test questions (OT) midterm Flashcards
[I give you a filled-out tableau.] What is the optimal candidate in this tableau and why?
look at which one has the least violation marks in the higher constraints
[I give you a filled-out tableau and a specific candidate.] Is there a way to re-rank these constraints so that this candidate wins?
should be self explanatory
[I give you a tableau that doesn’t have the violation marks, but I also give you constraints and their definitions.] Explain how these candidates violate or don’t violate each constraint.
should be self explanatory
[I give you a filled-out tableau.] Is there some other candidate not included on this tableau that would have been better than any of the choices shown here?
think about a candidate that wouldn’t break any of the constraints already broken by the top candidate
[I give you a schematic constraint ranking.] Explain what phonological pattern you would expect to arise from this constraint ranking and why.
Full Contrast:
F»M(c-s)»M(c-f)
F»M(c-f)»M(c-s)
Pos. Neutralization:
M(c-s)»F»M(c-f)
Allophony:
M(c-s)»M(c-f)»F
No Variation (or complete neutralization)
M(c-f)»F»M(c-s)
M(c-f)»M(c-s)»F
explain (with example) what no Variation (or complete neutralization) is and whats its schematic ranking would be. explain the schematic ranking
when you never see a certain segment surface.
CFM»CSM»F
CFM»F»CSM
no matter what is ranked below, if CFM is ranked first whatever segment is incurring a violation will never surface.
ex. if we have *NasalVowel as CFM which says assign a violation every time there is a nasal vowel, we will never see a nasal vowel surface.
explain (with example) what allophony is and whats its schematic ranking would be. explain the schematic ranking
allophony is when there are segments in complementary distribution, aka it is predictable because the sounds never occur in the same enviornment.
CSM»CFM»F
csm needs to be ranked highest because allophony is based on sounds occuring in specific enviornments. To account for these specific environments in place, segments often change from UR to SR to fit these environment requirements, therefore are often violating faithfulness.
explain (with example) what positional neutralization is and whats its schematic ranking would be. explain the schematic ranking
positional neutralization is when two sounds are mostly unpredictable (contrast) but are predictable (neutralize to context B) in one environment.
CSM»F»CFM
we need CSM ranked highest because without it, the sounds would just contrast. we need that one specific enviornment where the sounds are predictable.
explain (with example) what full contrast is and whats its schematic ranking would be. explain the schematic ranking
full contrast is when the sounds are fully unpredictable and are contrasting.
F»CSM»CFM
F»CFM»CSM
faithfulness is first no matter the other rankings because it is always faithful to the input. nothing is changing in terms of creating predictable environments in certain places.
ex. /cat/ and /bat/ in english are minimal pairs and are fully contrastive
[I give you description of a phonological pattern, something like “In Language X, voiced and voiceless stops can both occur word initially, but only voiceless stops can occur after [s].”] Explain what kind of ranking of standard OT constraints you would need to account for this pattern.
name the standard constraints
think back to factorial typology
rank them according to whether the pattern described is contrast, positional or full neutralization, or allophony
-in this example its describing allophonic variation, so the ranking would be
CSM»CFM»F
[I give you a set of phonological constraints for one of the datasets on the handout, like Yowlumne, Mokilese, or Black ASL.] Explain what the correct ranking of constraints would be for this dataset and why.
look for patterns of alllophony, full contrast, pos. neutralization, and no variation (full neutralization)
figure out ranking accordingly
Explain why it is that unpredictable information is stored in underlying forms, using at least one concrete example
For example, given we know that /pæt/ and /bæt/ (a minimal pair) exist in English, if only given the sequence /æt/ we can’t predict whether the first sound will be [p] or [b] because these sounds are contrastive in English.
we must include every segment of /pæt/ in the UR because there’s no way to tell otherwise what it will surface as.
In contrast, if needing to choose between [p] vs [ph] for the sequence /æt/ we do have markedness constraints making one candidate better than the other because these sounds are predictable in english.
Explain why putting faithfulness at the top of a constraint ranking will allow for surface forms that show phonological contrasts.
High-ranking faithfulness constraints ensure that distinctive phonological features remain intact. For example, if /b/ and /p/ are underlying forms, ranking faithfulness above markedness will maintain their contrast in the surface form, leading to outputs like [bɪg] and [pɪg]. If faithfulness was not ranked first these sounds may not contrast, and instead show positional neutralization.
