A contrastivist approach to the emergence of sound inventories Jade - - PowerPoint PPT Presentation

a contrastivist approach to the emergence of sound
SMART_READER_LITE
LIVE PREVIEW

A contrastivist approach to the emergence of sound inventories Jade - - PowerPoint PPT Presentation

jsandstedt@gmail.com jsandstedt.hcommons.org A contrastivist approach to the emergence of sound inventories Jade J. Sandstedt Humboldt University of Berlin 25. May 219 Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound


slide-1
SLIDE 1

A contrastivist approach to the emergence of sound inventories

Jade J. Sandstedt

jsandstedt@gmail.com jsandstedt.hcommons.org

Humboldt University of Berlin

  • 25. May 219

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

1 / 28

slide-2
SLIDE 2

Contrasts and features

Basic observation: ▶ size/shape of a language’s sound inventory ~ active phonological features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

2 / 28

(Hall 27; Dresher 29, 218; Mackenzie 213, 216; Iosad 217)

slide-3
SLIDE 3

Contrasts and features

(1) Contrastivist Hypothesis (Hall 27, p. 2) The phonological component of a language L operates only on those features which are necessary to distinguish the phonemes of L fsom one another

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

3 / 28

slide-4
SLIDE 4

Step 1: Identifz contrasts

(2) Sample of Chewa (N.31) vowel contrasts (Downing & Mtenje 217, ch. 3) a. túm- ‘send’ b. ɡɔ ́ n- ‘sleep’ c. phík- ‘cook’ d. tsɛ ́ k- ‘close’

☞ Acquired contrasts: {u, ɔ, i, ɛ}

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

4 / 28

slide-5
SLIDE 5

Step 1: Identifz contrasts

(2) Sample of Chewa (N.31) vowel contrasts (Downing & Mtenje 217, ch. 3) a. túm- ‘send’ b. ɡɔ ́ n- ‘sleep’ c. phík- ‘cook’ d. tsɛ ́ k- ‘close’

☞ Acquired contrasts: {u, ɔ, i, ɛ}

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

4 / 28

slide-6
SLIDE 6

Step 1: Identifz contrasts

(2) Sample of Chewa (N.31) vowel contrasts (Downing & Mtenje 217, ch. 3) a. túm- ‘send’ b. ɡɔ ́ n- ‘sleep’ c. phík- ‘cook’ d. tsɛ ́ k- ‘close’

☞ Acquired contrasts: {u, ɔ, i, ɛ}

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

4 / 28

slide-7
SLIDE 7

Step 2: Dene features

{u, ɔ, i, ɛ}

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

5 / 28

slide-8
SLIDE 8

Step 2: Dene features

{u, ɔ, i, ɛ} [labial] {u, ɔ} (non-labial) {i, ɛ}

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

5 / 28

slide-9
SLIDE 9

Step 2: Dene features

{ɔ, u, ɛ, i} [labial] {ɔ, u} [open] /ɔ/ (non-open) /u/ (non-labial) {ɛ, i} [open] /ɛ/ (non-open) /i/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

5 / 28

slide-10
SLIDE 10

What about acquisition?

Implication:

  • 1. Acquire segmental contrasts
  • 2. Dene features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

6 / 28

slide-11
SLIDE 11

What about acquisition?

Implication:

  • 1. Acquire segmental contrasts
  • 2. Dene features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

6 / 28

slide-12
SLIDE 12

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts

features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-13
SLIDE 13

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts

features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-14
SLIDE 14

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts

features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-15
SLIDE 15

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts

features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-16
SLIDE 16

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

▶ e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts

features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-17
SLIDE 17

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

▶ e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? ▶ e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts

features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-18
SLIDE 18

What about acquisition?

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?
  • 2. Once a contrast is acquired, how do language learners select their features?

▶ e.g. /ɛ/ vs. /i/ : [open], [close], [ATR], [RTR], something else? ▶ e.g. /ɛ/ vs. /ɔ/ : [labial], [dorsal], [coronal], something else?

☞ it is unclear how we go fsom contrasts → features

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

7 / 28

slide-19
SLIDE 19

Crux of the problem:

The top-down focus of contrastivist approaches: phonemic inventory features

* requires serious abstraction

seems to be the wrong direction

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

8 / 28

slide-20
SLIDE 20

Crux of the problem:

The top-down focus of contrastivist approaches: ▶ phonemic inventory → features

* requires serious abstraction

seems to be the wrong direction

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

8 / 28

slide-21
SLIDE 21

Crux of the problem:

The top-down focus of contrastivist approaches: ▶ phonemic inventory → features

* requires serious abstraction

▶ seems to be the wrong direction

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

8 / 28

slide-22
SLIDE 22

Outline

1

A bottom-up contrastivist approach Micro-cue model of acquisition Chewa test case: distinctions in lexical meaning and phonological behaviour

2 Conclusions

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

9 / 28

slide-23
SLIDE 23

A bottom-up contrastivist approach

Top-down limitations on sets of features are not the only way: ▶ we can go fsom the bottom-up by re-dening the Contrastivist Hypothesis

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

1 / 28

slide-24
SLIDE 24

A bottom-up contrastivist approach

(3) Correlate Contrastivist Hypothesis The phonemes of a language L are equal to the sum of features and feature co-occurrence restrictions which are minimally necessary for the expression

  • f phonological regularities in L.

☞ features

phonemes

not features phonemes

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

11 / 28

slide-25
SLIDE 25

A bottom-up contrastivist approach

(3) Correlate Contrastivist Hypothesis The phonemes of a language L are equal to the sum of features and feature co-occurrence restrictions which are minimally necessary for the expression

  • f phonological regularities in L.

☞ features

phonemes

not features phonemes

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

11 / 28

slide-26
SLIDE 26

A bottom-up contrastivist approach

(3) Correlate Contrastivist Hypothesis The phonemes of a language L are equal to the sum of features and feature co-occurrence restrictions which are minimally necessary for the expression

  • f phonological regularities in L.

☞ features → phonemes

▶ not features ↚ phonemes

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

11 / 28

slide-27
SLIDE 27

Same basic road map

Observe phonological patterns

tsɛ ́ k- vs. phík-

Generalise a feature for any observed distinction

[F] (emergent feature theory – Mielke 28; Iosad 217)

Presence/absence of a feature mechanically denes a contrast

[F] // vs. (non-F) //

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-28
SLIDE 28

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

Generalise a feature for any observed distinction

[F] (emergent feature theory – Mielke 28; Iosad 217)

Presence/absence of a feature mechanically denes a contrast

[F] // vs. (non-F) //

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-29
SLIDE 29

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

[F] (emergent feature theory – Mielke 28; Iosad 217)

Presence/absence of a feature mechanically denes a contrast

[F] // vs. (non-F) //

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-30
SLIDE 30

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

Presence/absence of a feature mechanically denes a contrast

[F] // vs. (non-F) //

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-31
SLIDE 31

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

Presence/absence of a feature mechanically denes a contrast

[F] // vs. (non-F) //

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-32
SLIDE 32

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

▶ Presence/absence of a feature mechanically denes a contrast

[F] // vs. (non-F) //

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-33
SLIDE 33

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

▶ Presence/absence of a feature mechanically denes a contrast

▶ [F] /a/ vs. (non-F) /b/

Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-34
SLIDE 34

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

▶ Presence/absence of a feature mechanically denes a contrast

▶ [F] /ɛ/ vs. (non-F) /i/

▶ Realised according to language-particular rules of phonetic implementation

‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-35
SLIDE 35

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

▶ Presence/absence of a feature mechanically denes a contrast

▶ [F] /ɛ/ vs. (non-F) /i/

▶ Realised according to language-particular rules of phonetic implementation

▶ ‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal

features contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-36
SLIDE 36

Same basic road map

▶ Observe phonological patterns

▶ tsɛ ́ k- vs. phík-

▶ Generalise a feature for any observed distinction

▶ [F] (emergent feature theory – Mielke 28; Iosad 217)

▶ Presence/absence of a feature mechanically denes a contrast

▶ [F] /ɛ/ vs. (non-F) /i/

▶ Realised according to language-particular rules of phonetic implementation

▶ ‘phonetics does not determine phonological destiny’, nor vice versa (cf. Hall 214; Anderson 1981; Hale & Reiss 2)

☞ speech signal → features → contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

12 / 28

slide-37
SLIDE 37

Outline

1

A bottom-up contrastivist approach Micro-cue model of acquisition Chewa test case: distinctions in lexical meaning and phonological behaviour

2 Conclusions

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

13 / 28

slide-38
SLIDE 38

What about acquisition?

Phonological acquisition: ▶ based on insights fsom Westergaard’s (29, 213, 214) model of micro-cues

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

14 / 28

slide-39
SLIDE 39

Acquisition via micro-cues

Assumptions: children are sensitive to ne linguistic distinctions generalise small pieces of abstract linguistic structure (‘micro-cues’)

e.g. a cue for OV word order is generalised as VP[DP V]

☞ Micro-cues accumulate in the course of language acquisition

the sum of which denes the linguistic grammar

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

15 / 28

slide-40
SLIDE 40

Acquisition via micro-cues

Assumptions: ▶ children are sensitive to ne linguistic distinctions generalise small pieces of abstract linguistic structure (‘micro-cues’)

e.g. a cue for OV word order is generalised as VP[DP V]

☞ Micro-cues accumulate in the course of language acquisition

the sum of which denes the linguistic grammar

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

15 / 28

slide-41
SLIDE 41

Acquisition via micro-cues

Assumptions: ▶ children are sensitive to ne linguistic distinctions ▶ generalise small pieces of abstract linguistic structure (‘micro-cues’)

e.g. a cue for OV word order is generalised as VP[DP V]

☞ Micro-cues accumulate in the course of language acquisition

the sum of which denes the linguistic grammar

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

15 / 28

slide-42
SLIDE 42

Acquisition via micro-cues

Assumptions: ▶ children are sensitive to ne linguistic distinctions ▶ generalise small pieces of abstract linguistic structure (‘micro-cues’)

▶ e.g. a cue for OV word order is generalised as VP[DP V]

☞ Micro-cues accumulate in the course of language acquisition

the sum of which denes the linguistic grammar

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

15 / 28

slide-43
SLIDE 43

Acquisition via micro-cues

Assumptions: ▶ children are sensitive to ne linguistic distinctions ▶ generalise small pieces of abstract linguistic structure (‘micro-cues’)

▶ e.g. a cue for OV word order is generalised as VP[DP V]

☞ Micro-cues accumulate in the course of language acquisition

▶ the sum of which denes the linguistic grammar

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

15 / 28

slide-44
SLIDE 44

Adapting the micro-cue model to phonology

Representational micro-cues: emergent, privative features and feature co-occurrence restrictions

e.g. [F], [G], and *[F, G]

Emerging inventories: micro-cues accumulate in the course of language acquisition

dening a set of contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

16 / 28

slide-45
SLIDE 45

Adapting the micro-cue model to phonology

Representational micro-cues: emergent, privative features and feature co-occurrence restrictions

e.g. [F], [G], and *[F, G]

Emerging inventories: micro-cues accumulate in the course of language acquisition

dening a set of contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

16 / 28

slide-46
SLIDE 46

Adapting the micro-cue model to phonology

Representational micro-cues: ▶ emergent, privative features and feature co-occurrence restrictions

e.g. [F], [G], and *[F, G]

Emerging inventories: micro-cues accumulate in the course of language acquisition

dening a set of contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

16 / 28

slide-47
SLIDE 47

Adapting the micro-cue model to phonology

Representational micro-cues: ▶ emergent, privative features and feature co-occurrence restrictions

▶ e.g. [F], [G], and *[F, G]

Emerging inventories: micro-cues accumulate in the course of language acquisition

dening a set of contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

16 / 28

slide-48
SLIDE 48

Adapting the micro-cue model to phonology

Representational micro-cues: ▶ emergent, privative features and feature co-occurrence restrictions

▶ e.g. [F], [G], and *[F, G]

Emerging inventories: ▶ micro-cues accumulate in the course of language acquisition

▶ dening a set of contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

16 / 28

slide-49
SLIDE 49

Adapting the micro-cue model to phonology

(4) Segment inventory dened by [F], [G], and *[F, G] cues Micro-cues Phonemes [F] /a/ [ ] /b/ [G] /c/ *[F, G] */d/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

17 / 28

slide-50
SLIDE 50

A working phonological learning algorithm

When parsing phonological input:

1

For every observed contrast/alternation

▶ generalise a unique representational micro-cue in the form of a feature [F] e.g. [phík-] vs. [tsɛ ́ k-] [F]

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

18 / 28

slide-51
SLIDE 51

A working phonological learning algorithm

When parsing phonological input:

1

For every observed contrast/alternation

▶ generalise a unique representational micro-cue in the form of a feature [F] ▶ e.g. [phík-] vs. [tsɛ ́ k-] = [F]

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

18 / 28

slide-52
SLIDE 52

Outline

1

A bottom-up contrastivist approach Micro-cue model of acquisition Chewa test case: distinctions in lexical meaning and phonological behaviour

2 Conclusions

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

19 / 28

slide-53
SLIDE 53

Chewa vowel contrasts and alternations

(5) Chewa contrasts and height harmony High phík-il- ‘cook’-appl. túm-il- ‘send’-appl. Mid tsɛ ́ k-ɛl- ‘close’-appl. ɡɔ ́ n-ɛ ́ l- ‘sleep’-appl.

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

2 / 28

slide-54
SLIDE 54

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [] phík-il- [] vs. (non-) contrasts/harmony

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-55
SLIDE 55

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [] phík-il- [] vs. (non-) contrasts/harmony

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-56
SLIDE 56

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [] phík-il- [] vs. (non-) contrasts/harmony

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-57
SLIDE 57

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [F] phík-il- [] vs. (non-) contrasts/harmony

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-58
SLIDE 58

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [F] phík-il- [F] vs. (non-F) contrasts/harmony

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-59
SLIDE 59

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-60
SLIDE 60

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [] phík-il- [] vs. (non-) contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-61
SLIDE 61

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [] phík-il- [] vs. (non-) contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-62
SLIDE 62

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [G] phík-il- [] vs. (non-) contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-63
SLIDE 63

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [G] phík-il- [G] vs. (non-G) contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-64
SLIDE 64

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [labial] phík-il- [labial] vs. (non-labial) contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-65
SLIDE 65

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [labial] phík-il- [labial] vs. (non-labial) contrasts c. ɡɔ ́ n-ɛ ́ l- [ɔ] vs. [u] [labial, (open)] túm-il- [labial, open] vs. [labial] contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-66
SLIDE 66

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [labial] phík-il- [labial] vs. (non-labial) contrasts c. ɡɔ ́ n-ɛ ́ l- [ɔ] vs. [u] [labial, (open)] túm-il- [labial, open] vs. [labial] contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-67
SLIDE 67

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [labial] phík-il- [labial] vs. (non-labial) contrasts c. ɡɔ ́ n-ɛ ́ l- [ɔ] vs. [u] [labial, (open)] túm-il- [labial, open] vs. [labial] contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-68
SLIDE 68

Generalising Chewa representational micro-cues

(6) Generalising Chewa vocalic representational micro-cues

Patterns Surface generalisations Micro-cue a. tsɛ ́ k-ɛl- [ɛ] vs. [i] [open] phík-il- [open] vs. (non-open) contrasts/harmony b. túm-il- [u] vs. [i] [labial] phík-il- [labial] vs. (non-labial) contrasts c. ɡɔ ́ n-ɛ ́ l- [ɔ] vs. [u] [labial, (open)] túm-il- [labial, open] vs. [labial] contrasts

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

21 / 28

slide-69
SLIDE 69

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-70
SLIDE 70

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-71
SLIDE 71

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-72
SLIDE 72

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-73
SLIDE 73

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-74
SLIDE 74

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-75
SLIDE 75

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-76
SLIDE 76

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-77
SLIDE 77

Dening inventories using a set of emergent features

(7) Chewa vowels dened by [open] and [labial] cues Micro-cues Phonemes [open] /ɛ/ [ ] /i/ [labial] /u/ [open, labial] /ɔ/

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

22 / 28

slide-78
SLIDE 78

Reinterpretation of the issues

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?

contrasts are mechanically dened by the emergent set of features

  • 2. Once a contrast is acquired, how do language learners select their features?

features predictably emerge when parsing linguistic input

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

23 / 28

slide-79
SLIDE 79

Reinterpretation of the issues

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?

contrasts are mechanically dened by the emergent set of features

  • 2. Once a contrast is acquired, how do language learners select their features?

features predictably emerge when parsing linguistic input

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

23 / 28

slide-80
SLIDE 80

Reinterpretation of the issues

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?

contrasts are mechanically dened by the emergent set of features

  • 2. Once a contrast is acquired, how do language learners select their features?

▶ features predictably emerge when parsing linguistic input

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

23 / 28

slide-81
SLIDE 81

Reinterpretation of the issues

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?

contrasts are mechanically dened by the emergent set of features

  • 2. Once a contrast is acquired, how do language learners select their features?

▶ features predictably emerge when parsing linguistic input

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

23 / 28

slide-82
SLIDE 82

Reinterpretation of the issues

Questions:

  • 1. How do language learners acquire contrasts in the absence of features?

▶ contrasts are mechanically dened by the emergent set of features

  • 2. Once a contrast is acquired, how do language learners select their features?

▶ features predictably emerge when parsing linguistic input

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

23 / 28

slide-83
SLIDE 83

Outline

1

A bottom-up contrastivist approach Micro-cue model of acquisition Chewa test case: distinctions in lexical meaning and phonological behaviour

2 Conclusions

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

24 / 28

slide-84
SLIDE 84

Conclusions

A bottom-up approach to the emergence of sound inventories:

  • 1. recapitulates contrastivist insights

upper bound on number of features correlation between sound inventories ~ features predicts a minimalist representational architecture

  • 2. provides an explicit account of how representations are acquired

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

25 / 28

slide-85
SLIDE 85

Conclusions

A bottom-up approach to the emergence of sound inventories:

  • 1. recapitulates contrastivist insights

upper bound on number of features correlation between sound inventories ~ features predicts a minimalist representational architecture

  • 2. provides an explicit account of how representations are acquired

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

25 / 28

slide-86
SLIDE 86

Conclusions

A bottom-up approach to the emergence of sound inventories:

  • 1. recapitulates contrastivist insights

▶ upper bound on number of features correlation between sound inventories ~ features predicts a minimalist representational architecture

  • 2. provides an explicit account of how representations are acquired

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

25 / 28

slide-87
SLIDE 87

Conclusions

A bottom-up approach to the emergence of sound inventories:

  • 1. recapitulates contrastivist insights

▶ upper bound on number of features ▶ correlation between sound inventories ~ features predicts a minimalist representational architecture

  • 2. provides an explicit account of how representations are acquired

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

25 / 28

slide-88
SLIDE 88

Conclusions

A bottom-up approach to the emergence of sound inventories:

  • 1. recapitulates contrastivist insights

▶ upper bound on number of features ▶ correlation between sound inventories ~ features ▶ predicts a minimalist representational architecture

  • 2. provides an explicit account of how representations are acquired

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

25 / 28

slide-89
SLIDE 89

Conclusions

A bottom-up approach to the emergence of sound inventories:

  • 1. recapitulates contrastivist insights

▶ upper bound on number of features ▶ correlation between sound inventories ~ features ▶ predicts a minimalist representational architecture

  • 2. provides an explicit account of how representations are acquired

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

25 / 28

slide-90
SLIDE 90

The broader picture

A highly simplied version of the model: Small, symmetric inventory and only two features

/ɛ, i, ɔ, u/ and [open]/[labial]

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

26 / 28

slide-91
SLIDE 91

The broader picture

A highly simplied version of the model: ▶ Small, symmetric inventory and only two features

▶ /ɛ, i, ɔ, u/ and [open]/[labial]

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

26 / 28

slide-92
SLIDE 92

The broader picture

What about?: asymmetric inventories (feature co-occurrence restrictions) interpretation of dierent kinds of segmental phonological patterns

e.g. locality variation (hierarchical organisation of features) e.g. grouped behaviour of features (geometric grouping into classes)

the formalisation of representational micro-cues extended handout online at jsandstedt.hcommons.org

☞ Sandstedt (218, ch. 2–3)

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

27 / 28

slide-93
SLIDE 93

The broader picture

What about?: ▶ asymmetric inventories (feature co-occurrence restrictions) interpretation of dierent kinds of segmental phonological patterns

e.g. locality variation (hierarchical organisation of features) e.g. grouped behaviour of features (geometric grouping into classes)

the formalisation of representational micro-cues extended handout online at jsandstedt.hcommons.org

☞ Sandstedt (218, ch. 2–3)

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

27 / 28

slide-94
SLIDE 94

The broader picture

What about?: ▶ asymmetric inventories (feature co-occurrence restrictions) ▶ interpretation of dierent kinds of segmental phonological patterns

▶ e.g. locality variation (hierarchical organisation of features) ▶ e.g. grouped behaviour of features (geometric grouping into classes)

the formalisation of representational micro-cues extended handout online at jsandstedt.hcommons.org

☞ Sandstedt (218, ch. 2–3)

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

27 / 28

slide-95
SLIDE 95

The broader picture

What about?: ▶ asymmetric inventories (feature co-occurrence restrictions) ▶ interpretation of dierent kinds of segmental phonological patterns

▶ e.g. locality variation (hierarchical organisation of features) ▶ e.g. grouped behaviour of features (geometric grouping into classes)

▶ the formalisation of representational micro-cues extended handout online at jsandstedt.hcommons.org

☞ Sandstedt (218, ch. 2–3)

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

27 / 28

slide-96
SLIDE 96

The broader picture

What about?: ▶ asymmetric inventories (feature co-occurrence restrictions) ▶ interpretation of dierent kinds of segmental phonological patterns

▶ e.g. locality variation (hierarchical organisation of features) ▶ e.g. grouped behaviour of features (geometric grouping into classes)

▶ the formalisation of representational micro-cues ▶ extended handout online at jsandstedt.hcommons.org

☞ Sandstedt (218, ch. 2–3)

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

27 / 28

slide-97
SLIDE 97

The broader picture

What about?: ▶ asymmetric inventories (feature co-occurrence restrictions) ▶ interpretation of dierent kinds of segmental phonological patterns

▶ e.g. locality variation (hierarchical organisation of features) ▶ e.g. grouped behaviour of features (geometric grouping into classes)

▶ the formalisation of representational micro-cues ▶ extended handout online at jsandstedt.hcommons.org

☞ Sandstedt (218, ch. 2–3)

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

27 / 28

slide-98
SLIDE 98

Thanks for listening!

Jade J. Sandstedt (Humboldt University of Berlin) Emergence of sound inventories

  • 25. May 219

28 / 28