Phonological Theory I
Phonological Theory I LING 200A
Popular in Course
Popular in Linguistics
This 6 page Class Notes was uploaded by Deborah Greenholt on Friday September 4, 2015. The Class Notes belongs to LING 200A at University of California - Los Angeles taught by K. Zuraw in Fall. Since its upload, it has received 96 views. For similar materials see /class/177855/ling-200a-university-of-california-los-angeles in Linguistics at University of California - Los Angeles.
Reviews for Phonological Theory I
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/04/15
Oct 14 2008 Class 6 Optimality Theory partI To do 0 Finish Prince amp Smolensky excerpt SQs due Thursday 0 Start working on beginning OT problem due Tuesday 1 The conceptual crisis Prince amp Smolensky p 1 Since Kisseberth 1970 constraints were taking on a bigger and bigger role But 0 What happens when there s more than one way to satisfy a constraint We need to prioritize the rules that could be triggered 0 Why aren t constraints always obeyed Relatedly what happens when constraints con ict What if one constraint wants to trigger a rule but another wants to block it We need a way of prioritizing constraints 0 Should a rule be allowed to look far ahead in the derivation to see if applying alleviates a constraint violation Or does the alleviation have to be immediate 0 Can a constraint be against making a certain type of change rather than against a certain structure 2 Prince amp Smolensky s solution Optimality Theory rulebased grammar with constraints OT grammar start with URinput from mental lexicon apply rules in sequenceiintermediate representation is known at all times apply all possible rules producing a large set of candidate outputs constraints may block or trigger rules constraints pick the best candidate look ahead is nonexistent or sketchy since the candidate outputs are all potential surface forms there is full look ahead to the end of each possible derivation interaction of constraints is nonexistent or sketchy constraints interact through strict J similarity to UR is the result of not applying too many rules and not having too many constraints similarity to UR is enforced by faithfulness constraints end with SRoutput send it to the phonetic system 3 Gen This is the function that creates the set of candidate outputs from the input One way to think of it1 apply all possible rules to the input any number of times Each underlying segment can be deleted or have its features changed extra segments can be inserted anywhere underlying segments can change their order 1 This is what PampS call anharmonic serialism except with a universal set of rules that s broad enough that the result is the all possible variants that PampS propose Ling 200A Phonological Theory 1 Fall 2008 Zuraw 2 Genab ab a b ba 1 ta at a6 m ab b ba tab atb abt eb aeb abe ib ob ap am Qta at ae ea i o 0 Why is the resulting set of candidates infinite assuming a finite alphabet of symbols 4 Constraints In standard OT we can think of each constraint as a function from a candidate output to a natural number the number of violations NOCODAbak 1 NoCODAtikpad 2 Alternatively we can think of each constraint Ci as imposing a strict partial ordering gti is more harmonic than with respect to CL on a set of candidates with the following additional properties 0 The ordering is strati ed If a k b and 7 hi then for any i gta i gt 7 too and for any j such that a gt j b gt j too We can say that ifa b and 7 2m a and b are of equivalent harmony 0 There exists some a such that there is no i gt a That is one or more candidates are the most harmonic there are not necessarily one or more least harmonic candidates though A strict partial ordering is transitive irreflexive and asymmetric 0 Transitive if a gt b and b gt c then a gt c 0 lrre exive a k a 0 Asymmetric If a gtb then b k a In Colin Wilson s targeted constraints variant of OT the stratification requirement is relaxed NoCODA tada bo badupi ta do i k ot lao takkat badkupit siklepbu 0 Why does assigning a non unique natural number 0 1 2 to each candidate meet the ordering requirements above o If you re ahead in the reading can you recall a case from PampS where numbers of violations weren t used 0 Why are there no least harmonic candidates for NOCODA Ling 200A Phonological Theory 1 Fall 2008 Zuraw 3 5 Eval Eval is a function from a set of output candidates and an ordered list Can of constraints to the subset of the candidates that is optimal typically that subset contains just one candidate EvalGeninputCon output or eg out1 out2 in the case of a two way tie Eval does this by taking the orderings imposed by the various constraints and assembling them into one giant ordering We can think of many ways this could be donestrict ranking is the mechanism used in standard OT for adjudicating harmony disagreements among constraints 6 Constraint interaction through strict ranking Strict ranking is like alphabetization to find the alphabetically earliest member of a list pick out all the members of the list that have the earliest first letteriand throw the rest of the list away from the new smaller list pick the members that have the earliest second letter etc Once a word is out of the running it can t redeem itself by eg having lots of as later on axiom axiate tab axicle caba banana azalea axolotl zabaglione baa Eval works the same way If you have n constraints 0 Pick out the candidates that tie for being best on the top ranked constraint CO discard the rest of the candidates 0 Move to the next constraint C1 and do the sameipick out the candidates that are best with respect to the set that is left Repeat for C2 C 0 Whatever candidates are still left at the end are tied for being the winner if you have enough constraints there is normally just one winner Q How can that be computable Wouldn t you have to go through an infinite list of candidates just to do the first step A For that reason most computational implementations of OT Ellison Eisner Albro Riggle represent the candidate set as a regular expression which is a finite way to represent a certain class of infinite sets For example aba is the set aa aba abba abbba abbbba These expressions can then be manipulated algorithmically either in a fairly literal translation of the above or by other means More declaratively a candidate a is optimal iff for any 7 and Cf such that b gtj a there exists some C such that i ltj ie C is higher ranked than C and a gt b In words for a to be optimal any candidate that does better than a on some constraint must do worse than a on some higher ranked constraint 0 Can you imagine some other ways that constraints could conceivably interact Ling 200A Phonological Theory 1 Fall 2008 Zuraw 4 7 Two types of constraint In pre OT approaches to constraints constraints were all markedness constraints they penalized certain surface structures such as CCC clusters So on first hearing about OT many people s second reaction the first was worrying about infinity was to wonder why if it s all about constraints every word isn t maximally