Top-Rated Free Essay
Preview

Phonology

Good Essays
1511 Words
Grammar
Grammar
Plagiarism
Plagiarism
Writing
Writing
Score
Score
Phonology
Phonology - How Speech Sounds Combine
Introduction to Linguistics for Computational Linguists

1

Speech Sounds
• Phonetics - Physical basis of speech sounds
– Physiology of pronunciation, perception – Acoustics of speech sounds

• Phonology - Patterns of combination of speech sounds
– Which sequences are allowed (phonotactics) – Effects of context on speech
2

Phonology
• Basic elements are phonemes. • Patterns of organization are phonology.
– – – – Structure of phoneme set Syllables, phonotactics (order of phonemes) Processes (adjustments in pronunciation) Rhythm, stress, tempo (not in this course)

• Phonological principles are psychological, sometimes with phonetic (physical) base
3

Other Phonology
• Stress, rhythm, intonation
– Stress: ’Verb und Nomen vs. Ver’bundnomen – Rhythm: Nicht! Aufhören! vs. Nicht aufhören! – Intonation: Ich bin der Nächste. vs. Ich bin der Nächste?

• Tempo, intensity (loudness) also • Emphasis (here) on segmental level
– Stress, rhythm, intonation are suprasegmental
4

Phoneme Inventory
• Structure in set of phonemes
– cross-classification in phonetic features – multidimensional matrix – place, manner, voice – [p,t,k] vs. [b,d,g] / [f,s,χ] vs. [v,z,-]

• Symmetry, but imperfect
– gaps (German voiced velar fricative) – crowding ([s, ] structurally close) – unique elements [l,R, ]
5

Phonotactics
• Phonotactics - allowable phoneme sequences
– reduce combinatorics of sequencing

• Which could be German?
– [frI ] [fstr t ] [kw t ] [kto] [χru t of] [kRil] [ u] [ptero]

• Preserved in “jargon” aphasia • Japanese allows only CV(n), i.e., consonsant followed by vowel perhaps followed by [n]
– Borrowings with final consonants, consonant clusters modified – [besiboru] ‘baseball’ ; [kurIsumasu] ‘Christmas’
6

Syllables
• (onset rime) = (onset (nucleus coda)) • Rime determines what rhymes (in one syll.)
– groß, los [os]; Rad, Tat [at]; Zahl, Kanal [al]

• Nucleus always vowel • Possible clusters largely determined by sonority

7

Sonority and Syllables
• Observation: mirror antisymmetry in consonant order in clusters in onset vs. coda
– [pl ts], [ lp]; [tr p], [fart]; [flai ], [h lft]

• Sonority - relative prominence
100 80 60 40 20 0 Vowels Liquids (l,R) Nasals Stops, etc. Sonority

8

Sonority in Syllable
• Sonority climbs toward peak, then declines

Prinz

p

r

I

n

ts

qualmt

k

w

a

l

m

t

9

Phonological Processes
• Compare Susi und Peter/Tom/Gabi spoken quickly • [zu.zi.m.pet R] / [n.t m] / [ gabi] • ‘und’ is pronounced [m/n/ ] • Similary /n/ in Es könnte dann passen/gehen • Sloppy?
10

Nasal Assimilation
• Speech requires lots of coordination • Nerves, muscles are preparing several segments ahead • Often we see effects in adjacent phonemes place n C

n adjusts its place of articulation to anticipate the following consonant

• Processes modify phonemes

11

Aspiration Revisited
• Recall from Phonetics lecture that voiceless stops are normally aspirated, i.e. voicing starts well after stop is released release • Unaspirated [b,d,g] stop voicing • Aspirated [p,t,k] -- note puff! stop voicing 12

Aspiration
• [p,t,k] unaspirated after [ ,s]
– [ p k] Speck, [ tIm ] Stimme, [ski] Ski

• We note aspiration with [ph], etc.
– Tücke/Stücke [th k ]/ [ t k ]

• [th], [t] are allophones (variants) of the same phoneme; likewise [ph], [p] • Since they are found in different contexts, they are in complementary distribution
13

Informal Rule Notation
• C[-voi,+stop] → Ch / . __ V • “Voiceless Stops become aspirated in the environment (/) after syllable begin (.) and before vowels”
– Tücke /t k / → [th .kh ] – phonemes → “are realized as” phones
14

Final Devoicing
• Auslautverhärtung
– lobe [lob. ] but lob! [lop] – blase [blaz. ] but blas! [blas] – steige [ taig. ] but steig! [ taik]

