chapter2c

2.3 Supra-segmental Elements
1. Stress
     Stress refers to the degree of prominence a syllable has. Stress syllable means that special emphasis is given to it in pronouncing a word in which the syllable occurs. Stress is the relative force or loudness with which a syllable is uttered. In some languages such as English the stress is phonemic, meaning that it can differentiate meaning, for instance, re`turn means coming back and `return means turning again, and in Russian /`muka/ 'torture' and /mu`ka/ 'flour'. In long words there is mid stress and weak stress besides the primary stress. The weak stress is usually not indicated. The word dictionary has primary stress on the first syllable, the second on the third syllable, and in animation the primary stress in on the third syllable and the second on the first. The symbols for stress in words:
a. single inverted comma for primary stress (');
b. back single inverted comma for mid stress (`);
c. small letter v (a short curve) on the syllable for unstressed syllable n(v). We can divide the stress into free stress and fixed stress. In free stress the stress can occur on different syllables depending on the word. In fixed stress the placement of stress is restricted to one particular syllable. Hungarian for instance, assigns the stress to the first syllable, Turkish to last syllable, Polish and Indonesian to the penultimate (second from behind) syllable. In these languages the stress is predictable. We can write the Hungarian phonological rule for stress as this:
Sy ??-> [+stress] / ## ? .
In Spanish the stress lies on the last syllable when the word ends with consonants, except /n/ and /s/, and on the second from the end if the word ends with vowels or /n/ and /s/, for instance, se�or [s?`�or] 'sir', se�ora [s?`�ora] 'madam' , and se�orita [s?�o`rita] ' miss' (Sanyoto, 1995).
Exercise: Where are the primary and the mid stresses of the following words?

secretary separation aviation amendment
perpendicular academic intellect parenthesis
fundamental absolutely socialized caterpillar
For phrases or sentences, the stress can be divided into four, they are:
' for primary stress
^ for secondary stress
` for third stress
v for weak stress
The phrase intellectual curiosity has the first stress on o, the second on lec, and the third on cu. These are more examples:
remarkable invention
academic procession

Patterns of Stress
1. ' `   : dining room, bluebird, high-school, sidewalk,
           drugstore, bookstore;
2. ^ '   : sick nurse, poor house, working man, fat man
           etc.
         : qualifier + adjective: very good, more quickly,
           rather small, etc.
           verb + noun object: They bought apples, We are             making kites
The phrase short story can have ' ^ or ^ ' but different in meaning. So is flying planes.
2. Length of Sound/Duration
In some languages the length of sound is phonemic, meaning that it gives different meaning of the utterance. In Bugis and Skow the length of sound seems to be phonemic.

3. Pause
When we speak we make some kinds of pause. To differentiate that a syllable belongs to word A not to word B, we make a pause between word. Some other kinds of pause are pause between syllables, between phrases or clauses or between sentences. Samsuri for instance uses the symbols (+) for pause between syllables, (/) for pause between words, (//) for pause between phrases or clauses, and (#) for pause between sentences. These are some examples: ke/meja and ke+me+ja, ban+tuan and ban/tuan, it+swings and its+wings, a+nice+man, an+ice+man, [nait+reit] and [nai+treit] (night+rate, nitrate), #ka+mi/sa+ngat/bu+tuh/ ban+tuan/tuan#

4. Tone and Intonation
The normal pitch of the voice of the speaker is /2/, called mid. Pitch /1/ which is called low, is lower two or three notes below /2/. Pitch /3/, high, and pitch /4/, extra high. In Vietnamese for instance the pitch is phonemic.
ma 'ghost' (said on a level tone)
ma 'cheek' said with a short rise that begins on a high
              tone)
ma 'tomb' (said with a long rise that starts on a low tone)
ma 'rice' said with a short rise starting on a low tone)

The commonest intonation contour in English is /(2)/31 /. The parentheses indicate that the /2/ will occur if there are syllables before /'/, but not otherwise.
The three kinds of clause terminal are fading, rising, and sustained. Fading or falling means a rapid trailing away of the voice into silence; both the pitch and volume decreases rapidly. Rising means a sudden, rapid, but short rise in the pitch; the volume does not trail off so noticeably, but seems to be comparatively sharply cut off. Sustained means a sustention of the pitch accompanied by prolongation of the last syllable of the clause and some diminishing of volume.
Intonation contour is the combination of two, three or four pitch phonemes and one clause terminal. It can be illustrated by digits, as follows:
2          3 1
I bought a book           (statement)

2              3 1
We drove to the lake      (statement)

2         3 1
The car is ready          (statement)

2      3 3
Are you there             (yes/no question)

2          3 1
What's your name          (wh question)

2         3 1
Go to your room           (imperative)

2          3 3 2   3 1
Do you want tea or coffee    (question with or)
2             3         1      3 3
John has a new piano teacher    Who   (What did you say?)

