EXPLORING THE USE OF MODAL AUXILIARY VERBS IN CORPUS OF CONTEMPORARY OF AMERISCAN ENGLISH (COCA)

Abstract. This paper deals with the frequent use of modal auxiliary verbs in Corpus of Contemporary of American English (COCA). The modal auxiliary modal verbs mentioned as the data are can, could, will, would, may, might, shall, should, and must. Each data collected are taken from Corpus of Contemporary of American English (COCA) in https://corpus.byu.edu/coca/. The research method is descriptive analysis; the present writers describe every data related to their function in each clause. The uses of corpus-based findings of this study have shown several valuable insights. Firstly, the frequency and ranked order of modal auxiliary verbs found in the Corpus of Contemporary American English (COCA) has been revealed.  From the nine modal auxiliary verbs that classified into ‘central modals’, it is collected 6,008,840 data. From the total number mentioned, the verb would (1,301,269) is the most frequently used, then it is followed by the verb can (1,228,608), will (1,118,515), could (885,835), may (494,144), should (435,903), might (300,258), must (223,936), and shall (20,372). Secondly, this study has revealed the elements following each of modal auxiliry verbs found in the Corpus of Contemporary American English (COCA). This is the important information on the use of modal auxiliary verbs and their structure in real language.Abstrak.  Tulisan ini mengenai frekwensi penggunaan verba bantu modal pada Corpus of Contemporary of American English (COCA). Verba bantu modal yang digunakan sebagai data adalah can, could, will, would, may, might, shall, should, dan must. Setiap data yang  digunakan, dikumpulan dari pada Corpus of Contemporary of American English (COCA) pada  https://corpus.byu.edu/coca/. Metode penelitian yang digunakan di dalam riset ini adalah metode deskriptif; para peneliti menggambarkan setiap data yang berhubungan dengan fungsi pada setiap klausa. Penggunaan korpus di dalam penelitian ini menunjukkan beberapa hal. Pertama, menggambarkan frekwensi  dan urutan penggunaan verba bantu modal di dalam Corpus of Contemporary of American English (COCA). Dari sembilan verba bantu modal yang dikelompokkan ke dalam ‘central modals’ terkumpul 6.008.840 data. Dari keseluruhan data yang terkumpul, verba bantu would (1,301,269) sebagai verba bantu yang paling sering digunakan, diikuti oleh verba bantu modal can (1,228,608), will (1,118,515), could (885,835), may (494,144), should (435,903), might (300,258), must (223,936), dan shall (20,372). Kedua, tulisan ini juga mendeskripsikan unsur-unsur yang mengikuti verba bantu modal yang digunakan di dalam Corpus of Contemporary of American English (COCA). Hal ini merupakan informasi penting mengenai penggunaan verba bantu modal dan strukturnya di dalam pengunaan bahasa yang sebenarnya. 


INTRODUCTION
Many researchers do their research on Linguistics in most languages including English. It is limited research on specific Linguistic topic such as verb, especially, in modal auxiliary verbs. It is reported that Verstraete (2001), Hsieh (2005), and Mukundan & Khojasteh (2011) did their research on modal auxiliary verbs. Verstraete's research discusses the system of Modal Auxiliary in English. He concerned the difference of subjective and objective modality on English modal auxiliaries. Hsieh did his research on Chinese language concerning modal verbs and modal adverbs based on Semantic source.

