THE SEMANTIC PROSODY ANALYSIS OF ‘INCREASE’ IN COVID-19: A CORPUS-BASED STUDY

Increase is a neutral word that is generally perceived as a positive word, were synonymous with positive words such as rise, grow, expand. However, when resorting to dictionary definitions, increase has been found to possess negative words such as 'increase the violence'. This study investigates how 'increase', generally often categorized with positive words, interacts with neighboring words to achieve particular meanings and how particular forms of several words can change a word to negative, positive and neutral connotation. This study is analyzed on a descriptive qualitative research design because corpus linguistic research needs to be interpreted more with qualitative considerations. The data are collected using a corpus-based approach, Covid-19 corpora, and the data analysis using semantic prosody based on Stubbs,s (2001) theory. The result of the data showed that word increase in Covid-19 corpora have negative meaning if it is collocated with several word such as risk, rate, levels, number, significantly and associated and followed by negative word but it will have positive meaning if it is located with word expression.


INTRODUCTION
Increase is a neutral word which generally perceived as a positive word. According to Collins English Online Dictionary, Longman Online Dictionary, Word Reference Online Dictionary, word Net online dictionary, and LEXICO online dictionary (2020) defined this word "to become or to make greater in size, amount, or degree. Word "greater" caused good meaning if it is collocated with a positive word, for example increase his productivity, but it will have a different case if it collocates with a negative word such as increase of violence. It is not only shown in the definition but also in the thesaurus. This word is usually synonymous with rise, grow, expand, etc. All of those words can be categorized as words that have a positive meaning.
Then what if the word "increase" is synonymous with escalate such as escalate in crimes, it will cause a negative meaning. In this case, the collocate of a node word is very important, it will cause the semantic meaning changes in sentences, especially in positive and negative in semantic prosody. Related to this issue, the author interested to know what will happen to the meaning of this word if it is located in Covid-19 data, whether this word still had positive meaning or negative meaning. As all people in the word know that nowadays people are scared by the existence of this virus. Based on this situation, there are so many researcher try to analyses Covid-19 in different way, but almost all of them only focused in health perspective and only a few of experts do that analysis in linguistic perspective.
There have been some of previous studies from experts attempted to discuss about the neutral word in collocation and semantic prosody in corpus linguistics. One of them is conducted frequent, with relatively few of them bearing an affectively positive meaning. However, a speaker/writer can also violate a semantic prosody condition to achieve some effect in the hearer/reader, for example irony, insincerity, or humor can be explained by identifying violations of semantic Prosody Louw (1993).
Another researcher, such as Partington (1998) defined semantic prosody as "the spreading of connotational coloring beyond single word boundaries". In this definition, semantic prosody is more strongly associated with connotation. Stubbs (2002) and Hunston (2002) expanded the notion of semantic prosody by suggesting that in addition to collocating with positive or negative groupings of words, lexical items can also collocate with semantic sets. According to Hunston (2002), "A word may be said to have particular semantic prosody if it can be shown to co-occur typically with other words that belong to a particular semantic set". In addition, semantic prosody is not only one word; semantic prosody can be two words or phrases, as Schmitt & Carter (2004). Both individual words and phrases can have semantic prosodies. The table below shows the example data from some of the researchers.  Louw (1993Louw ( , 2000 Stubbs (1995, 1996, 2001a, 2001b) (2004) bordering on

METHODOLOGY
Since the aim of this study to investigate the collocate increase as a verb through COVID-19 corpora. This study is analyzed based on a descriptive-qualitative research design.Kwary & Arum (2011) stated that corpus linguistic research needs to be interpreted more with qualitative considerations.
The object of this study is all content word of the article in COVID-19 corpora which is collocated with word increase. According to BBC (2020) content word are words that have meaning such as risk, significantly, levels, expression, number, associated, rate, & production. All data are collected by using a corpus-based approach. COVID-19 corpora by using Sketch Engine tools it can be visited at https://www.sketchengine.eu/ for searching the most collocates word increase which will appear on COVID-19 and classified by using into some dictionaries such as Longman Online Dictionary into a group of function word and find the meaning.
There are several procedures to collect the data completely. For the first step, the author tries to found the word which frequently collocates with a node increase as a verb in COVID-19 corpora by using Sketch Engine, one of the corpus query tools. It was found 248.288 words collocate with increase. After that, the researcher selects top 10 words which most frequently collocate with a node increase, grouping and classifying into word that includes as the content of word (risk, significantly, levels, expression, number, associated, rate, & production) and ignoring word that includes as a function of a word (an, & with) since this study is about semantic prosody. Finally, the researcher chose 8 words collocates with a node increase to be analyzed.
The next part will deal with the procedure for identifying the semantic prosody of increase based on Stubbs (2001) theory, how increase collocates with other words and changes the word and the meaning of the word, from neutral word to a positive or negative word.