unmarked o In ruleconstraint theories what prevents every word from coming out baba or whatever the least marked word is o For those who know OT or have read ahead how do PampS prevent every word from coming out baba Markedness constraints are constraints on the output they could be thought of as requiring articulatory ease or perceptual clarity or some other natural type of unmarkedness The simplest ones can be defined by the structural description that they ban voice CG You can and should give a constraint a helpful mnemonic name like NOCODA for CLS as long as you clearly define the constraint somewhere A good constraint definition should make it clear not just what is banned but how the number of Violations is assessed Faithfulness constraints are constraints on the relationship between the input and the output the standard ones require similarity but we can imagine other possibilities PampS use PARSE roughly don t delete and FILL roughly don t insert but this has been superseded by McCarthy amp Prince s correspondence approach which you ll learn about in more detail in 201 MAX X don t delete X e g MAX C MAX V DEF X don t insert X eg DEF C DEF V IDENT F don t change a segment s value for the feature F People often have a hard time at first with IDENT F The most common confusion is thinking it means don t delete a segment that is F The next most common is thinking it means don t alter a segment that is F e g by changing its values for some other feature G 8 Exposition the tableau In a utopic future which may not be that far away we will all check our analyses with software that evaluates the infinite candidate set In the meantime we illustrate our analyses for the reader with a tableauj showing a finite subset of candidates that have been chosen to demonstrate aspects of the constraint ranking The danger here is obviousiwhat if you didn t think of some important candidate 2 Or maybe they are just arbitrary and learned by speakers in response to whatever cards history has dealt them Or maybe both natural and unnatural constraints are possible but learners treat them differently This is a matter of current debate See Elliott Moreton s WCCFL paper to appearisee his webpage for a very clear overview 3 French for table The singular tableau is pronounced tablo in French theeblou in English The plural tableaux is pronounced tablo in French theeblou or theeblouz in English Ling 200A Phonological Theory 1 Fall 2008 Zuraw 5 This tableau shows a ranking argument we have two candidates that differ in that NOCODA prefers a the winner whereas DEF V prefers b If that s the only difference between the candidatesithere is no other constraint that prefers a over bithen NOCODA must outrank gtgt DEF V atka NOCODA DEFV C a at9ka b atka Parts of the tableau I input I output candidates I constraints highest ranked on left I asterisks 7 exclamation marks These three don t add any new Shadmg information but are there for the convenience of the reader I pornting finger or you can use an arrow 9 How do I know which candidates and constraints to include in my tableaux Here is a procedure that usually works reasonably well I Start with the winning candidate and the fully faithful candidate I If the winning candidate i the fully faithful candidate I Add the markedness constraintgs that rule out the fully faithful candidate I Add the faithfulness constraints that the winning candidate Violates I Think of other ways to satis the markedness constraints that rule out the fully faithful candidate Add those candidates and the faithfulness and markedness constraints that rule them out You have to use your judgment in deciding how far to take this step I If the winning candidate 2 the fully faithful candidate then you are probably including this example only to show how faithfulness prevents satisfaction of a markedness constraint that in other cases causes deViation from the underlying form I Add that markedness constraint I Add one or more candidates that satisfy that markedness constraint I Add the faithfulness constraints that rule out those candidates 0 Let s try it for atka gt ateka 0 One of the candidates below is unnecessary in arguing for the constraint ranking Why A candidate is harmonically bounded if it could not win under any constraint ranking General tips for reading scholarly articles Note I put this handout together for an undergraduate course when some students expressed frustration about trying to read a research article You may already have your own strategies for reading difficult material if they are successful don t feel you have to follow these suggestions They come from my own experience and from a niceithough repetitious and fmickyibook that I recommend Van Doren Charles amp Mortimer Adler 1972 revised edition of 1940 original How to Read a Book Touchstone 0 Before you read preread 0 Can you guess from the title what the article will be about What position it will argue for o If there is an abstract read it carefully o Read the introduction carefully then skip ahead and skim the conclusion or summary if there is oneithis will help you know where the body of the paper is going when you go back to read it o Often the introduction ends with a preview of how the paper will be structureditake note of this and ip through the paper to see all the section headings From what you learned in the abstract introduction and conclusion try to predict how each section will fit in with the main argument 0 Engage with the article To repeat from above when prereading you re trying to form an idea before you really start reading the article of what it will be about what problem it tackles what it will argue and how each section will contribute to the argument Then when you re reading keep checking against your expectations For example say that a section of an article is arguing that Arabic epenthesis is postlexical and therefore invisible to stress Then the author says that some more examples will be given that argue for epenthesis coming late As you look at each example try to see what the problem would be if epenthesis came early and whether the problem goes away if epenthesis is late When you see a set of data before you read the author s analysis try to see what s going on for yourself Sketch out as much of an analysis as you can and note what the tricky aspects are Then compare what you got to what the author says After each example each subsection and each section pause to consider what its importance has been to the article s main argument It helps to write on the article if it s your own copy I Underline the major points or use a vertical line in the margin if the passage is too long to underline Circle key words or phrases or new terms Put asterisks in the margins with a brief note to mark the 5 or so most important points Put numbers in the margins to mark a sequence of points in developing an argument For example if an author says There are several or even better 0 O O O O Ling 200A Phonological Theory 1 Fall 2004 Zuraw lam39n
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'