2             3         1      3 1
John has a new piano teacher    Who   (Who is the teacher?)

2           3   1    1 2
Why are you washing   Jane

2                      3 1    1 2
What did you put on the table Mellon

I just ate it. /231/ What? /33/ (Slightly rising intonation??? i.e. Repeat your statement)
I just ate it. What? /32/    (Slightly falling intonation??? i.e. What did you eat?)
The caterpillar. What? /34/ (Extra high intonation ??? Surprise or disbelief)

2.4 Modifications
Sounds or phonemes can be modified by nasalization, prenasalization, palatalization, labiolization, backing, fronting, raising, lowering, aspirating. Nasalization means making the sound nasalized, letting the air go out through the nasal cavity. Prenasalization means nasalizing the initial sound of a word. Palatalization means palatalizing certain phoneme by inserting the sound [y] after the phoneme. Labiolization means labiolizing certain sound or phoneme by inserting the sound [w] after the phoneme. Backing means producing the sound by moving the articulator a little to the back. Fronting means producing the sound by moving the articulator a little to front. Raising or lowering means producing the sound by raising or lowering the articulator. Aspirating means producing the sound by giving the sound of aspiration after the phoneme. Rounding means changing the producing the sound from unrounded to rounded. The other modifications are fortis and lenis. Fortis means that the sound is articulated strongly or tensely while lenis means not articulated strongly (laxly).
Examples:
nasalization : [an] or []
prenasalization : [mbanyak]
palatalazation   : [kyai]
labialazation    : [kwai]
backed           : [d 3]
front            : [d 4]
lowered          : [e6 ]
raised           : [e5 ]
glottalized       : [?]
fortis           : inverted comma under the phoneme symbol
lenis            : [] (number 3 facing to the right) under the phoneme
Shift Signs: 3, 4, 6, 5, , :, ::,
(diacritics) raised, lowered, backed, fronted, short, long, longer

2.5 Morphophonemic Changes/Phonemic Changes
When two morphemes are combined a change is likely to occur in one (both of) them, it becomes more like its neighbor in some respect. This process is called morphophonemic change. There are some terms referring to these changes; they are assimilation, synthesis, simple consonant change, addition, loss of phoneme, stress shift, metathesis, epenthesis, and epithesis.
Assimilation is the process of phoneme becoming more like the neighboring phoneme when two morphemes are combined. When the change becomes more like a preceding phoneme, it is called progressive assimilation, and if it becomes more like a following phoneme, it is called regressive assimilation. The example of progressive assimilation are gitti 'he went' in Turkish (Gleason, 1961:84) which comes from git and ?di, and cabs [k�bz] in English in which the voiceless /?s/ becomes /?z/. The phoneme /d/ in di becomes /t/, and the phoneme /s/ becomes more like /b/, that is voiced consonant. The example of regressive assimilation is the prefix {in?} becomes /im?/ when it is combined with possible. The process is that /n/ becomes bilabial /m/ which is more like bilabial /p/, the first phoneme of possible. When the change becomes more different from the neighboring phoneme, it is called dissimilation; for instance ignoble, in which the prefix {in?} becomes /ig?/. It is clear that in that process the phoneme /n/ becomes /g/ that is not more like /n/ or /p/.
Synthesis is the fusion of two consonants becoming another phoneme which differs from both phonemes such as moisture which comes from moist and ure. The final phoneme /t/ of moist is combined with the initial phoneme /y/ of ure to become phoneme /c/.
Simple consonant change is the change of consonant or consonants in combination of two morphemes to form another morpheme; for instance, permissive which comes from permit and ?ive. The phoneme /t/ is changed to /s/.
Addition of phoneme is the addition of a phoneme or phonemes to the new combination; such as longer and solemnize which come from long and er and solemn and ize. On the former the phoneme /g/ is added and on the latter the phoneme /n/ is added.
Loss of phonemes is the missing of a phoneme or phonemes in the combination of two morphemes, such as the prefix {?n} in illegal, in that the phoneme /n/ is dropped.
Stress shift is the change of stress from one syllable to another in combination of two morphemes, such as in syntactic which comes from syntax and ?ic. The stress moves from the first to the second syllable. In this process sometimes there is a process of gradation, that is, the change of vowel because of the change of stress; for instance, symbolic which comes from symbol and ?ic. The vowel /;/ in /sImb?l/ becomes /o/ in /sImbolIk/.
   Voicing is the change form a voiceless consonant to a voiced consonant; such as in West Dani the morpheme ara? 'his/her' plus kom 'taro' become aragom 'his taro' and ana? 'my' plus pelir 'bat' become anabelir 'my bat'.

Only phonemic changes:
Metathesis is the reverse of phoneme at morpheme borders. it means that there is a process of changing places; for instance, the stem poy and the suffix ?pa become popya instead of poypa (Elson Pickett, 1961:44). It appears that /p/ and /y/ change places.
Epenthesis is similar to addition that is the insertion of an extra sound within a word, such as the /p/ you may hear in something, or the /t/ in sense, and in Indonesian jumlah and kapak are sometimes pronounced jumblah and kampak.
Epithesis is the addition of an extra consonant to the end of a word. It occurs after a final /n/ or /s/. Sometimes people pronounce the word 'drown' /draund/ or 'once' /wanst/ (Stageberg, 1967). In Irian Jaya some people pronounce the word 'rangka' /rangkah/ the word 'bawa' /bawah/.
Other phonemic changes are umlaut, ablaut, and vowel harmony (Verhaar, 1999). Umlaut is the change of a vowel to become higher because of a higher vowel following the vowel, such as /a/ in /hand/ 'hand' becomes /e/ in /behendig/ 'skilled' in Dutch because of the higher vowel /i/ of the /ig/. Another example is the sound of /a/ in gast [gast] 'guest' in German becomes /E/ in gaste [gEst;] (the plural form). Ablaut is the change of a vowel to become another vowel, such as 'sing' becomes 'sang' and 'sung'. Vowel harmony means the conformation of a vowel with the vowel preceding it, such as /atlar/ 'kuda', /odalar/ 'kamar' /evler/ 'rumah'. The vowel in /lar/ is conformed with the preceding vowel /a/ and the vowel in /ler/ is conformed with the preceding vowel /e/.

2.6 Finding Phonemes To find out the phonemes of a language, first of all the linguist must collect the data in form of a list of basic words. The list should be at least 800 words. After collecting the data, the analyst can follow this procedure to analyze them.

1. List all the sounds (consonants and vowels) found in the data.
2. Draw the phonetic charts of the consonants and the vowels.
3. Circle those sounds that are phonetically similar in the chart.
4. List the sounds that are not phonetically similar, for the time being they are considered separate phonemes.
5. List ambiguous segments and ambiguous sequences; that is, the segments that can be vowels or consonants and the sequences that can be a unit or a sequence.
6. List the syllable patterns that contain unambiguous vocoids and unambiguous contoids, and the patterns that contain unambiguous sequences.
7. Interpret the ambiguous segments and sequences based on the unambiguous segments and unambiguous sequences (non-suspicious predominant structural patterns)
8. Rewrite the data according to the interpretation.
9. List the suspicious pairs, that is, those which are phonetically similar, or those which are most likely sub-members of a single phoneme, such as:
   [i] and [I] [m] and [n]
   [E] and [e] [n] and [?]
   [�] and [a] [n] and [�]
   [u] and [u]    [w] and [y]
   [a] and [?]    [l] and [r]
   [o] and [?] [c] and [j]
   [p] and [b]
   [t] and [d]
[k] and [g]
   [s] and [z]
   [f] and [v]
   [k] and [x]
   [x] and [?]

