LIN229 Final Exam Chapter Reading Notes.docx

11 Pages
297 Views
Unlock Document

Department
Linguistics
Course
LIN229H1
Professor
A.Kochetov
Semester
Summer

Description
LIN229 Final Exam Chapter Reading Notes Based on “Understanding Phonology” Third Edition by Gussenhoven and Jacobs Chapter 2  Places of articulation:  Pharyngeal: root of tongue articulates with pharyngeal wall  Dorsal: dorsum of tongue articulates with roof of mouth (if the front of the tongue articulates then this is called palatal). If it is the back then it is called velar.  Coronal: when the crown articulates with the teeth, alveolar ridge, or forward part of hard palate. Retroflex is with the tip curled back.  Labial: if lips articulate with each other (bilabial) or with upper teeth (labiodental).  Types of constriction:  Obstruents: leads to friction: plosives, fricatives, and affricates  Sonorants: no friction: nasals an approximants  Approximants have a light or near-contact of articulators, the air stream being so weak that no friction is produced. Sounds with a partial occlusion Iike [l] are called laterals.  Long consonants are called geminates  Secondary articulation occurs in a few cases:  Labialization: lips are rounded during articulation of consonant  Palatalization: front of tongue is raised during pronunciation  Velarization: back of tongue is raised (e.g. dark l)  Pharyngealization: root of tongue is retracted back  Double articulate: e.g. w  Nonpulmonic segments:  Clicks: velar closure plus closure elsewhere (called velaric airstream mechanism)  Implosives:  Ejectives  Two types of feet: left and right dominant. Commonly they are disyllabic,  In left dominant first syllable is stressed and rest aren’t  In right dominant the rightmost syllable is stressed.  Usually in English if there are two feet the first gets primary stress and second gets secondary stress (e.g. celebration)  You can also have monosyllabic feet (e.g. syntax, where there is both primary and secondary stress but only two syllables) Chapter 3  Speech and languages prefer distinctions that are easy to perceive and produce. Harder to produce voiced plosive than voiceless because takes more air pressure, so more voiceless than voiced plosives are found. Opposite is with nasals –easier to voice than to not voice.  We want to maximize contrasts so if we nasalize one vowel use this process with other vowels too  Low cost contrasts are thus frequent We find many similarities are found in all languages. We can say this is innate but maybe it’s just the circumstances the led to these similarities made sense in how languages were made up.  According to UPSID: most consonants: 141, least:11.  Languages differ in syllable structure; lowest complexity is a (C)V syllable.  The peak (onset peak coda), may be complex – long vowels or dipthongs  The peak can sometimes be filled by a consonant  The presence of a complex constituent implies the present of less complexity. E.g. language with 2 consonants in coda will allow one.  Universals and implicational relationships:  All languages have syllbles, and all inventories can be split into consonants and voowels  All languages have voiceless stops  Near universals: only Hawaiian doesn’t have a t. only 2 languages don’t have sonorants. Coronal place of articulation is most common  Unusual segments tend to occur in larger segment inventories  Unusual segments are also less common in languages that have them  Unusual segments are phonologically more complex (harder to pronounce)  Languages build up their inventories in an orderly way. If one has X it probably has Y. this is an implicational relationship.  No voiceless nasals without voiced ones  No z without s  No front rounded vowels without front unrounded and back rounded  NONE of these are absolute, only statistically likely  Number of consonants and vowels are positively correlated as well as complexity in syllable structure and tone structure  Best inventories are those where contrasts are maximally distinct with least articulatory effort  Plosives are more common than fricatives  Voiceless plosives are more common than voiced  Voiceless fricatives are more common than voiced  Front rounded and back unrounded are less common than opposite (the opposite are intermediate between the two in the bottle analogy, where [i] is like a full bottle and [u] is like an empty one. Since they are intermediate they are less perceptually salient).  