Dixon 2002:450-453). Only one of these languages, Minjungbal, was described in the nineteenth century (Livingstone 1892). Another comparably small group of about a dozen Pama-Nyungan languages make a two-way gender distinction in third person pronouns (Dixon 2002:461). A disproportionate number of these are among the few Australian languages that were grammatically described in the pre-contemporary era. They are Hunter River Lake Macquarie language/Awabakal (henceforth HRLM) described by Threlkeld (1834); Diyari, described by four Lutheran missionaries between 1868 and 1899 (Flierl 1880; W. Koch 1868; Reuther 1981; Schoknecht 1947); Minjungbal, described by Livingstone (1892); Pitta Pitta, described by Roth (1897); and Kala Lagaw Ya, described by Ray (1893).

Grammars written in the classical European tradition employing the framework and schema of Traditional Grammar (see Koch 2008:87) discuss the grammatical category Gender within an initial chapter dedicated to the word-class Nouns. Gender is presented alongside the two nominal inflectional categories, Number and Case (see for example Ramshorn 1824 and Gildersleeve 1895). In recognition of the lack of grammatical gender in Pama-Nyungan languages, some grammarians abandon the traditional category altogether (Taplin 1867, 1872; Ridley 1875; Günther 1892). Others state that the language has no gender (Teichelmann and Schürmann 1840:4; Meyer 1843:10; Taplin 1880:7; Kempe 1891:2; Strehlow n.d, n.page). Lutheran missionary C. Schürmann for instance, who co-published the second grammatical description of an Australian language with Teichelmann (Kaurna 1840) and a second description of a related South Australian language shortly after (Barngarla 1844) noted:

In accordance with the other Australian dialects no distinctions on account of gender have been discovered in the Parnkalla language.
(Schürmann 1844:3)

But the term Gender is used within Traditional Grammar to describe both the category held by nouns with which other word classes agree, as well as the lexical marking for biological gender (e.g. Gildersleeve 1895:10-11; Ramshorn 1824:19-32). Consequently the category Gender is maintained in a body of early Pama-Nyungan grammars of languages with no system of gender. This group of grammars, which interestingly do not include the works of German speaking missionaries (Teichelmann and Schürmann 1840; Meyer 1843; Schürmann 1844; Kempe 1891; Günther 1892; C. Strehlow 1908, n.d.), give lexical pairs which refer to different genders of the same type: husband/wife, daughter/son, male kangaroo/female kangaroo etc. (Threlkeld 1834:10; Livingstone 1892:6; Roth 1897:15; Mathews 1907:324). T. G. H. Strehlow also gives substantial discussion of lexical gender under the heading ‘Absence of Gender’ (1944:59), explaining that the language ‘has to add’ adjectival modifiers denoting male and female to the names of species. He does this despite recognizing that “[t]he Aranda nouns know no distinctions of gender” (Strehlow 1944:59). R. H. Mathews similarly fills the prescribed category with a description of adjectival modifiers. The fact that “exponents of nominal Gender were recognisable members of another part of speech was irrelevant” (Koch 2008:192).

It was not until the 1930s, that watershed decade in Australian linguistic thought (McGregor 2008:9), that linguists in Australia began to systematically distinguish between natural and grammatical gender. In 1937, and probably in response to the newly described structures of non-Pama-Nyungan languages from Australia’s north-west with large and pervasive gender systems (e.g. Elkin & Capell 1938; Kaberry 1938; Love 1938), Capell writes:

It is very doubtful whether the idea of sex which English-speaking people read into “gender” is in the minds of these people at all.

Here he first proposes the use of the term ‘noun classes’ instead of ‘gender’ to describe the category in Australian languages:

“Gender” is a method of classification which may be only a matter of grouping words with a similar ending or beginning into one group, and I suggest that it would be easier to speak of noun-classes in Australian just as in Bantu.

The term gender is used in this present discussion to refer to both noun class systems and pronominal gender systems, which are taken to exhibit grammatical gender on the grounds of agreement because “the control of anaphoric pronouns by their antecedent (the girl … she) [is seen] as part of agreement” (Corbett 2013). Note that this usage differs from that used by Dixon (2002:452).