10. Decide whether the suspicious pairs are two different phonemes or they are the allophones of a phoneme by:
1. examining the data for minimal pairs of the suspicious pair; if these exist, they are separate phonemes. Two words are a minimal pair when between the words there is only one different sound/phoneme, like [kotak] and [katak]. The difference is only on the second sound, in the first word it is [o] and in the second it is [a].
2. examining the two sounds for complementary distribution, if they exist, they belong to one phoneme. Two sounds are in complimentary distribution when the two sounds never occur in the same position. If one occurs initially, the other never occurs initially, but at the end or in the middle. Or one occurs after nasal, the other never occurs after nasal. Another term for this is mutually exclusive.
3. examining the data for analogous environments of the two sounds. If this is the case, he must check further to determine whether they are one phoneme or separate phonemes by making a hypothesis and trying to refute it (Pike, 1947:74). Two sounds are in analogous environment when between the two words there are only two different sounds, like /badai/ and /gapai/. The differences are the first sound and the third sound, three sounds are the same two are different.
Other things that are necessary to do are:
1. Describe each phoneme based on 1. articulator and point of articulation, 2. manner or articulation, 3. voiced /voiceless (for consonants), 4. the position of the
    tongue, 5. the height of the tongue /lower jaw, and 6. for vowels, the shape of the lips. The analyst should also describe the distribution of phonemes in syllables and words.
2. Describe the process of morphophonemic changes, if any.
3. Describe the clusters that occur in syllables and in words of the language.
4. Describe the syllable patterns of the language, and the distribution of the syllables in words.
5. Propose the orthography of the language based on the phonemes and other factors, such as the national language.
Exercise
Decide whether [x], [?], and [h] are one phoneme or three or two phonemes, and decide whether [m] and [n], and [r] and [n] belong to one phoneme or two phonemes, based on the data below. To solve this problem, you must follow this procedure:
1) collect words containing those sounds (suspicious pairs) by arranging them in three lines, first occurring initially, in the middle, and finally, 2) if the pairs occur mutually exclusive they are one phoneme of some allophones, if they occur in the same position by showing minimal pairs or analogous environment they are separate phonemes.
A. Data of Tobati
Some words of Tobati
[x?n`.sin]     'ash'
[xOh]          'exist'
[`xa.na.tUk]   'child'
[xO�]          'dog'
[x?n.`tEn]     'how'
[`xum.brEk]    'shoulder'
[`x?mbo]      'bad smell'
[s?Ans]        'gnat'
[pAr.`?oi]     'heron'
[r?Or]         'carcass'
['s?Ar]        'rock'
[s?On]         'right'
[ya.`?e.tu]    'stand'
[`wa.?At]      'walk'
[m'?ai]        'big'
[o.?i]         'seed'
[`ma.?au]      'flying fish'
[rEh]           'fence'
[nEh]          'mine'
[`tnEh]        'body'
[`we.nAh]       'bed'
['mOh]         'mountain'
[mtAh]         'owl'
[rwAh]         'road'
[s?.`sEh]       'little brother'
[xrOh]         'little sister'
[nAh]          'stick for digging'

[mAn] 'bird'
[nAn] 'wave'
[mOn] 'woman'
[nOn] 'steal'
[mAs] 'corpse'
[nAs] 'sago'
[mau] 'yawn'
[nau] 'fresh water, not sea water'
[nAr] 'nest'
[rAr] 'inside'
[niu] 'coconut'
[riu] 'outside'
[nOr] 'big white snail'
[rOr] 'yesterday'

B. Data Karo
According the data below, decide whether
1) /t/ and /d/ are a) the same phoneme b) are separate phonemes
   2) /k/ and /g/ are a) the same phoneme b) are separate phonemes
   3) /k/ and /?/ are a) the same phoneme b) are separate phonemes
   4) /a/ and /A/ are a) the same phoneme b) are separate phonemes

/dalu/ 'male'
/kari/ 'later'
/talu/ 'to lose'
/g?rtA?/ 'bridge'
/kuda/ 'horse'
/gari/ 'handcuff'
/tasA?/ 'cook'
/k?rtA?/ 'thin'

English has 24 consonants and 10 vowels and Indonesian has 22 consonants and 6 vowels (see the charts).