We see that inventories are built as if segments ad elements to sets of segments (e.g. p t k + voiced  b d g)  Why do we have system gaps? E.g. no g above. Could be because g is less salient and harder to pronounce. Chapter 4  Loanwords can show us a lot about the phonology of a language (nativization)  Hawaiian has the syllable structure (C)V(V)  In the process of nativization speakers need to:  Interpret each segment in terms of their own segment system  Make sure they don’t break their own syllable structure  These take place at two levels. The first is perceptual (choosing right segment) and second is operative (adhering to constraints)  So each segment has input-> perceptual->operative (p.47)  Note that the actual segment might only change in the operative letter, e.g. p.48 Chapter 5  When different sounds correspond to the same sound on the surface we have neutralization of a contrast.  Two kinds of allophony: one is a particular segment or segment group context. Another is structural position like coda or onset of syllable.  If not allophones then the segments are contrastive, and can form minimal pairs.  Allophones are in complementary distribution (in order to be allophones the two segments have to be similar in articulation).  There are two levels of representation: underlying and surface. Three good reasons:  Economy: easier to store/retrieve with rules than with specific entries in lexicon  Without two levels you couldn’t express relatedness of morpheme alternates  Many generalizations are valid at a level other than the surface level (p.66)  Surface forms that contradict a phonological generalization are called opaque (nontransparent)  This can happen when a rule has failed to apply but its structural description is met at the surface form (understand this!!)  When rules produce segments where such segment already appear normally we have neutralization rules  How to choose the underlying segments? Try to make the rules simple. Not good idea to choose a neutralized segment, because we don’t want to lose contrasts.  When a rule is applicable but produces the same result it is called a vacuous application. Chapter 6  Distinctive features are the elements by which we can refer to natural segment classes  This is the motivation to actually have these features (i.e. the fact that we can refer to natural segment classes with them)  Natural segment classes are groups that come up again and again in rules, as opposed to groups of random segments which don’t usually show up together. We see time and time again that natural segment classes are phonetically similar. Naturalness condition: distinctive features must have a phonetic definition.  Three requirements of distinctive feature system: 1)Capable of characterizing natural segment classes 2)should describe all segmental contrasts in the world languages 3)Should be definable in phonetic terms  With binary features we can define natural classes with the + feature and with the -. With unary features (e.g. LABIAL), we can’t say that there is a natural class that is –LABIAL. Features  Major class features:  +/- consonantal: segments have a constriction somewhere along the vocal tract which is at least narrow enough for a fricative. (stops, affricates, fricatives, nasals, laterals, and [r]) (- are vowels, glides and those with constriction in the larynx)  +/-sonorant: distinguishes obstruents from sonorant consonants and vowels. + are produced with equal air pressure behind and in front of the constriction. –have significant constriction or no constriction in the vocal tract. + are all vowels, glides, liquids, and nasals. – are stops fricatives, affricates, and laryngeal segments.  +/- approximant: those that have a constriction in vocal tract which allows free oral escape of air. + are vowels and non-nasal sonorants (liquids)  Note: the difference between vowels and glides is commonly assumed to be a difference of phonological (specifically, syllable position) structure, and not phonological content, but we use the term +/- syllabic informally.  Laryngeal features:  +/- voice: + are vowels, glides, sonorants, voiced obstruents. – are voiceless obstruents.  +/- spread glottis: audible friction in glottis. +is aspirated segments and glottal fricatives.  +/-constricted glottis: vocal cords are tense and drawn together. +are laryngealized vowels and larynealized sonorant consonants, glottalized obstruents and implosives.  Manner features:  +/-continuant: lack central occlusion in vocal tract. - are stops, nasals, affricates, laterals  +/-nasal  +/-strident:relevant for obsturents only (p.79) also used to differentiate affricates from stops.  +/-lateral: central tongue contact in oral cavity  [l] behaves ambiguously w/ respect to +/-cont feature.  [h] does not participate in rules referring to cont. In fact, laryngeal segments are not specified for either manner or place features.  Place features:  LABIAL: articulated with the lips  +/-round  CORONAL: raised crown of tongue (from dental to prepalatal [j])  +/-anterior: articulates with alveolar ridge or more forward  +/-distributed: +articulate with a constriction that articulates a relatively great distance along vocal tract.  