The pronominal gender systems of HRLM, Diyari, Minjungbal, Pitta Pitta and Kala Lagaw Ya were all well described in the early sources. The framework of Traditional Grammar anticipated a natural gender distinction in 3sg pronouns and had schema and terminology in place to convey it. It is worthwhile quickly noting that while pronominal systems of gender are typically motivated by biological gender in Australia and elsewhere, the nature of this distinction differs cross-linguistically. The two-way gender distinction in both Pitta Pitta (Blake 1979:192) and Diyari is contemporarily described as distinguishing feminine/non-feminine, rather than feminine/masculine:

[G]ender is determined by natural sex distinctions. Feminine [gender is marked on] all animates whose reference is distinctly feminine … and is the marked term
(Austin 2013:64)

Roth successfully conveys this split when describing Pitta Pitta:

[There are] two forms of gender – one for masculine and neuter, the other for the feminine.

The Lutheran missionaries’ descriptions of Diyari however, do not elucidate the nature of the semantic divide and simply give ‘masculine’ and ‘feminine’ labels to the two pronominal classes.

Separate paradigms of different case forms of 3sg pronouns headed ‘masculine’ and ‘feminine’ are presented in the first grammatical description of an Australian language, HRLM (Threlkeld 1834) and in Diyari (Flierl 1880; W. Koch 1868; Reuther 1981; Schoknecht 1947). Grammars of Minjungbal (Livingstone 1892:6); Kala Lagaw Ya (Ray & Haddon 1893:125) and Pitta Pitta (Roth 1897:2) present gendered forms in a nominative paradigm. Of incidental interest here is missionary C. Strehlow’s surprise in discovering that Arrernte makes no pronominal gender distinction. Strehlow, who was already well acquainted with the structure of Diyari when he embarked on the study of Arrernte, assumes that pronominal gender is a universal linguistic category. He anticipates that Arrernte should, like Diyari and Indo-European languages, make a pronominal gender distinction:

Strangely gender of the personal pronouns cannot even be seen in the third person. era serves to indicate he, she and it. Era pitjima as well as meaning he comes also means “she comes” and “it comes”.
(Anon n.d)

Although all these early sources give 3sg gendered forms, they are not always presented as Gender. Threlkeld, for instance, does not appear to conceive that HRLM has gender, despite presenting masculine and feminine paradigms: “there are a few distinctions of gender in certain nouns, but not generally” (1834, p. 10).

C. Schürmann, who was well acquainted with Threlkeld’s work, also views HRLM language as having no gender when referring to “other Australian dialects” (see above) and T. G. H. Strehlow describes English as having undergone “a complete loss of gender” (1944:60).

Livingstone’s discussion of Minjungbal Gender is entirely confined to lexical gender and the naming male and female pairs:

There are two ways the feminine is distinguished from the masculine- either by a different word or by adding the termination –gun …

Feminine and masculine forms of 3sg pronouns are not presented as Gender. Nor are the language’s four noun classes. Livingstone’s ‘Classification’ of Minyug (Minjungbal) nouns and adjectives (1892:4-5) tabulates the agreement of adjectives with four classes of noun. Current analysis of Minjungbal noun classes (Crowley 1978:43-45) is based entirely and somewhat tentatively on Livingstone (1892), which is the only source for this Bandjalang variety. As Livingstone describes Minjungbal, nouns appearing without an adjective, which would normally be suffixed to show noun class agreement, may optionally take the noun class suffix themselves (Crowley 1978:43). The feminine members of the gendered lexical pairs presented by Livingstone are marked with the female noun class suffix –gun. It is curious that Livingston does not present his four Minyung ‘Classifications’ as Gender, since the four noun classes are, as Livingstone describes them, semantically determined largely by the masculine/feminine, animate/inanimate oppositions underlying Indo-European gender systems (Kurzová 1993:61).

Pama-Nyungan deictic systems and early accounts of gender

The Lutheran missionary grammars of Diyari describe gender also under the heading ‘Demonstrative Pronouns’. Under this heading is presented masculine and feminine paradigms of the ‘demonstratives’ showing inflection for case. The forms they include are said to be “constructed as per the third person of the personal pronoun to which the syllable pini or para is attached” (W. Koch 1868). Koch’s ‘syllables’ are members of
“a complex set of deictic suffixes used with pro-nouns, locationals and predicate determiners” (Austin 2013:13), and the forms he gives are currently analysed as being the gendered 3sg pronouns suffixed “with one of a set of deictic suffixes which indicate relative location … –parra ‘there’ (there) seems to indicate location away from the speaker” (Austin 2013:64).

