LING 210 - Phonemes and Allophones

which brackets do phonetics use and which brackets do phonemics use?

[ ] → phonetics

/ / → phonemics

what are the two levels of representation?

concrete and abstract

_________ are abstract mental representations of sound.

phonemes

WHY does the English phoneme /p/ cover [p], [p̚], and [pʰ]?

because /p/ treats these other sounds as the same underlying unit as it doesn’t change the meaning of the word in the English language.

T or F: phonetics are concrete

true

T or F: phonetics are abstract

false, they’re concrete

T or F: phonemics are concrete

false, they are abstract

T or F: phonemics are abstract

true

the concrete level of representation covers ______ _______ material

surface phonetic

the abstract level of representation covers ________ _________ material

underlying phonetic

the concrete level relates to allophones or phonemes?

allophones

the abstract level relates to allophones or phonemes?

phonemes

what is “the underlying organizing system that dictates how a phoneme will be phonetically realized?”

The phonological representation. → abstract

_______ are abstract phonological unites underlying the surface sounds of a language

phonemes.

T or F: phonemes are pronounceable by itself

false, they are abstract and cannot be pronounced by themselves without context. They are simply the symbol for the phonetic sounds that underly the spoken language.

if you change a phoneme, you change what?

you change the meaning → phonemes carry meaning

T or F: every language has its own inventory of phonemes

True.

are p and b phonemes or allophones in English?

phonemes, as the different letters (phonemes) change the meaning of the word.

are p and pʰ allophones or phonemes in English?

they are allophones, as the difference in the letters don’t change the meaning in English…they might in different languages though.

________ are the predictable, concrete manifestations of phonemes.

allophones.

T or F: allophones are phonetically different

true, they have different phonetic properties, but they have the same underlying phoneme

how are allophones “different sounds” but have the same underlying phoneme?

for example, [p] and [pʰ] make different sounds but they still mean the same thing in English.

why might native speakers of a language treat allophones as the same sound?

because they often can’t even hear the phonetic difference…they’re used to what words sound like and don’t necessarily think twice about it as these phonetic differences are non-meaningful to the word spoken.

whats the difference between minimal pairs and contrastive distribution?

• Minimal Pairs: Focus on specific word pairs as concrete examples to demonstrate that two sounds are phonemically distinct. They provide direct evidence of contrastive distribution but are not required for it.

• Contrastive Distribution: Describes a more general relationship between two sounds across a language. It doesn’t rely on specific pairs and can be established without finding minimal pairs.

what are minimal pairs?

sets of words which differ by just one sound, or one articulatory component, but which are different lexical items (have different meanings)

are minimal pairs phonemic?

yes, because since they differ by just one sound, they create distinct meanings.

ex: ‘mat’ and ‘bat’ → [m] and [b] were the only sounds to change, but they have very different meanings.

what does it mean when you say [b] and [m] contrast?

that they are allophones of separate phonemes /b/ and /m/

T or F: there’s no such thing as minimal pairs that aren’t pure.

false, there are “near minimal pairs” which may find two sounds in identical immediate phonetic environments, but don’t necessarily use the same lexical items.

what are near minimal pairs?

when we do not find 2 lexical items differing by only one speech sound, we may find the 2 sounds in an identical immediate phonetic environment.

ex: mission and vision

[mɪʃən] and [vɪʒən]

what if you can’t find a minimal pair that differs by ONLY ONE speech sound?

you have near minimal pairs

are mission and vision minimal pairs or near minimal pairs?

These words are near-minimal pairs because they differ primarily in the initial consonants /m/ and /v/ but also have a slight difference in the medial consonants /ʃ/ (in mission) and /ʒ/ (in vision). Although they differ in more than one sound, these differences are still close enough to illustrate the phonemic contrast between /m/ and /v/ at the beginning.

what are some other classifications that show different types of contrast?

• place of articulation

  • /pɪn/ ‘pin’ /tɪn/ ‘tin’ /kɪn/ ‘kin’

• state of vocal cords

  • /pɪn/ ‘pin’ /bɪn/ ‘bin’

    Hindi: ‘rice’ bhāt /b̤aːt/ ‘talk’ bāt /baːt/

• vowel nasalization

  • French bas /bɑ/ ‘low’ banc /bɑ̃ / ‘bench’

• stress

  • Spanish paso ‘step, I pass’ pasó ‘he/she passed

• tone

  • Kono /kͻ́ͻ́/ ‘to mature’ /kɔ̀ɔ̀/ ‘rice’

• vowel length

  • Kinyarwanda /guˈsiβaʰ/ ‘to erase’ /guˈsi:βaʰ/ ‘to be absent’

what is contrastive distribution?

when 2 sounds can occur in the same environment but yield different meanings

contrastive distribution characterizes _________, while complementary distribution characterizes ________

contrastive → phonemes

complementary → allophones

whats the difference between contrastive and complementary distribution?

• contrastive

  • allows 2 sounds to occur in the same environment

  • phonemes

  • creates two words with different meanings

• complementary:

  • do not allow 2 sounds to occur in the same environment

  • allophones

  • predictable according to the environment

how come allophones are predictable and phonemes not?

since phonemes can occur in the same environment, we can’t predict which sound occurs in the same spot (ex: /_ɛt/ → could be p, b, d)

since allophones don’t occur in the same environment, it’s easier to predict which sound will be in the word (ex: [_ɛt] → [pʰ] as aspirated stops occur word-initially)

what is free variation?

when 2 sounds can occur in the same environment but yield the same lexical item

• its an optional variation that does not affect meaning

• ex: [mæt] and [mæt̚] can have different allophones and occur in the same environment with the same meaning

what are the 3 ways speech sounds are distributed that ultimately affect their meaning?

• contrastive distribution

• complementary distribution

• free variation

free variation distinguishes allophones or phonemes?

allophones

complementary distribution distinguishes allophones or phonemes?

allophones

concrete distribution distinguishes allophones or phonemes?

phonemes

how do we link the two levels of representation?

with rules!

what is the format used for rules?

A → B / X_Y

(A turns into B when in this situation)

why is the following not the best interpretation of rules?

underlying phonemic form + phonological rules = surface phonetic form

its an easy way to understand as it's interpreted as the og form + rules = new phonetic form, but having a model that represents a relationship (rather than a product) shows less of a process for production or interpretation of speech sounds, and more of a model of a relationship between the cognitive organization of speech sound and its production.

what will the rule for the following be? /i/ → [ĩ] / _/m/

[+syllabic] → [+nasal] / _[+nasal]

(any syllabic segment becomes nasalized if it occurs before a nasal segment)

what are the 2 levels of representation?

[concrete] → phonetic, and

/abstract/ → phonological

provide an example of the following rule:

[+ syllabic] → [+ nasal] / _ [+ nasal]

/o/ → [õw] / _/n/

what are the 7 heuristics/rules of thumb for choosing the underlying representation for a rule?

1. be “phonetically natural” or similar

2. be the same as one of the allophones

3. be the “simplest”

4. be the form with the widest distribution

5. capture native speaker intuition

6. represent a natural process

7. fit the wider phonological pattern of the language

what does “phonetically natural” mean in regard to choosing the underlying form?

you want to choose the phoneme that is most similar to the allophones

ex: /p/ would be the best option as its phonetically similar to [p] [pʰ] and [p̚]

what does ‘the same as one of the allophones’ mean in regard to choosing the underlying form?

you want to use an underlying form that is also one of the allophones used in the rule

ex: /p/ would be best as the allophones in a rule might be [p]…

what does “simplest” mean in regard to choosing the underlying form?

you want to choose the most basic phonetic form for the underlying form (so no diacritics present) so its the best representation

ex: /p/ is the best form for allophones [p] [pʰ] and [p̚], as there’s no diacritics present and its the simplest form of the three allophones

what does ‘the form with the widest distribution’ mean in regard to choosing the underlying form?

you want to choose an underlying representation that is present in the most environments (out of the other phonemes you’re looking at). this way it tells us that its more common and is the better option of the two to choose as the underlying form.

ex: if [p] is in 5 environments, and [b] only in 2, /p/ would be the best option as an underlying form.

• you could then just make the rule: -vc bilabial stop → +vs / …

what does ‘captures native speaker intuition’ mean in regard to choosing the underlying form?

you want to choose the form that comes most naturally to the native speaker.

ex: if the native speaker articulates a /p/ w no aspiration majority of the time, you probably wouldn’t have aspiration in the underlying form

what does ‘represents a natural process’ mean in regard to choosing the underlying form?

you want to choose the underlying form that represents what commonly happens in such environments

ex: assimilation, substitution, aspiration, etc.

what does ‘fit the wider phonological pattern of the language’ mean in regard to choosing the underlying form?

you want to choose the underlying form that best represents the phonological patterns of the language at hand.

ex: you wouldn’t say /pʰ/ is the underlying form for english when /p/ is only aspirated when at the beginning of a syllable or word. in addition, its not aspirated when beside an s. therefore, you would choose /p/

is /p/ the best underlying representation for [p] [pʰ] and [p̚]?

YES, because:

1. its “phonetically natural”? yes

2. it matches an allophone? yes

3. its in its “simplest” form? yes

4. its the widest distribution? we dont know

5. it matches native speaker output? we don’t know

6. it represents a natural process? idk

7. its the widest phonological pattern in the language? probably