• C → C[-voice] / __ #,

– where ‘#’ is a word boundary – or morpheme boundary? --See lecture on Morphology. – or syllable boundary? Wagner [w g.n R] / [w k.n R]
15

Release
• /p,t,k/ may be unreleased finally • Morgen geht Peter weg. [geth] or [get-] • Alternative pronunciations arising through optional processes also allophones, said to be in free variation
16

Releasing
• Stops before other stops are normally not released
– [ kt], [kIpt] Akt, kipt – IPA [ k-t], [kIp-t]

• Third allophone of /p/: [ph,p-,p]
– in complementary distribution and/or free variation

• C[+stop] → C- / __ C [+stop]
17

Finding Phonemes
• To determine phonemic inventory, linguists analyze all (apparent) cases of complementary distribution and free variation • Earlier seen as part of automatic (discovery) procedure, which is infeasible.
– But still standard procedure
18

Nasalisation
• Vowels before nasals ([n,m, ]) are pronounced with velum lowered
– eng [ ] vs. Eck(e) [ k] – tilde normally above the nasalized symbol

• V → V~ / __ N, where N is [n,m] or [ ]
– allophone in complentary distribution

• French uses nasalized vowels contrastively
– beau [bo] vs bon [bo]
19

Intrusive [t], etc.
• Why does Benz [b nts] rhyme with Jens [j ns]? • ∅ → t / n __ s • Sims [zIms] or [zImps]? Example with [ ]?
– Hamster [h mp.st R], des Lamms [l mps]

• See Tim und Struppi detectives Schultze und Schulze, English Thomson and Thompson
20

Whispered Sonorants
• [l,R,w,j] are sonorants (likewise nasals) • Pronounced after voiceless stops, they are also voiceless (because of aspiration)
– Prinz [pRInts], Tratsch [tRat ], Quatsch [kw t ], Klauen [klau n]

• S → S / Ch __
21

Weak Syllables
• Second, unstressed syllables followed by sonorants
– Boden [bo.d n] or [bod.n] – Sattel [z t. l] or [z t.l] – Butter [b t. R], [b t R], or [b t ]

• •

R→

R/

__ #, where ‘#’ signifies end of word
R

– kleinere [klain R ], not [klain

],
22

S → S/ __ #

Nasals in Weak Syllables
• Lappen [l p.m], loben [lob.m], kommen [kom.m] • Boden [bod.n], Ratten [R t.n], lassen [l s.n] • packen [p k. ], sagen [z g. ], singen [zI . ] • Same reduction to syllabic sonorant plus assimilation in place. • Can create unusual pronunciations!
– kommen [kom:], rennen [r n:], singen [zI :]
23

Reduced Nasals after Labiodentals
• Laufen [lauf.n], [lauf.m], [lauf. ]; • Löwen [l v.n], [l v.m], [l v. ]; • [ ] is a labiodental nasal (i.e., shares place of articulation with [f,v]) • [ ] is a syllabic version
24

Velar/Palatal Fricative
• [ ] (velar fricative) also pronounced palatally
– Aachen, [a . n], Bach [b ], Buch [bu ], Loch [l ] . R] – ich [i ], echt [ t], Bücher [b . R], Löcher [l





/ V[+front] __

– complementary distribution, allophones

• But liebchen [lib. n], Mädchen [met. n] -Why palatal? • → / V[+back] __ ?? (with as basic)
25

Phonemic Analysis
• liebchen [lib. n] palatal, not after front V • → / V[+back] __ but Kuhchen/Kuchen • Frauchen [fRau. n] vs. Rauchen [Rau . n]
– Near-minimal pair: with → rule, no account
• we don’t wish to say that -chen is [- n] since phonological processes work on phonemes, and (under → rule), is the phoneme





preferable n] has no because of the syllable (‘.’)
26

• [fRau.