The missionaries’ analysis is astute. They perceive a linguistic construction foreign to the Traditional Grammatical framework and are challenged to find terminology to describe the forms. In arriving upon the label ‘Demonstrative’, the missionaries borrow schema from Traditional Grammar and use it to label and present the deictic forms that are perceived to have comparable functions. Hence gendered Demonstratives in Diyari.

Roth similarly describes a functionally parallel set of Pitta Pitta deictic suffixes as marked for gender (1897:2). He does not however classify these as demonstratives. The terminology he uses is notably similar to that employed in analysis of the language seventy years later (Breen & Blake 1971:77):

Each number has three persons, the third having two forms of the gender —
one for the masculine and neuter, the other for the feminine. Furthermore, both
genders have additional inflexions in the form of suffixes according as the person
or object referred to is either (a) close up in front, or at side of, (b) close up at
the back of, or (c) anywhere yonder, at some distance away from — the person

Three genders are sometimes analysed in early descriptions of Australian languages. Threlkeld presents HRLM as having the three-way 3sg gender distinction of Indo-European languages. In applying Traditional Grammar to the structure of HRLM, Threlkeld unwittingly assumes categorical universalism, despite himself stating that:

the arrangement of the grammar now adopted, is formed on the natural principles of the language, and not constrained to accord with any known grammar of the dead or living languages. The peculiarities of its structure being such, as to totally prevent the adaptation of any one as a mode.
(Threlkeld 1834:10)

In addition to the masculine and feminine, Threlkeld shows three separate neuter pronominal paradigms. His analysis is the earliest attempt to wrestle with the representation of the grammaticalisation of location in Australian languages. Forms in the first paradigm are translated as “this, present”, those in the second paradigm as “that, at hand” and in the third, “that, beside”. It is beyond the scope of this piece to tabulate the forms given by Threlkeld against their contemporary analysis (Lissarrague 2006:34-36). Suffice to say that Threlkeld declines many of the deictic demonstratives listed by Lissarague within his three neuter third person paradigms in an ad hoc manner. He does however, perceive that the inserted forms are not functionally equivalent to the European neuter pronominal forms:

The Neuter pronouns are inexpressible in English in consequence of the locality of the person being included in the word used as a pronoun.
(Threlkeld 1834:21)

The grip that Traditional Grammar held on the early description of Australian languages has projected the neuter class of pronoun into the structure of HRLM, even though Threlkeld recognizes that the forms do not perform the same function as that motivating the paradigm in Indo-European languages. Like the later missionaries’ descriptions of Diyari, Threlkeld attempts to accommodate the perceived foreign deictic structures within the schema of Traditional Grammar. While the Lutherans use the word class Demonstrative to convey the structure, Threlkled employs the ‘vacant’ schema of neuter third person pronouns.

Three genders in early analyses of Diyari

The first three Lutheran grammars of Diyari (Koch 1868; Schoknecht 1947; Flierl 1880) also give a third gender class:

[W]e shall see when discussing the pronoun, all three genders, masculine, feminine and newter [sic], appear in the language.
(Schoknecht 1947:1)

The Lutherans’ analysis of three genders in Diyari is unique within both early and contemporary Pama-Nyungan analyses. Their neuter category however, meets the definitional criteria of agreement. Evidence for the neuter class is taken from the form of the “neuter interrogative pronoun”, minha “what?”, which is distinct to the form of the “personal interrogative pronoun”, warli “who?”. Nouns select different interrogative/indefinite pronouns according to their animacy. In Diyari, the interrogative/indefinite minha “covers the set of common nouns with non-human reference” (Austin 2013:58). “The interrogative-indefinite pronoun ‘who?, someone’ rang[es] over the class of pronouns and nouns with human reference” (Austin 2013:67). T. G. H. Strehlow also approaches this analysis when describing the Arrernte interrogatives: ngwenhe “who” and iwenhe “what” as “particularly interesting because it is here that we find the one solitary relic of gender-division still extant in modern Aranda” (1944:59), and as the “one solitary glimpse of gender-differentiation afforded by this native language” (1944:98).

