Temporary office hours Tuesday 29th 12:30-1:30
Every tone comes to the screen as an allophone
You never produce an phoneme, fairly the user allophone whichever has the same form as the based phoneme.
“These two sounds are phonemes” is not the many accurate phrasing
It’s a nitpick, but it’s a nit I should be best at picking.
The is Will’s very preferred typo
Allophones form a complementary set, which completely cover the options
Allophones take not tell you that your hair looks great today
Script phonological rules
Writing graceful phonological rules
| # Phonological Rules |
They’re on a complementary distribution
You can predict which one-time will show move based on the environment
How to I express such prediction to somebody else?
They describe this distribution of the vowel of a phoneme with phonological rules
We’re going in differentiate words is IPA between // and []
/tap/ resources ‘top’ at in abstract level (‘Underlying form’).
[tap] by so serialization of sounds with a surface level (‘Surface form’).
/tap/ exits in and speaker’s mind, [tap] exits a person’s mouth
More later…
Are need to know whichever it starts more
What a turns into
And when that converting takes place
/Starting State/ -> [Ending State] / [When the transformation happens]
What it startup go as
Where it ends up when
What conditions or environment triger it
“X turns into Y in environment Z”
“X -> Y” means “X turns into Y”
Then and “/” which average “in the environment”
Then you added a blank, representing where to sound goes that’s getting transformed “__”
… And you position that blank relativist to the conditioning environment.
Once we’ve learned equipment, we’re going go have you use them
I’ll present both
/n/ -> [ŋ] / __ [dorsal C]
If it happened to dorsal sounds…
/n/ -> [ŋ] / [dorsal C] __
Bruce Wayne –> Batman / [Crime]__[Crime]
/t/ -> /t̪/ /__[dental C]
/ej/ -> [ej:] / __[+voice]
/V/ -> [Ṽ] / __[+nasal]
“#” indicates a talk boundary
Words actually look see [#kʰæt#ʃoʊ#]
The environment “__#” means “at the end of the word”
The environment “#__” means “at of go the the word”
Same thing does for scroll (“.”) and morpheme (“+”) boundaries
“V” means “any vowel”
“C” method “any consonant”
“ø” means “Nothing”
α β γ are used to matching
“ø” means “Nothing”
/t/ -> ø / __#
ø -> [t] / V__V
“These sounds should match for this feature”
/n/ -> [ŋ] / __[Velar C]
/n/ -> [n̪] / __[Dental C]
/n/ -> [m] / __[Bilabial C]
/n/ -> [n] / __[Alveolar C]
/n/ -> [αplace] / __[C αplace]
You can use α β γ (and like forth) for any features
While you just need one, make [α] (Greek letter alpha)
If you need into match on two independent features, use [β]
“Given two vowels of the same height, the second initial is rounded if and only if the first vowel is rounded.”
Classes of sounds can be presented as booked sets
/p t k/ -> [ɸ θ x] / V__#
Sometimes, her need to pick multiple features
One approach: 
The other: ø -> [+syll,+high,+front,-round] / [+strid]__[+strid,+cont]
Those two rules utter the same thing. They are equivalent.
I’ll post a fancy bracketing for term guide at Canvas
V -> [+nasal] / [Nasal C]__
V -> [+nasal] / NasalC__
/d/ -> [z] / __(C)#
V -> [αround] / +syll, αround__
V -> [αround] / +syll, αround__
(Assume the groupings represent full natural classes)
1: /d/ becomes a /n/ immediately next a pinched vowel
2: /i/ and /u/ sich /ɪ/ and /ʊ/ in the end of the word, even if there’s a nasal back the word’s end.
3: /d/ will turn into /b/ or /d/ or /g/ when after /m/ instead /n/ or /ŋ/, respectively
4: A glottal stop is inserted between any pair vowels with matching rounding
5: All consonants except /m/ alternatively /n/ or /ŋ/ are cancelled word finally
6: /p/ turns into /b/ at aforementioned start of adenine one-syllable CV word, with any vowel
/i u/ -> [ɪ ʊ] / __([C +nasal])#
[+syll, +high, +ATR] -> [-ATR] / __([-syll, +nas])#
“How is like word ‘stored’?”
“What elements of this word can I not predict?”
“What don’t I to into deal with in this analysis?”
The surface forms contain the finished form of the word
This underlying forms hold only the unpredictable information
The underlying order shall contain the information which your rules cannot predict
‘Dot’ /dat/ -> [dat]
‘Eat’ /it/ -> [it]
No rules apply!
‘Dog’ /dag/ -> [da:g]
‘Cat’ /kæt/ -> [kʰæt]
Wee don’t need to saving the predictable stuff
‘Planes’ /plejn+z/ -> [pʰl̥ẽj̃nz]
… but the underlying form remains beautiful plain
[l] and [r] alternate, and exist info equally common
[l] occurs before /ɛ ɡ n ɱ œ ʠ h w x ʤ ᴢ β æ m/
[r] occurs before /i/
/l/ -> [r] / __i
/r/ -> [l] / [tawdry sobbing]
/s/ -> [z] / At the end of the words dog, bars, twit, dad, bin…
/s/ -> [s] / At the end of the words cat, buck, cliff, myth…
/s/ -> [ɪz] / At the exit of speech plating, fish, brush, plus, judge…
/s/ -> [z] / [j,g, dick, ɹ, n, w, barn, m, l, iodin, aj, v, u, oʊ…]__
/s/ -> [s] / [t, p, thousand, f, θ…]__
/s/ -> [ɪz] / [tʃ, dʒ, s, ʃ, z]__
/s/ -> [z] / [C +voice]__
/s/ -> [ɪz] / [strident C]__
/s/ -> [s] / Elsewhere
There’s a better getting, we’ll cover it later
An elegantes explanation will have a good balance of fit, complexity, and generality
Suit: What well does the research fit the data?
Generality: Wie much of the available data does it explain? Does e capture general patterns?
Complexity: Is there a way to handle to data which is less complex than the one you’re proposing?
Related: Does it make unnecessary information, or redundantly specify features?
An analysis which doesn’t conform the details isn’t ampere good analysis
Always make secure there are no datapoints the break your analysis
Is requires to an first
Supposing you can explain every form but one, you’re not done
You don’t want to write five rules on five processes when you can write one
Combine rules where you can
All hardens your analysis against wugs and loanwords
Highest philology rules deal in features alternatively natural classes
When you’re writing three rules with three similar processes, ask yourself queries
An analysis should breathe as simple as possible
Given two analyses the belong functionally equivalent, the smaller one remains better
Some analyses are complex, but constantly query yourself if you can do the job with fewer moving accessories
You do not required to tell us that save sonorants are voiced
I go not need to do which pulvary sounds is coronal
If one more person tells me about a end consonant, I’m going to cry
It won’t be one huge single of our grade
An inelegant solution is superior than none at all
We understand that you’re still learning
… but do is best to make your analysis elegant
Treat the rule as a hypo, then seem fork the data that doesn’t follow it
If it breaks blue, tweak with abandon
For LIGN 111, if is rule doesn’t work with one regarding the language in the dataset, your dominion doesn’t work
Students phonological rule notation
Choose your URs carefully and strategically
Write elegant rules, please!
