Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Phonetics and Phonology

No description
by

Kristen Ware

on 17 January 2014

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Phonetics and Phonology

Larynx
Acoustic
Phonetics
The Sounds of English
Auditory
What are
sound waves
and how are they
produced during speech
?
How do we
talk about
sounds?
Why don't we just use the alphabet?
the International Phonetic Alphabet
The anatomy of speech
Consonants v. Vowels
Consonants
Vowels
Monophthongs
Diphthongs
Tongue height
Tongue advancement
Lip rounding
Tenseness
Parameters
Transcription
Phonology
Phonotactics
Phonemes
Phonological processes
The study of the production and comprehension
of speech sounds.
Three main areas:
Acoustic
Auditory
Articulatory
The study of the production of speech sounds in the vocal tract.
Voicing
Airstream
Place
Manner
For more, see 'Articulatory' below
Coordinated movements of the various articulators in the human mouth and throat produce the many sounds of human language.
The sounds of human language fall into one of two main categories:
There are two kinds of vowel sounds: monophthongs and diphthongs
Monophthongs are
single
vowel sounds. The tongue remains in one position during the articulation of the vowel. They are classified with respect to four parameters
How close the tongue is to the roof of the mouth during articulation of the vowel
How far forward or backward the bulk of the tongue is during articulation of the vowel
The shape of the lips during articulation of the vowel
Whether or not the tongue is in an 'extreme' position of the mouth. If mid-central is the neutral position of the tongue, a
tense
vowel means that there is a bigger change in the tongue position from this neutral spot (so, the high vowels [i] and [u] are tense).
Lax
vowels, on the other hand, involve little movement from this neutral position. Most vowels in English are lax.
Phonetics and Phonology
...couldn't we?
Benefits:
1. It's based on etymology (word origins)
2. It's fairly standard across dialects of English
3. (most importantly) we are familiar with it already
But!
The English alphabet is no longer straightforwardly related to the sound system. Characters can represent various sounds, or no sounds, or sounds can lack characters to represent them.
The alphabet is problematic for several reasons:
There may be a mismatch between the number of sounds and letters
One letter may represent multiple sounds
One sound may be represented by multiple letter combinations
And!
What about sounds from other languages?
a. Combinations of letters may represent a single sound
(
sh
oot, rou
gh
, ei
th
er, na
ti
on, c
oa
t)
b. 'Silent' letters may represent no sound at all
(autum
n
, resi
g
n,
w
rath, hol
e
, s
w
ord,
k
ni
gh
t)
c. Sounds that occur may not be represented by any letter
(
cute
is 'kyute', not 'coot';
use
is 'yous' not 'ooze')
Xhosa
, a Bantu language spoken in various African nations, has sounds called
'clicks'
that are made by sucking air
into
the mouth.
...Which letter of the alphabet would we use to represent these sounds?
Introducing
(Because of these problems, linguists use the IPA to talk about sounds)
Forget about spelling!
Focus on
sounds
The IPA (International Phonetic Alphabet) is a
universal
system for describing the sounds of all human language.

Each IPA symbol represents only
one
sound.

