 Welcome everybody to today's department of serenal. Today I'm glad to welcome Ellen Smith from UCL, she's a teaching fellow of that, and she'll talk about language change and language contact on the Bougainville Islands of Papua New Guinea. Thank you. Thank you very much Alex, thank you everybody for coming along. So today I'm going to talk about contact reduce change in Papua Panna which is a highly endangered Australian language of Papua New Guinea. So the data for this research comes from my PhD project which I carried out at the University of Newcastle in Australia. It was a documentation project where I created community materials and then my thesis was a grammatical description and I also looked at some language contact issues as well. This project was funded by ELDP. I had two field work trips, 11 months in total and I collected about 60 hours of audio and video recordings. Some were elicitation sessions, some were texts. That's all been annotated and then various other types of data like field notes. So that's the data that this research is based on. The aims of this particular presentation are to look at the consequences of language contact in the Papua Panna community. First I'll just talk a bit about Papua Panna and the contact situation and then look at contact-induced change in clause orders, obliques and possession and looking here at contact-induced change due to contact with Papua languages. Then I'll summarise that and look at related languages and how Papua Panna shows similar changes to some other related languages and what the implications of that might be. Finally, I just want to present a little bit of preliminary research on contact-induced change but this time looking at the influence of Tok Pison rather than Papua languages before concluding. Papua Panna is spoken on Bougainville Island which is up here in Papua New Guinea. Here you can see the top of Australia and Indonesia. It's spoken on the northeast coast so here's, sorry that's not very clear on this one, here's Bougainville Island here. If we zoom in to around about Coy Coy there and then zoom in even further you can see Coy Coy at the top and then some of these other villages. So there's six villages in total. Some of them are just small clearings and some of them are spread around rivers or roads or whatever and in total there's a population of about 510 people in those six villages. But of those 510 people only 106, actually now 104 because two have died, 106 are fluent speakers. So Papua Panna is an Austronesian language. In Papua New Guinea 27% of its languages it's got about 840 odd languages in Papua New Guinea and 27% are Austronesian and 73% are Papuan. The Austronesian language family has about 1200 languages in it spreading from Madagascar in the east all the way to Easter Island, sorry, Madagascar in the west all the way to Easter Island in the east and from sort of Hawaii and Taiwan in the north down to New Zealand in the south. There's about 230 million people who speak these languages. Papua Panna is one of them. Within Austronesian language family there's a group, the Oceanic group and within that go sort of down the subgroups you get to North West Solomonic which is the group that Papua Panna belongs to. Oceanic languages are spread across Micronesia, Melanesia and Polynesia and Boganville situated in the Melanesian area of the Pacific. The North West Solomonic languages are spoken in Boganville. You can see the ones in pink here are North West Solomonic languages and in the North West Solomon Islands so you can just see the tip here of one of the Solomon Islands and then they go further south. The purple languages here are Papuan languages so you can see there's quite a lot of diversity just on this one very small island. As I mentioned before then Papua New Guinea's got 840 languages which is 13% of the world's population. Sorry, 30% of the world's languages and 6 to 7 million people which is 0.1% of the world's population so it's got a massive amount of linguistic diversity. In Boganville there's about 20 languages both Papuan and North West Solomonic but only 175,000 people so multilingualism is the norm in this region. In terms of the Papua Panna community before white settlement contact was very minimal there was only really contact for trade or intermarriage with Rottacus speakers which you can see here if Papua Panna is here the Rottacus area completely surrounds the Papua Panna area. But then when the Europeans settled and introduced missions and schools and started translating the Bible and so on the Papua Panna people had contact with English with Toppisson and with other local languages. Due to work in plantations there's a lot of plantations both North and South of the Papua Panna villages due to the Papua Panna speakers working in those plantations and people coming from all over the island to work in them there was further contact. In the 1990s there was a big civil war in Boganville called the Boganville Crisis this caused a massive amount of displacement with people fleeing into the mountains or being placed in care centres and then there's been a certain degree of overpopulation and people have sort of their land size decreased for a while but then they became very overpopulated so they had to then spread back out again and there's been increased travel to population centres for things like work, meetings, markets, church and so on as well as increasing media and technology so all of these factors here have led to even more contact with Toppisson and with other local languages. So we've got more contact outside of the community and more intermarriage inside of the community so if we look at the languages that are spoken within those six villages we find that six of them are North West Solomonic languages another six are Papua languages and another five are languages from various different places in Papua New Guinea so there's quite a number of different languages represented in the villages. In terms of language endangerment Papua Pana is likely to be one of the world's sort of 7000 languages which won't exist in 100 years time only 17% of the village population of fluent speakers is only two child first language speakers these two boys here and there is some other transmission but it's very one way children will just reply back in Toppisson. Papua Pana is used in the home occasionally in church maybe for the odd song and a little bit in school but in school it's a bit tricky because there's only one Papua Pana speaking teacher the children of various ages with different L1 backgrounds there's temporary materials and it's sort of taught or the lessons I've observed anyway it's taught more as a second language along with English so the teacher uses Toppisson to teach Papua Pana and Papua Pana is sometimes used in maths and science for numbers or naming certain things like plants so yeah it's definitely endangered and it's endangered because of the shift to Toppisson as Creol Toppisson is a lingua franca both in and outside of the community it's economically advantageous literacy is useful whereas Papua Pana is a small minority language so it's less prestigious and the fact that there's been this increased contact over the last 100 years or so has meant that the speaking numbers have sort of been diluted as it were which has possibly weakened their sense of identity and so on and also the lack of materials doesn't really help either so it's just kind of a little bit of an introduction as to why the language is endangered it's endangered because not because of contact with Papua languages but because of contact with Toppisson ok so if we look now one of the other consequences is this contact induced change which Hain and Coutivert describe as a linguistic behaviour that differs from that of earlier generations and has been influenced in some way by language contact and this is a major factor in the development of atypical linguistic features in Austronesian languages in the Melanesian area Ross talks about how we have sort of this sustained bilingualism lead speakers to adapt the structures of their emblematic language so the sort of recipient or replica language to the model of the intergroup