- avoids positional neutralization of phonological contrasts (e.g., merging /b/ and /p/ into a single sound) are avoided.
Explain how you decide what the underlying form of a morpheme is when doing an Optimality Theoretic analysis.
-lexical optimization:
assume input form is the same as output form if you don’t have contradicting evidence
OR if that doesn’t work:
based on which sound is predictable vs which one is not
-the unpredictable one will be the sound that has the harder environment to describe, ie. it’s the “else” environment. this will be UR because the UR stores unpredictable information
Explain why the sequence CVCV will be predicted to be syllabified as CV.CV and not CVC.V even in a language that does allow CVC syllables
Principle of onset maximisation, which reinforces the universal syllable structure of CV.
*look at week 6/7 syllable ranking chart. it occurs the least amount of violations
[I give you a sequence of consonants and vowels, like CVCCVC.] Explain how this sequence would be syllabified and why.
look at 6/7 syllable ranking chart
it seems 2 diff syllabic versions have the same amount of violations, maybe its dependent on language
[I give you the dataset for Yowlumne or Attic Greek.] Explain what phonological process this language employs to guarantee ‘good’ syllabification and how you know, using at least one concrete example.
Describe the largest syllable shape
then describe how Yowlumne epenthesis and Attic Greek deletes
Explain one phenomenon that an OT-style approach to phonology can better handle than a rule-based approach. [Note: we briefly touched on that today, but we’ll come back to it in later classes, so don’t worry too much if you’re not yet sure how to answer this question.]
OT style can make cross-linguistic predictions (typological predictions, ie. what is possible vs what is not possible in a language)
eg. Syllable structure predictions
every language has CV structure
if a lang allows complex onsets then it allows simple onsets
If lang allows complex codas, then it allows simple codas
Rule based makes predictions for only language specific phenomena
Explain some of the key similarities and differences between a rule-based and a constraint-based approach to doing phonology.
similarities:
both use phonological features
both describe how to get from UR to SR
differences:
rules are lang specific
temporally ordered (ie. they apply one at a time in a specific order)
rules cannot be violated
SR is a consequence of the rules
OT
constraints are language universals; langs just rank their importance differently
no temporal ordering of constraints application, all apply at once
constraints will be violated
SR is a goal orientated approach. If we want the SR to have certain qualities, we must rank the constraints accordingly
Explain the difference between Faithfulness and Markedness constraints.
[I give you an example of a constraint and its Kager-style definition.] Re-define this constraint to be clearer and explain why the new version is clearer.
markedness looks at the output only and seeks to make things match the constraint
faithfulness compare the input with the output and resists any changes in order to keep the input and output as identical as possible
[I give you an input and an output form.] Explain which elements are in correspondence between these two forms and what has changed between them.
look at the subscript numbers
it could be deletion
change in linear order
maybe there was a voicing change
[I give you one of the languages in #4, #6, or #8 on the Week 6-7 handout.] Explain what kind of syllable structure this language allows and what phonological process (if any) it uses to enforce that syllable structure.
Ex. #4 allows CV as maximum syllable type, this is achieved through epenthesis
/pao/ becomes [pa.to]
[I give you one of the languages as described in #4, #5, #6, #7, #8, or #9 on the Week 6-7 handout.] Explain what the constraint ranking of Onset, NoCoda, Dep-IO, and Max-IO would have to be to get a language like this.
For the language in #4 on the handout, the ranking would be Onset, NoCoda, Max-IO»_space; Dep-IO
This is because the language requires epenthesis to maintain a CV syllable structure, so Dep-IO has to be ranked last
[I give you a particular constraint ranking for the four constraints Onset, NoCoda, Dep-IO, and Max-IO.] Explain what type of language this would end up giving you (in terms of its syllable structure and phonological processes used to enforce that syllable structure).
taken from example 7, Onset, DEP-IO»_space; Max-IO»_space; No coda would get us the cvc structure via deletion rather than epenthesis since it is ranked higher than MAX
then for 4-7 repeat for any cvc data that
if a lang doesn’t delete codas, rank it the lowest
if a lang epenthesis, rank dep above max
if a lang deletes, rank max above dep
for 8-10, use a 2x2 tableau maybe