Concluding Remarks

Pama-Nyungan Gender systems were generally very well described in early sources, although not usually under the heading “Gender”. Many authors demonstrate particularly sound understanding of the language, e.g. Livingstone’s (1892) description of four noun classes, and Koch (1868) and Roth’s (1897) descriptions of deictics on gendered 3sg pronouns in Diyari and Pitta Pitta respectively. Gender was often described innovatively. The schema from Traditional Grammar was borrowed and applied with altered reference, sometimes in ways that might seem surprising from a modern linguist’s point of view or even misleading, such as Threlkeld’s (1834) description of neuter pronouns. A.P. Elkin’s (1938) observation that the significant material held in the early grammatical sources of Australian languages is detectable only to “the seeing eye” (ibid.:10) and after a process of “careful sieving” (ibid.:9) applies to the assessment of the early description of gender in Pama-Nyungan languages.

It would be interesting to further investigate the cause of the discrepancy between German speaking and English speaking grammarians in including lexical gender as Gender. Preliminary investigation suggests that nineteenth century grammars of Latin written in German present Gender very similarly to grammars of Latin written in English. Perhaps the influence is just the relative pervasiveness of the grammatical gender systems in the authors’ mother tongues.


Anon. n.d. ‘Die Grammatik der Aranda-Sprache’, Lutheran Archives, Unpublished typescript. Typist unknown, probably copied from unpublished C. Strehlow manuscript written prior to 1908.

Austin, P. 2013. A grammar of Diyari, South Australia, (digital edition). First published 1981, Cambridge University Press, Cambridge.

Blake, B. 1979. ‘Pitta-Pitta’, in R.M.W. Dixon & B.J. Blake (eds.), The handbook of Australian languages, vol. 1, John Benjamins, Amsterdam, pp. 183-242.

Breen, J.G. & Blake, B.J. 1971. The Pitta Pitta dialects, Linguistic Communications 4, Monash University, Melbourne.

Capell, A. 1937. ‘The structure of Australian languages’, Oceania, vol. 8, no. 1, pp. 27-61.

Corbett, G.G. 2013. ‘Number of genders’, in M.S. Dryer & M. Haspelmath (eds.),
The World Atlas of Language Structures Online, Max Planck Institute for Evolutionary Anthropology, Leipzig. (Available online at http://wals.info/chapter/30, Accessed on 2014-10-21.)

Crowley, T. 1978. The middle Clarence dialects of Bandjalang, AIAS, Canberra.

Dixon, R.M.W. 2002. Australian languages, Cambridge University Press, Cambridge.

Elkin, A.P. 1938. ‘The nature of Australian languages’, in A.P. Elkin (ed.), Sudies in Australian Linguistics, 3 edn, Australian National Research Council, Sydney.

Flierl, J. 1880. ‘Dieri Grammar’, Lutheran Archives.

Gildersleeve, B.L. 1895. Latin Grammar, 3rd edn, Macmillan, London.

Günther, J.W. 1892. ‘Grammar of the Wiradhari in NSW’, in J. Fraser (ed.), An Australian language as spoken by the Awabakal, the people of Awaba or Lake Macquarie (near Newcastle, New South Wales) being an account of their language, traditions and customs, by L.E. Threlkeld, Part IV Appendix D, Government Printer, Sydney, pp. 56-120.

Kaberry, P. 1938. ‘Notes on the Languages of East Kimberly’, in A.P. Elkin (ed.), Sudies in Australian Linguistics, 3 edn, Australian National Research Council, Sydney.

Kempe, H. 1891. ‘A grammar and vocabulary of the language spoken by the Aborigines of the Macdonnell Ranges, South Australia’, Transactions of the Royal Society of South Australia, vol. 14, pp. 1-54.

Koch, H. 2008. ‘R.H. Mathews’ schema for the description of Australian languages’, in W.B. McGregor (ed.), Encountering Aboriginal languages Studies in the History of Australian linguistics, Pacific Linguistics, Canberra, pp. 179-218.

Koch, W. 1868. ‘untitled ‘, Barr Smith Library Rare Books and Special Collections.

Kurzová, H. 1993. From Indo-European to Latin: The evolution of a morphosyntactic type, Current Issues in Linguistic Theory, vol. 104, John Benjamins Publishing, Amsterdam.

Lissarrague, A. 2006. A salvage grammar and wordlist of the language of the Hunter River and Lake Macquarie, Muurrbay Aboriginal Language and Culture Cooperative, Nambucca Heads.

Livingstone, H. 1892. ‘Grammar and vocabulary of the dialect spoken by the Minyug people on the north-east coast of New South Wales’, in J. Fraser (ed.), An Australian language as spoken by the Awabakal, the people of Awaba or Lake Macquarie (near Newcastle, New South Wales) being an account of their language, traditions and customs, by L.E. Threlkeld, Part IV Appendix A, Government Printer, Sydney, pp.1-27.

Love, J.B. 1938. ‘An outline of Worora grammar’, in A.P. Elkin (ed.), Studies in Australian linguistics, Oceania Monograph 3, Australian National Research Council, Sydney, pp. 112-124.

Mathews, R.H. 1907. ‘The Arran’da language, Central Australia’, The Proceedings of the American Philosophical Society, vol. 46, pp. 322-339.

McGregor, W. 2008. ‘Introduction’, in W. McGregor (ed.), Encountering Aboriginal languages: studies in the history of Australian linguistics, Pacific Linguistics, Canberra, pp. 1-34.

Meyer, H.A.E. 1843. Vocabulary of the language spoken by the Aborigines of the southern portions of the settled districts of South Australia preceeded by a grammar, James Allen, Adelaide.

Ramshorn, L. 1824. Lateinische Grammatik, F.C.W. Vogel, Leipzig.

Ray, S.H. & A.C. Haddon. 1893. ‘A study of the languages of Torres Straits, with vocabularies and grammatical notes’, Proceedings of the Royal Irish Academy, vol. 4, no. 3, pp. 119-373.

Reuther, J.G. 1981. ‘Three Central Australian Grammars. Diari Wonkanguru Jandruwanta. 1899, Part A Volume V Translated by Hercus, L.A. and Schwarzschild, T., edited by Hercus, L.A. and Breen, J.G. Additional notes on Diari by Austin, P. (1973)’, AIAS microfiche, Australian Institute of Aboriginal Studies, Canberra.

Roth, W.E. 1897. Ethnological studies among the north-west-central Queensland Aborigines Edmund Gregory, Government Printer, Brisbane.

Schoknecht, C.H.M. 1947. ‘Grammar of the language of the Dieri Aborigines ‘, Missionary Carl Schoknecht, Killalpaninna Mission 1871-1873 : selected correspondence compiled by Alan Carl Schoknecht [and] Colin Paul Schoknecht, A. & C. Schoknecht South Oakleigh, Victoria.

Schürmann, C.W. 1844. A Vocabulary of the Parnkalla language spoken by the natives inhabiting the western shores of Spencer Gulf. To which is prefixed a collection of grammatical rules hitherto ascertained by C.W. Shürmann [sic], George Dahane, Adelaide.

Strehlow, C. 1908. ‘Einige Bemerkungen über die von Dr. Planert auf Grund der Forschungen des Missionars Wettengel veroffentlichte Aranda-Grammatik’, Zeitschrift für Ethnologie, vol. 5, pp. 698-703.

Strehlow, T.G.H. 1944. Aranda phonetics and grammar, Oceania Monograph 7, University of Sydney, Sydney.

Taplin, G. 1867. ‘Vocabulary and grammar of the languages of the Aborigines who inhabit the shores of the lakes and lower Murray’, Barr Smith Library, University of Adelaide.

Taplin, G. 1872. ‘Notes on a comparative table of Australian languages’, The Journal of the Anthropological Institute of Great Britian and Ireland, vol. 1, pp.84-88

Taplin, G. 1880. Grammar of the Narrinyeri tribe of Australian Aborigines, E. Spiller, Adelaide.

Teichelmann, C.G. & C.W. Schürmann. 1840. Outlines of a grammar, vocabulary, and phraseology, of the aboriginal language of South Australia, spoken by the natives in and for some distance around Adelaide, R. Thomas, Adelaide.

Threlkeld, L.E. 1834. An Australian grammar: comprehending the principles and natural rules of the language, as spoken by the Aborigines in the vicinity of Hunter’s River, Lake Macquarie &c., New South Wales, Stephens and Stokes, Sydney.

How to cite this post

Stockigt, Clara. 2014. Early Descriptions of Gender in Pama-Nyungan Languages. History and Philosophy of the Language Sciences. https://hiphilangsci.net/2014/11/12/early-descriptions-of-gender-in-pama-nyungan-languages

Tagged with: , , , , , , , , ,
Posted in 19th century, 20th century, Australia, Field linguistics, Grammars, History, Linguistics, Syntax, Typology

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: