Zompist Bboard Again

It was brought to my attention by a post on the Reddit linguistics subforum that several more resources are now available for Iau, the Lakes Plain language of New Guinea with six phonemic consonants, eight phonemic vowel qualities, and eight phonemic tones that has been the object of wonderment for conlangers and linguists alike.

First, the Iau people themselves now have a website. The website features a brief history of the Iau, an explanation of its orthography, texts, and recordings. The texts include various stories and translations of Genesis, although none are glossed or translated into English. The recordings include readings from Genesis and various songs and a sermon (some of the songs are just instrumentation, without any speech).

In addition, a draft Iau grammar sketch by Janet Bateman is now available. Some of us may be familiar with Bateman's work on Iau verb morphology, which details how the eight tones of Iau encode a fine-grained aspectual distinction on verbs, including through use of tone clusters.

But the additional detail provided in the grammar sketch is quite remarkable and I think will further enhance the reputation of Iau among conlangers and language enthusiasts.

The part I want to focus on has to do with the verb morphology: As it turns out, Iau verbs are not just monomorphemic segmental stems overlaid with tonal aspect morphemes - despite virtually all words being monosyllabic CV stems, Iau verb stems show definite (but at times semantically obscure) internal morphological structure, being made up of a C morpheme and one or more V morphemes compounded together.

You may have heard of Kalam, another New Guinean language that Bateman compares Iau to:

In an article about Kalam, a Papuan language with a closed set of 90 verbs, Pawley (1987:336-338) says these 90 verbs are composed of fewer than 30 verb roots, all having very broad or abstract meanings. These verb roots are strung together in serial verb constructions to comprise 90% of all verbs occuring in Kalam texts. In Kalam, these strings of generic verbs are used to indicate lexically complex events such as `hunt', `carry', `send to jail'. There are also many ongoing studies in Southeast Asia concerning the phenomena of “complex verbs”, light or vector verbs, serial verbs etc.

In Iau, monosyllabic segmental verb stems likewise seem to be composed of strings of consonant and vowel root morphemes which combine to form stems with broad abstract meanings that are furthur [sic] defined from context, from number of arguments for the predicate in the context, and also by the aspect of the super-imposed aspect tone morphemes. This is illustrated by Iau verbs formed with the consonant “b” below.

Something Bateman doesn't explicitly state here is that, while in Kalam the vast majority of all verbs consist of serial verb constructions, in Iau single CV verb roots appear to be quite common, at least from my cursory examination of some of her glossed sentences in later portions of the grammar, probably making up the majority of verbs.

Here are the Iau verbs formed with the consonantal root b-:

ba³ to come (process); to shoot, kill, throw at (a rock at someone)
ba⁹ come
be⁹ to paint
bi³ to arrive
baui³ to arrive at /reach
ba⁶ to come to
da⁸ba⁶ to bring to
bai⁶ beach /dock a canoe
bae⁵ (sun /light) shine in on
bu⁵ burn down
bui⁵ cut down (tree)
bo⁵ act of moving to a sitting /squatting position on the ground
boe⁵ to lower into a hole, to lower to the ground

Bateman says:

Although the individual lexical definitions of these verbs beginning with the consonant ‘b’ vary widely, note that they all share common spatial and directional motion orientation between the participants and the action. All the “b” verbs listed above have to do with motion towards a goal location which results in contact or action on/at that goal-location. The vowels mark various viewpoints of this action.

Below this, she discusses how the second or third vowel in a stem is usually a morpheme indicating "transitivity, that is, the degree of effectiveness, the degree of completion of the action, or some limitation on the action or affectedness of the undergoer."

Here is an example of a vocalic verb morpheme:

The vowel i indicates "motion /travel into and through /throughout a location or path," and when it appears as the second or third vowel in a diphthong, it indicates "do totally".

z⁹i⁹ go [I assume the tone number after the z is a typo?]
bi³ arrive
di³ hit
fi³ swell/grow
si⁹ push

da⁶ dip in water vs. dai⁶ cross river
ba⁹ come vs. bai³ (sun’s rays) shine in on

Most Iau nouns are made up of different roots than verbs, but Iau verb and verb roots can also be used to derive nouns, sometimes via tone modification:

Verb u⁹ ‘to descend, come down from /out of’
Noun u⁹ki⁹ ‘a strap, a rink, a hook, ie hangs down /descends from something’
Noun u⁸ ‘vomit, saliva’ ie something that comes down out of the mouth

Verb Root o to come out from something, to float to the surface
Verb o⁹ to give off light; to shine; to be lit
Noun o⁹su⁹; rays of light; what is given off /shines from the sun, a lamp, a fire
Adj o³tu⁹ daylight the time when it is light /the sun is giving off light

Verb Root o to come out from something, to float to the surface
Noun o⁷ sandbar sticking up in the river; sand, gravel
Noun (u⁸) o⁴ a splinter from a board /or on a board
Noun o⁷su⁹ leaf

I will leave it at that; you can see more examples of all of this in the grammar sketch.

Now, one thought I had while reading this was, surely some would argue that some degree of skepticism is warranted here? There are only six consonants in Iau, so it's inevitable that there will be large numbers of verbs with related meanings that share the same initial consonant. One could dream up some similar analysis of English verbs if you wanted: bar, base, ban, bounce, bound - ah, we see here a verb root b- with the very general meaning of "to put/move down", as in a barrier or a foundation, or an object that rebounds.

However, there is nothing suspect about Bateman's previous work on Iau, and despite the typo-ridden nature of this draft, I think the amount of examples provided makes a good case for this analysis.

So, now it's time to let your mind run wild with ideas for an oligosynthetic language made up of monosyllabic consonant-vowel compounds...

Looks interesting! Though despite my love for Kalam, I’m not sure how good a comparison it is here. Perhaps it would be more relevant to compare to Abui, which has a comparable set of monophonemic generic roots — b ‘join’, d ‘hold’, m ‘be in/take’, a ‘be at’, etc. — which also combine with aspectual suffixes and personal prefixes to form verbal words.

That being said, I’m not convinced this analysis is entirely correct: the verbal meanings just seem too distinct. e.g. according to her analysis, b+u+i should be ‘motion to+move down with affectedness+totally’, but the gloss is ‘fell a tree’ — there’s certainly some correlation there, but it’s not wholly compositional.