“[aIt] flapping”

background-image: url(theme/3_Wydzial_Anglistyki_better-removebg-preview.png), url(theme/WA_logo_xaringan_small.png), url(theme/ncn-logo-poziom-en.png)
background-position: 35% 95%, 50% 92%, 80% 92%
background-size: 17%, 3%, 28%

---
class: clear, center, middle, hide-logo

.font200[.ipa[/aɪt/] sequences that sometimes flap end up with less dramatic F1 trajectories than those that never flap]

Accents 2021 :: December 10th, 2021

---
layout: false
class: center, middle, keep-h1-up

# PRICE raising

## Before voiceless obstruents

## *sight* .ipa[[ʌi]] (raised)

]

## Elsewhere

## *side* .ipa[[aɪ]] (not raised)

]

---
class: clear

background-image: url("media/anae_map10.37_ayraising.png")
background-size: contain

.footnote[[🔗 **Map 10.37, ANAE**](https://www.degruyter.com/document/doi/10.1515/9783110167467/html?lang=en)]

]

---

## *wide* .ipa[[waɪd]]

]

## *white* .ipa[[wʌit]]

]

`\(~\)`
`\(~\)`

.ref[Examples from: [🔗 University of Washington/Northwestern University Corpus](https://depts.washington.edu/phonlab/projects/uwnu.php)]

---
class: clear

---
class: clear

background-image: url("media/price_vd_formant_tracks.png")
background-size: contain

---
class: clear

---
class: inverse, center, middle, keep-h1-up

# PRICE raising vs. flapping

### *write* :: raising

### *ride* :: no raising

### *wri*.ipa[[ɾ]]*er* :: ???

### *writer* vs. *rider*

---
class: middle, keep-h1-up

# PRICE raising vs. flapping

## Flapping rate

## Hypothesis: Raising is negatively correlated with flapping rate

(Cumulative Context Effect, cf. .ref[Raymond et al. 2016])

---
class: font180

# Method

- Data: Buckeye .ref[(Pitt et al. 2007)]
    - 40 speakers, stratified for age (< 30, > 40) and gender (female, male) from Columbus, OH
    - Midland (Labov, Ash & Boberg 2006) dialect area
    - Hand-corrected allophonic transcriptions

- Analysis: 
    - Corpus querying in Labb-CAT <img style="float:left" src="media/labb-cat2.png" /> (Fromont & Hay 2012)
    - **22,184** word tokens with PRICE retrieved
    - 9 measurements (F1 & F2) per vowel taken (Praat integrated with Labb-CAT)
    - <svg aria-hidden="true" role="img" viewBox="0 0 581 512" style="height:1em;width:1.13em;vertical-align:-0.125em;margin-left:auto;margin-right:auto;font-size:inherit;fill:#76232F;overflow:visible;position:relative;"><path d="M581 226.6C581 119.1 450.9 32 290.5 32S0 119.1 0 226.6C0 322.4 103.3 402 239.4 418.1V480h99.1v-61.5c24.3-2.7 47.6-7.4 69.4-13.9L448 480h112l-67.4-113.7c54.5-35.4 88.4-84.9 88.4-139.7zm-466.8 14.5c0-73.5 98.9-133 220.8-133s211.9 40.7 211.9 133c0 50.1-26.5 85-70.3 106.4-2.4-1.6-4.7-2.9-6.4-3.7-10.2-5.2-27.8-10.5-27.8-10.5s86.6-6.4 86.6-92.7-90.6-87.9-90.6-87.9h-199V361c-74.1-21.5-125.2-67.1-125.2-119.9zm225.1 38.3v-55.6c57.8 0 87.8-6.8 87.8 27.3 0 36.5-38.2 28.3-87.8 28.3zm-.9 72.5H365c10.8 0 18.9 11.7 24 19.2-16.1 1.9-33 2.8-50.6 2.9v-22.1z"/></svg> Mixed-effects Generilized Additive Modeling (GAM) with `mgcv` (Wood 2017) in R (R Core Team 2021)

---
class: clear

---
class: clear

---
class: clear

---
class: clear

---
class: clear

---
class: clear

---
class: clear

---
class: clear

---
class: inverse, center, middle

# Hypothesis 1️⃣

### Degree of raising is highest in words that never flap (*rights*), lower in words that variably flap (*__bite__ it*, *__bite__ this*), and lowest in words that almost always flap (*biting*)

---
# Result 1️⃣

---
class: inverse, center, middle

# Hypothesis 2️⃣

### In the non-flapping environment, the degree of raising is higher in words that never flap (*rights*) than in words that variably flap (*__bite__ this*)

---
# Result 2️⃣

---
# Result 2️⃣

]

]

---
class: font180
# Conclusions

- Result 1️⃣: most raising in words that never flap, least raising in words that usually flap

➡ flapping disfavors raising, phonetic voicelessness favors raising

--
    
- Result 2️⃣: words that variably flap show less raising than words than never flap, **even in non-flapping environment**

➡ frequency of occurrence in favoring (here: non-flapping) environment influences the extent of raising **even outside that environment**

--
    
*Result 2 requires rich storage: The phonological representation of words is shaped by their frequent phonetic realization.*

---
class: middle, hide-logo, keep-h1-up, font150

# Thank you!

**In all environments:**

👉 Raising is negatively correlated with flapping rate

**In non-flapping environment only:**

👉 Words that variably flap show less raising than words than never flap

👉 This Cumulative Context Effect supports rich storage

]

]

<hr>

This research was supported by an NCN grant no. 2017/26/D/HS2/00027

]
]