Case of Polynesian Linguistics 203 10/8/2010 Polynesia Polynesia - - PowerPoint PPT Presentation

Linguistic Reconstruction: Case of Polynesian

Linguistics 203 10/8/2010

Polynesia

Polynesia

Polynesian Migration

Polynesian Languages

• Little contact with outside languages, so we

expect little blending

• No writing system before Westerners, thus no

written documentation of earlier language(s)

• How can we reconstruct the proto language?

Polynesian Languages

• All have significant similarities not shared with other

languages.

• cognates – words descended from a common source.

Tongan Samoan Tahitian Maori Hawaiian gloss manu manu manu manu manu ‘bird’ ika iʔa iʔa ika iʔa ‘fish’ kai ʔai ʔai kai ʔai ‘to eat’ tapu tapu tapu tapu kapu ‘forbidden’ vaka vaʔa vaʔa waka waʔa ‘canoe’ fohe foe hoe hoe hoe ‘oar’ mata mata mata mata maka ‘eye’ ʔuta uta uta uta uka ‘bush’ toto toto toto toto koko ‘blood’

(Table 13-1)

Sound Correspondences

• We look for sound correspondences to identify

what sound was in the proto language

Sound Correspondences

• What correspondences can we find here?

Tongan Samoan Tahitian Maori Hawaiian gloss manu manu manu manu manu ‘bird’ ika iʔa iʔa ika iʔa ‘fish’ kai ʔai ʔai kai ʔai ‘to eat’ tapu tapu tapu tapu kapu ‘forbidden’ Tongan Samoan Tahitian Maori Hawaiian m- m- m- m- m-

• n-
• n-
• n-
• n-
• n-
• k-
• ʔ-
• ʔ-
• k-
• ʔ-

t- t- t- t- k-

• p-
• p-
• p-
• p-
• p-

Sound Correspondences

• What correspondences can we find here?

Tongan Samoan Tahitian Maori Hawaiian gloss vaka vaʔa vaʔa waka waʔa ‘canoe’ fohe foe hoe hoe hoe ‘oar’ ʔuta uta uta uta uka ‘bush’ taŋi taŋi taʔi taŋi kani ‘to cry’ Tongan Samoan Tahitian Maori Hawaiian v- v- v- w- w-

• k-
• ʔ-
• ʔ-
• k-
• ʔ-

f- f- h- h- h-

• h-
• Ø-
• Ø-
• Ø-
• Ø-

ʔ- Ø- Ø- Ø- Ø- t- t- t- t- k-

• ŋ-
• ŋ-
• ʔ-
• ŋ-
• n-

Sound Correspondences

Tongan Samoan Tahitian Maori Hawaiian m- m- m- m- m-

• n-
• n-
• n-
• n-
• n-
• k-
• ʔ-
• ʔ-
• k-
• ʔ-

t- t- t- t- k-

• p-
• p-
• p-
• p-
• p-

v- v- v- w- w- f- f- h- h- h-

• h-
• Ø-
• Ø-
• Ø-
• Ø-

ʔ- Ø- Ø- Ø- Ø-

• ŋ-
• ŋ-
• ʔ-
• ŋ-
• n-
• Here are all the consonantal sound correspondences

from Tables 13-1 and 13-2 in the book.

Proto Polynesian Sound Inventory

• Which sound is the proto sound?
• General rule: it is the form requiring the least change

Tongan Samoan Tahitian Maori Hawaiian proto-sound m- m- m- m- m- *m-

• n-
• n-
• n-
• n-
• n-

*-n- t- t- t- t- k- *t-

Proto Polynesian Sound Inventory

• We can represent this as follows:
• In Hawaiian, */t/ became /k/. This is shown

with the following rule: *t > k

Proto Polynesian Sound Inventory

• What do we do in the following case?

Tongan Samoan Tahitian Maori Hawaiian proto-sound m- m- m- m- m- *m-

• n-
• n-
• n-
• n-
• n-

*-n- t- t- t- t- k- *t-

• ŋ-
• ŋ-
• ʔ-
• ŋ-
• n-

Proto Polynesian Sound Inventory

• What do we do in the following case?
• In Hawaiian, /n/ and /ŋ/ have merged.

Tongan Samoan Tahitian Maori Hawaiian proto-sound m- m- m- m- m- *m-

• n-
• n-
• n-
• n-
• n-

*-n- t- t- t- t- k- *t-

• ŋ-
• ŋ-
• ʔ-
• ŋ-
• n-

*-ŋ-

Proto Polynesian Sound Inventory

• What if we reconstructed *-n-
• We would have to explain the following:
• *-n- > -n-
• *-n- > -ŋ-

(in the same languages)

Tongan Samoan Tahitian Maori Hawaiian proto-sound m- m- m- m- m- *m-

• n-
• n-
• n-
• n-
• n-

*-n- t- t- t- t- k- *t-

• ŋ-
• ŋ-
• ʔ-
• ŋ-
• n-

*-n-

Proto Polynesian Sound Inventory

• Caveats:

– A sound may have changed the same way in all daughter languages; if so, it came from a phonetically similar sound

• What might /m/ come from in such a case?

– Borrowed words might have sounds not in native words, or not in the same position

• In Tongan, *s > h; however, /s/ is found in the loanword

/sikaleti/

Proto Polynesian Sound Inventory

• Caveats:

– A language may have borrowed a word while related dialects did not. – A language may have borrowed a word from a related dialects.

Subgroups

• Some daughter languages are more closely

related than others. Why?

Language A /sahag/ Language P Language X *h > ∅ *s > h /saag/ /hahag/ Language Q Language R Language Y Language Z | *a > o *g > k | /saag/ /soog/ /hahak/ /hahag/

Subgroups

• Tongan retains /h/, unlike the other languages.
• The other languages retain /l/ or /r/, unlike Tongan.
• Niuean shares these similarities with Tongan. They

belong together in a subgroup of Proto Polynesian.

Subgroups

Reconstructing Words

• To find proto forms, we look for reflexes (i.e.

cognates) in all subgroups.

• Most plausible form based on sound

correspondences is considered the proto form.

Tongan Samoan Tahitian Maori Hawaiian proto-form vaka vaʔa vaʔa waka waʔa Tongan Samoan Tahitian Maori Hawaiian proto-sound

• k-
• ʔ-
• ʔ-
• k-
• ʔ-

v- v- v- w- w-

Reconstructing Words

• Assume we had found the following sound

correspondences:

• What should we reconstruct?

Tongan Samoan Tahitian Maori Hawaiian proto-form vaka vaʔa vaʔa waka waʔa Tongan Samoan Tahitian Maori Hawaiian proto-sound

• k-
• ʔ-
• ʔ-
• k-
• ʔ-

*-k- v- v- v- w- w- *v-

Reconstructing Words

• Assume we had found the following sound

correspondences:

• What should we reconstruct?

Tongan Samoan Tahitian Maori Hawaiian proto-form vaka vaʔa vaʔa waka waʔa *vaka Tongan Samoan Tahitian Maori Hawaiian proto-sound

• k-
• ʔ-
• ʔ-
• k-
• ʔ-

*-k- v- v- v- w- w- *v-

Reconstructed Vocabulary Cultural Clues

• Words that can be reconstructed tell us about

the culture and origins of the ancestors of current language speakers.

Reconstructed Vocabulary Cultural Clues

• Many terms related to ocean

= people lived near water

• Topographic features found on large volcanic islands

= people probably didn’t live in atolls or raised coral islands

Reconstructed Vocabulary Cultural Clues

• Many terms related to sea animals
• Not as many related to land animals

Reconstructed Vocabulary Cultural Clues

• Bats and owls don’t exist in Tahiti, Easter Island or

the Marquesas; probably wasn’t original homeland

Reconstructed Vocabulary Cultural Clues

• Snakes only found east of Samoa; homeland

probably was not west of Samoa

Reconstructed Vocabulary Cultural Clues

• Pigs, though not native, existed on Polynesian islands, except

for Niue, Easter Island; also New Zealand (Maori)

Reconstructed Vocabulary Cultural Clues

• Proto-Polynesian word for ‘owl’ *lulu
• Hawaiian word for ‘owl’ pueo
• Marquesas have no owls. Ancient Polynesians

likely went to Marquesas, lost word for owl

• ver time, then went to Hawaii and needed a

new word for owl.

Reconstructed Vocabulary Cultural Clues

Reconstructed Vocabulary Cultural Clues

• Many words for fishing, cultivating.
• Three words for hunting:

*fana ‘to shoot with a bow’ *welo ‘to spear’ *seu ‘to snare with a net’

Summary

• Reconstruction looks for sound correspondences

in cognates of related languages to arrive at the proto sounds.

• Reconstruction of vocabulary relies on cognates

and sound correspondences.

• Not only does historical linguistics help

understand language change, origins and relations, it also provides clues about past cultures.

Reconstruction: The Comparative Method

• Requires a number of related languages
• Based on assumption that sound change is

regular

Reconstruction: The Comparative Method

• 1. Compile a set of cognates and eliminate

borrowings

• 2. Determine sound correspondences
• 3. Reconstruct a sound for each position
• 4. Once sounds correspondences are set up,

you can reconstruct proto forms.

Reconstruction: Choosing Which Sound to Reconstruct

• 1. Make a sound correspondence chart.
• 2. Do all languages have the same sound in a

particular position?

– if yes, reconstruct this sound; if not, continue to step 3.

Tongan Samoan Tahitian Maori Hawaiian proto-sound m- m- m- m- m- *m-

• n-
• n-
• n-
• n-
• n-

*-n-

Reconstruction: Choosing Which Sound to Reconstruct

• 3. Are there any sets of sound correspondences

like the following? (if not, go to 4)

• Reconstruct *-i- in ‘pitchfork’, and *-e- in

‘strawberry’

A B C gloss siza sesa siza ‘strawberry’ sizu sisu sizu ‘pitchfork’

Reconstruction: Choosing Which Sound to Reconstruct

• 4. Is one type of sound change more natural

than another? (if not, go to 5)

• -b- is slightly more frequent, but the change

VbV > VpV is less natural than VpV > VbV

A B C D E proto-sound

• p-
• b-
• b-
• p-
• b-

*-p-

Reconstruction: The Comparative Method

• Common sound changes:
• voiceless sounds become voiced between vowels and before

voiced consonants

• stops become fricatives between vowels
• consonants become palatalized before front vowels
• difficult consonant clusters are simplified
• difficult consonants are made easier (e.g. loss of aspiration in

stops)

• ral vowels become nasalized before nasals
• fricatives other than [h] become [h]
• [h] deletes between vowels
• clusters of vowels are broken up by consonants

Source: Language Files 10: Materials for an Introduction to Language and Linguistics. Anouschka Bergmann, Kathleen Currie Hall and Sharon Miriam Ross (eds.). The Ohio State University Press. Columbus, OH.

Reconstruction: Choosing Which Sound to Reconstruct

• 5. Use Occam’s Razor: the simplest solution is

the most likely

Reconstruction: Practice

• Find the sound correspondence for the vowels

in the following data:

English German Dutch Swedish Gloss [mæn] [man] [man] [man] ‘man’ [hænd] [hant] [hant] [hand] ‘hand’

Reconstruction: Practice

• What sound correspondences exist in the data

below?

Reconstruction: Practice

Mandarin Hakka gloss tɕin kim ‘zither’ la lat ‘spicy hot’ mɔ mɔk ‘lonesome’ lan lam ‘basket’ tɕi gip ‘worry’ lan lan ‘lazy’ pa pa ‘fear’

• What sound correspondences exist in the data

below? ([tɕ] is a voiceless, palatal affricate)

Reconstruction: Practice

Mandarin Hakka Mandarin Hakka tɕin kim tɕ- k- la lat

• n
• m

mɔ mɔk l- l- lan lam

• Ø
• t

tɕi gip m- m- lan lan

• Ø
• k

pa pa tɕ- g-

• Ø
• p
• n
• n

p- p-

Reconstruction: Practice

Mandarin Hakka proto-sounds tɕ- k-

• n
• m

l- l- l-

• Ø
• t

m- m- m-

• Ø
• k

tɕ- g-

• Ø
• p
• n
• n
• n

p- p- p-

Reconstruction: Practice

Mandarin Hakka proto-sounds tɕ- k-

• n
• m
• m

l- l- l-

• Ø
• t

m- m- m-

• Ø
• k

tɕ- g-

• Ø
• p
• n
• n
• n

p- p- p-

Reconstruction: Practice

Mandarin Hakka proto-sounds tɕ- k-

• n
• m
• m

l- l- l-

• Ø
• t
• t

m- m- m-

• Ø
• k
• k

tɕ- g-

• Ø
• p
• p
• n
• n
• n

p- p- p-

Reconstruction: Practice

Mandarin Hakka proto-sounds tɕ- k- k-

• n
• m
• m

l- l- l-

• Ø
• t
• t

m- m- m-

• Ø
• k
• k

tɕ- g- g-

• Ø
• p
• p
• n
• n
• n

p- p- p-

Reconstruction: Practice

proto-sound Mandarin Hakka proto-form k- tɕin kim kim

• m

la lat lat l- mɔ mɔk mɔk

• t

lan lam lam m- tɕi gip gip

• k

lan lan lan g- pa pa pa

• p
• n

p-

• Reconstruct the proto-forms of the words:

Reconstruction: Practice

Mandarin Hakka proto-forms tɕ- k- k-

• n
• m
• m

l- l- l-

• Ø
• t
• t

m- m- m-

• Ø
• k
• k

tɕ- g- g-

• Ø
• p
• p
• n
• n
• n

p- p- p-

• Describe the sound changes:
• Hakka:

– none

• Mandarin:

– Bilabial -m became alveolar -n – (voiceless) stops > Ø – initial (velar) stops > palatal affricates

(velar stops > palatal affricates in front of [i])