language or they speak the intergroup language with the emblematic phonology then what can happen is this emblematic language sort of survives in its restructured version and Ross calls this metatipi a change in morphosyntactic type which occurs when speakers are bilingual and restructure the morphosyntactic constructions of one of the languages on the model of constructions from the other language this is also known elsewhere as grammatical replication structural borrowing, pattern replication Matras and Cycle talk about pattern replication where you've got patterns of distribution of grammatical and semantic meaning and a formal syntactic arrangement moddled on an external source and they sort of have pattern replication and matter replication where you're replicating the actual form and substance which I'll come on to later on so in Papa Panna we find that there's a mixture of right and left headed structures and the question is is this evidence of content induced change if we're following Thomas and Steps of identifying content induced change first of all we need to look at the language as a whole which I've done in my grammar then we need to identify a source language and identify that there's been intense contact which I've just described earlier on so source languages are these Papa languages I think and shown that there's been intense contact over time particularly with the language Roticus but also with Buin and a language called Matuna because those speakers would have come up to the Papa Panna area to work in the plantations there's a lot more movement around the island now as well so third step is to identify shared structural features which I'm saying are verb final clause orders post positions and pre post possesses then we need to prove that those proposed interference features didn't exist in the receiving language in Papa Panna before contact but because there's no documentation of Papa Panna until the last few years I can't do that we haven't got records of Papa Panna from over a hundred years ago so instead we need to examine the canonical oceanic morphos and tectic features that are found in many North West Solomonic languages and illustrate how Papa Panna conforms to or diverges from that canon and the fifth step is to prove that these interference features were present in the source language before contact and again that's quite difficult because we haven't got records of Papa Panna languages from a hundred or so years ago but if we look at the constructions in the North Bougainville language Roticus and in the South Bougainville languages Buon and Motuna constructions that are similar between those languages we can sort of take those as typical of the Papa Panna languages of that region and these languages are considered part of the East Papa Panna film even though the North and South Bougainville languages are not sure completely related looking at clause orders first in North West Solomonic languages the pragmatically unmarked clause order is verb initial but the order of the postverbal subject and object varies so we could have VSO order or VOS order there's a preverbal argument position for pragmatically marked arguments which is called T and that might be the topic or it might be the focus and then we end up with this pragmatically marked order of TVX where T is the pragmatically marked argument and X is the unmarked argument so if we look at Benoni which has spoken in the West we can see here that this third person singular pronoun is the topic in the first line there so he killed a pig and we have this TVX order but then in the second line because this third person singular is no longer the topic it occurs postverbally and we have VOS order so that's in Benoni another North West Solomonic language the clause initial topic is obligatory and that means that all clauses are verb-medial so there is some variation between North West Solomonic languages but regardless of whether the topic is optional or obligatory the verb always precedes unmarked arguments apapapana however we have a bit of variation we've got SOV and SVO orders and these are the pragmatically unmarked clause orders I tried my hardest to find some kind of pattern to explain this but there just isn't one across a variety of genres a range of speakers and there's no grammatical motivation or markers, it's just variation so in two you can see we've got the SVO order so he refused his brother, his big brother and in the second one we've got the SOV order we've got you, my wife maybe you don't like her the pragmatically marked clause order in apapana you've got this initial topic so of course if the subject is topic we end up with the same orders we have either SOV or SVO but if the object is topic we've got OSV order as we do here so us, nobody saw us so we've got the SVO order which is a bit like North West Solomonic but where does this SOV order come from cos that's definitely not North West Solomonic so if we look at the Papa languages we find that SOV is the basic clause order although it's possible to have post-verbal argument in some languages so in Rattacost we've got this verb final SOV order as in number five we can also have OVS and SVO under certain conditions then in Buin we've got SOV as preferred order again we can have OSV dependent but that's something to do with the animacy hierarchy but if both arguments are equal in animacy we've got SOV SOV basic clause order for both of these languages and similarly with Matuna well with Matuna there's a lot of variation but SOV is the most frequent and the least pragmatically marked as in number seven here and then Nagarisi limited data but again shows verb final clause order and in Nassioi again it's usually SOV basic clause order so perhaps that's why we've got SOV order in Papana okay moving on to obliques then so in Northe Solomonic languages oblique arguments and adjuncts are expressed with prepositions and there's a small number of them which introduce participants with a wide range of semantic roles for example in Buinoni we've got these three here so moll expressing location goal source, gynife instrument and mat on mate for a component and this shows this one here so I shall go with Barnaw but me at Barnaw with preposition followed by the noun phrase okay so that's typical Northe Solomonic and in Papana we've got prepositions as well we've got te which can mark location either in space or time it can mark goal it can mark source instrument possession so in nine we've got goal with this one and in ten you can see instrument with tenor torara so you cut the tree with an axe then we've got this weird one which I'm going to come back later which is not a very typical preposition in Northe Solomonic languages and this means until so we've got here we've got here so he will sleep until tomorrow this can only occur with a certain number of these location nouns like tomorrow or Friday or something like that okay so so far that's typical of Northe Solomonic we've got two prepositions and there's a small number of them and they introduce participants with a wide range of semantic roles so that's typical but then we've got this post position tomana which expresses accompaniment as in number 12 so we've got we left the old women there with an old man and this isn't typical to have a post position this post position is also a bit strange because it occurs between the head noun and then if you've got a possessive prepositional phrase like teiaia or if you've got a relative clause this possessive prepositional phrase and relative clause has always come after the noun but you would expect tomana to occur outside of that noun phrase so you'd have nabau, car, cal, teiaia tomana but it doesn't instead it occurs between the head noun and the post modifiers but there's only a few examples of this and usually speakers use another construction this applicative commissitive construction which I can't get into today so only a few examples but it shows that it's occurring not where we'd expect it to occur so it's tactically inside the noun phrase and it's not the head of a postpositional phrase so it suggests it's not a fully fledged post position yet possibly it's grammaticalised as a commissitive marker from the additive marker 2 so we've got this example here nascuro tomana ie so munuera so the school where we're sitting to and then he continued the sentence so here we've got the head noun and then we've got a relative clause and we can see that the 2 is in the middle here which follows this same pattern so possibly this tomana is actually grammaticalised from the additive marker 2 the question is is this evidence of lexical calceng so lexical calceng is where the meaning range of a lexical item in the replica language is matched to the meaning range of the item in the model language until the two vocabularies are readily inter-translatable so let's have a look and see what the papyrn languages do so in papyrn languages as we might expect given they've got the final clause order obliques are expressed by case and clitics and post positions and case suffixes so eroticus here shows you an example of a case and clitics sorry that should be the equal sign there and then we've also got a post position in this one here but if we have a look at eroticus the post positions in eroticus we also find we've got this this one here taporal which Robinson glosses is also to and with but in the data examples Robinson always glosses it as also but it's often found introducing adjunct noun phrases and Robinson describes it as an oblique marker so in this example it's translated as with so Beryfasiri always wants to play with the other children so possibly what we've got here is we've got this taporo in eroticus meaning to and also meaning with and then we've had lexical, cow king where the sort of meaning of Tomana has been extended from to to a commissative marker under the sort of the Papuan model these examples here just show you the case suffixes in Buin and in Motuna and then in Nagavisi and Nassau we've got case suffixes and post positions respectively so Papuan languages have all got sort of case and clitics or suffixes or post positions whereas the North West Solomonic languages don't okay moving on then to possession there's two constructions we have to talk about in North West Solomonic languages the first is the direct construction and this is used for nouns which are inalienable so body parts locative parts and kin terms and you get the possessum so the possess noun has a possess of suffix attached to it and if there is a possessor noun phrase it will always occur it will always be post post it will always be after the possessum but it doesn't have to be there so for example here in Benoni we've got Numana meaning his hand and then we've got Ken there so we could just say Numana meaning his hand or we could say Numana Ken meaning Ken's hand I guess when it's something like less needs to add the pronoun there but it's this optional possess mp will always occur after the possessum the other construction in North West Solomonic languages is the indirect construction and this is used for inalienable nouns so that's all other nouns removable body parts kin that have been acquired through marriage and this construction is a bit more complicated we've got a particle possessive classifier usually for consumable or non consumable or general possession and then that classifier or particle has a possessor suffix added to it and then that whole construction which should be like this here in Kubikota that whole construction proceeds the possessum the possess noun and then if we have a possessor noun phrase again that will be post post so in Kubikota we can see the consumable possessive classifier followed by the possessor suffix and then the possessor but there's no possessor mp in that one in Roviana we can see that as is often the case in North West Solomonic languages the possessive particle is not synchronically segmentable so here we just have none of that and we can't split it up into a classifier and the suffix the Roviana example also shows that if we do have a possessor mp that will follow the noun here so here we've got a t-a-o-e that person okay so those are the North West Solomonic constructions the Papapano one for direct possession is used also for inalible nouns there's a variation as to which nouns are considered inalible and aliable so in Papapano inalible nouns are body parts some bodily products locative parts and some kin terms but not all kin terms the constructions exactly like the North West Solomonic one where you get a possessive suffix attaching to the possessor so in fact it would be exactly the same as ben only for hand numana but the difference is that if you have got a possessor noun phrase it can either be pre-posed as in 24 so nainu na matana the house's door or it can be post-posed as in 23 etamana nusia okay so that shows some variation there which we I mean this one obviously is like the North West Solomonic pattern but this one isn't the indirect construction marks inalible nouns which is basically all other nouns that aren't inalible and for this construction we have a possessor proclitic attaching usually to the well depending on the noun class it might attach to the possessor more to some other pre-modifier so there's no classifiers in Papapano there's no distinction between consumable non-consume or anything like that and these possessor proclitics or possessor particles are not segmentable into a possessive constituent and a suffix in the same way that we could do it with Cwbacodau they don't co-occur with specific and non-specific articles so what that means is that if there's no other there's no other article or numeral or anything before the noun then what happens is this proclitic attaches to the head noun but only if we've got the class one I've sort of got three four noun classes, we've got personal noun class and all the personal nouns are directly possessed so we don't need to worry about them and then we've got class one and class two and then location nouns so we've got a class one noun and it's singular possessor proclitic attaches directly to the head noun as in this example here so that's a bit strange because then that means it's actually direct and headmarking but if it's a class two noun like usia we find that the possessor proclitic attaches to this marker al so we get amiel amiel usia so we've got our teacher and our child there if there was some other article there like this one here this plural article it would attach to the plural article so we'd get amiel bau en a bau adolpe usia okay because of this we have we find that the possessive number is always marked and the possessive class is marked in the singular so from this construction we can tell not just the number but also the noun class of the possessor the other thing important thing is like the direct construction we have variation in the pre-posed and post-posed let's call possessor noun phrases so in this one here in 26 we've got an al al sinoni so my husband and we've got this pre-posed possess MP whereas in this one we've got en a bau adolpe usia and the possess MP is post-post which is typical of North West Olymonic but 26 isn't typical so let's see what oh one more thing about Papa Panna then so then we also have the possessive pronoun paradigm and this paradigm marks noun class so possessive pronouns can function as noun phrase arguments or as noun phrase predicates if the possessum referent is singular and class one the pronoun consists of the possessor proclitic al in this case followed by aata so we have alata but if the possessum referent is class two we have possessor proclitic and al so we get amel and so then that just looks exactly like this one here but for the class one we don't get amelata va mamamata like that so it's almost a bit like we've got these possessive pronouns and they're the same as the possessor particles that go before the possessum except if it's class one where this aata kind of gets knocked off and what's remaining just attaches directly to the possessum if the possessum referent is plural then the pronoun consists of the proclitic and vao so we've got amel as we did a bit like here with enabal so one thing I'm wondering here is whether or not we've got evidence of grammatical calc where we've got changes in the semantic divisions of grammatical categories because we don't have classifiers any more in Papua Paranae whereas a lot of North West Solomonic languages do have possessive classifiers and so we've lost that distinction but gained this distinction of marking noun class in these possessive pronouns which is unusual looking at the Papua languages their possessors are pre-posed so presumably that's where Papua Paranae gets its pre-posed possessors from so in Roticus we have indirect constructions with three constructions and in this one these two we've got those possessors this one here is the most common where we've got the possessor followed by the possessum and then this possessive marker here and it shows it in this example 31 in Bowen we've got direct and indirect constructions and in both of them the possessor is pre-posed so here it's attached directly to the possessum and here to this genitive marker and in Mautuna we've got a variety of constructions as well but mostly the possessor proceeds the possessum and in Nagobisi in Naseoi we've got two constructions and again if you've got a possessor MP it will be pre-posed so Roticus so we've got these pre-posed possessors if we've got a possessor MP but then we also find that Roticus and Mautuna their possessum noun class is marked by a pronoun or a classifier so in Roticus here this dummy pronoun agrees with the possessum in number sorry in gender sorry Robertson calls it gender but noun class so this dummy pronoun agrees with the possessum noun class and similarly in Mautuna we have this pronoun here marks the possessum noun class and then in Nagobisi in Naseoi the possessum number is marked by the possessive particle so here this possessive particle has the possessum number marked on it so this is a little bit like Pappapanna where the number and noun class of the possessum is marked so to summarise then the clause order in Pappapanna is typical of North West Solomonic and Oceanic languages where the SOV order isn't typical and instead the SOV order is like the Pappawin clause orders and obliques the prepositions, the two prepositions of Pappapanna are typically North West Solomonic but the post position isn't and instead the post position is more like the Pappawin post positions and we also get this situation where Tomanna has probably grammaticalised as the post position under the influence of Roticus then in possession we've got the post-post possessor MP is typical of North West Solomonic but the pre-post possessor is typical of Pappawin languages the direct construction is perfectly typical in both form and semantics but the indirect construction in Pappapanna differs somewhat from the canonical oceanic structure because the proclitics are not independent particles sometimes they're not segmentable Pappapanna is like Pappawin languages because there's no classifiers and it's like Roticus and Martuna because it's marked for noun class when they possess some singular so if we look outside of Pappapanna now to some related languages we can see that these changes are similar to those found in the Pappawin tip linkage in southern New Guinea in Takia in northern New Guinea and also in Mono, Taurau and Urawawa which are North West Solomonic languages so Taurau is spoken just south along the coast from Pappapanna Mono is spoken should I go back up to the map up here I can show you Monoalu is spoken here and then we've got Taurau here and Urawawa was spoken here but it's now extinct so Evans and Palmer carried out some research into these languages and found similar kinds of changes so in Urawawa, Mono and Taurau we've got SOV order and some other orders as well but the unmarked order is SOV in these three languages we've got a mixture of prepositions and postpositions as well which is like Pappapanna but there's some differences to Pappapanna because ablative is marked differently whereas in Pappapanna locative, alative and ablative all the same there's adpositions stacking in Urawawa and Mono under the influence of Nagovisi and there's an optional alative in Urawawa under the influence of Nasiw so they've got some things in common in terms of their contact induced changes a bit different and possession there's no classifiers in all four of the languages then in Urawawa, Mono and Taurau we've got prepose possessive noun phrases but in Pappapanna we have a mixture Pappapanna marks the noun class of the possessive if it's singular Urawawa marks the plurality for possessives and Taurau marks possessives being singular so there's some similarities there's some differences there but we can see that Pappapanna has a lot more mixture and it's a bit more conservative it's retained some of the North West Solomonic features more so than the other three languages so is this a common change or is it an independent change there's a common change we could have shared innovations of a common ancestor which might suggest sort of be evidence for internal subgrouping so Pappapanna is actually in Ross's 1988 paper he placed Pappapanna in this group here the Nehan North Bogenville subgroup which includes the languages of Northern Bogenville and Booker Island just at the top but Bill Palmer's recently found some sort of similarities in lexicon and syntax which raised the possibility that actually Pappapanna is related more closely to Urugawa which is here and and possibly even Mono and Toro and I think the sort of similarities and contact induced grammatical change kind of provide further support for that hypothesis the other idea is that these are actually independent changes that parallel other oceanic languages whose speakers have or had contact with speakers of typologically similar Pappapanna languages so in the Pappapanna's hip and Tachyr languages they also have the final clause order post positions and pre post possessors and then we can't say that they're in the same subgroup they're definitely not all northeasolomonic so perhaps also the differences that we saw between Pappapanna Mono, Toro and Urugawa might suggest that the languages have independently undergone metatipii on the model of languages belonging to the same linguistic area and indeed post positions and pre post possessors correlate with verb final clause orders cross linguistically so possibly what's happened is we've got Pappapanna contact which has led to verb final clause orders in Pappapanna and in Mono, Toro and Urugawa and then that change has led to related shifts reaching different degrees of completion in each language the other idea is that it's a mixture of both and that's in fact what Evans and Palmer argue that the similarities and differences in Mono, Toro and Urugawa show a history of interrelated changes some of which are likely to have occurred before the break up of the ancestral speech community and others that occurred independently and this might account for why Pappapanna displays more of a mix of left and right headed typology so possibly Pappapanna split off from the common ancestor at a point when that change from left to right headed structures hadn't it wasn't complete Further evidence for that is some of the speakers were telling me how they had come up they'd migrated up from the Solomon Islands sort of in the early 1900s it got up all the way to where tepory is where the main village is, Pappapanna village and then after that other speakers moved back down south again so possibly they've all come up together when they are all one group and that change hasn't yet been completed and then because of then the split and the migration and people moving further south the Pappapanna people have been left behind and developed in their own way so in this community possibly this shift hasn't reached the same degree of completion that was halted or perhaps it was reversed because then what happened in the Pappapanna community was other tribes came down from the north of Bougainville from the Tayop speaking areas which are north-west Solomonic languages and joined the villages which possibly meant that change that was occurring perhaps got reversed back in the direction of north-west Solomonic the north-west Solomonic pattern or it could be that it did reach this same degree of completion but because of the recent shift to Toppissin which is SVO, Order and Left-Headed there's been further contact induced change more recently perhaps back in the direction of this canonical oceanic pattern okay so since we're talking about Toppissin there we might wonder alright so there's been contact with Pappapanna languages historically over time and that's led to contact induced change but it hasn't led to language shift but then the contact with Toppissin has led to language shift so surely you know has there there must have been some kind of linguistic changes as well or we might expect there to be so let's have a look so earlier on I mentioned this preposition Yang or Yenna and basically it's just a bit too morphologically complex to be a preposition a typical North West Solomonic preposition so I'm wondering with this Yang or Yenna whether it shows evidence of lexical calking under the influence of Toppissin so lexical calking is where the meaning range of a lexical item in the replica language is matched to the meaning range of the item in the model language till the two vocabularies are inter-translatable so I think that this preposition has likely been lexicalised likely to be a lexicalised form of the modal verb Yang or Yenna and the third person singular postverbal subject indexing Yenna Yang or Yenna is a verb meaning be able and it occurs inside of what's called a verb complex that's basically the verb along with participant reference markers and tense aspect mood and so on but doesn't include the arguments themselves so Yang or Yenna inside of this verb complex with these postverbal subject indexing like this and it requires a clause or complement so here we've got this complement on now wak or nai postverbal subject indexing here's second person singular which matches our subject and our subject marker here as well also Yang or Yenna Yang or Yenna is a lexicalised clause level adverb indicating ability so this time we've got Yang or Y and then we've got Yenna which is usually the third person singular so here if it was Aya third person singular we'd have Yang or Yenna there so this Yang or Yenna has sort of been lexicalised as an adverb so we've got Jerry Yang or Yenna e atunye so Jerry can attack the pig and this occurs outside of this verb complex here but then we've also got Yang or Yenna as that preposition I mentioned earlier on meaning until if we have a look at top person we find this word inap is also a verb expressing ability so as in doki no bin inap long kismim so the dog wasn't able to get him or oli no inap win long mipela they're not able to beat us and inap is a preposition denoting until as in 37, emigol inap long friday so possibly what's happened in Papa Panna is this word Yang or Yenna which means be able its meaning has been extended to this preposition until under the influence here of top person because we don't find this other north west solemonic languages also possibly there's evidence of lexical kelking with this complimentiser of Vosia where Vosia introduces finite clauses licensed by desiderative or propositional attitude or knowledge verbs like dislike think and know we get reasoned verbial clauses introduced by a visi meaning because but also they can be introduced by a Vosia similarly with purpose of verbial clauses introduced by Tanawa which means so that but they can also be introduced by a Vosia as in this one here a Vosia tal si pa pa si nur i si myri o obotu so they said compadol and then they mean so that the canoe rope breaks quickly so a Vosia is becoming more of a general subordinator perhaps when people can't recall the specific subordinator so in sort of text recordings where people are just sort of freely telling stories they'll often just use a Vosia and it took a fair bit of elicitation sometimes to get these other words to come out they weren't a visi into now weren't that common so possibly a Vosia is sort of just taking over as this general subordinator because of the fact that this language is dying out and people are forgetting some things or it could be under the influence of top pissen where the subordinator also has quite a range of meanings like this way thus therefore and so we also find a lot of matter replication so that's direct replication of morphemes and phonological shapes from a source language so it's the formal substance or matter so previously I've been talking about pattern replication and now this is matter replication the question is is this propagated or on the spot loans so sometimes we find that the top pissen word is used when there isn't a pappapanna equivalent perhaps because it's some kind of foreign concept like army or playing cards so top pissen is used there which I think is perfectly common in the world's languages that happens all the time so here we've got bulk army for the army we've got play cards and then we've got kiki here for kick sometimes this is adapted phonologically so in top pissen it's actually kick, pelai and cat so it's been adapted to the pappapanna phonology where syllables are always open and where for example there's no phoneme in pappapanna so they've got rr there instead sometimes the speaker just sort of mixes them so in this example here they've got so sixpela which is top pissen sixpela o tau wita so sixpela o how many and then they use the top the pappapanna first and then switch to the top pissen so there's no way you can say that they've used the top pissen because there isn't a pappapanna equivalent because they've used the pappapanna first in the area in Nangananga one month and then they've said it in top pissen one pelamun and some other examples we find the opposite where the top pissen ones use first perhaps because they're trying to recall the pappapanna one and this one comes from a speaker who was not particularly so she says she keeps repeating herself and then eventually she gets to the pappapanna meaning twin so one of the things I want to look at instead of my research at the moment is which let's call domains are the most affected by this matter replication how phonologically integrated are some of these top pissen loans is there any variation according to gender or age or proficiency is there any variation amongst tech genres and also this question of whether sometimes we end up with English loans and are those direct from English or are they direct from top pissen because more and more people are speaking English now in the towns children are learning it from a much earlier age in school so it's hard to know then sometimes has it come directly from English has it come from English via top pissen as top pissen gets more and more sort of decrealised so in conclusion and the main question was what are the consequences of language contact in this community what is their evidence of contact to do language change yes there definitely is we find pattern replication so we've got this metatipi that Ross describes in his words metatipi we've got right-headed structures and not typical of north-westal and monarchy languages but they resemble Papua languages we've got evidence of let's call calking with this post position tommano which is grammaticalised from the additive market too and it's got the same meaning range as roticus topo or toporo we've also got let's call calking under the influence of top pissen with iangoyena and then we've got grammatical calking because we've had a neutralisation of classes of indirect possessive relations and this introduced possessive noun class distinction and these things resemble Papua languages and on top of that we've also got matter replication so there's many top pissen and English leximes used and some are more phonologically integrated than others a question is are they propagated or are they on the spot so we can say that given the degree of contact with Papua languages and with top pissen I think these changes are all due to contact Papua languages have influenced morphosyntax and some semantics top pissen has influenced language use considerably causing language endangerment but it's also influenced mathematics and the lexicon and the other implication of these findings is that the similarities in contact induced changes between Papua Panna and Mono Torda on Loava suggest that they've got more of a shared history than the sort of current subgrouping suggests but can the mixture be attributed to shift back in the direction of left headed typology due to top pissen so we've got these similarities that suggest to shared history but Papua Panna obviously shows much more of a mixture between left and right headed typology why is that, is it because the Papua Panna group sort of was split off at an earlier stage or is it because top pissen has actually sort of influenced the morphosyntax more than it seems but it's really hard to tell that because the top pissen is also left headed so if we have had this shift back in the direction of left headed it's hard to know if that's due to top pissen or whether that was just there all along thank you very much Matana any questions did you look at the the yes I did particularly when I was looking at the clause orders I tried to see if if there was any correlation between for example SOV clause order and the speakers linguistic repertoire so some of the speakers spoke Rottacus for example did that mean that if I had a speaker that spoke Papua Panna and Rottacus were they more likely to use SOV clause orders or not and there wasn't any sort of pattern there it would have been nice if the ones that spoke Papua language were the ones that produced the SOV clause orders but it wasn't the case so we had speakers who didn't speak Papua Panna and they also produced SOV clause orders it's a really nice talk thank you for the really clear presentation I think we always have to keep in mind that it's not languages that are in contact with people who are in contact the question Julia just asked was part of a bigger question which is that this kind of research has to be supported by ethnographic research not only looking at at repertoire but also looking at personal histories marriage patterns ideologies about maybe that the Papua Panna are more ideologically conservative than those other guys so you don't have to start talking about shifts back and forth degrees of shift or change can actually be influenced by ideologies about conservatism and so on and making up complex systems this group migrated and then they split and all this that's one possible story but there may be another story which actually has to do with transmission and language use and political views on language so go back and do a full ethnography I did do quite a considerable amount of that research but I didn't have time to get into all of that today but I basically mapped the family trees of all 500 people figured out how they're all linked up together and that kind of started out because I was trying to figure out how many speakers are there but you can't just go from like hut to hut and say how many of you speak Papua Panna because there'd be some people up in the gardens or some people in town or some people away in the city or something so I was like I'm never going to be able to go and knock on every you can't knock on every door and count the speakers because it changes every day so then I started trying to draw all their family trees and it just ended up in this massive thing I basically figured out how about 500 people were related who was married to who who spoke which languages what their kids spoke so yeah I did look at that as well that's based on what they told you that they spoke or based on your observations a bit of both, yeah sometimes based on my observations and sometimes based on their observations of either themselves or of each other okay because I couldn't meet all 500 people that's exactly the sort of stuff that really needs to be there to support the more structural stories yeah and that's why when I was trying to figure out why is there this variation and I had then got all of that data there where I could go and look at a speaker and then look at their family not only what do they speak which is data that you collect anyway when you record a speaker so who were they married to and who were their parents or who were their children married to and so on fantastic I think in addition to that there's also who do they speak the fishman kind of questions who do they speak what language to and when the sort of Milroy network kind of stuff as well I think mostly they all from what I observed just use top pissing most of the time and then Papa Panna if there's a group of people speaking and there was nobody else there who spoke if they all spoke Papa Panna then they'd use that but then if I was there then they'd sometimes just switch to top pissing because I was there as well but yeah I think even in meetings they sorry yeah exactly but I think generally they just all use top pissing because there's just I guess more likely that there's going to be somebody there that doesn't speak top doesn't speak Papa Panna because only 17% of the population speak it then most of the time there's going to be someone there that doesn't speak it so yeah that's why it's yeah that kind of all links into another presentation for another day about why is it endangered and what do they think of the language and so on yeah and perhaps I mean as Pete was saying about them being conservative to do with ideology it's a bit conflicting with their ideology because on the one hand they talk about how Papa Panna is really important part of their culture you're not really a Papa Panna person unless you speak the language but then on that their actual linguistic behaviour and practice doesn't reflect that attitude and then with top pissing they talk about how terrible top pissing is how it's just bad English, how it's a rubbish language but then it's really useful and they use it all the time so what they say and what they do are just pretty different that's a synchronic account you don't know what it was 500 years ago no no I think oh didn't hear what you said I have a question to follow on from that one 500 years ago I think they were still down in the Solomon's and then I think when they did move up to where they are now I think yeah if there was well there was multilingualism but with the Papa Panna languages and this is the thing they sorry when did all this happen I think like late 1800s groups moved up and there was some kind of fight with some chief or something like that there's different accounts of why they moved but I'm pretty sure the Papa Panna came up with the Taurau and Oduava speakers and there's also when I said about the tribes coming down from the north I kind of got that idea from things that people had said and how all these tribes came they named the tribes and these ones joined us later they've also got photographs from the 30s and the caption read something like oh the huts on the right show the northern style of huts and the huts on the left show the southern style so there's also this evidence that there had been in the early 1900s a mixture of two tribes Is that a Papa Panna gathering you're showing us in this picture here? No there's not that many speakers this is one of the gatherings at a local not mission station it's a place called Wakanai it's like one of the there's an airstrip there and a couple of stops and like little nurses station So how many languages would you guess we are looking at when we look at this crowd here? Definitely Roticurs there'd be Papa Panna speakers there half dozen or something like that because this particular picture was taken it was a gathering of all the schools I think sorry this was like years ago this picture so I'm trying to remember it was the picture was taken just here around in this area so I'd imagine there'd be like Papa Panna Roticurs possibly Teop speakers maybe some Torau speakers so it was kind of the schools in that sort of district there Do the Papa Panna people all basically know each other or are they likely to meet strangers and not know which language to use? No they all know each other Yeah or they'd know off each other there's only like a hundred of them Would they be able to produce something like that as a sketch map themselves how much knowledge is there of the situation as you depict it there a sketch map of what of the which people are aware of whether you know them as which languages I don't know actually