RESULT AND DISCUSSION 4.1. The collocation of Increase in Covid-19 Corpora
In this research, increase was chosen to be the node word to analyze, the span was -5+5. The observed frequency of increase in Covid-19 corpus was 248.288 words that are categorized as content words (nouns, verbs, adjectives & adverbs) and function words (aux verbs, prepositions, articles, conjunctions, and pronouns). All the words within the span were called collocation. The result on this corpus, point out increase mainly collocate with quantity words such as number, levels, rate and so on but in some case, it collocates with a quality word such as expression, concentration, sensitivity, etc. After that the researcher chose the top 10 words collocates with increase to be analyzed that is a risk, significantly, levels, expression, number, associated, an, rate, production, and with. Since this study aims to analyze semantic prosody where talked about meaning, then the researcher determines to classify word that included as the content of word (risk, significantly, levels, expression, number, associated, rate, & production) to analyze and as the limitation of this research, because according to BBC (2020) content word are words that have meaning. It will help the researcher to see sense and the meaning of the word itself during analyzing proses. The data in table 3 above clearly showed that among all the collocates of increase, the word risk enjoyed the highest value. The total frequency of word risk was 14372. From all the hits, it shown that mostly a node increase appeared with word risk, so the typically collocation was increase/increasing/increased risk or increase the risk. After that, word significantly take at the second place with total frequency was 11191, followed by word expression with total frequency was 10559 at the third place, then at the fourth place taken by word levels with total frequency was 10363, followed by word numbers was 8655, associated was 8040, rate was 5150, and the last place taken place by word production with total frequency was 4881. The concordance below if the result of the collocate of a node increase in Sketch Engine tools.

Semantic prosody of increase in Covid-19
The result from the data showed that a node increase that is generally perceived as positive word turns into negative words due to the negative collocation of that word. Evidence from the Covid-19 corpus showed that the word increase may have negative semantics prosody and that this has a significant influence upon its collocational behavior. The table below showed the evidence of alteration in semantic prosody from neutral to negative, a node increase collocate with word risk. The picture above showed that all of a node increase collocates with risk always have a negative meaning. According to Longman Online Dictionary Online (2020), word risk means 'the possibility that something bad, unpleasant, or dangerous may happen'. Such as data showed in the sentence 'The relative risk by time increased steadily from January 20' and 'An increasing number Figure 3. Concordance of increase collocates with the number As shown in figure 3 above. It is one of some evidence of semantic prosody of a node increase that collocates with word number and, followed by negative words such as case, word case in this corpus refer to the coronavirus case. In this phenomenon, the word increase is generally mentioned as positive word turn into a negative word which has negative meaning caused the behavior of that word. The data above showed that increase which collocates with number and stays in negative word behavior will have a negative meaning. For example: a. In December of 2019, the increasing number of patients with pneumonia, b. Since December 2019, an increasing number of atypical pneumonia cases, c. The number of cases has increased rapidly but laboratory diagnosis is limited. All of the data above showed in Covid-19 corpora, the three of those sentences have a negative meaning caused as the behavior of negative words or the collocates with negative meaning. In Longman Dictionary Online (2020) word patients defined as 'someone who is receiving medical treatment from a doctor or in a hospital', the word explained the people ill. The data above pointed out that word levels collocated with increase and followed by a negative word such as patient will have a negative meaning in this corpus. It was the evidence of semantic prosody changed from neutral word become negative word caused by the collocate around that word. All of the data above showed that increase has negative meaning if it is word collocates with risk, rate, levels, and numbers in COVID-19.
On the other hand, in this corpus, a node increase has a positive meaning when it is collocated with word expression. Again, that case of semantic prosody caused by the collocates of a positive word. Here the picture below showed the evidence on Covid-19 corpora. All of the data above showed that word increase in Covid-19 corpora have negative meaning if it is collocated with several words such as risk, rate, levels, number, significantly and associated but have positive meaning if it is located with word expression. For example, in this case, for the collocates word number, we have to look at the word that can effect this word: In December of 2019, the increasing number of patients with pneumonia. A node increase collocated with word number that refers to the patient, and according to the Longman English Dictionary patient is a negative word. As like Stubbs (2001) stated that some words have predominantly negative prosody, a few have positive prosody, and many more words are neutral or mixed in this respect. If the collocates that a node word attracts are mostly of strong negative semantic characteristics, the node word bears a strong negative semantic prosody. If the collocates are mainly positive words, then the node word is endowed with positive semantic prosody. If both positive and negative collocates exist in the context, the node word can be said to bears a neutral or mixed semantic prosody. Regarding what stubs stated about semantic prosody, the researcher is interested in finding how word increase with positive meaning can turn into a negative word.
The researcher found that the semantic prosody of word increase in this study is evidence of Stubbs (2001) stated that semantic prosody in the corpus-based analysis.

CONCLUSION
The aim of this study is to investigate how 'increase', generally often categorized as positive words, interacts with neighboring words to achieve particular meanings and how particular forms of several words can change the word to negative, positive and neutral connotation.
The result of data showed that word increase in Covid-19 corpora have negative meaning if it is collocated with several words such as risk, rate, levels, number, significantly and associated but have positive meaning if it is located with word expression. Finally, at this conclusion, the advantages of doing this study are to improve the knowledge in the corpus linguistic field, especially about collocation and semantic prosody and hope that the result of this study may be beneficial to other researcher and it can be a reference and give a contribution in this study. Hoping in the future, there will be more researchers in this interesting and important field focused on the different approaches, and the researchers can do more research related to this issue.