Béu : Chapter X
Word generation
In the word generating program ;-
17 % chance that a word will start with a vowel
83 % chance that it will start with a consonant.
10 % chance that the second consonant will be lʔ ... lh
50 % chance that the second consonant will be ʔ ... l
31 % chance that the second consonant will be nʔ ... nh
9 % chance that the second consonant will be sʔ ... sh
Complement clause ... relative clause
Originally I had the relativizer and the complementiser the same ... tà
MAYBE WE SHOULD CHANGE THE RELATIVE CLAUSE MARKER TO "SÀ". THIS IS BECAUSE DUE TO THE FREE WORD ORDER IN BEU, "THAT" OF A COMPLEMENT CLAUSE DOESN'T ALWAYS FOLLOW THE VERB AS IN ENGLISH, BUT CAN FOLLOW A NOUN ??
kolape is when a clause is modified slightly and functions either as an adjective modifying a noun (a relative clause) or as a noun being one of the core arguments in another clause. kolape saidau is a relative clause construction which works in pretty much the same way as the English relative clause construction. A relative clause is a clause that qualifies a noun. It is introduced by a special particle, tà in béu. In English it is usually "that" but a number of other words can also be used.
Personel possessives
In the above section we learnt how to say "mine", "yours", etc. etc.. But how do we say "my", "your", etc. etc.
Well these words (which would be considered adjectives in the béu linguistic tradition) are represented by infixes. The table below shows how it works.
my coat | kaunapu |
our coat ("our" includes "you") | kaunayu |
our coat ("our excludes "you") | kaunawu |
your coat | kaunigu |
your coat (with "you" being plural) | kauneju |
his/her coat | kaunonu |
their coat | kaununu |
xxxx own coat | kaunitu |
It can be seen that the infixes are the same as the plain pronouns, but the order of the consonant and vowel are swapped over.
There could also be another entry in the table above. That is the infix -it- (this is the possessive equivalent of the reflexive pronoun tí (see above). It is probably easiest to explain -it- by way of example;-
polo hendoru kaunitu = Paul will wear his coat (To be absolutely specific "Paul will wear his own coat")
polo hendoru kaunonu = Paul will wear his coat (To be absolutely specific "Paul will wear someone else's coat")
A thing to note is that you can not insert an infix into a monosyllable word. You could not say *glapa for "my woman" but would have to say glá nà pà
Hi-lighting a question topic
In the last section we saw how to focus on a particular element in a normal declarative statement. There is a special way to focus on a particular element in a YES/NO question..
If you want to query a particular element in the clause and not the clause as a whole, you stick ʔói on to the element that you want to query. For example ...
1) glà timpori ʔói báus = Did a man hit the woman ? (I thought that he had kicked her)
2) glà timpori baus ʔói = Was it a man that hit the woman ? (I thought it was a boy)
3) glà ʔói timpori baus = Was it a woman that the man hit ? (I thought it was a ladyboy)
And indeed a whole noun phrases can be brought into interrogative focus this way. For example ...
sá báu jutu dè ʔói timpori jene = was it the big guy that hit Jane. .... (notice the inclusion of dè when you do this, an item that would not normally to included)
But it is not possible to focus on an element within a noun phrase.
The topic marker "wa" and the discourse strategy of dropping the topic.
I THOUGHT THAT THIS WAS A PRETTY NEAT IDEA FOR A TIME. IT ALLOWS PRONOUN DROPPING BUT FOR THIS PRONOUN DROPPING, YOU MUST GET RID OF ALL AMBITRANSITIVE VERBS (SUCH AS "TURN TO". I DECIDED THAT I WANTED AMBITRANSITIVE VERBS, HENCE THIS WHOLE IDEA OF PRONOUN DROPPING USING "WA" WOULD NOT WORK.
English has what Dixon calls a S/A pivot construction. What that means is you can drop the A argument or the S argument if it is the same as the A argument or S argument in the previous clause. For example ;-
1) You can drop the A if it is the same as the S in the previous clause ... John saw Mary: John laughed => John saw Mary and laughed
2) You can drop the A if it is the same as the A in the previous clause ... John saw Mary : John hit Bill => John saw Mary and hit Bill
3) You can drop the S if it is the same as the S in the previous clause ... John entered : John sat down => John entered and sat down
4) You can drop the S if it is the same as the A in the previous clause ... John entered : John saw Mary => John entered and saw Mary
A small number of languages have a S/O pivot. That is you can drop the S argument or the O argument if it is the same as the S argument or O argument in the previous clause. (the Australian language Dyirbal is one example of this type of language).
Anyway, the above is just some side-information that I am giving you. béu has what I call a declared pivot construction. The "pivot" (or topic) in a discourse must be stated and from that point on all reference to that "pivot" is dropped, until a new "pivot" is declared.
You declare the topic by affixing wa to it when it is in S, A or O function. If it is in A function that the topic is declared then the s (ergative marker) is dropped. (However in the clause in which you declare a pivot can not have any dropped arguments ... if it is a transitive verb in the clause, and there in no argument with the ergative marker, then you can work out that it must be the argument marked by wa which is the A argument). From then on the topic is dropped until a new topic is declared. For example;-
1) giant.wa destroyed the castle on the hill
2) Then ø came down into the valley
3) There ø met a dwarf doing good works
4) The dwarf turned ø to stone
5) Dwarf.wa then climbed the mountain
6) ø gave succour to the people from the castle ...
It is the rule that the topic must be dropped. if the topic appears in a peripheral roll (pilana 1-> 14) then that pilana is attached to the verb.
For example ;-
1) Last night I saw Thomas
2) Thomas.wa (or o.wa) was very drunk
3) Mary had given.ye a bottle of Chevas Regal
How does this system mesh in with passives ? Particles that appear between clauses ? Particles that change the subject ?
You can see from 4) above, that this just doesn't work if you have labile verbs. In English "turned" is called a labile verb (ambitransitive is another name for this). That means it can be used in a transitive clause and in an intransitive clause. Foer example ;-
1) The dwarf turned the giant to stone ... transitive
2) The dwarf turned to stone ... intransitive
Some Rubbish
..... gaza ... the copula of existence
The copula complement of gaza ia always a noun or a noun phrase. It is how you say "there is ... "
gaza is similar to sàu in that it takes the 9 verb modifiers but 3 of them are wildly irregular. It is the same 3 tense/aspect forms that are irregularin the sàu copula. Namely ;-
*gazora => ʔá meaning "there is"
*gazori => ʔái meaning "there was"
*gazoru => ʔáu meaning "there will be"
Actually while theoretically gaza can have the full range of modifiers enjoyed by a normal verb, in reality all forms other than ʔá, ʔái, ʔáu are extremely rare. Occasionally you come across the "infinitive" gaza.
There is no word that corresponds to "have". The usual way to say "I have a coat" is "there exists a coat mine" = ʔá kaunu nà pà
Internal possessives are not allowed in the nouns introduced with gaza. That is, you can not say *ʔá kaunapu, but must say ʔá kaunu nà pà (I have a coat)
As I said above, gaza always comes with one noun. If it comes with an adjective, then that adjective can be considered a noun (well this is one way to look at it)
Also note that when the noun is a noun as opposed to an adjective, ??? , it is always indefinite.
pona = cold (an adjective), ponan = coldness (a noun)
ʔá ponan = "it is cold"
ʔá pona paye meaning "I feel cold" (word for word ... "there is coldness to me")
There is fixed word order : it is always gaza followed by the noun or NP.
The three irregular forms have their own negative marker. ya is stuck on to the end of the copula.
ʔaya ponan = "it is not cold"
Note that the word ʔaya (there is not) and ʔaiya (there was not) are very close to each other phonetically. However the middle part of the second word takes twice as long as the middle part in the first word : they are phonetically quite distinct.
The particles lói (probably) and màs (maybe) normally, come before the verb that they qualify. However the 3 irregular forms of gaza really like to come clause initially. Hence lói and màs immediately follow the verb.
ʔáu lói ponan = It will probably be cold
Also the evidentials are affixed to the wild forms, just as normal.
ʔaunya lói pona = They say it will probably not be cold
ʔaunya.foi lói pona = Do they say it will probably not be cold ?
geu
gèu is an adjective if it comes immediately after the copula* sàu. For example báu rì gèu => The/a man was green. (if you wanted to put a substantive after sàu, you would stick aja "one" in front of it).
gèu is also an adjective if it comes immediately after a noun i.e. báu gèu dí => This green man
In other positions gèu represents a substansive noun**.
*gèu is a qualitative noun if it comes immediately after the copula of existence gaza. For example ʔá pona => It is cold ... or ... ʔá pona paʔe => I am cold
**Well actually in one other position géu represents a qualitative noun ... after the "copula of existence" (just to make things complicated) Now how can we tell if the unmodified gèu is representing an adjective or a substansive noun. Well we can tell by its position with respect to other elements in the clause.
And above we see one more possibility. In the above two examples you can get to the "G" form from the "A" form by a regular process. With mapa/mapau this is not possible. So it appears that this word has two base forms ("A" and "G") and this word would have two entries in a dictionary.
From NOTEPAD
LIMBAWA ... SHIT
We have two words meaning "one" (maybe also "two")
ABA = one. WEBA ... Once. KIBA .... Unique
HEU ... One. KIHEU ... Lonely. WEHEU ... Alone
WEHE ... First
All nondaua are either nouns, verbs, or adjectives.
BARON top houseN ... They say I am on top of the house BORIS vicinity doorN ... I saw her near the door MAS BIRU game ... Maybe you will be at the game MO BIRU game ... You will not be at the game
BREAK can just happen be volitional be accidental
Find word for ... protagonist, modifier, wild irregular, tame regular, noun, verb, adjective, pronoun, proof
KYU ... time KWI ... when ?? KYUDHI … now
To give and to receive ... kyé and bwò
kyé means to give and bwò means to receive or get.
Normal usage
1) jonosA kyori jeneye toiliO = John gave a book to Jane "or" John gave Jane a book
Linguistic jargon ... In the Western linguistic tradition, Jane is called "the indirect object"(IO). Quite an unfortunate term I think as it is human 99% of the time, hence hardly what you would normally call an object.
2) jeneA bwori toiliO (jonovi) = Jane got a book (from John)
O.K. the above is the usage normal usage of kyé and bwò. They sort of describe the same action but from two different perspectives.
Causative construction
You can replace the cwidau with a gomua in 1) and you get a causative construction.
kyari jonoye dono = I made john walk
or you can alternatively use the form kyari jono dono (in which case jono dono is considered a gomuaza)
kyari jono* timpa jene = I made John hit Jane ... (in which case jono dono jene is considered a gomuaza)
jonos kyori pà solbe moze = John made me drink the water
*béu tries and drops all arguments that can be known without being specified. Now in the above example timpa is a transitive verb and usually has an A argument and an O argument. In the above example, if jene was dropped from the semo (but of course understood from context), then we could have a form kyari jonoye timpa
Passive construction
3) jonosA timpori jeneO = John hit Jane
4) jeneS bwori timpa (jonovi) = Jane was hit ... jeneS bwori jono timpa
4) is the passive equivalent of 3) ... used when the A argument is unknown or unimportant.
You can replace the cwidau with a gomua in 2) and you get a passive construction.
jene bwori dono = Jane was made to walk ?????
bwò a verb that is intransitive=>a passive causative ??
bwari dono = I was made to walk .... this is called a "causative construction" in linguistic jargon.
(pás) bwari solbe moze (jonovi) = I was made to drink the water (by John)
moze bwori solbe (jenevi) = The water was drunk (by Jane)
Called the passive construction in linguistic jargon ... It is used when the A argument is unknown or unimportant.
flompe is an intransitive verb
kyé jene flompe = to make Jane trip
kyé jono kyé jene flompe = to make John to make Jane trip = to make John trip Jane
BIA ......... a copula to much
Ah well ... three copula's are just too many
bià means "to be at"
For example polo bori london = Paul was in London
polo borta london = Paul has been to London
auto bora lence = The car is in the street.
pele boru nambo = Paula will be at home
bià is the rarest of the copulas and has no irregular forms.