The symbols are organized in the IPA chart according to their articulatory properties.
Diphthongs are like 'double' vowel sounds. The tongue moves during the articulation of these vowels. They are classified according to the sequence of vowels that constitute them, as the video explains.
The study of the sound system
Phonology involves studying the rules that govern the sounds identified in phonetics
.
These rules come in two varieties:
Rules governing possible sound sequences (=Phonotactics)
Rules governing which sounds carry a meaning difference (= Phonemes)
The syllable
The syllable consists of two main parts: the onset and the rhyme, which is further divided into the nucleus and the coda. These parts are organized hierarchically as shown.
Most phonotactic constraints are defined at the level of the syllable
The nucleus is the '
heart
' of the syllable. It is the
only required component
. Vowels only appear in the nucleus, never the onset or coda.
Syllable
Onset
Rhyme
Nucleus
Coda
The rhyme contains the nucleus and the coda. In emphatic or elongated speech, this is the portion of the syllable that is lengthened or that receives extra stress. The onset is never affected in this way.
The coda is an
optional
component of the syllable. In English, when there is a coda, it may be comprised of one, two, three, or, in very rare cases, four consonants. If there is more than one consonant in the coda, it is said to be
complex
.
The onset is an
optional
part of the syllable. In English, when present, the onset may contain one, two, or three consonant sounds. When more than one consonant appears in the onset, it is said to be
complex
.
1. A nucleus must be present in every syllable.
2. When there is no vowel, one of a few select consonants may appear in this position: [m], [n], [l], and [r]. These are then called syllabic consonants.
Constraints on nuclei
Constraints on onsets
1. Liquids, glides, and affricates may never appear as the first sound of a complex onset.
2. If there are three consonants in the onset, they obey the following structure:
[s] + [p]/[t]/[k] + [l]/[r]/[w]/[j]
3. Two stops may never appear in the same onset.
Constraints on codas
1. Glides ([j] and [w]) never appear in codas.
2. Two obstruents in the coda must share the same value with respect to voicing.
3. [h] is never allowed in the coda.
English speakers are sensitive to these constraints on sound combinations. We can distinguish between sequences that are possible (though non-existent) words of English and sequences that are simply impossible in English.
Phonotactics and Accents
Other languages do not allow as many syllable types as English does. In Hawaiian and Indonesian, for instance, complex onsets and codas are impossible. Syllables in these languages are limited to the following:

Hawaiian
CV, V
Indonesian
CV, V, VC, CVC
English learners who speak these languages will often delete sounds when confronted with complex onsets/codas.
Other languages also have different constraints on which sounds may appear together.
Spanish
Does not allow [sp] or [sk] onsets. In Spanish words, these sounds are syllabified separately:
'escuela' = [es.kwe.la]
'espejo' = [es.pe.jo]
English learners from these language backgrounds may add sounds to 'correct' foreign combinations in English.
Spanish-speaking English learners, for instance, will add an [e] vowel to the beginnings of words like 'school' and 'speak'. This breaks the [sp]/[sk] sequences into two syllables:
[skul]
[spik]
[es.kul]
[es.pik]
Assimilation
There are certain changes that take during natural speech. When speaking fast, for instance, sounds can blend together or be omitted, and in careful speech, sounds may be inserted or emphasized.
Dissimilation
Insertion
Deletion
Strengthening
Weakening
Metathesis
Two sounds become more like each other in terms of their articulatory parameters.
Two sounds become less like each other in terms of their articulatory parameters.
A sound is inserted into the sequence.
A sound is deleted from the sequence.
A sound is made stronger (by increasing its sonority or by making it longer)
A sound is made weaker (by decreasing its sonority or by making it shorter)
The order of sounds is reversed or otherwise jumbled

Articulatory
ban
bannen
'summon, curse'
tele
phone

phon
ics

phon
etics

phon
ology

bhanati
'he speaks'
Middle English
Sanskrit
Old Norse
Old High German
Greek
Latin
banna
'prohibit'
phanai
'say'
fari
'speak'
Modern English
phone
'voice'
Germanic
Proto-Indo-European
bannan
'command'
bha-
'speak'
MRI video showing the vocal tract in action
Aharon Dolgopolsky
speaking reconstructed Eurasian language, circa 13,000 BCE
Old Norse