Mukundan & Khojasteh's research is on Malaysian
English textbooks concerning modal auxiliary verbs. They found that the distinctions ofEnglish language textbooks and real language use. Besides the three previous research mentioned, Kennedy (2002) did his research related to British National Corpus (BNC) as the source of the classification of modal verbs. He did his research to identify the distribution of modal verbs in it. He offered nine different types of pattern consisting modal verbs and he argued that the nine different patterns found in British National Corpusrelated to 14 different types of modal verbs. Further he concluded that the modal verb will accounts for almost 23% of all modal tokens, followed by would, can, and could. It is found thatthe modal verbcan frequently found in spoken texts. From the previous research described, many researchers do their research related on modal auxiliary verbs, but it is only one research on the frequent use of modal auxiliary verbs in English, especially in British National Corpus (BNC). This paper is a preliminary research, and the aims of the research are to describe the frequency and the order of the rank of modal auxiliary verbs found in the COCA and to describe the elements following each modal auxiliary verbs found in the COCA.
Based on the two aims of the research mentioned, the present writers collected and identifiedthe data taken from the COCA in https://corpus.byu.edu/coca/. The data taken are limited on modal auxiliary verbs; they are can, could, will, would, may, might, shall, should, and must later called central modals. In Quirk et al. (1992), followed by Biber et al. (1999) and also mentioned by Kennedy (2002) that the 'central' modals are considered to be will, would, can, could, may, might, shall, should, and must or sometimes called as auxiliary modals. In line with Quirk et al. (1992), Kennedy (2002) mentioned in the nine central modals is found a small group of 'marginal modals'. The marginal modals areought to, need to, used to and dare which can act like modals and sometimes seem like main verbs. There is another group of expression known as 'semi-modals', which function like modals -(had) better, (have) got to, be about to, be going to, and be supposed to. This present paper only focuses on the nine 'central modals ': will, would, can, could, may, might, shall, should, and must, and excludes both 'marginal modals' and 'semi-modals'.

RESEARCH METHOD
The research method applied in this research is descriptive analysis. The corpus that is referred as the dataare obtained from COCA. It is reported in https:// corpus.byu.edu/coca/ that this COCA corpushasabout 560 million wordsof text and 20 million words whichwere collected from the year 1990-2017.The corpus aredivided into spoken, popular magazines, newspaper, fiction, and academic texts.
The data obtained in this present paper are limited on the clauses containing modal auxiliary verbs and every data collected is described one by one. The each data has different function from one to another, since it has different structure of each.
The modal auxiliary verbs used in this research are nine modal auxiliary verbs which categorized into 'central modals' as keywords. The nine keywords as the modal auxiliary verbs (will, would, can, could, may, might, shall, should, and must)are inputted one by one to application of COCA in https://corpus.byu.edu/coca/. Jones & Walter (2015) offered the definition of corpus, they argued that corpus are searchable collection of texts electronically stored while the text could be spoken or written in various length, although generally, it would be longer than a single utterance or a single written clause. The corpus could be measured by the number of words or the number of tokens found in every utterance or written clause.
After collecting the data from COCA, the present writers identified 6,008,840 corpus one by one to classify the types of verbs and the element following the verb. Every keyword, as one of modal auxiliary verbs, is identified of each function in each clause. Then the collected corpus are presented as the data of the present research. After identifying, analyzing, and describing the data, the present writers counted and compared the frequent use of the modal auxiliary verbs in COCA. Then the present writers described each function of every keyword and describe the element following the verb.