It is often supplanted by sàu ... but if this happens a locative particle must be suffixed to the noun (the noun after the copula). For example ;-
polo rì london.pi = Paul was in London
auto (rà) lence.la = The car was in the street (literally "on the street")
pele mò rù namboʔe = Paula will not be at home
béu shows the imperfective aspect by prefixing the verb with the particle bai (see the section on Serial Verb Construction, to find out the origin of this particle)
... -kun or kun or kunta
Affixed to gamba only.
I passed my exams by cheating
Positive and negative
Above we have used -ya to generate a negative meaning. This form is used in two other situations to give a negative meaning. In aiya meaning "no" and in kya meaning "don't". However there is also 2 situations where -ya or -ia have a positive meaning ... in fanfia (as oppopsed to fanfua) and in the verbal aspect -y
in kunjua (as opposed to kunja and in umutu as opposed to mutu. This is just the way things are.
Said by the philosopher Kantu Banpoʃi, necessary to balance out the ying and yang elements for every minute of the day. However there is another school of thought that says that that is a load of balls and that Kantu Banpoʃi had his head up his own arse.
... -dis or dis or dista
Affixed to gamba only.
I passed my exams without cheating Two consonants can appear together at the beginning and middle of a word. The various combinations that are allowed at these two positions are stated later (see juzmi). Only two consonants are allowed word finally. These are n and s.
The vowels ia and ua can only occur in the final syllable of a word. If a suffix is added, making either ia or ua occur in a non-word-final syllable, then they must change to ya and wa respectively. However these changes can occur only in certain circumstances, depending on the consonant to the left of the y or w (refer to the table in the juzmia @aba section to see what combinations are acceptable). If the change to ya or wa is not allowed, then they both change to a simple a.
If the "modification" is something solid (something you can touch) then the form gwomo would be used. It is actually hard to draw the line between when gwoma should be used, and when gwomo should be used. But the linguistic usage falls just to the gwoma side of the line. Hence we talk about the béu verb having 9 gwoma instead of 9 gwomo.
Word structure "nandau"
All nandau are what are called "content words"⁕ (LINGUISTIC JARGON). They are words like "house', or "run" or "beautiful" that have a definite meaning embedded in themselves.
Each pyabu is defined by 3 juzmia.
juzmi can be translated as "gesture", "a definite movement given a meaning by socially agreed convention", it also is used for the three parts that define a pyabu.
The three parts are juzmi @aba (the first gesture), juzmi @iga (the second gesture) and juzmi @oda (the third gesture).
The rule for determining what is a nandau and what is not (and by definition "what is not" => yadau), is that there must be one, and ONLY one jwavo in the three gestures.
jwavo = "molecule made from more than one element" or "consonant cluster" or "diphthong"
⁕A small number of yadau are also "content words". Invariable they are very common words. For example dunu "brown" or hiaᴴ "red".
⁕⁕⁕It is thought that when multiplication tables were invented, a name for each "entry" was sought. The adoption of pyabu came about thru analogy to a fishing net (multiplication tables are called "multiplication nets" by the way). The word later spread to 1D systems (i.e. items on a list) and to 3D systems (well the nandauli is one example)
⁕⁕⁕⁕By the way kyamo = "molecule made from only one element" or "geminate" or "long vowel" (where long vowels contrast with short vowels to produce minimal pairs)
.... the first element "juzmia @aba"
There are 37 juzmia @aba. Some of them are "kolta" (consonants in this case) and some of them are jwavo(meaning consonant clusters in this case). All the juzmia @aba are "complex sounds"(consonant or consonant clusters).
.... the second element "juzmia @iga"
There are 9 juzmia @iga. Some of them are kolta (vowels in this case) and some of them are jwavo (diphthongs in this case). All juzmia @iga are "simple sounds"(vowels or diphthongs).
The juzmia @iga order is e, eu, u, au, a, ai, i, oi, o
.... the third element "juzmia @oda"
There are 58 juzmia @oda. Some of them are "single sounds" (consonants) and some of them are jwavo (consonant clusters in this case). All the juzmia @oda are "complex sounds"(consonant or consonant clusters).
Most yadau are what are called "particles" in linguistics. These are the short words such as "the", "to", "because" that impart meaning to the nandaua around them, or specify the relation between two nandaua, or add a certain nuance/meaning to the whole utterance.
Examples of yadau are foi that is cliticized to the end of the first word of a sentence (thereby turning the sentence into a question). And mo which goes directly in front of a verb and negates the whole utterance. All the pronouns are also yadaua. All affixes⁕ also.
All words that are not a nandau are either yadau or yauyadau. yadau are mono-syllabic and possess either a high tone or a low tone. yauyadau are poly-syllabic and have neutral tone.
⁕In béu an affix is called a "part yadau" (as opposed to all the non-affixes which are called "whole yadau")
The juzmia @oda order is l@, lm ... ln, lh, @, m ... n, h, n@, ny ... mw, nh, s@, zm ... zn, sh
If you look in the nandauli⁕⁕ (dictionary) you will get a form such as hend-. This is what is also called a pyabu.
Actually the original meaning of pyabu⁕⁕⁕ was "knot". It's meaning then spread to "entry" (in a ledger for example) or "item" (in a list for example). Then it spread to such forms as hend-. If you add a tail to a pyabu you get a nandau. For example henda = "to wear" is a nandau, or hendo = "an item of clothing'" is also a nandau
A list of the 12 colours
black | àu |
white | ái |
red | hìa |
green | gèu |
yellow | kiʔo |
light blue | nela |
dark blue | nelau |
orange | suna |
brown | dunu |
pink | celai |
purple | helau |
grey | lozo |
..... Three infixes for Verb => Verb
béu has a three process for generating new verbs from existing verbs. These three processes can be done to any verb.
.... -el-
If you split a verb and insert el between the final vowel (of the gamba) and the rest of the word, you give the added meaning of "to begin", "inception" or "to start off". For example ;-
sàu = to be
selau = to become
bìa = to be at
belia = to arrive at
doika = to walk
doikela = to start to walk
logo doikorwi = Roger used to walk ...
logo doikelorwi = Roger used to start walking ...
gazelari = I was born
à rì kiʔo = it was yellow ... remember that rì is an irregular form. The regular form would be *sori.
à lori kiʔo = it became yellow ... selau is irregular. If it were regular we would have the form *à selori ki@o
So there are thee irregular verbs in béu (well if you count selau as a different word from sau) ... sàu, bìa and selau.
.... -ow-
If you split a verb and insert ow between the final vowel of and the rest of the word, you get the meaning that you are making somebody else do the verb. For example ;-
ò timpiri = you hit him
(pás) gís ò timpowari = I made you hit him ???
A gamba form exists for this construction also. For example;-
doikowo = to make (somebody) walk
gasowa has the special meaning "to give birth" and doesn't mean "to create".
.... -ay-
If you split a verb and insert ay between the final vowel of and the rest of the word, you get the meaning that the verb is being attempted. For example ;-
selbaru à = I will drink it
selbayaru à = I will try and drink it
By the way, in béu to get a progressive meaning we use a Serial Verb Construction (SVC) ... báu bài kludora = The man is writing ... báu = man, bìa = to stay
By the way, in béu to get a passive meaning we use a Serial Verb Construction (SVC) ... toili gài kludorta = The book has been written ... toili = book, gùa = to undergo .... (toili gài kludorta is this right ?)
Actually we can make a really biy SVC and have toili bài gài kludora = The book is being written.
Index
- Introduction to Béu
- Béu : Chapter 1 : The Sounds
- Béu : Chapter 2 : The Noun
- Béu : Chapter 3 : The Verb
- Béu : Chapter 4 : Adjective
- Béu : Chapter 5 : Questions
- Béu : Chapter 6 : Derivations
- Béu : Chapter 7 : Way of Life 1
- Béu : Chapter 8 : Way of life 2
- Béu : Chapter 9 : Word Building
- Béu : Chapter 10 : Gerund Phrase
- Béu : Discarded Stuff
- A statistical explanation for the counter-factual/past-tense conflation in conditional sentences