TABLE 3
INDONESIAN CONSONANT CHART

A'lator&Place/Manner of Arti Bila-bial Labio dental ApikoDen-tal ApikoAlveo-lar ApikoAlveo P'tal BladeAlveoP'tal Fron-to Pa-latal DorsoVelar Uvu-lar Pharyngeal Glot-tal
Stops     Voiceless     Viced pb td     kg   ?
Affricates     Voiceless     Voiced      cj   x h
Fricatives     Voiceless     Viced f sz �
Nasals m   n   � ?
Lateral    l
Trill    r
Semivowels w      y

Table 4
INDONESIAN VOWEL CHART

Part of the to-ngue raised, form of the lips/ height of jaw FRONTunrounded FRONTrounded CENTRALUnrounded CENTRALrounded BACKunrounded BACKrounded
High i     u
Low High
High Mid e ?   o
Low Mid
High Low
Low   ?

TABLE 5
ENGLISH CONSONANT CHART (Gleason, 1961)

A'lator&Place/Manner of Articulation Bila-bial Labio dental ApicoDen-tal ApicoAlveo-lar ApicoAlveo P'tal BladeAlveoP'tal DorsoVelar Glot-tal
Stops     VoicelessVoiced pb   td   kg
Affricates     Voiceless     Voiced      cj h
Fricatives     VoicelessVoiced fv �� sz �?
Nasals m   n   ?
Lateral    l
Semivowels w   r y

Table 6
ENGLISH VOWEL CHART

Part of the to-ngue raised, form of the lips/ height of jaw FRONTunrounded FRONTrounded CENTRALunrounded CENTRALrounded BACKunrounded BACKrounded
High i     u
Low High I     U
High Mid e ?
Low Mid      o
High Low � A
Low   ?

2.7. Phonotactics
Phonemes of a language have the system of sequence, they cannot combined disorderly. Certain phonemes can only be followed certain phonemes. This system is called phonotactics. Each language has its patterns of syllable. Some languages have simple patterns and some have complex ones. Sequence of consonants in one syllable is called consosnant cluster and the sequence of vowels is vowel cluster (diphthong). Phonotactics covers the sequence of consonant and vowel, of consonant and consonant, vowel and vowel either in a syllable or word. In English for instance, only the phoneme /s/ can be followed by two other consonants for three?phoneme cluster, such as /str/ in structure, strike, /spr/ in spring, /skr/ in scream. It means that three?phoneme clusters occur only where the initial /s/ followed by voiceless stop and followed by /r/.

Canonical form is a sort of generalized phonemic shape in terms of the number of kinds of syllables (Hockett, 1958:264). It can also be defined as the high frequency of phonemic shape of words in a language. This illustrates the general shape of the words. English for instance has two favoured canonical forms, a single stressed syllable, and unstressed second syllable. The examples of the single stressed syllable are girl, sing, jump, and man. The examples of the unstressed second syllable are hammer, settle, bottom, and button.
A syllable is a word or part of a word pronounced with a single, uninterrupted sounding of the voice; unit of pronunciation, consisting of a single sound of great sonority (usually a vowel) and generally one or more sounds of lesser sonority (usually . A syllable consists of an onset, peak, and coda. The peak and the coda can be called the core, so one syllable consists of an onset and a core. The word /beg/ consists of the onset /b/, the peak /e/ and the coda /g/. And /eg/ is the core. A syllable can be open or closed. Open means that it ends in a vowel, and closed means it ends in a consonant. Besides, there are also what we call light syllable and heavy syllable. A syllable with a core which consists of a short vowel which is not stressed is light, and that with a core which consists of a long vowel (V, VV, or VC) or stressed vowel is heavy.
Each language has certain syllabic patterns, Indonesian for instance, has V, VC, KV, and KVK while English has much more complex, since it has also CCV, CCCVC /strAk/'struck', CCCVV /strei/ 'stray', CCCV /stro/ 'straw' etc. A description of phonology should explain the distribution of the syllabic patterns, initial, middle, and final.
When someone is writing a phonology of a language, he should find all the sounds available in the language, then he should analyze the sounds which ones belong to one phonemes and which ones belong to separate phonemes. He must describe the phonemes and their distribution in words and syllable. Next he should describe the syllable patterns of the language and the phonotactic and canonics of the language. To decide the syllable patterns must be based on the unambiguous patterns. Besides he needs to propose the suitable orthography for the language, if it is not yet provided.

References: Payne (1980), Stageberg (1984), Pike (1947),
             Hockett (1955), Hockett 1958, Brewster (1977),
             Gleason (1961), Chaer (1994), Hyman (1971)