English Modal Auxiliary Verb
Verb is the important parts in some languages including English. In the English clause, it always has a verb since the verb is the main part of clause. So that, every clause in English has, at least, one verb. The one verb in a clause is called main verb (sometimes called full verb or lexical verb) while the two verbs or more could be a main verb and the othersare auxiliary verb(s). The auxiliary verb could be modals auxiliary verbs and primary verbs or sometimes called as semi auxiliary verbs as described by Quirk et al. (1992) that verbs is one of the class of words. Related to the function of each, the verbs could be classified into three different major categories. The three categories are named full verbs or is known as lexical verbs, the second one is primary verbs, and the last is modal auxiliary verbs. The full verbs or lexical verbs in English are such as buy, make, and go while the primary verbs are be, have, and do (the three verbs could be full verbs as be in They are singers, have in They have a car, and do in They do it). The modal auxiliary verbs as the focus of the present research are will, would, shall, should, can, could, may, might, and must. To understand the definition deeper, in line with Sujatna (2012) who described the classification of the verbs above in the following Picture 1. Kennedy (2002) argued that verbs constitute about 20 per cent are in written texts and about 8 per cent of all verb forms are typically constitute modal verbs. Modal auxiliary verbs have the important role in clauses, including English clauses, in line with Manaf (2007) in Mukundan & Khojasteh (2011). Mukundan & Khojasteh (2011) reported that Manaf (2007) argued that one of the important part of grammar and semantics of most languages including English is modal auxiliary verbs. He added that modal auxiliary verbs contribute to the semantics communication.
To understand more about modal auxiliary verbs, the present writers describe the characteristics of modal auxiliary verbs as argued by (Watkins, 1990); (1) go before not in negative sentence (e.g. He would not come.), (2) go before the subject in interrogative sentences (e.g. May I go home?), (3) go before forms of be and have in compound tenses (e.g. He may be coming or He may have come), and (4) do not have Infinitive or Gerund forms (we cannot say to must or *mighting). Mukundan & Khojasteh (2011) in line with Biber et al. (1999) explained that some particular modal verbs show dissimilarity for their combinations since the majority of modal verbs take place with marked voice or aspect. They also reported that the modal auxiliary verbs can, could, should, and must are commonly found in passive construction while may, might, should, and must are the most commonly used modal with perfect aspect construction.
Besides Biber et al. (1999) and Mukundan & Khojasteh (2011) who reported the usage of modal auxiliary verbs, the following Table 1 is the modal verb phrase structure mentioned by Kennedy (2002) in British National Corpus (BNC).
Modal auxiliary verbs have own function of each modal in every sentence, in this paper, English sentences as Biber et al. (1999) mentioned that according to their meaning modals and semi-modals could be classified into three different major categories. The modal auxiliary verbs can, could, may, and might express permission/ possibility/ ability while the modal auxiliary verbs must and should express obligation/ necessity, and modal auxiliary verbs will, would, and shall express volition/ prediction.

Distribution of Modal Auxiliary Verbs in the COCA
Modal auxiliary verbs discussed in this present paper is 'central' modals; they are nine central modal verbs. The central modal verbs are will, would, can, could, may, might, shall, should, and must. From the collected data, totally the present writers found 6,008,840 data. The examples of the data are described in the following. non-intentional offenses as abandoning a regime of differential punishment would almost always lead to a significant reduction in punishment for offenders of nonintentional crimes

Picture2. The usage of modal auxiliary modal in COCA
From the total number mentioned, the following is a description of the distribution of the nine modal auxiliary verbs in COCA as represented in the following Picture 3. It should be done.
They will be arriving soon.
He might have gone. You must have been hungry. It might be being sold tomorrow.
He could be being disturbed.
It should have been written.
He must have been lying.
He might have been being blackmailed. They must have been being careless.

(i) Modal Auxiliary Verb: would
Verb would is also one of the modal auxiliary verbs and it is reported that it is the most frequent use in COCA. From the data collected, it is found 1,301,269 data from the total 6,008,860 data. The present writers figure out the elements follow the verb would as described in the following.

(ii) Modal Auxiliary Verb: can
Verb can as one of the nine modal auxiliary verbs found in the data as the second frequent use in COCA. The number of the verb can usage is 1,228,608 data taken from the 6,008,860 data as the total number of data found in COCA. The following are the examples of the data containing modal auxiliary verb can and the elements follow the verbcan.
can + V can + Adv + V can + not + V ... differential punishment in many circumstances. # Therefore, to the extent that differential punishment can be justified at all, it can only be justified in reference to these.... #By punishing criminal offenders, the state can simultaneously accomplish two forms of deterrence.16 First, punishing an offender.... ... and her conduct-rather than on the occurrence of harm outside of the offender's control-they can not provide adequate justification for the practice of differential punishment.
The examples of the data cointaining modal auxiliary verb can above shows that the verb can only has three different types of elements. The types generally divided into two types; the statement can + V (full verb or lexical verb) and can + Adv + V (full verb or lexical verb); and the other is negative can + not + V (full verb or lexical verb).

(iii) Modal Auxiliary Verb: will
Verb will is one of the nine of modal auxiliary verbs. This verb is the third frequent use in COCA. The number of the verb will usage is 1,118,515 data from 6,008,860 as the total number of data. The following are the examples of the data cointaining modal auxiliary verb will and the elements follow the verbwill. will + V #However, we will show that victimfacing justifications for punishment are not available for every instance of criminal misconduct....
will + not +V It assumes offenders are optimistic, which would imply that nonintentional offenders envision that they will not cause any harm, and would thus look primarily to the penalty for the....
will + Adv + V Indeed, non-intentional offenders will generally be worse off by virtue of having caused an unintended harmful result-i.e.,.... It is shown from the examples above that the modal auxiliary verb would has seven different types of elements following the verbwould. Generally, the types of elements could be identified as the statement and negative forms. The statements, firstly, would only (there is ellipsis process); secondly, would + V (full verb or lexical verb); thirdly, would + Adv + V (full verb or lexical verb); fourthly, would + PP (Prepositional Phrase); and lastly, would + will + not + Adv + V will + PP ... the Model Penal Code (" MPC "), which stated: # Juries will not lightly find convictions that will lead to the severest types of sentences unless the....
... at 1589 (" It is statistically inevitable that those who have caused harm will on the average have created higher risks, in terms of circumstances of which they.... It is shown from the examples above that the modal auxiliary verb will has five different types of elements following the verbwill. Generally, the types of elements could be identified as the statement and negative forms. The first category has three types: will + V (full verb or lexical verb); will + Adv + V (full verb or lexical verb); and will + PP (Prepositional Phrase) while the second category has two types: will + not + V (full verb or lexical verb) and will + not + Adv + V (full verb or lexical verb).

(iv) Modal Auxiliary Verb: could
Verb could is also one of the modal auxiliary verbs and the present writers found that it is the fourthmodal verb frequent use in COCA. From the data collected, the number is 885,835data from the total 6,008,860. The present writers figure out the elements follow the verb could as described in the following. could could + V could + Adv + V could + not + V ...more than ten times the number that a force of 2,500 Second World War fighter-bombers could "). # 25. Id. at 441 (describing the " speed.... ...punished at all, a man considering committing a crime, such as murder, could rest assured that either .... ...clearly articulating such views, crimes for which society is taken to be the victim could also include (1) particularly heinous offenses, such that the " public at.... That is, if Ray's bullet had not killed Dr. King, he could not have been convicted of murder, no matter how evil his intentions; alternatively....
From the examples above, it is described that there are four types of the elements following the verbcould. The category is divided into two general types: statement and negative form. The statement are, firstly, the verb could only since there is an ellipsis process; secondly, could + V (full verb or lexical verb); thirdly, could + Adv + V (full verb or lexical verb) while the negative form is could + not + V (full verb or lexical verb).

(v) Modal Auxiliary Verb: may
Verb may is one of the nine of modal auxiliary verbs. This verb is the fifth frequent use in COCA. The number of the verb may usage is 494,144data from 6,008,860 as the total number of data. The examples above describes that all the data found are in the statement forms. There are three different types of elements following the modal auxiliary verb may. They are: may + V (full verb or lexical verb); may + Adv + V (full verb or lexical verb); and may + PP (Prepositional Phrase).

(vi) Modal Auxiliary Verb: should
Verb should is also one of the modal auxiliary verbs and it is the sixth frequent use inCOCA. From the data collected, the number of the verb should usage is 435,903data from the total 6,008,860 data. The present writers figure out the elements follow the verb should as described in the following. From the examples above, it is described that there are four types of the elements following the verbshould. The category is divided into two general types: statement and negative form. The statement are, firstly, should + V (full verb or lexical verb); secondly, should + Adv + V (full verb or lexical verb); thirdly, should + NP (Noun Phrase) while the negative form is should + not + V (full verb or lexical verb).

(vii) Modal Auxiliary Verb: might
Verb might is one of the nine of modal auxiliary verbs. This verb is the seventh frequent use in COCA. The number of the verb might usage is 300,258data from 6,008,860 as the total number of data. The followings are the examples of the data cointaining modal auxiliary verb might and the elements follow the verb. might + Adv + V might + V ...unauthorized entry into a building with intent to commit a felony therein,? one might reasonably think that the intended felony, and not the unlawful entry, is the.... ...severely as those convicted of vehicular homicide currently are.34 Of course, such a regime might prove unsatisfactory to certain jury members-because they would think it too lenient on offenders that....
There are only two types of elements following the modal auxiliary verb might, as described in the examples above. The two types are both statements: firstly, might + Adv + V (full verb or lexical verb) and secondly might + V (full verb or lexical verb).

(viii) Modal Auxiliary Verb: must
Verb must is also one of the modal auxiliary verbs and it is the eighth frequent use in Corpus of COCA. From the data collected, the number of the verb must usage is 223,936data from the total 6,008,860. The present writers figure out the elements follow the verb must as described in the following. must + V Moore, in order to avoid the " trap " of moral skepticism, we must abandon the idea that causal determinism is incompatible with legitimate attributions of culpability and.... must + Adv + V As Stephen Morse and others have pointed out, the system of criminal law must necessarily presuppose a "compatibilist" 71 view with regard to free action. must + not + V ...condition punishment on whether harm befalls those victims. # The importance of this point must not be overlooked: while the reader may not accept as legitimate the....
From the examples above, it is described that there are three types of the elements following the verb. The category is divided into two general types: statement and negative form. The statements are, must + V (full verb or lexical verb) and must + Adv + V (full verb or lexical verb) while the negative form is must + not + V (full verb or lexical verb).

(ix) Modal Auxiliary Verb: shall
Verb shall is also one of the modal auxiliary verbs and the verb shallis the lowest frequent use in COCA. From the data collected, the number of the verb shall usage is 20,372data from the total 6,008,860 data. The present writers figure out the elements follow the verb shall as described in the following.
shall + V shall + not + V shall + Adv + V ... of imprisonment is to be imposed, in determining the length of the term, shall consider the factors set forth in section 3553(a) to the extent that they are applicable.... Provided further, That the limitations contained in this section for the commencement of suits shall not bar any suit against the United States brought hereunder within one....
... force of history: " In running over the pages of our history, we shall scarcely find a single great event of the last seven hundred years that has not....
There are three types of element following the modal auxiliary verb shall. It is divided into two general types: statement and negative forms. The statements are, shall + V (full verb or lexical verb) and shall+ Adv + V (full verb or lexical verb) while the negative form is shall + not + V (full verb or lexical verb).
The following is the Picture 4 as a picture of the types of element following the nine modal auxiliary verbs found in COCA.

Picture 4. The Numer of Elements Following Modal Auxiliary
Verbs in COCA Picture 4 describes that from the nine modal auxiliary verbs, the modal auxiliary verbwould has the most various element following the verb would while the modal auxiliary verb will becomes the second, could and should becomes the third, can, may, must, and shall becomes the fourth, and the modal auxiliary verb might becomes the least types of modal auxiliary verb that has elements following the verb.

CONCLUSION
After collecting, identifying, and analyzing the data, the present writers,finally, could describe two different findings. The findings from the research uses of corpusbased have shown several valuable insights. The first conclusion is the frequency and the order of the rank of modal auxiliary verbs found in the COCA has been revealed. The data describes how many times modal auxiliary verbs are used in the COCA have been exposed to these modal auxiliary verbs in various degrees. The second conclusion is this study has revealed the elements following each of modal auxiliry verbs found in the COCA. The modal auxiliary would has the most various element following the verb and might has the least. As an addition, from the data analyzed, it is also reported that the modal auxiliary verb would is not only has various number in frequency (as the highest frequency) but also has various number in the elements follow the verb (as the highest element types). Based on the result of the research, those are the important information on the use of modal auxiliary verbs and their structure in real language since the data are collected from corpus.