Why are there Phonological Processes?
• Speech is hard -- 2 wd./sec. (~ 10 phon/sec) • Signals reach muscles at different speeds, which then contract in varying times
– long nerve pathway to intercostals (in chest) – velum is slow compared to tongue

• Some processes simply make speech easier
– nasal assimilation, velar/palatal alternation [ / ], nasalization of vowels
27

Why Phonological Processes?
• Speech is also hard to understand • Some processes make sounds more distinct
– aspiration – exaggerated release of final stops
• Halt! [h lth] Gut! [guth] (compare [gut-])

– vowel lengthening under emphasis
• Das tue ich nie [d s tu i ni:]

28

How “Real” is Phonology?
• Could patterns be accidental? • Speakers apply native phonology even when dealing with unknown material
– inventory – phonotactics – processes

• Evidence in generality, application to foreign material (accents/mishearings), even errors
29

Inventory
• We tend to hear/pronounce foreign languages as composed of “our” sounds
– Eng. / then/thin pronounced [z/d;s/t] by French, Germans – German front rounded vowels hard for English, Russian, Spanish speakers [y, , , ] – Spanish trilled [r], English retroflex [r] hard for French, German speakers Sp. perro, Eng. Red – Initial [h] hard for French speakers • Eng. home [om]; Germ. Haus [aus] – Japanese r/l “Conglaturations on Erection!”
30

Phonotactics
• We find it hard to pronounce sounds out of place -- even if they exist in our language
– Dutch [s …] used to detect Germans: Scheveningen – Sp. has sequences [prjeto] (prieto), [kljente] (cliente), [krwel] (cruel) -- Eng./Ger. tend to mispronounce [prijeto], [klijente], [kruwel] – Vietnamese initial [ ] hard for Europeans • Ngu Van Thieu [ u ...] “simplified” to [nju ...]

31

Processes
• We mishear/mispronounce by using native language processes in foreign languages
– French hear English/German as “nasal” and vice versa
• nasality in unexpected places Fr. [bo] Eng. [bon]

– English/German aspiration interferes w. French
• French [p] , Eng./German [b] similar VOT’s ; • French accent in English ‘You pig!’ [ju big]

– German final devoicing in English
• ‘Child’ [t ailt], ‘could’ [k t]
32

Psychological Reality
• Processes apply where they were never heard
– foreign speech – errors ‘tip of the slung’ [thIp …] – nonsense words
• “’Twas brillig and the slithy toves” … [thovz]

33

Bigger Picture
• Processes here tip of iceberg
– [k] in Kind [kInd] further front than in Kuh [ku]

• Constant anticipation, perserverance
– Effect of consonant on formants [ku] vs. [tu]

• Creates redundancy in signal
– enable understanding even when perception lags

• Shifts information to acoustically prominent elements (vowels)

34

Phonology
• Patterns of combination of speech sounds
– Inventory of basic sounds – Which sequences are allowed (phonotactics) – Processes -- effects of context on speech

• Emphasis (here) on segmental level
– Stress, rhythm, & intonation are suprasegmental

• Phonology involves imposition of structure
– seen in novel applications (foreign words, nonsense words, and even errors)
35

Phonology
• Phonological processes serve to ease production and perception of speech • Even apparent production-simplifying processes may enhance redundancy, ensuring perception.

36

You May Also Find These Documents Helpful

  • Satisfactory Essays

    Rhyme- recognizing & producing words with the same ending sound. Ex: What word rhymes with word cat? Or Does cat rhyme with the word hat?…

    • 188 Words
    • 1 Page
    Satisfactory Essays
  • Satisfactory Essays

    Spanish 104

    • 616 Words
    • 3 Pages

    -Some words ending in a consonant different from n or s are NOT stressed on the final syllable.…

    • 616 Words
    • 3 Pages
    Satisfactory Essays
  • Good Essays

    Psychology Study Guide

    • 1729 Words
    • 7 Pages

    Due Date Rd. Pg.410-Pg.417 1. How learning different sets of phonemes make it difficult for people of one language to produce phonemes of another language? 2. How many morphemes in the word cat?…

    • 1729 Words
    • 7 Pages
    Good Essays
  • Good Essays

    Wangkajunga

    • 649 Words
    • 3 Pages

    As stated in the Fletcher and Butcher’s paper (2014), one of the typical patterns of allophony…

    • 649 Words
    • 3 Pages
    Good Essays
  • Better Essays

    The process of learning to read is not considered to be an innate developmental function of the brain and therefore it requires explicit teaching of phonemic awareness, phonics, fluency, vocabulary and comprehension (Department of Education, Science and Training, 2005). When all of these components are taught together children develop an understanding of the relationship between the sounds in spoken language, the letters and letter combinations that make up written words and their meanings (Emmitt, Hornsby & Wilson, 2013). This essay identifies the key characteristics of emergent readers and describes a range of strategies used by educators to enhance the process of learning to read.…

