LECTURE 1

MENTALISM
General approach: the idea is not to prescribe what is correct and what is not, rather we would
like to understand what is happening in the vocal tract and the mind of the speaker and why is
that happening

- Speaking is an automatic but deeply regular/systematic activity and we want to

discover and understand these regularities (unknown knowns, habits, mental
knowledge) of native speakers so that as L2 speakers we get closer to their speaking
habits (what to do and when, or in which context)

Segmental phonetics (what we know from Phonetics 1)

- Organs of speech
- Description/classifications of sounds
- articulations (what they do to produce particular sounds)
- acoustics (how individual sounds look like on acoustic record – sound waves,
spectrogram)
- Use Praat as a tool for working on your pronunciation, particularly in short words out
of context

Beyond segments (this semester)

- Allophonic rules
- dark vs. clear [l], nasalization of vowels, aspiration, unreleasing plosives,
devoicing…
- Syllable (cont’d)
- Word stress
- Coarticulation (mostly across the words)
- Prosody (accents and boundaries and their absence)
- How do all these affect the realization of individual segment

SYLLABLE
- We know how to count them

Syllable boundaries
- How do we determine where one syllable ends and the other starts

We use syllables in processing what we hear.

 - The role that syllables have in speech processing
- An experiment where you are listening to words and have to press the button when
you hear a particular syllable for example ’ba’
- ‘ba’ is recognized faster in the word ‘balance’ and slower in the word ‘balcony’ =
- ‘bal’ is recognized faster in the word ‘balcony’ and slower in the word ‘balance’ =

- When we listen to people, in our minds we already organized the speech that come to
our ears into syllables and we have some preconceptions about this

Syllables participate in speech errors

Spoonerism
- Segments
- “a phonological rule”  “a phonological fool” (consonant preservation)
- “fill the pool”  “fool the pill” (vowel reversal)
- Syllables
- “unanimity of opinion”  “unamity of opinion” (syllable deletion)
- “Stockwell and Schacter”  “Schachwell and Stockter” (syllable reversal)
- Speech errors seem to obey a “structural law of syllable place”
- initial syllables interact with initial syllables, medial with medial, and final with final
- initial consonants of syllables (onsets) interact with onsets, vowels with vowels, and
final consonants of syllables (codas) with codas
- phonotactics tend to be respected (mad  had, dam x dah – since /h/ does not occur
at the ends of syllables in English
= Syllables are involved in speech planning

Syllable structure obeys language-specific ‘rules’

- Play, stop
- Most varieties of English have at least 24 consonants. Hence, there should be 24x24
initial 2-consonant clusters = THIS IS NOT TRUE = there is a very limited number of
2-consonant clusters in English
 - How are we going to decide which 2-consonants clusters are okay and which not
- is there a pattern?

Sonority = amount of sound while loudness is constant

- how far would the sound carry (/a/ has more energy than /i/) -hierarchy is connected
to the openness of the vocal tract – if it is open it will carry further than when it is
closed

Intermediate summary
Syllables are cognitively real divisions of speech
- we use them in speech processing (syllabic priming)
- speech production
- speech errors
- allophonic rules – shortening, nasalization, dark [l], aspiration, devoicing… all refer
to syllables
- must have some structure
- can be observed through independent general features (sonority) and language-
specific patterns

SYLLABLE STRUCTURE
- every syllable must contain a syllabic segment
- the nucleus (peak) of the syllable –can be a vowel or a syllabic consonant
- onset = any consonant or sequence of consonants preceding the nucleus
- coda = any consonant or sequence of consonants following the nucleus
- the nucleus is said to form a unit with the coda called the rhyme/rime

English syllable phonotactics

- Phonotactics refers to which segment goes with which segment in a specific language
(not a word starting ps, rk in English…) = patterns that guide which sounds combine
with which sounds
- Any vowel can form a peak of a syllable
- Some consonants can be peak as well (l,n – people, button)
 - In English syllabic consonants can be only found in unstressed syllables

ONSETS
0-3 consonants (0 = syllables that starts with a vowel /up/)
- Single consonants
-
- CC cluster
- s + C = s + any consonant = speak, smile, stop…
- C + [l, j, w, ɹ] – C is not nasal (ml), absent some homorganic cluster (same place of
articulations – dl, tl, pw)
- [s] vs [ʃ] – *sriek vs. shriek but slick vs *shlick
- CCC clusters = combination of 2 most used clusters
- s + [p,t,k] + [l, j, w, ɹ] = spray, skrew

CODAS
0-4
- Single consonants
- [j, w, ɹ, h] not possible (r is part of a vowel – diphthongs)
- CC cluster
- [m, n, ŋ, l, ɹ, s] + C = but not homorganic mb, ŋg
- C + [s, z, t, d, θ]
- CCC clusters = combination of the two CC patterns
- CCCC = extremely rare and often simplified in real speech

Syllable boundaries
- Respects language phonotactics
- e.g. sa.ndy is not possible because /nd/ is not possible onset in English
- Follow (not strictly) tendencies for sonority
- onset maximization = where there is a choice, always assign as many consonants as
possible to the onset, and as few as possible to the coda. However, remember that
every word must also consist of a sequence of well-formed syllables.
- Affect allophonic realization
- in the fist we expect an aspiration on /t/ and
in the second one we do not

Full structure of an English syllable

PHONOTACTICS IN ACTION
- Borrowings
-
- English plural and past tense

-
- schwa insertion or deletion?
- Slovak
- why do we say v skle, v hmle but not *v vode, *v Zvolene (we insert o – vo =
because of the syllable phonotactics, because of what syllables can be together)
- how about v/vo dverách or v/*vo tvári, vo/v kostole?

Onsets >> Codas

They do not have identical way in language
Onsets are more important then codas
- Repetitions with increasing frequency
- [pi] vs. [ip] = you can keep saying pi for much faster frequencies, but ip at some
point switches to pi
- The coordination of oral and glottal gestures in onsets is very stable across various
frequencies, not in coda
- Onset position is more perceptually salient
- ask  aks (in AAVE) but not *ski  ksi
- Phonological contrast in coda re often neutralized (spodobovanie)

Where are syllable boundaries

- How to syllabify falter, bottle, camera?
- a good test is the instruction? “Say each syllable twice”
- fal.fal-ter-ter, bot-bot-tl-tl, cam-cam-me-me-ra-ra
- Ambisyllabicity = the idea that some consonants might belong to both coda and the
onset of the syllable

SYLLABLE WEIGHT
- The concept needed for the discussion of stress
- Recall the difference between tense and lax vowels
- correlates with duration (tense vowels is longer than lax) = those that are short are
kind of lighter and those that are long are heavier
- tense is both long monophthongs and diphthongs (a is longer than i, e)
- no contents monosyllabic English words with lax vowels and no coda (none words
that end with lax vowels)

-
-

- Think of mora as a unit smaller than a syllable – that are heavy syllables (2 moras) and
light syllables (1 mora)

- Duration
- Intensity
- Pitch