DORSAL: articulated with bunched dorsum. All vowels are DORSAL.  +/- high  +/-low  +/-back. +back are velar or uvular. ç is interpreted as a fronted velar.  +/-tense (refers to tenseness of vowels). (sometimes we see +ATR and +RTR instead)  RADICAL: articulated with the root of the tongue.  Many vowels have more than one place feature, and are complex.  In many Germanic languages the vowels divide into a set of lax and tense vowels  Sometimes features will be redundant (e.g. p.87) e.g. in some languages +round predicts +back for vowels since there are now –back +round vowels. Or lots of times +son predicts +voice so +/-voice is redundant in sonorants (but said to be contrastive) in obstruents.  Underspecification theory is a body of work devoted to the question of whether redundant features should be included in the underlying representation of morphemes.  Contrastive undespecification says that features are specified only in segments that contrast for the features concerned. Chapter 7  Segments are represented by feature matrices. A morpheme is a string of feature matrices.  Beginning and end of morpheme is indicated by a boundary symbol: word internal (+) and word boundary (#). (2 #s are inseted between words, one for each word).  General format of a rule given on p. 94: FOCUS STRUCTURAL CHANGE/ CONTEXT  FOCUS+CONTEXT= STRUCTURAL DESCRIPTION (SD)  Rules apply within words. If it applies across boundaries you need to include #.  If a rule applies ONLY in a + boundary, this needs to be included.  No need to specify unnecessary features (pay attention to not including +voice in focus of p.84 (5)).  Add C as shorthand for [-syll]. V is used for [+syll, -cons+, i.e. a vowel. Superscript n means “n or 0 more”. Syllable boundary can be described by )σ  The brace ({) is used out of convenience. It means “either…or…”  We used variable feature values for assimilation rules (p.96)  We put optional stuff in parentheses (e.g: (##) for optional word boundary.  “regressive voicing” – applies based on later segment.  Transformational rule format (p. 97) helps for rules like metathesis.  is used for deletion (c.f. p.99 (14)) or to mean “nothing”.  Rules are ordered because otherwise you would have to assume every rule applies to the underlying representation which not only complicated every rule but makes each rule have to restate rules.  The order of rules is constant and transitive (p.100).  P.101 shows another rule format that is useful.  To represent the rule for only one type of morpheme we use the notation on p.102 (22)  A dash (-) is used for nonapplication. “vac.” Is used for vacuous application. Chapter 8  If order in which rules apply is predictable from properties of the rules then the rule order is intrinsic.  The elsewhere condition says that when one rule applies to a subset of the forms that another rule applies to, the general rule is blocked from applying to that subset.  Arguments for extrinsic rule ordering include two dialects that differ just by their rule ordering. (e.g. p.113)  Feeding order: if rule A increases the number of forms to which rule B can apply.  Counterfeeding order: if rule A increases the number of forms to which rule B applies, then B is a counterfeeding order.  Bleeding order: if rule A decreases the number of forms to which rule B can apply.  NOTE: counterfeeding rules don’t increase the number of rules to which B applies. Bleeding rules decrease the number of rules to which B can apply.  If rule A decreases the number of forms to which rule B can apply, then BA is counterbleeding.  Based on a principle of maximal rule application, feeding and counterbleeding orders are the natural rule orders.  From the viewpoint of maximum rule transparency, bleeding and feeding are the natural orders.  In counterbleeding, something happens but it’s not transparent why. In counterfeeding, something doesn’t happen and it’s not transparent why not. Chapter 9  Two levels of representation not enough, we postulate an intermediate level called lexical representation. This corresponds to native speakers’ intuitions – what we see in dictionaries.  In dictionary we see phonemic representation but for example, “looked” is transcribed with t, not d, while aspirated and unaspirated are considered the same.  Maybe intermediate level has neutralizing rules (which produce segments already in inventory) and next level produces allophonic rules? Can’t be right – some rules produce both, and these would have to be split up into subrules for these levels to hold true (e.g. p 132)  P.133 gives reason why we need another level of representation  Intermediate
More Less

Related notes for LIN229H1

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit