Minggu, 16 Desember 2012
Jumat, 22 Juni 2012
SENTENCE COMPREHENSION AND PRODUCTION
CHAPTER II
DISCUSSION
A. Sentence Comprehension
1. What is comprehension?
Comprehension is two common senses. In its narrow sense it denotes the mental processes by which listeners take in the sound uttered by a speaker and use them to construct an interpretation of what they think in the speaker intended to convey. More simply, it os the building of meanings from sounds.
Comprehension in its broader sense, however, rerely ends here, for listeners normally put the interpretations they have built to work. On hearing an assertion, they normally extract the new information it conveys and store that information in memory. On hearing the question , they normally search for the information asked for and compose an answer. On hearing an order or request, they normally decide what they are supposed to do and do it.
The Two Processes
This preliminary tour through comprehension in both of its sense suggests that it devides naturally into two somewhat distinct areas of study. The first will be called the construction process.it is concerned with the way listeners construct an interpretation of a sentence from the speakers’ words. The second area of study will be called the utilization process. It is concerned with how listeners utilize this interpretation for further purposes-for registering new information, answering question, following order, registering promises, and the like.
It would be a mistake, however, to think of the contraction and utilization process as truly separate. People listen because they want cooperate with speakers-to register the information they are offering, answer their questions, and carry out the request. Their goal is to discover how they are expected to utilize speakers’ sentences. This goal should motivate and guide comprehension from beginning to end-from identifying words through building interpretations to utilizing those interpretation.
2. The Contruction Process
a. Underlying Representations
Propositions were shown to consist of a verbal unit plus one or more nouns. In the basic types shown in Table 2-1, each proposition is expressed as a verbal unit (walks, is handsome, is a bachelor, hit, is in, and gave) plus one or more nouns (John, Bill, Paris, Fido and Mary). To be more precise, each of these propositions is a predication about one or more entities.
TABLE 2-1
Propositional Functions
Examples of Six Simple Sentences and Their Corresponding Propositional Functions
| SIMPLE SENTENCE | SENTENCES WITH VARIABLES | PROPOSITIONAL FUNCTIONS |
| John walks. John is handsome John is a bachelor John hit Bill John is in Paris John gave Fido to Mary | x walks x is handsome x is a bachelor x hit y x is in y x gave y to z | Walk(x) Handsome(x) Bachelor(x) Hit(x,y) In(x,y) Give(x,y,z) |
entities : walking, handsomeness, bachelorhood, and the like. The proper nouns John, Bill, Paris, Fido and Mary can just as well be replaced by variables such as x, y, and z, as shown in the middle column in table 2-1. From there it is a short step to denote what is predicated as logicians might, as propositional function, which are shown in the right-hand column in Table 2-1.
Propositions
Propositional functions have an important notational advantage. Once the arguments of a propositional function have been filled in, the result is a proposition. Walk(John), for example, is a proposition that specifies all the information, except tense, that goes into the sentence John walks. propositions are very different from sentences. Whereas the sentence John walk is something one can pronounce, the proposition Walk(John) is not. Walk(John) represents meaning of John walks and nothing more. To put in another way, whereas there is temporal significance in the fact that John comes first and walks second in the sentence John walks, there is no temporal significance to the order of walk and John in the proposition Walk(John). The order of the arguments in Give(John, Fido, Mary) spesifies only how they function in the action of giving. The first argument John, is agent; the second, Fido, is the object; and the third, Mary, is the indirect object.
b. Constructing Interpretations
In the construction process, listeners must take a linear string of words and from it construct a hierarchical arrangement of propositions. They would be helped tremendously in building the propositions if they could work out the surface structure of each sentence first.
Constituents
The surface structure of a sentence, on the other hand, divides up into phrases and subphrases called constituent. They are implicit in the function and order of the words. Constituents are potentially very useful. Once listeners have identified a constituent, they can decide on its underlying propositions with little or no reference to the rest of the sentence.
If listeners are to have enough time to build proposition from each constituent, however, they must retain in memory its phonologicalrepresentation-its verbatim content. Constituents large than single words are generally complete at the ends of noun phrases, clauses, and sentences. So once listeners have built the underlying propositions of a constituent these point, they could safely purge their memory of its verbatim content. No longer useful, that content will probably only interfere with the next batch of words put into memory. It is a reasonable hypothesis, then, that listeners’ verbatim memory for a constituent should fade rapidly after it has been processes.
Preliminary Outline
These preliminary remark suggest that listeners construct the underlying representation for a sentence in roughly four steps:
1) They take in the raw speech and retain a phonological representation of it in “ working memory “
2) They immediately attempt to organizethe phonological representation into constituents, identifying their content and function.
3) As they identify each constituent, they use it to construct underlying propositions, building continually on to a hierarchical representation of propositions.
4) Once they have identified the propositions for a constituent, they retain them in working memory and at some point purge memory of the phonological representation. In doing this, they forget the exact wording and retain the meaning.
3. Surface constituent
a. The conceptual unity of constituent
Constituent have a clear consceptual reality. The definition alone attest to this. A group of words to be replaceable by a single word with an equivalent function, it must have a conceptual coherence. Levelt presented people with sentences such as The Boy Has a dollar. People were to judge which two words, (boy and a) it were most connected in the given sentences. And (boy and a) which at least closely connected.
b. Constituent as Aids in Perception
Listeners may immediately try to isolate the constituent in the speech they hear. That should be greatly aided. To pick an obvious device, pause between each major constituents should help considerably. Few psychologists have studied whether isolating constituent by pauses help listener or not. Gray and Torrey (1966). Isolated constituents for readers and found that it help. They presented people with a prose passage in a machine that exposed the print line by line, with each line visible for only a fixed periode of time.
c. Constituent in working memory
Listeners have isolated constituent, they should hold them verbatim as constituents in working memory until they have no more need for them, until they have used them to construct the underlying propositions. There is relatively good evidence that a sentence is divided into constituents in working memory. People would listen to a tape-recording of a sentence such as : “the polite actor thanked the old woman who carried the black umbrella”, immediately after they would hear a “probe word” e.g polite and be asked to give the next word if the probe and its response are from the same constituent
d. Constituent in the construction of proposition
When sentences are divided into constituent by pauses or printed lines, listeners appear to understand more easily because these breaks help them do what they eventually do anyway. That are isolate and identify each constituent.
Bever and Garet have argued in an extensive series of studies that constituent play a role in the perception of speech. They argued that in perception genuine units tend to preserve their integrity by resisting interruption from other sources. So, if the constituent is a basic unit in speech perception, it should resist interruption by a click from an extraneous auditory sources. Imagine that listener while attending to speech in one ear, incidentally hears a click in the other. If the click falls at a constituent b=u74ndary, it doesn’t interrupt the constituent either before and after it and should be heard as occurring at that boundary. But if it comes in the middle of a constituent, it should be perceived as falling nearer one of the constituent boundaries than it actually does, since the constituent tends to resist interruption.
e. Real time processing
Listener have limited capacity for processing what they hear in the time available. Their ability to deal with speech breaks down if it too fast, in a language they know imperfectly, or in competition with another conversation. Their limited capacity shows up even in normal speech, for sentences are quite uneven in the demands they make. Imagine hoe listeners build up a semantic representation for “the old man lit his awful cigar”. They take in and isolate the first constituent, the old man, but while trying to construct propositions from it, they have to take in the reminder, lit his awful cigar. If for some reason they find it difficult to construct propositions from the old man, their ability to take in lit his awful cigar will suffer. They may delay isolating the constituents from “lit his awful cigar” until they are finished with “the old man”.
There is sample evidence that a difficulty I processing at one point in a sentence slow down processing immediately after that point. People were asked to listen to a sentence, knowing they would be responsible for what it meant afterwards, and simultaneously listen for the sound “b” in one of the words. They were to press a button as quickly as possible when they heard the “b”. if processing was difficult immediately before the “b”, they should have taken longer to detect the “b” than otherwise.
4. Syntactic Approaches To The Construction Process
In the syntactic approach, listeners are assumed to use the surface features of a sentence in coming to its interpretation. they identify sounds, words, and larger constituents and from them build and connect propositions in an interpretation for the whole sentence.
In the semantic approach; on the other hand listeners are assumed to work from the interpretation a sentence must be conveying. They then actively search for sounds, words, and constituents that satisfy these assumptions and expectations.
The older and more thoroughly investigated of these of these two approaches is the syntactic, so it will be taken up first. One reason for its appeal is its emphasis on the goal-directed nature of comprehension-listeners comprehend in orde to discover how they are meant to use a sentence.
Building Up Constituents
The major proposal in the syntactic approach as developed by Bever (1970), fodor and Garrett (1967), kimball (1970b) and others, is this. Listeners have at their command a battery of mental strategies by which they segment sentence into constituents, classify the constituents, and construct semantic representations from them.
As listeners identify constituents, they must not only locate them, but also implicitly classify them – as noun phrases, verb phrases determiners and the like.
They identify the word as the , classify it as a determiners (det), and note that it is the first word in a noun phrase (NP). They further note that this NP must be the first constituents of a clause or sentence.
They can therefore set up three labeled constituents, each one inside the next, like this :
[[[the]det… ]NP… ]s
Here each constituent is encloses in “labeled brackets” instead of the simple unlabeled parentheses used so far. The right-hand bracket around a constituent has a subscript that labels the constituent.
So [..] det is a determiner
[…] NP is a noun phrase, and […] s is a sentence or clause.
Use Of Function Words
Function words – determiners, preposition, conjuction, pronouns, quantifiers, and the like – may play a crucial role in the strategies for segmenting speech into constituens.
Kimbal (1973) has proposed the following strategy :
Strategy 1:
Whenever you find a function word, begin a new constituent larger than one word.
This strategy reflects the observation that function words are very reliable clause to constituent structure because in English each one signals the beginning of a major constituent. The preposition to, at, in, on, etc. mark beginning of preposition phrases, not noun phrases, verb phrases, or mecessarily sentence.
Function words usually signal not only the beginning of a new constituent, but also the end of the previous one. In the man was leaving, was signals the beginning of a verb phrase and the end of the earlier noun phrase thus :
[[[the]det [man]N]np[[WAS]AUX…]vp]S
Sometimes function words signal the presence of other larger constituents as well. in the man in the parlor, the preposition in tells the listener to close off the man as a noun phrase an begin a prepositional phrase, like this :
[the man] NP [[in] prep… ] pp
But it also signal the information of a larger noun phrase encompassing both major constituents formed so far, like this :
[[the man] NP [[in]prep… ] pp ] np
In rare cases, a function word does not mark the end of the previous constituent.
Leving is a so called discintinous constituent.
Evidence For Strategy 1
Fodor and Garrett (1967) noted that some function words in English are optional. Compare 14 and 15.
14. the pen which the author used was new
15. the pen the author used was new.
These two sentence have roughly the same meaning; it is just that the relative pronoun which hs been deleted in 15. They argued that this delection should impair comprehension.
Which in sentence 14, they know by strategy 1e that they are beginning a clause.
In sentence 15 this should delay its identification and the construction of its identification and the construction of its interpretations. The prediction is clear. Sentence 15 should be more difficult to take in than sentence 14.
Anticipating Content Words
Function words should also help listeners classify neighboring content words. On encountering a content word - according to the syntactic approach – listeners must identify its function and place it within the appropriate constituens.
Most content words cannot be identified unambiguously.
Example : Man : noun (the man)
Verb (to man a ship)
Strategy 2:
After identifying the beginning of a constituent, look for content words appropriate to that type of constituent.
Example :the marks the beginning of a noun phrase, it should lead listeners to expect a noun that “heads” the noun phrase. E.g. man in the old man.
Strategy 1, strategy 2 consists of a series of specialized strategies.
Use Of Affixes
Listeners may also be helped by suffixes and prefixes. Most such affixes can be added to only one or two kinds of content words. They can be used to pick out what kind of content word a word is.
Example : -ing, and –ed mark a word as a verb; -ly usually marks a word as an adverb; and –y, -ic, -al, and –ive usually mark words as Adjective.
Suffixes that mark concrete and abstract nouns include –tion (descrition) –ity (activity) –ness (rudeness), and –er(believer). The siffix –s marks nouns as plural (the bears sleep) and verbs as singular (the bear sleeps).
Strategy 3:
Use affixes to help decide whether a content word is a noun, verb, adjective, or adverd.
Use Of Verbs
Content words themselves limit what can occur around them.
Example, the verb often specifies whether there should be one, two, or three noun phrases with it, as in the following sentence :
1. The man slept
2. The man hit the ball
3. The man put the dog into the house
Slept in 1 requires only one noun phrase – its subject – though it may have more. Hit in 23 requires at least a subject and an object, two noun phrases. Put in 3 requires a subject, na object, and a location, three noun phrases.
Strategy 4:
After encountering a verb, look for the number and kind of arguments appropriate to that verb.
Fodor, Garrot and Bever (1968) noted that verb also limit the syntactic form of the object, or complements, they can occur with.
Example : requires a noun phrase that is not a complement :
1. John hit [the doctor]NP
On the other hand, belived allows in addition several types of complements :
2. John believed [the doctor] NP
3. John believed [that the doctor had left]s
4. John believed [the doctor to have left]s
Hit, listener can restrict theis search to noun phrases that are not complements, but on encountering believed, they cannot. fodor and his associates found evidence that sentences like 2 were more difficult than ones like 1.
By the syntactic approach, Strategy 1 through 4 seem not only sensible, but even necessary if listeners are to make full use of syntactic information.
Memory Capacity
Some sentences seem intrinsically difficult to understand :
1. I gave the dream I had cherished all my life about living in ullapool during herring season up.
2. The squid the blowfish the shark ate ate’ tasted awful
3. The man offered a hundred dollars for the bottle of 1962 vintage Mouton Rothschild is my uncle.
In 1, the discontinuous constituent gave up is interrupted by too much speech. In 2, there are too many confusing self-embedded relative clauses. And in 3, by the time the listeners research is it is too difficult to infer that who was had been deleted from between the man and offered. What three sentences have in common is that they tax the listener’s memory.
Strategy 5:
Try to attach each new word to the constituent that came just before.
When this strategy works, listeners are saved from searching memory any further back than the least constituent. It is affects only how listeners deal with discontinuous constituents.
To see how strategy 5 works, consider :
1. The dog that was rabid came from Ney York
2. The dog came from New York that was rabid
3. The dog bit the fox that was rabid.
Imagine the dog that was intended to be rabid. In 1, the relative clause that was rabis is directly preceded by the dog, so strategy 5 immediately gives the intended interpretation. In 2, listeners first try to attach that was rabid to New York , reject that on semantic grounds, and then attach it to the dog. In 3, listeners try to put that was rabid with the fox, find it makes sense, and never arrive at the intended meaning with that was rabid attached to the dog.
Two Principles Of Memory Capacity
Kimball also proposed a “principle if fixed structure” and a “principle of two sentences.” According to the first, once listeners have parsed speech into constituents one way.
This principle explains why 1 and 2 are so hard.
1. The horse raced past the barn fell
2. The dealer offered two dollars for the painting refused to sell.
In 1, listeners are led by strategy 7 to identify [the horse raced past the barn]s as a clause. But fell isleft over, they have to go back in memory and repearse the sentence as [the horse [that was raced past the barn] fell].
In 2, listeners get into the same trouble and have to repearse it as [the dealer [who was offered two dollars for the painting] refused to sell].according to Kimball’s principle of two sentence, it is very costly in memory capacity to parse more than two constituents labeled S at a time.
Clause
Although strategy 1 can be used to isolate and classify a constituent as a clause, there are probably additional strategies for identifying just how that clause function in the current sentence.
Strategy 6:
Use the first word (or major constituen) of a clause to identify the function of that clause in the current sentence.
Strategy 7:
Assume the first clause to be a main clause unless it is marked at or prior to the main verb as something other than a main clause. This strategy distinguishes main from non-main clause.
Historical Evidence
In arguing for strategt 7, Bever and Langendoen (1971) appealed to the history of relative clauses in English.
Building Underlying Propositions
Syntactic approach, once listeners have isolated and labeled constituents, they are still no through. They must use these labeled constituensts to construc an interpretation for the sentence – a hierarchy of underlying propositions.
Evaluation Of The Syntactic Approach
The syntactic approach has it good and its bad points. Its strongest appeal is that it accpunts for certain difficulties that arise from syntactic sources-from self-embedding, missing relative pronouns, discontinuous constituents, and the like.
Problems
The syntactic approach has problems in principle. In this approach listeners are assumed to make heavy use of function words, and prefixes and suffixes.
5. Semantic Approaches To The Construction Process
a. The reality and cooperative principles
The reality principle is concerned with the substance of a sentence,the ideas being talked about,and the cooperative principle is concerned with the way these ideas are expressed.Listeners use the cooperative principle to interpret sentences in the belief that the speaker is trying to tell the truth,and use sentences clearly and unambiguosly.
The consequences of the cooperative principle are potentially for reaching too,for assumptions help listeners come to the interpretation the speaker intended.
b. Making sense of sentences
The listeners assume,in line with cooperative principle,that speaker intend their sentences to make sense.
For example : if listeners had heard the words flower,girl,picked and red without any other syntactic information,they could guess that two propositions were involved: the girl picked the flower and the flower was red.
They could make the same guess on hearing The flower the girl picked was red and then attempt to confirm the guess by finding constituents to possible constituents.
Stolz (1967) ,by asking people to parapharase doubly self-embedded sentence of two-kinds:
v The vase that the agency hired dropped broke on the floor.
v The dog that cat that girl fought scoled approached the colt.
hese sentences,however so difficult that they are treated more as jigsaaw puzzle than as sentences.
Storhen and Nelson (1974) found that children aged two and three intrepreted the sentences:
The simpler sentences:
· The cat chased the mouse
· The mouse was chased by the cat
· The mouse chased the cat
· The cat was chased by the mouse
These children apparently assumes very strongly taht what they were told ought to make sense – cats chase mice and vice versa. They gave asumption precedence over any syntactic information. By ager four or five,children interpreted all four sentences correctly taking account of the syntactic information.
A demonstration by Fillenbaun (1971,1974) brings home the reality principle the most clearly of all. In one study people were given sentences to pharaphrase intheir own words. Among them were “ perverse” sentences like :
· John dressed and had a bath
· John finished and wrote the articled on the weekend
· Don’t print that or I won’t nsue you
Normalized the pervese sentences
· If you print that, I’ll sue you
Anticipating Constituents
Listeners probably have specialized strategies for searching for constituents,especially for anticipating ones to come. Fodor (1971). Schank (1972), winograd (1972) and others, for example, have suggested that listeners center requitments. Put,for example, requires three noun phrases an agent to do the placing.an ob ject to be placed,and location to receive the object.
· Jhon put the book on the table (expresses the three place proposition) Put ( john, book , table)
The assumption is that listeners can quickly identify these words, determine their propositional functions,and narrow in on constituents that fit the missing arguments.
c. Tying Sentences to Context
Listeners are especially anxious to make sense of a sentence in the circumnstances they are in at the time. From the cooperative principle they assume that the speaker has made his utterance relevant to use the ongoing discourse.
For example, that the speaker has used definite noun phrases like the general to refer to enties they know.
For illustration, imagine that someone had said A followed immediately by B:
A. Claire and Kent climbed Mt. McKinley last summer
B. She photographed the peak,and he surveyed it.
d. Use of Word Order
In Lingustics, word order typology refers to the study of the order of the syntactic constituents of a language,and how different languages can empploy differnt orders. Correlation between orders found in different syntactic subdomains are also of interest. The primary word orders that are of interest are the constituent order of a clause – the relative order of subject,object, and verb,the order ofd modifiers (adjectives,numerals,demonstratives,possessives,and adjuncts) in a noun phrase,and the order of adverbials.
The order of words and clause within sentences has considerable constituency,both in English and in may other languages,and listeners could make good use of that consistency.
B. Sentence Production
As in the discussion about utterance production, the utterance is produced in three phases. There are conceptualization phase (planning of what will be said); formulation / grammatical encoding phase (giving categories to the mental lexicon) and articulation phase (the frame is formed to be a sound).
| Breatng |
Pause and Tongue Slips
Based on the picture, there are two kinds of thing used to conclude the mental process when we are uttering something. Those are pause and errors. Then, there are two kinds of error which are error of tongue slip and error of Aphasia.
a. Pauses
An ideal utterance process is implemented in a fluent utterance, since the utterance is begun until the end of utterance. However, we can’t always make such ideal utterance. Generally, the harder topic of speaking, the more number of pauses will retrieve.
1) Kinds of Pauses
People usually pause for a while, whether for breathing or for the other necessaries. There are some reasons why people pause. First, people pause for a while to seek for the ‘missing’ word to continue their utterance which has been started. Then, they pause for a while to seek for the ‘forgotten’ word to continue their utterance. And the last, they try to be aware of public or listener effect of their utterance.
Therefore, there are some kinds of pause as follows.
· Silent pauses; speakers stop and they are silent. They continue after finding the searched words. E.g. There is …. (behind your room).
· Filled pauses; e.g. There is …. Ghost behind your room. From this example, the speaker pauses for a while to retrieve the word ‘ghost’.
This pause can be filled by some words, like oh, ah, well, say or some expressions, like that is, or rather, I mean and well.
2) Place of Pauses
Hesitation pauses doesn’t locate in haphazard place. There hasn’t been a convention about the precise location. However, there some places that are agreed by the experts (Clark & Clark 1977: 267), those are grammatical juncture, another constituent border and before the first word in the constituent.
Grammatical juncture is a place to plan both frame and first constituent of the sentence that will be uttered. In the border of one constituent and the other one, people can also pause, because people plan the detail of the next constituent here. For the third place, people usually pause before the first word in the constituent.
b. Errors
Error in speaking can be caused by tongue slip and Aphasia disease.
1) Tongue slips
Tongue slip is a phenomenon in utterance production in which the speaker’s tongue slips, so that the words produced aren’t the meant words. There are two kinds of tongue slip, error of selection and error of assembling.
a) Error of Selection
There are three kinds of error of selection. Those are error semantic selection type, malapropism and blends.
In error semantic selection type or Freudian slips, people retrieve the words that obviously aren’t the wanted words. Errors of semantic selection generally occur in the whole words which come from the same semantic field. E.g. carrot and cabbage.
Malapropism is viewed as a woman who wants to be seemed as a high-class one by using elegant words. However, the form of words is similar, but it is an error. E.g. allegory with alligator, ravishing and ravenous.
The last is blends. Blends appears if people are in a hurry and they take one or half syllable from the first word or the other half of the second word. Then they become one part. E.g. slightest and least become sleast, explain and expand become expland.
b) Error of Assembling
Error of assembling is a form of error in which the chosen words are right, but their assembling are wrong. One of these forms is transposition case. In this form, people move the word or sound in one position to another position.
E.g. I need a gas of tank → I need a tank of gas
I caked my bake.
The second type is anticipation error. Speaker anticipates that it will appear a sound, then that sound is uttered as a replacement of the exactly sound.
E.g. Bake my bike.
The last is perseveration error. It’s also called repetition, and it is the opposite of anticipation. In anticipation, the error occurs in the beginning, but in the perseveration, error occurs in the end.
E.g. pulled a tantrum → pulled a pantrum
2) Aphasia
Aphasia is a disease of speaking in which people can not speak well because there is a disease in they brain. This disease generally appears because the people get a stroke, that is a half of their brain is lack of oxygen so that that part is flawed.
3) Units in Tongue Slip
The units of tongue slip are distinctive feature, phonetic segment syllable, word and constituent that is larger than word.
a) Error of Distinctive Feature
It happens when the part that slips is not a phoneme, but it’s only in the distinctive feature of that phoneme. However, this error rarely occurs.
E.g. clear blue sky → glear plue sky
b) Error of Phonetic Segment
This error is more common to happen. It’s caused by the number of feature is more than one.
E.g. with this ring I thee wed → with this wing I thee red
left hemisphere → heft lemisphere
c) Error of Syllable
In this case, an exchange almost occurs of first consonant in one syllable and first consonant in another syllable.
E.g. harp-si-cord → carp-si-hord
a-ni-mal → a-mi-nal
d) Error of Word
This error occurs if the exchange of place occurs to the word.
E.g. Tank of gas → gas of tank
Go for broke → broke for go
2. Tip of the Tongue and Latah
People sometimes don’t remember one word that may not be used in a long time. This symptom is called tip of the tongue (lupa-lupa ingat). There are some patterns of this symptom which is usually followed by people. Those are:
· The number of syllable is always true
· The beginning sound is also true
· The final result of error is similar with the exactly word
Another unique symptom is latah. This is a linguistic act in which someone spontaneously and unconsciously says some words when she is surprised. The characteristics of latah are:
· Exist in southeast Asian
· The doers are almost women
· The words that are said spontaneously and unconsciously often related to the sex
· Can only repeat the only one word
Those four characteristics have never been researched so far, and they need to be examined one day.
3. The Uttering Process
After all of the processes of conceptualization have been done, then it’s time for articulation. Articulation refers to the process of how make the utterance into the sound that will be comprehended by interlocutor.
The steps of how to articulate the utterance can be described in this example. Say, there is a Holland man that sees a cigarette and wants to say the word for that thing. The lexical access will be started from the concept construction for this thing that will be symbolized CIGAR (X) till the phonetic form /siɸar/. These are the steps.
· Lemma retriever receives concept CIGAR (X) and retrieves the syntactic category “nominal”, non-neutral gender, and singular form.
· Then, the “bundle” is sent to word-form encoding. In morphological encoding phase, the things that related to the morphology, such as singular form and gender category, are processed become morpheme /siɸar/ not /siɸare/
· In phonological encoding phase, this morpheme is formed to be its sound, its syllable, and its word stress.
· Result of phonological encoding is sent to the part of phonetic encoding to be processed of its phonetic features.
· The result of it is sent to the last phase that is articulation program.
4. Sentence Articulation
After the words of sentence candidate have been processed and will be uttered, the area of Broca commands the korteks motorik to “start working”. If the first word in the sentence that will be uttered is rokok, so the korteks motor will give instruction such as:
· Vocal cord, be ready for vibrating!
· Tongue, adhere your tip to the alveolar and vibrate several times!
· Vocal cord, together with the tongue vibrates, you vibrate!
· Uvula, adhere to the throat in order the air doesn’t go out through the nose.
The description above actually doesn’t reflect the actual circumstance. That description apply that every sound of the words seem as a whole sound and it’s not influenced by the sounds around them. In fact, the quality of sound is changeable influenced by the sound in front or behind it. In other word, the articulation process for sound is depended on the apparatus condition when we are uttering that time. The speed of utterance also influences this process. The faster someone speaks, the less time he own to process all of the instructions.
5. How Errors Occur
Due to the speed of utterance or due to other reasons, words or sentences can be uttered in error way. Fromkin (1973) describe it in the following example.
| Actual utterance A weekend for MANIACS | Uttered A maniac for WEEKENDS |
From the example above, the capital letter has primary stress and the Italic word has secondary stress. The steps are as follows.
· Speaker decides to make constituent that refers to the time.
· Speaker builds the syntactic skeleton to this constituent.
{Indefinite article + nom +(2nd stress)+ prep + nom + plural +(1st stress)}
· Speaker then accesses the needed lexical forms, which are weekend and maniac to be a nominal column. However, he makes an error, and makes them in the wrong column.
{Indefinite article + maniac + (2nd stress)+ prep + weekend + plural + (1st stress)}
· Speaker form those words in to a form of sound with its affix and stress.
A + maniac + for + WEEKEND + [z]
· Each word is specified according to the distinctive feature of each.