Publishing Partner: Cambridge University Press CUP Extra Publisher Login

New from Cambridge University Press!


Revitalizing Endangered Languages

Edited by Justyna Olko & Julia Sallabank

Revitalizing Endangered Languages "This guidebook provides ideas and strategies, as well as some background, to help with the effective revitalization of endangered languages. It covers a broad scope of themes including effective planning, benefits, wellbeing, economic aspects, attitudes and ideologies."

New from Wiley!


We Have a New Site!

With the help of your donations we have been making good progress on designing and launching our new website! Check it out at!
***We are still in our beta stages for the new site--if you have any feedback, be sure to let us know at***

Review of  A grammar of Komnzo

Reviewer: John Mansfield
Book Title: A grammar of Komnzo
Book Author: Christian Döhler
Publisher: Language Science Press
Linguistic Field(s): General Linguistics
Language Documentation
Subject Language(s): Wára
Issue Number: 30.3888

Discuss this Review
Help on Posting

The Yam language family of Southern New Guinea has generated significant interest in recent years for its complex and unusual morphological structures (e.g. Evans 2015; Carroll 2016). Doehler’s grammar of Komnzo adds substantially to our knowledge of these languages, being the first published reference grammar for the Yam family. A key point of interest in Komnzo is the elaboration of distributed exponence (Matthews 1972), making this a promising source of data for morphological theorists. Phonologists will find interesting material in the prolific vowel epenthesis. Other highlights are the rich social, environmental and sociolinguistic background provided, and the use of fully contextualised example sentences. The grammar is also paired with an accessible online corpus.

Chapter 1 explains the background and plan of the book, and also provides a generous amount of information about the Farem people who speak Komnzo, their homeland and their recent history. Komnzo is spoken by some 150-250 people, most of whom are multilingual in neighbouring languages. They are a predominantly agricultural people, with yam cultivation playing a particularly important role in both subsistence and ritual. This gives rise to a fascinating senary number system used for ritual yam counting, and the creative use of yams to punish men who mistreat their wives. Doehler also provides valuable sociolinguistic details, including discussion of multilingualism and language attitudes. Linguistic diversity runs deep in Komnzo mythology, with the ancestor-hero Kuramonggo said to have heard many voices from a tree, which he then chopped down to gradually releasing people speaking different languages. The branches and upper parts were other peoples’ languages, and when he got to the base he found his own language (p. 36).

Chapter 2 describes the phonology, which is distinguished by extensive use of schwa epenthesis. Following previous work on the unrelated Papuan language Kalam (Blevins & Pawley 2010), Doehler argues that the phonemic representation of Komnzo words may be quite deficient in vowels (e.g. yrakthkwa ‘he put on top’), or in some cases lacking vowels altogether (e.g. mnz ‘house’). The phonetic form of these words uses predictable schwas to provide missing syllable nuclei, i.e. [jərakəθkʷa, məⁿts], though Doehler argues that syllables are not defined in lexical representations, noting that there is some free variation in syllabification, e.g. mrn-en ‘family-LOC’ → [mərənen ~ mərnen]. Komnzo may thus contribute important data to theories of syllabification.

Chapter 3 gives an overview of the word classes, thus serving as an entrée to the morphological and syntactic descriptions that make up the remainder of the book.

Chapter 4 describes nominal morphology, predominantly case clitics. There is a large number of cases, which Doehler divides into ‘core cases’ that are complements of verbs or possessed nouns (ERG, ABS, DAT, POSS), and ‘semantic cases’ that are not required in this way (LOC, ALL, PURP etc). Case marking involves a grammatical distinction for animacy, with animate case suffixes inflecting for number, while inanimates are underspecified.

Chapter 5 is on verb morphology, which is one of the most complex and typologically interesting dimensions of Komnzo. The verb hosts prefixes, suffixes and clitics on both edges, while the stem also indicates aspectual distinctions in its ‘restrictive’ versus ‘extended’ forms. Stem alternations are also notable for the unpredictability of their formal relations (e.g. rfitf ~ rfitfak ‘answer’, garf ~ gar ‘break’, mg ~ ru ‘shoot, spear’) (cf. Ackerman & Malouf 2013). TAM categories are marked distributionally, combining prefix, stem and suffix markers. The number categories of participants (singular, dual, plural) are also marked distributionally, as well as dual/plural neutralisation in various forms. For example, in the following form Actor agreement is marked on two different suffixes, with different degrees of number specification, and one of which is also associated with IPFV aspect. This aspectual value, in turn, is marked on the stem, a prefix and a suffix, each of which also contributes other lexical or grammatical meanings:

‘They hold him away.’ (p. 177-179)

Komnzo person agreement tends towards a binary distinction of 1 vs 2|3 (as in the example above), while gender agreement is a rich though relatively straightforward system in which gender of both actor and undergoer can be specified.

Verbs may be inflected in a range of valency templates, some of which involve prefix agreement with a single participant, while others use prefix and suffix to agree with two participants (including a non-referential ‘dummy’ prefix in middle verbs). Komzno verb stems are a large closed class with 380+ attested members, but they show great flexibility in filling diverse valency templates, so that each stem may cover a great range of event types. For example, Doehler shows that the verb migsi ‘hang’ can appear in five different valency templates, producing meanings such as ‘He is hanging’, ‘Something is hanging from him’, ‘It hangs itself up’, ‘She hangs him up’ and ‘She hangs it up for him’ (p. 188).
A distinct subclass of ‘positional’ verb stems can only take a single argument, but can be inflected for an additional number category ‘large plural’. The large plural category is marked by an unexpected combination of dual and singular markers:

woz y-ræs-thg-n
bottle 3SG.M-erect-STAT-DU
‘Many/all bottles are standing.’

Chapter 6 describes the TAM system, some of which overlaps with the verb affixation system. Particles also provide more specific TAM categories. Distributed exponence is again a major theme here, and as Doehler remarks, it is striking how the TAM system produces so many grammatical categories from a relatively small set of grammatical markers, by assigning a value to almost every possible combination.

Chapters 7 to 9 describe the syntax of noun phrases, simple and complex clauses respectively. There is a relatively fixed NP structure, though most modifiers (e.g. adjectives) can go on either side of the noun. Clause structure is also relatively fixed, with verb in final position, and [Actor Undergoer Verb] being the preferred order where two arguments are present. However, arguments are freely elided (especially objects), and verb-only clauses are common. Light verb constructions also play a major role, allowing a limited set of verbs to combine with ‘property nouns’ that denote many events. Yet more flexibility is found in a copula verb that can fill both single-argument and double-argument templates, covering the semantic ground of both ‘be’ and ‘do’ in other languages.

Chapters 10 and 11 describe information structure and lexicon respectively, thus providing some bonus topics that are not always found in grammars. Information structure is frequently marked by particles (including komnzo ‘only’), but also to a great degree by using irrealis verbs for backgrounded events. This section also brings out some of the typologically unusual properties of perfectivity in Komnzo. The chapter on lexicon gives rich examples of sign metonymy among natural species names, as well as interesting details on the lexical semantics of landscape.


The Yam family provides a fascinating and relatively new addition to our knowledge of human language, and Doehler’s comprehensive description of Komnzo is therefore an important publication. This work has many enjoyable and informative features. It is comprehensive and detailed in its coverage of phonology, morphology and syntax. The description is richly exemplified, with examples often containing interesting cultural vignettes, always carefully contextualised by the author. The grammar also has valuable ‘add-ons’, especially the social, geographic and sociolinguistic background in the first chapter, which is a substantial research achievement in its own right.

The presentation of linguistic examples is a constantly evolving documentary technique, and one in which accuracy, brevity and legibility are often at odds. Like many contemporary grammars, Doehler’s record of Komnzo is greatly enriched by accompanying texts, including linked online texts. These provide fully contexualised and naturalistic examples, which are used to illustrate most of the grammatical phenomena. On the other hand, the length of these examples sometimes make it hard for the reader to quickly spot the feature under discussion. I would therefore suggest that there is still a role for truncated, decontextualised and even elicited example sentences or words, as concise illustrations of grammatical phenomena.

Distributed exponence creates its own challenges for grammatical glossing. This book tends towards non-segmentation of verbs, and purely conventional affix glosses, to represent word-forms in which meaning cannot be neatly attributed to some morphological elements. For example, TAM-sensitive agreement prefixes are labelled using empty Greek symbols like wo- 1sg.α, kw- 1sg.β, since the prefixes are responsive to TAM, but don’t align neatly with particular categories. Personally, I would have preferred more semantic glossing (e.g. 1sg.IRR), even if not strictly accurate, since it would at least provide a mnemonic. Furthermore, if morphological components are only to be given semantic glosses where they consistently align with a particular meaning, this might require a vast amount of material to be glossed with meaningless Greek symbols. To pick but one example, in the northern Australian language Murrinhpatha, certain stem forms are usually associated with SG subject agreement, but in specific morphological contexts are instead associated with DU subject agreement (Mansfield 2019). The question is, how consistent does an asssociation have to be in order to merit a semantic gloss?

In any case, Doehler is to be commended for taking an explicit and principled approach to grammatical glossing, as Komnzo highlights some of the fundamental limitations of standard segmentation methods.


Ackerman, Farrell & Malouf, Robert. 2013. Morphological organization: The low conditional entropy conjecture. Language 89(3). 429–464.

Blevins, Juliette & Pawley, Andrew. 2010. Typological implications of Kalam predictable vowels. Phonology 27. 1–44.

Carroll, Matthew J. 2016. The Ngkolmpu Language, with special reference to distributed exponence. Canberra: Australian National University. (PhD thesis.)

Evans, Nicholas. 2015. Inflection in Nen. In Baerman, Matthew (ed.), The Oxford Handbook of Inflection. Oxford, UK: Oxford University Press.

Mansfield, John. 2019. Murrinhpatha morphology and phonology. Berlin: DeGruyter.

Matthews, P.H. 1972. Inflectional morphology: A theoretical study based on aspects of Latin verb conjugation. Cambridge: Cambridge University Press.
John Mansfield is a Lecturer in Linguistics at the University of Melbourne, specialising in Australian languages. Mansfield’s research interests include language change, morphology, phonology and the relationships between linguistic structure and social structure.