    • 1588 Words
    • 7 Pages
    Better Essays
  • Good Essays

    The United States is one of the most diverse countries in both linguistic and cultural aspects. Increasing rates of Asian populations have been observed in the past decades. Many Asians live in Spanish-Cantonese speaking environments and learn English as a second language. To date, there is no study on the phonological process in Cantonese-Spanish bilinguals learning English as an L2. De Houwer (1995, as cited in Law and So, 2006) and Romaine (2001, as cited in Law and So, 2006) indicated that numerous studies on bilingual development in different language combinations have found, that bilingual and monolingual language development differ qualitatively and quantitatively.…

    • 997 Words
    • 4 Pages
    Good Essays
  • Better Essays

    Spanish and English share several consonantal phonemes including the stop sounds /p/, /t/, /k/, /b/, /d/, /g/, the nasals /m/, /n/, the fricatives /s/, /f/, the liquid /l/, the glide /w/, /j/, and the glottal /h/. Despite these common phonetic symbols, it is important to note that there are differences in voicing, aspiration, and precise place of articulation for many of these sounds that result in acoustic differences (p.8).…

    • 1132 Words
    • 5 Pages
    Better Essays
  • Good Essays

    Figurative language

    • 1117 Words
    • 5 Pages

    There is alliteration in the repetition of the s sound at the start of “silken” and “sad”. There is assonance in the repetition of the ur sound in “uncertain”, “purple”, and “curtain”. And there is consonance in the repetition of the s sound in the words “uncertain” and “rustling”.…

    • 1117 Words
    • 5 Pages
    Good Essays
  • Satisfactory Essays

    The processing of recognizing and responding to the meaning embedded in spoken words is defined as speech recognition. Phonemes are series of corresponding sounds part of each letter of the alphabet. When a computer recieves input from speech recognition, it has to break down a word into the different phonemes to determine what word was being said. Likewise, if a whole sentence or phrase was said, the computer has to work to find the different starting and ending points of each phoneme, while also recognizing points of silence to indicate different words. Sound is captured in analog form and is then transformed into digital form by method of digital sampling, and the resulting digital pattern is compared with a library of patterns corresponding to known phonemes. There are…

    • 508 Words
    • 3 Pages
    Satisfactory Essays
  • Satisfactory Essays

    Phonology

    • 296 Words
    • 2 Pages

    Phonology is the study of the speech sounds and sound of words in a language. It is also concerned with the way words are pronounced in a language. Each language has its own phonology. From a child's point of view, the business of phonology is figuring out how to produce those sounds that are necessary for making meaning. Infants know the sound of language before their first word. The most amazing part is babies learn from way before in utero (Siegler, 2005).…

    • 296 Words
    • 2 Pages
    Satisfactory Essays
  • Powerful Essays

    Phonics For Research Paper

    • 2148 Words
    • 9 Pages

    Systematic phonics instruction clearly identifies a carefully selected and useful set of letter-sound relationships and organizes the introduction of these relationships into an instructional sequence that is easy to follow. The instructional sequence will usually include the relationships between the sounds associated with single letters, as well as with larger units of written language or even spelling patterns. Systematic instruction also provides students with a number of opportunities to practice the relationships being studied.…

    • 2148 Words
    • 9 Pages
    Powerful Essays
  • Good Essays

    g. [frut] h. [pritʃər] i. [krak] j. [baks] k. [θæŋks] l. [wɛnzde] m. [krɔld] n. [kantʃiɛntʃəs] o. [parləmɛntæriən] p. [kwəbɛk] q. [pitsə] r. [bərak obamə] s. [dʒɔn məken] t. [tu θaʊzənd ænd et]…

    • 762 Words
    • 4 Pages
    Good Essays
  • Good Essays

    Setswana Phonology

    • 1223 Words
    • 5 Pages

    When Batswana communicate, they use a combination of sounds which enable them to convey various ideas with each other. Each of these sounds consists of words which are made out of discrete speech sounds known as phonemes. This study of phonemes and other phenomena related to the sounds of a particular language is known as phonology. In order to gain a full understanding of the Setswana language, it is necessary to carefully analyze the inventory and structure of its consonants and vowels.…

    • 1223 Words
    • 5 Pages
    Good Essays
  • Good Essays

    phonetics and phonology

    • 548 Words
    • 3 Pages

    When we talk about how phonemes function in language, and the relationship among the different phonemes – when, in other words, we study the abstract side of the sounds of the language – we are studying a related but different subject that we call phonology.…

    • 548 Words
    • 3 Pages
    Good Essays
  • Powerful Essays

    The Language Lab focuses on the production and practice of sounds of language and familiarises the students with…

    • 4498 Words
    • 48 Pages
    Powerful Essays