Languages of the World - Southeast Asian Languages

<lang>Thai</lang><br><ldata>Tai-Kadai - Thailand</ldata>

Z-Hot feat Zero - From the deepest corner

---

### Fact of the Day

Will's out of town next week!

---

### Will's out of town next week!

* Skype office hours Thursday are cancelled.

* Will will be slower to respond to emails

* Niloo will talk Austronesian Monday

* Nick Williams will talk Indonesian languages Wednesday

* I'll bring you back some awesome acoustics!

---

### Administrative Notes

* Exams will be handed back Friday in lab

* HW4 and HW5 are posted.  Enjoy!

* <danger>HW4 is due Friday!</danger>
	
---

<img class="r-stretch big" src="humorimg/assignmentcannotleave.jpg" alt="$txt"> 
	
---

# Southeast Asian Languages
### LING 1020 - Will Styler

---

### Today's Agenda

* Hmong Mien

* Tai-Kadai

* Austroasiatic

* Complexity in Language

---

## Southeast Asia is a Language Area

* This means many shared features across the unrelated langauges.

---

Put differently...

---

### Today, we're going to see a *lot* of Serial Verbs, Isolating morphology, and **Suprasegmentals**

---

### Which of the following is *not* a suprasegmental?

A) Level Tone

B) Vowel Nasality

C) Vowel Length

D) Retroflex Consonants

E) Contour Tone

---

### Which of the following is *not* a suprasegmental?

A) Level Tone

B) Vowel Nasality

C) Vowel Length

<correct>D) Retroflex Consonants</correct>

E) Contour Tone

---

Let's check out the families first!

---

# Hmong-Mien

---

---

### Hmong-Mien Languages

* The "H" is a lie.  Pronounced "Mong"

* Originally called a part of Sino-Tibetan

* Around 8 million speakers across Southeast Asia

* Denver and Boulder have a sizable Hmong-speaking population

---

### Hmong-Mien Features

* Isolating Morphology

* Serial Verb Constructions

* Numerical Classifiers

* Tonality

* **Voiceless nasals!**

* ... and lots of other fun consonants!

* **Suprasegmental-heavy**

* Tone, Nasality, Breathy, Creaky

---

### Isolating Morphology

Been there, done that...

---

---

### Serial Verb Constructions, Numerical Classifiers, Tonality

Been there, done that, no T-Shirt visual pun

* Sorry!
---

### Voiceless Nasals

* (and lots of other fun consonants!)

---

<lang>Danashan Miao</lang><br><ldata>Hmong-Mien - China</ldata>

<img class="r-stretch big" src="lotwimg/danashanhmongcons.png" alt="$txt"> 
---

### Suprasegmental Heavy

---

### Contrastive Nasality

Producing vowels with nasal airflow

* This is totally what I'm going to Indianapolis to talk about!

---

<lang>Hmong</lang><br><ldata>Hmong-Mien - China</ldata>

---

### Tone!

* Mandarin has 5 tones

* Hmong has 7 (8 in some dialects)

* Hmong also has breathiness and creak

---

<lang>Hmong</lang><br><ldata>Hmong-Mien - China</ldata>

---

<lang>Hmong</lang><br><ldata>Hmong-Mien - China</ldata>

### Hmoving on!

---

---

# Tai-Kadai

---

---

### Tai-Kadai Languages!

* Around 100 million speakers, across 95 languages

* Most commonly spoken languages are Thai and Lao

* Thai has around 60 million speakers
	
	* Lao has around 3 million speakers

---

### Features of Tai-Kadai

* Isolating Morphology

* Serial Verb Constructions

* Numerical Classifiers

* Tonality
 	
---

(Seeing a pattern?)

---

<lang>Thai</lang><br><ldata>Tai-Kadai - Thailand</ldata>

### Now that that's all Thai'ed up...

---

# Austroasiatic

---

---
### Austroasiatic Languages

* Some people call these "Mon Khmer"

* ... but that's not fair to the Munda sub-branch
	 
* Vietnamese is the most widely spoken (66 million speakers)

* Khmer (16 million), Santali (7 million) and Mundari (1.5 million) are the next most common
	
* The Munda branch is *very* different due to language contact.

* Pereltsvaig handles this well :)

---

---

<danger> Austroasiatic are *not* the same thing as "Austronesian" or "Afroasiatic" languages</danger>

---

### Austroasiatic:Mon Khmer Features

* Isolating Morphology

* Serial Verb Constructions

* Numerical Classifiers

* Tonality

* **"Wow, there's something to this "Language area" thing!"**

---

### About those serial verb constructions

---
<lang>Vietnamese</lang><br><ldata>Austroasiatic:Mon Khmer - Vietnam</ldata>

*tôi muốn đi về nấu nướng cho vợ tôi*

* I want go return cook roast give wife me

* I want to go home to cook for my wife.

---

<lang>Vietnamese</lang><br><ldata>Austroasiatic:Mon Khmer - Vietnam</ldata>

### Wrapping up Southeast Asia

* Sino-Tibetan, Austroasiatic, Hmong-Mien, and Tai-Kadai

* Tonality, Isolating, Numerical Classifiers, Serial Verb Constructions

* These occur throughout the language area!

* Lots of awesome suprasegmentals

* ... and very short words
---

Now, let's zoom back out.

* ### I want to talk about one of the reasons I *love* Language

---

# A Love Letter to Complexity

---

### Language is a deeply complex thing

* (because the world is deeply complex, and language has to describe it!)

---

### How do languages get nasality/tone/breathiness?

* They move complexity around!

---

</tr>
</table>
</center>

---

### "You know, nasal consonants are lame.  Let's get rid of them."

---

(Languages get rid of final consonants all the time, actually, and nasals are *very* commonly removed.)

---
<center>
<table>
  <tr>
    <th><h2>da
	</h2><img class="r-stretch" src="img/optimus.jpg" alt="$txt"></th>
    <th><h2>dad
		</h2><img class="r-stretch" src="humorimg/rooster.jpg" alt="$txt"></th>
    <th><h2>dag
		</h2><img class="r-stretch" src="humorimg/surprisedkoala.jpg" alt="$txt"></th>	
    <th><h2>da
		</h2><img class="r-stretch" src="img/highlighter.png" alt="$txt"></th>

</tr>
</table>
</center>

<danger>What's just gone wrong?</danger>

---

### "Well, maybe we can't remove nasals completely!"

* Let's leave just enough to maintain the contrast!
---

</tr>
</table>
</center>

* <correct>Contrast Preserved!</correct>

---

### "Man, I'm tired of final consonants altogether!"

* Let's trash those!

---

---

### "Uh, well, let's... uh... use tone!"

---

(There are tiny little aerodynamic properties of consonants that nudge languages towards particular tones.)

---

<correct>Problem solved!</correct>

---

## When you lose a contrast in one place, you need to replace it!

---

This reveals a fundamental truth of Language

---

We want our language to be...

* # 1) Simple enough to be usable

* # 2) Complex enough to be clear

---

### Language is all about finding the balance between these two extremes!

---

## We can't let "simple" win...

---

### New Language!

* Consonant Inventory: /b/

* Vowel Inventory: /a/

* Strict CV syllables

* <danger>What's the problem with this?</danger>

---

<center>
<table>
  <tr>
    <th><h2>ba
	</h2><img class="r-stretch" src="img/optimus.jpg" alt="$txt"></th>  
    <th><h2>baba
		</h2><img class="r-stretch" src="humorimg/rooster.jpg" alt="$txt"></th>  
    <th><h2>bababa
		</h2><img class="r-stretch" src="humorimg/surprisedkoala.jpg" alt="$txt"></th>	  
    <th><h2>babababa
		</h2><img class="r-stretch" src="img/highlighter.png" alt="$txt"></th>  
  </tr>
</table>
</center> 
---

The average English speaker knows around 30,000 words

---

## ba(x 30,000)
<img class="r-stretch big" src="humorimg/waterbuffalo.jpg" alt="$txt">

---

---

### Let's go more complex!

* Consonant Inventory: /b, d/

* Vowel Inventory: /a, i/

* Strict CV syllables

* Four possible syllables (ba, bi, da, di)

* To get 30,000 words, the longest would be 8 syllables

---

### More complex still!

Consonant Inventory: /b, d, g/

Vowel Inventory: /a, i, u/

Strict CV syllables

Nine possible syllables (ba, bi, bu, da, di, du, ga, gi, gu)

* To get 30,000 words, the longest would be 5 syllables

---

### What if we added tone?

Consonant Inventory: /b, d, g/

Vowel Inventory: /a, i, u/

Strict CV syllables

* High, Low, Rising, Falling Tone

* 36 possible syllables (ba, bi, bu, da, di, du, ga, gi, gu in all four tones)

* To get 30,000 words, the longest would be 3 syllables

---

### What about the other extreme?

* All 83 consonants on the IPA chart

* All 28 vowels on the IPA chart

* Strict CV ordering

* 2324 Unique Syllables

* With just two syllables, we get 5.4 *million* words

---

### There's more to the story!

* Some combinations aren't desirable

* Strict CV isn't that common

* Morphemes can help!

* Listeners can't handle 83 consonants and 28 vowels

* ... and they don't need to!

* But the moral of the story?

---

## Adding phonetic complexity reduces word length

* You're trading "complex sounds" for "shorter words"

* ... and tone helps with this!

---

## What about grammatical complexity?

---
### All languages can express everything

* For a language to be a language, you must be able to express any concept

* There's no such thing as a "simple" language
	
	* ... but that doesn't mean it's *easy* to express everything

* If you remove a method of expressing something, you'll need to make up for it *somehow*

* **You never reduce complexity, just move it around.**

---

### You cannot use tense anymore to mark time.

* What happens?  How can we cope?

---

### Temporal Arguments!

* "I walk to class yesterday"

* "I walk to class now"

* "I walk to class usually"

* "I walk to class tomorrow"

* No complexity is lost

* No simplicity is gained

---

### What if we removed pre/postpositions?

* Case marking

* Allative case - Handles "to" (and so on)

* This is what Finnish does!
	
* Serial verb constructions

* pig bring enter house = "bring the pig inside the house "
	
	* This is what Vietnamese does!

* Anything else?

---

## Complexity is never removed, just moved!

---

Here's a real-life example!

---

## Imagine you're a doctor!

---
"Keep records for *every patient*, but we won't pay you to write!"

* This forces *simplicity* and *economy*

---

"If your records are unclear, I'll sue your pants off!"

* This forces *clarity* and *precision*

---

The result?

---

> s/p lap appy conv. open, Lungs c/ausc, A&Ox3

* "Patient status-post laparoscopic appendectomy converted to open, lungs are clear to auscultation, the patient is alert and oriented x 3"

* "The patient has had a laparoscopic appendectomy which, due to a problem, was converted to an open appendectomy.  Her lungs weren't producing any concerning sounds when I listened, and she's alert and knows who she is, where she is, and what time it is."

---

## Medical records are hellscapes of inference

* cf. [Styler et. al. 2014](http://www.transacl.org/wp-content/uploads/2014/04/47.pdf)

---

### Other Gems from the Medical World

“Resected Invasive Grade 3 Adenocarcinoma (AJCC 7th PT4N1bMX) in tubovillious adenoma.”

* “We biopsied the colon, the results were negative”

* “Noted postoperative scarring.”

* “History of Pneumonia, Asthma, h/x diverticulitis, MS”

---

### When you leave out complexity, the listeners need to fill it in.

* ... and this is why nobody understands what doctors say!

---

# So...

---

### Language is complex

* Although we want language to be simple, it needs to be clear!

* You can have simple sound systems, or short words.  Not both.

* Removing complexity in one place adds it back in elsewhere!

* Doctors do terrible, dirty, and nasty things to language

* ... but it's because they had to!
	
---

### Are you more interested in...

a) Language and Language Family Specific Material (like the first part of today)

b) Material describing how Language works (like the second part of today)

---

## Next time: Japanese, Korean, and "Altaic"

<correct>Read Pereltsvaig Section 11.3</correct>

---

<huge>Thank you!</huge>