English
Germanic

Proto-Indo-European
circa 4,000 BCE
Sanskrit

Latin

Greek

Main Language Families of the World
Delve deeper
https://en.wikipedia.org/wiki/Language_family
Consonants and vowels have different articulatory and acoustic properties.
Articulatory properties
Consonants
involve considerable obstruction of the flow of air in the vocal tract. When we talk about how consonants are articulated, we can reference
where
in the mouth this obstruction takes place (place of articulation) or
how much
it is obstructed (manner of articulation). For
vowels
, on the other hand, there is almost no obstruction of the flow of air, so things like the 'place' or 'manner' of obstruction are irrelevant. Instead, we talk about what the tongue is doing when we produce a vowel: how
far back
is it? how
high
is it?
This means that we use different
parameters
to talk about consonants and vowels (listed to the right).
Consonants
Airstream
Voicing
Place
Manner
Vowels
Tongue height
Tongue advancement
Tenseness
Lip rounding
Acoustic properties
For more, see 'Consonants' (left)
'Vowels' (right)
In general,
vowels
are acoustically more prominent than consonants. Vowels are characterized by acoustic phenomena called
formants
, the 'peaks' in the speech stream.
Consonants
are always less prominent, and voiceless consonants, in particular, have a low acoustic profile.
Syllabic properties
Consonants and vowels also occupy different positions in the syllable.
Consonants
appear in the syllable onset or coda;
vowels
appear only in the nucleus.
For more on the syllable, see this section in Phonology
As mentioned before, consonants involve obstruction or constriction of the flow of air in the vocal tract. When we talk about these sounds, we make reference to four parameters: airstream mechanism, voicing, place, and manner.
*For more, see the Xhosa video under 'How do we talk about sounds?'
The
airstream mechanism
refers to the way airflow is created in the vocal tract. All English consonants are
pulmonic
. English consonants are also
egressive
, meaning they involve air being expelled from the lungs (ingressive sounds--when air is sucked in--are what we call
clicks
*)

Voicing
describes the activity of the vocal cords. It is a binary parameter, meaning it has only two values: voiced and voiceless. When the vocal cords are vibrating, a
voiced
consonant is produced; when they are not vibrating, a
voiceless
consonant is produced.
The
place of articulation
of a sound is where in the mouth the obstruction or constriction of air flow occurs.
Bilabial
: The two lips come together
Labiodental
: The lower lip and the top teeth come together
Interdental/Dental
: The tongue comes between the teeth
Alveolar
: The tongue touches the alveolar ridge, just behind the teeth
Postalveolar
: The tongue touches the soft palate further back in the mouth
Velar
: The tongue touches the back of the mouth
Glottal
: The mouth is neutral, and air escapes from the glottis
The
manner of articulation
of a sound describes just how much the airflow is obstructed.
Stop/Plosive
: The articulators come together completely, stopping the flow of air altogether
Nasal
: The velum is lowered, preventing air from resonanting in the oral cavity altogether. Air resonantes instead in the nose
Tap/Flap
: Similar to a stop, but shorter and weaker.
Fricative
: The articulators get close together, but they don't completely touch, causing frication
Approximant
: The vocal tract is narrowed at the place of articulation, but not enough to produce a turbulent air stream.
Lateral approximant
: An approximant produced when air flows over the sides of the tongue instead of down the middle.
Affricate
: The articulators come together and then release, creating almost a stop+fricative in short succession

Click here to watch (with captions) on YouTube (in new browser window)
MRI video showing the vocal tract in action
Action at the larynx
to control the opening
and closing of the vocal
cords.
Velum raising and'
lowering.
tongue
moving
lips opening
and closing
Air move up from the lungs
through the trachea.
nasal cavity
The classification of speech sounds
Transcribing speech using the IPA
Describing the articulatory, acoustic, and auditory properties of sound (how they are produced and perceived)
The Speech Chain
Activity
Answer the questions
in Part One (Sound and Meaning) in the Comprehension Activity.
Delve deeper
Folow the green arrow for a closer look at Indo-European. We'll return to this in a later unit.
Delve
Deeper
What we are looking at in this set of lessons is the phenomenon of speaking and hearing, or the SPEECH CHAIN of linguistic communication, which makes it possible for us to share concepts with others by means of intricate manipulations of the airstream flowing out of our lungs. Using our vocal cords, and lips, and tongue, and other parts of our mouth, we cause this airstream to create sound waves in a variety of precise patterns that travel rapidly through the air and cause a series of mechanical responses in the ears and brains of the people we're talking to, such that they can reconstruct the meanings of the words and sentences that are encoded in the patterns registered in the sound waves. Among other things, you should come away from this set of lessons with an appreciation of why it is that a linguist who studies the historical development of languages would see the connection between the words “ban” and “phone,” because of the way they are alike, in terms of both their meanings and their sound structures.
Transcript
for this step
Classic from AT&T
Activity
Answer the questions
in Part Two (Acoustic and Auditory) in the Comprehension Activity.
Activity
Answer the questions
in Part Three (TS Watt poem) in the Comprehension Activity.
Clicking on this image of the IPA will take you to a website where you can click on these symbols and hear how they are pronounced!
Click here!
Clicking on this image of the ear will take you to a website with a short reading about auditory perception. This reading covers the points listed here in more detail.
Click here!
Clicking on this image of the ear will take you to a website where you can learn more about the structures in the ear by clicking on them.
Click here!
Clicking on this image of the vocal tract will take you to a website where you can click on each articulator and learn more about it.
Click here!
Click here!
Clicking on this image will take you to an interactive website where you can select a consonant and see an image of how the vocal tract moves when you pronounce that sound.
Click here!
Just like in a wikipedia page, clicking on various terms in this image will take you to the corresponding wikipedia pages. If you hover over a linked term (like 'fundamental frequency') you will see that the color changes slightly, indicating that it is clickable.
Click here!
Answer the questions
in Part Five (Vowels) in the Comprehension Activity.
Consonants
What to focus on
The three videos you just watched discuss almost all the possible sounds of human languages.
You only need to focus on the subset of these sounds that appear in English
Vowels
Note on terminology:
The
website with the moving vocal tract diagrams uses the following terminology:
[r] and [l] are liquids, not approximants/lateral approximants
[w] and [j] are glides, not approximants
All of these terms are acceptable in this course; you can call them approximants or glides/liquids on homework and exams.
Note on the 'r':
The official IPA symbol for the English pronunciation of the 'r' sound is the upside down 'r', as shown in this chart. Often, for typographical convenience, we will simply use the regular 'r' in this course.
Activity
Answer the questions
in Part Four (Consonants) in the Comprehension Activity.
Activity
A key part of our study of phonetics is knowing how to do phonetic transcriptions, which are representations of words using symbols from the IPA.
Where possible, try to lengthen the sound you're interested in as you pronounce it (ie, 'paaaage', 'paaaa...')
The charts below represent a great go-to list for how the IPA symbols are pronounced. (Other great examples of the words, complete with video clips of native English speakers pronouncing the words, can be found by clicking this image)
Help with transcriptions
Rachel is a superstar internet English teacher who walks viewers through the pronunciation and transcription of many English words. Her explanations are very clear and linguistically accurate.
A sample of one of her videos is included here, but we urge you to check out her YouTube channel for more transcription examples. Just click on her logo above!
Clicking on Rachel's logo will take you to herYouTube channel, where you'll find TONS of GREAT examples of transcriptions!
Click here!
Delve deeper
This is on a somewhat tangential topic. But it's pretty interesting. This guy's a master of the larynx. Click on it to watch.
Watch on YouTube. Opens in an external window.
Clicking on this image will take you to an interactive website where you can select a vowel and see an image of how the vocal tract moves when you pronounce that sound.
Click on this image to go to the original wikipedia page and learn more about the airstream mechanism.
Click here!
The video talked about some places and manners of articulation that English doesn't make use of (such as pharyngeal, uvular, and trill). Since our focus in this class is on English, you should concentrate on those places and manners listed in the chart in this frame. This chart contains only the consonant sounds of English. For the most part, this chart is a copy of the official IPA chart we saw in the "How to talk about sounds" section earlier. But, we've made some important changes: 1. The sounds that are not used in English have been deleted. 2. The voiced bilabial approximant [w] has been added. 3. The palatal affricate sounds have been added. The three sounds we have added appear on the website with the moving vocal tract diagrams referenced earlier in this prezi. Clicking on the smaller image in this frame will take you back to that same web site if you would like to look again in more detail at the articulatory properties of English consonants.
Transcript
for this step
As the second video on vowels makes clear (about 3:00 in), the cardinal vowels are
idealizations of human vowel sounds. Languages have sounds that are approximations of these cardinal vowels, and these can vary. The high front vowel [i] in English, for instance, is slightly lower than the [i] in French (even though the same symbol is used). As it turns out, the vowel system of English, represented in this diagram looks rather different from the cardinal vowel chart because its vowels are sometimes rather distinct from these idealizations. You'll notice that the feature tense, discussed earlier in this prezi, is a key feature in the English vowel system. It distinguishes between many pairs of vowels (the high front vowels, the high back vowels, the mid front vowels, and the mid back vowels).
Transcript
for this step
The English [e] and [o] vowels are often pronounced as diphthongs.
Activity
Answer the questions
in Part Six (Phonotactics) in the Comprehension Activity.
The period indicates a syllable boundary
Activity
To see this, answer the first three questions of part 6 (Phonotactics) in the Comprehension activity.
Listen up!
Put your headphones on--there's audio with this step!
Listen up!
Put your headphones on--there's audio with this step!
Listen up!
Put your headphones on--there's audio with this step!
Click here!
The movement of the vocal cords
Visualization of a sound wave
Transcript
for this movie
The difference between a phoneme and a phone is really no different from the way humans categorize things in general. The word ‘shoe’, for instance, applies to all kinds of different physical objects: high-heeled shoes, tennis shoes, flip flops, and even those weird-looking shoes that fit each toe like a glove. Though physically different, we would look at any one of these things and say yes, that thing is a shoe. We would not, however, look at something like this, a sock, and say that it is a shoe. Similarly, we recognize something as the letter ‘a’ no matter what font it appears in. All of these symbols count as the letter ‘a’. We would not, however, suggest that this symbol is an ‘a’ –because it is, of course, a ‘b’. Even though there are physical differences between the a letters and the shoes, we can recognize them as belonging to some single, abstract category. And notice furthermore that this category is abstract. There is no single thing that is ‘the’ letter a or that is ‘the’ shoe; from these examples here, which shoe would we pick as the representative of shoe? And which font would we pick as the representative of the letter a? Picking between these choices would be entirely arbitrary—’letter a’ and ‘shoe’ are labels for a category, for a group of things. The same thing happens with speech sounds. For the word ‘lip’, we may hear a difference between a pronunciation with an unaspirated p, lip, and one with an aspirated p, lipH, but we recognize both sound sequences as communicating the word ‘lip’. Contrast this with ‘lib’. Here, suddenly, we’ve used a sound that does NOT count as a p sound. It is distinctly a b sound. So while aspirated and unaspirated p belong to some single category, the ‘b’ does not belong to this category. And we label this category with the abstract name ‘phoneme p’, represented as a p in angled brackets.
To conclude, let’s review the technical terminology that the rest of our study of phonological will build on. The phoneme is the abstract sound that exists in our head. Like ‘shoe’ or ‘the letter a’, it is a label for a category of sounds. The phone is the physical realization of a sound. One phoneme may be expressed by more than one phone, as with aspirated and unaspirated p or nasal and oral a.

Human perception
In phonetics, we learned about the different sounds of human languages and how they are produced. In the study of phonology, we turn from studying the physical properties of these sounds to looking at the way these sounds are perceived by the human brain. Our brain actually only hears a portion of all the sounds we produce. It groups similar-sounding phones (or sounds) together, interpreting them as instances of the same sound. So for instance, there is only a very subtle difference between the oral low back vowel [a] and the nasal low back vowel [a]. There is a physical difference: in the oral a, the air resonates in the oral cavity, a, but in the nasal sound, the air resonates in the nasal cavity, a. Alternating between the two a sounds I can tell that there is an acoustic difference, but my brain interprets both of these as variations of one and the same [a] sound. Similarly with the voiceless bilabial stop [p]. At the end of the word, I can pronounce this sound as ‘lip’, ‘lip’, or as ‘lipH’, ‘lipH’, with a slight puff of air at the end of the [p] sound. While there is clearly a difference between the [p] sounds in these different pronunciations, my brain registers both as one and the same [p].
In other words, while my mouth produces two subtle variations of the [a] or the [p], the oral [a] and the nasal [a] and the aspirated [pH] and the unaspirated [p], my brain registers both variants as the same sound. The sounds I actually produce are called phones, and the sound my brain hears, the sounds that exist only in my head, are called phonemes.
To understand exactly what is going on here, it is important to know a little about how humans perceive the world around them. Interestingly, how we experience the world is only partially determined by the actual physical stimuli in the world. Our experience is in large part determined by our brain. To see this, consider these two images. For each, there is only one physical stimulus: there is only one image. But, our brain can ‘see’ each of these images in one of two ways. The image on the left can either by seen as two faces or as a vase. The image on the right can either be seen as the profile of an old lady or as an image of a young woman with her head turned away. In both cases, there is only one physical stimulus, but two images are perceived because of the role the brain plays in interpreting the physical stimuli we receive from the world.
This process can go the other way. Just like we can produce two mental states from one physical stimulus, we may also produce only one mental state from two different stimuli. That is, our brains may ‘trick’ us into perceiving two stimuli that are actually physically distinct as the same thing. For instance, this shade of blue and this shade of blue may look like the same color when seen separately, but shown next to eah other, we can see that they are two distinct stimuli. This same phenomenon happens with speech sounds. While we can hear the difference between an oral and nasal [a] when they are produced one after the other, we cannot perceive this difference when these sounds are produced as part of a chain of natural human speech. The two physical sounds are called phones, and the single sound we perceive is called the phoneme
To understand the concept of phonemes, it is important to first understand a little bit about how humans perceive and categorize the world. This is the topic of this video.
Phonemes are the sounds our brains 'hear'. This is a relatively small subset of all the sounds (phones) we actually produce. To determine whether two sounds are phonemes or phones, we use minimal pairs, the topic of this video.
We’ve seen that our brain may interpret two phones, two distinct physical sounds, as instances of one and the same sound, called a phoneme. Now we’ll look at how we can determine which phones are interpreted as the same phoneme in a language. Languages group phones in different ways. For example, while in English, the oral a and nasal a belong to the same phoneme a, for French speakers, these two sounds are always perceived as distinct. For French speakers, in other words, the oral a and nasal a are as distinct as the p and b sounds in English—French speakers never hear these two sounds as the same like English speakers do. Similarly, aspirated and unaspirated p are perceived as the same sound in English, but in Thai, these sounds are always distinct. In Thai, the unaspirated p is a phoneme and the aspirated p is a phoneme. So how do we know that this is the case?
Determining which sounds represent distinct phonemes in a language is one of the key tasks in the study of phonology. To do this, we look for minimal pairs. Minimal pairs are two words that (a) mean different things and (b) are phonetically identical with the exception of one sound. This means that they have the same number of sounds and those sounds are in the same order, the only difference is one sound. Dad and bad is an example: the words mean different things, and they are phonetically identical: both have three sounds in the same order, the only difference is the first sound. These first sounds, then, represent different phonemes: the d phoneme and the b phoneme.
Transcript
for this video
Let’s look at another example in more detail. Consider the phonetic sequences lit and lip. Are these a minimal pair? Well, these words both have the same number of sounds, three, and these sounds are in the same order: the l comes first and then the vowel, i. One sound is different: at the end of the word is either a t or a p. And, these words mean different things: the first sequence is the word ‘lit’, as in I lit the candle, and the second is lip, as in the upper lip on my face. This means this is a minimal pair, which in turn means that the sounds that are distinct in these words, the t and the p, are different phonemes.
Take one more example: lip and lipH. Are these a minimal pair? They do have the same number of sounds, three, these sounds are in the same order, and one sound, the final p sound, is different: in the first it is unaspirated, and in the second it is aspirated. BUT these words do not mean different things: both are ways of saying the word ‘lip’. They are not, then, a minimal pair. This means that the two different p sounds are representative of the same phoneme, as we have seen.
A word of caution is in order. When you are trying to find minimal pairs, it is important that you consider the phonetic transcriptions of the words you are looking at. You can’t tell if something is a minimal pair just by looking at how it is spelled. The words reed and bead, for instance, don’t look like a minimal pair for the r and b sounds because the spelling of the vowel sound is different. But they actually do form a minimal pair: when we transcribe them, we see clearly that the vowel sound is the same, I, and hence, the words must be a minimal pair.
We say in phonology that phonemes are the sounds of a language that can bear a meaning difference. Words that form minimal pairs are identical with the exception of one sound and yet they have different meanings. We say, then, that this meaning difference comes about because of the use of that different sound.
If two sounds don’t differentiate meaning, there is no reason to care that they are different sounds. Over time, we stop hearing them as distinct sounds. This is the case with the two versions of p in lip, lip and lipH: these two p’s don’t affect the meaning of this or ANY word in English, so our brains slowly stop paying attention to the acoustic difference between them.
As I mentioned in the beginning of this video, languages vary in what they count as phonemes. For English, the aspiration on the p sound is irrelevant: there are no minimal pairs with these sounds. But for Thai, this difference is highly important. In Thai, we can identify minimal pairs with aspirated and unaspirated p, as this example shows.
Of course, it makes sense that phonemes should be different in other languages. Languages differ dramatically in the ways they categorize the world. These two objects, for instance, have different names in English: armchair and sofa. In Portuguese, however, both of these items are called a sofa. Similarly, these three creatures have different names in English: rat, mouse and gopher. But in Japanese, a rat and a mouse are members of the same category, and a gopher is part of a separate category. The labels languages and cultures give to objects in the world can vary, and there’s no reason this can’t extend to language. The labels languages and cultures give to objects in the world can vary, and there’s no reason this can’t extend to language. Just like we have different rodents, we have different p sounds, for instance, and while English groups them together, calling them the phoneme p, a language like Thai keeps them separate, calling them the phonemes p unaspirated and p aspirated.
Our brain stores words based on the phonemes it takes to produce them. This is a matter of efficiency, because, as we know, each phoneme is associated with multiple phones. If words were stored based on their phonetics, we would would end up with multiple representations of each word to accommodate the various phones that could be used to pronounce the word. The blueprint for the sounds of words, then, are phonological structures. Of course, phonemes are things that only exist in the brain—they are abstract (just like there is no single thing in the world that represents ‘shoe’). Phonological structures, too, only exist in the brain. This means that when we actually pronounce a word, we produce a phonetic representation of the word. And since there are sometimes many phones for each of the word’s phonemes, there may be many ways to pronounce the word.
The pronunciations of words may vary from speaker to speaker or even for the same speaker at different times. Speakers from different dialects of a language may pronounce the same word differently, and speakers who are learning a language, whether as their first or second language, may pronounce a word differently from fluent speakers. A single speaker may pronounce a word differently at different times if he is speaking especially fast or slow.
Transcript
for this step
When two sounds in the phonological structure assimilate in the phonetic structure, they become more like each other. This happens often in fast speech. When asking someone if they want to come to the movies with you, for instance, the t sounds at the end of want and the beginning of to blend together, becoming an n sound. This is exactly like the sound preceding it, so we say that assimilation has taken place. Dissimilation represents the reverse: when two sounds that are next to each other become less like each other when they are pronounced. This happens for instance in Greek: two stop sounds one after the other in the phonological structure are often pronounced as a sequence of a fricative and a stop when speaking quickly. The Greek word for seven, epta, becomes efta, for instance, and the word for building, ktizma, becomes xtizma. In both of these words, two sounds that have the same manner of articulation are pronounced as a sequence of sounds with two different manners of articulation. This is an example of two sounds becoming less like each other, which is dissimilation.
An entire sound may be inserted into a phonological structure when it is pronounced. This often happens in English when there is a sequence of a nasal consonant followed by a voiceless stop. In strength, for instance, a k sound appears that isn’t present in the phonological representation, and in hamster, a p sound is inserted. As you might expect, the opposite can also happen: a sound may be deleted in the pronunciation of a word. This happens all over the place in fast speech, and especially with the voiceless glottal fricative sound, h. The most natural pronunciation of ‘I like her’ for instance, has the h deleted, I like ‘er.
The order of sounds in a phonological representation may also become jumbled; this is called metathesis. Children learning English, for instance, are prone to pronounce things like spaghetti or animal with some of the consonant sounds reordered, pskedi and aminal. Strengthening is another phonological process; when this takes place, a sound from the phonological structure is made louder or longer so as to increase its acoustic ‘strength’. Aspiration is one example of a strengthening process in English. Aspiration adds a puff of air to a voiceless stop like p, t, or k: as voiceless stops, these sounds are usually very short and quiet. The puff of air increases their duration and thus makes them relatively ‘stronger’ and easier to perceive. The reverse may also happen: a sound from the phonological structure may be made weaker. The flapping of the voiceless stop t, which you’ll see in discussion section, is one such example.
As I mentioned before, speakers of different dialects may pronounce words differently. Usually, however, we don’t assume that this variation is the result of pronouncing a certain phonological structure in a different way. Instead, linguists assume that these speakers have different phonological representations of these words in their brains. But, we can still compare these phonological processes and see that the processes we’ve just discussed are taking place. This isn’t to suggest that one dialect’s phonological structure is ‘right’ while the other’s is ‘wrong’—they are just different. We’ll talk much more about dialectal differences later in this class, and these terms will be important when we get there.
When we talk about these different pronunciations, we usually assume that there are different phonetic structures for the same phonological word. The phonetic realizations of phonological words vary in systematic ways. In particular, we talk about seven phonological processes: assimilation, dissimilation, insertion, deletion, metathesis, strengthening, and weakening.
Activity
Answer the questions
in Part Seven (Phonemes) in the Comprehension Activity.
Activity
Answer the questions
in Part Eight (Phonological Processes) in the Comprehension Activity.
Click here to watch (with captions) on YouTube (in new browser window)
Transcript


The vocal cords (or vocal folds), which are housed in the larynx can be manipulated in a variety of ways that are important in the generation of speech. Most importantly they can be made to vibrate rapidly, due to being drawn closely together so as to close off the airflow and then being forced slightly apart due to the build up of air pressure beneath them. They then open slightly and close again over and over again and thus vibrate, in much the same way that the reed on a clarinet or saxophone does when a musician forces air through the small opening between the tip of the reed and the edge of the mouthpiece.
This vibration of the vocals cords cause the voicing or phonation that is responsible for the sound in speech. Sound waves are generated in the larynx and are

We talk about things moving at the speed of sound. We think of an echo as a sound that goes off into the distance and then circles back to us. But, what is it that's moving when sound TRAVELS. We talk about speech as if there's some substance that goes from one person to another. In their classic book "Metaphors we Live by" George Lakoff and Mark Johnson claim that our concept of language is shaped by what they call the conduit metaphor: We put our concepts into words, pack lots of thoughts into fewer words, and other people get them from us, or let them go in one ear and out the other.

But, in fact there is no physical substance that is transferred through speech. The physical phenomenon is simply a wave phenomenon. Particles of air are affected by our vibrating vocal cords. The way in which they are affected is that they come closer together and then move apart quickly and when that happens the particles of air that are next to them do the same thing and there is a ripple effect, kind of like with a slinky. The particles of air are the substance through which a wave is manifested. When I speak, the air around me doesn't actually travel to your ear. Rather, patterns of air pressure, or moments of compression of particles followed by moments of rarefaction or expansion are replicated in the surrounding particles of air and this pattern is replicated, in chainlike fashions, literally at the speed of sound, forming a longitudinal wave. So, the particles of air in your eardrum compress and rarefy at the same speed and in the same complex pattern that occurs at the edges of my lips. Remarkably, the pattern is maintained as air wave gets transformed into a fluid wave when it gets to your inner ear. The video clip in the next step gives you a brief everything you need to know about what goes on inside the ear such that we are able to reconstruct the configurations and movements in the vocal tract of the speaker we are listening to, based almost entirely on the nature of fluctuations in air pressure on the tiny membrane that we call the ear drum.
Full transcript