that's an interesting question probably not they probably have like a vague idea like they've got their six villages and they're spread out there and then they'd probably know that if so and so's great niece works as a nurse in town or something like that but they definitely haven't got any records of who's where or anything because only few of them can write so they might have I'm just sort of interested in how much the structure you're showing it's a sort of artifact of the people coming in and saying well now we see what you're doing and how much it is actually a political reality um in the world I mean it's not what I'm familiar with at all but it's finding quite difficult to get a grip what this map is you mean well yeah I mean it's somebody's imagination about something yes that's right whether it's a schedule if you saw a map of I don't know the counties of England by and large and they know where they are and what they are not really I don't think no I don't think they really would like they know they're on you know that Surrey's in the south and Essex is in the east yeah I don't think that they would know it because well they definitely don't have anything they don't have books they don't have paper what's this supposed to be anyway because nobody lives in these these are not delineated places where people live right what this map here I need to go to the other maps to show you so this map here is showing the language areas so well this one's from 2011 so this one's pretty recent this is from Robinson's Grammar so this is showing where the different northwist onmonic languages and papyrint languages are located so say if you took the Roticus area here I don't I mean there's definitely I think you get speakers all around this area but it's not necessarily that every inch of that area has a speaker in it because it's a mountainous area but you probably get some of them over here and around here dotted around in this area but it's not going to be fully concentrate you know every inch of that yeah exactly yeah it's not like in this area you've only got Roticus speakers you're going to have maybe or potentially Papapano speakers have married in or people have come up from Buin and they've met or people have gone to town they've met there, they've got married and then they've moved back to that area so it's not all of these language areas are going to have language contact mixture in them because people marrying in and out the question is what do you mean by language area it's an imagined area it's some kind of political ideological or a language defined area I guess these areas are showing where the language originates, which area the language originates from except you just said that Papapano and the others don't originate there at all true but it's not your fault but obviously's fault no it's just a whole problem I guess Nick is pointing to as well which is these maps are pretty fictional in many ways I think say with Buin what it's trying to show is that the Buin speakers originated in this area and that the most dominant language in this area is Buin Papapano I think when I'm saying it's come down from this area here and moved up that's their ancestral language the ancestral group has moved up all up here a group of people have got left behind and their language has then developed into Papapano and then the other people have moved down and their ancestral language has developed into Torau and then into Urawaba and I'm not saying that when they lived down here they spoke Papapano when they lived down here they all spoke one language they've moved up and after they got split up their language has developed into three different languages but as in the Papapano area here there's a lot of language mixture in the Papapano villages as I've shown before lots of Intermaritan what there was about 18 different local languages represented in just those six villages so I think you'd get the same situation all over Bougainville but perhaps not to the same degree because saying some of the Rottacus villages nearby pretty much the whole village spoke Rottacus as their first language and yes they might be multilingual or they might be people that have married in but most people would speak Rottacus as their first language Papapano villages that's not really the case because Papapano is so endangered Thank you very much for your talk it's very interesting and I've been involved in a project looking at languages in the Solomon Islands and in fact we are intended to describe language contact but we find that it's maybe just as difficult as it is for you because with not having access to historical data or only partial it's very difficult in fact to establish who influences who and these areas where people are very multilingual but also where the history of movement is so complicated so I think your description of the changes certainly I can't comment very much on the obliques but what we've looked at the word order seems to correlate with what we found with language like papapano and I do think about that with that post position to mana maybe the influence maybe that's what papapano has always done and Rottacus has been influenced by papapano well I guess I'm going on numbers Rottacus is a massive language and I highly doubt that any of those speakers speak papapano because the papapano people told me that when they had contact with them I said oh so what language did you speak together because this was before top person had come along and they said oh well we spoke Rottacus because there was more of them than there were of us so I'm just making educated guess that the papapano speakers were multilingual in Rottacus as well but the Rottacus speakers probably didn't speak papapano so I'm assuming that the influence has come from Rottacus but it could well have been the opposite direction you never know how does that explain the history of English where the majority population of English speakers are a minority of French which is heavily influenced yeah true so maybe well yeah exactly so the French yes the minority but they had power but then in the papapano community Rottacus has seen to be more prestigious because there's more of them and papapano is not prestigious because there's so few in number because there's so few of them they're not seen as a powerful group but I'm guessing that it's not just been totally peaceful because moving around there have been power relationships or something like that I don't know what happened when they got to this area here I know that I think there was some kind of dispute why they left the Solomon Islands and moved up but I don't know what happens when they actually got here whether that or there was a fight or what I know from New Caledonia that when the French displace some people basically in the traditional system there isn't any spare land it's all allocated to some town so presumably there must have been some kind of conflict regarding any of those yeah I don't know except for the Rottica speakers and mountain dwellers so whether they would have been on that stretch of land on the coast I'm not sure yeah I don't know that and there's no record when I was interested in the 500 years thing I'm not sure where that came from but it's just generally where people tend to be in line of of colonial influence and the package contact but the reason I was interested in the timescale was whether it's worth looking at age-related differences particularly with regard to the fact that you've got not many kids learning I mean that would be interesting to compare in my parent time study yeah that would be I'm very impressed because for one person it's a lot of huge amount of work you've done because it's not common it's a bit of all these different I just saw Andrew shaking his head when you said 500 family trees and a grammar on top well because the main purpose the main reason I went there was to document the language and write a grammar but because I'm so interested in language contact I just couldn't help myself by looking at this other stuff so my thesis ended up being a grammar with then as much language contact stuff as I could do yeah that was all completely additional but I think it's definitely given there's such interesting things exactly and I can kind of explain them now Can I ask how old the earliest materials are which study this stuff there's a word list from the 60s as 1960s 1960s I think there's a word list from the 1960s and that's it the second generation of scholars scholars there's really not a lot there and the last yeah and it's a shame because once these two boys grow up their brothers are not going to marry each other are they and pass on to their kids so I think if they are still speaking it when they're older then the likelihood of them passing it on I think is quite slim Is it kind of paturing your men and paturing your men and you're supposed to have them now I don't know actually I think that kind of well because now everyone will just speak top piss and they won't learn Papua Panna I think it's much linear with land and stuff yeah with the lands that follows that line but with languages in terms of which language you're meant to learn when you get married I don't know because they just will all speak Papua Panna because the immigrants come in and I think you know maybe they might have that attitude of what's the point nobody speaks it there's not enough people to speak it with so kind of top pissing has taken over but I know like the previous generations the lady I stayed with her father spoke something like five languages so they've historically been a multilingual community and so one of my questions was why can't they just add top piss into their repertoire they've historically speak all these different languages but obviously then that gets complicated they've added that to the repertoire but because Papua Panna's lost prestige and usefulness then they've dropped dropping that which is a shame any other questions here? We've still got time so first of all I don't know anything about this area I'm afraid so when did top pissing become a serious lingua franca in this for these guys? Well it would have top pissing would have sort of come in where Europeans settled and set up plantations so that's late 1800s early 1900s but as a serious lingua franca last few decades I think I think it varies in different parts of Papua New Guinea but definitely in Bogenville I think the Bogenville crisis has sped up the process because that meant there was all this displacement loads of groups of people that might not have had contact before were suddenly placed in these government run care centres where top pissing was dominant so I think that's massively sped up the process so in parts of Papua New Guinea because there's a huge mine out there and there was a civil war and the government came in there's massive civil war over this mine so I think that has sped up the process in Bogenville but in different parts of PNG a top pissing is less of a creel but definitely in Bogenville I'd say it's more of a creel because children are growing up speaking that as their first language Thanks Raj I mean in the same kind of ballpark as a lot of these other questions I was wondering if you're able to kind of make any kind of educated guess about in the period before top pissing was a serious kind of threat as it were how do you think the the contact between I guess it's mainly between Aberpanna and Rottacus how do you think that was mainly mediated you know from what you've said I mean, yeah I do think So when I was looking at all the different family trees and finding out what languages everyone spoke it seems like the older generation sort of 40s and 50s upwards can all speak at least one other local language so their multilingual and their parents who've died also had five languages in their repertoire that kind of thing so I think that's how I think before top pissing came along they were just multilingual in one or two other local languages like Teot was quite a popular one maybe in Torral and then also Pappapanna and Torral for example are very similar so sometimes even if you didn't say if a Pappapanna person married a Torral person they might not necessarily have to speak each other's language to be able to understand each other because there's a certain because they're so similar I think even if you didn't fully get fully fluent you could probably understand so yeah I think before top pissing came along they were just multilingual in their other local languages and looking at the sort of looking at which so I looked at all the speakers that were fluent but then also recorded which speakers were kind of semi fluent or had some understand or could partially speak the language and then which speakers kind of had some passive understanding and it's mainly the 30 year olds upwards that speak Pappapanna fluently and then you get some speakers in their 20s a couple who speak it fluently but not many and so it's probably not really a coincidence that the last sort of fluent speakers were born in what the mid 80s and that's when the crisis started so basically the point at which intergenerational transmission was disrupted is the exact point that the Bougainville crisis started so I think that's if you're going to name one factor as the most important factor then maybe you could say it's the Bougainville crisis The reason I asked about how it's mediated and stuff is I'm interested in this question of where the kind of different context scenarios are more likely to trigger different types of context-induced change or changing different domains so there's this idea like from Plancoots and Winford if you have a situation where people are so the agents of change are dominant in the recipient language right which I think is it's always an oversimplification but that seems to be from what you've said seems to be likely to be the case here everyone who spoke Papapano or whatever it was 100-200 years ago or nearly everyone will have learned some of these other languages but would Roticus speakers who'd been there since forever whose community had been there a long time would they have learned Papapano? So there have been some of that but basically you're looking at what Plancoots calls recipient language agentivity and exactly this kind of abstract morphosyntactic change is what supposedly you don't expect in that scenario but I've never really kind of believed that anyway so you know I'd like to the trouble is you can never really be sure about what the agentivity was but it looks like the agentivity was recipient language agentivity and therefore it looks like another piece of evidence against the idea that when you have abstracts and tactic change it has to be source language agentivity and I have one kind of slight frivolous question which is does it have kind of stress this language similar to English and is in the way they pronounce their language name is stress on the third syllable and do they always pronounce all three of those paths because it seems Yeah they do Papapano Papapan and then the stress of the third because you know your bottom door there'd be a kind of at least one of those would be haplologised out I want to hear them say it now Yeah Papapano Papapano I can do it I don't know you can do it but when you say it every day it'd be bitcha and cha-cha-ra original language in Australia Any other questions It's going over English They have got the word Pana and Papana just reduplicated Papapano isn't Oh so it's got some kind of it's got some structure to it No I don't think it does No but so I've done some other research recently on reduplication and Papapano has multiple reduplications so one of the questions that's come up was is this an example of multiple reduplication where you've got two reduplicans pa pa and then the root Pana but I can't know because Pana kind of means side I think you can have Papana but I don't I don't know what Papapano would be because when you've got multiple reduplication it expresses a bitchal aspect so I don't It's not used in Yeah Maybe I don't know It would be really nice for my talk on reduplication if Papapano was an example of multiple reduplication but I don't think it is Because of that the language is actually full in practice of tokens of three identical syllables in a row Yeah So that explains it's always been impossible Any other questions? Yeah