A Corpus-Assisted Critical Discourse Analysis of the Syrian Refugee Crisis in Jordanian Newspapers

The present study employs corpus linguistics and critical discourse analysis to investigate the attitudes of Jordanian news towards the Syrian refugee crisis. The corpus of the research, which consists of 10140 articles (Word types: 103170 and Word tokens: 1956589), were taken from the Petra news agency between 2016 and 2018. Antconc Tools Version 3.4.4w was used to analyze the data. The study used corpus statistical tools of collocates and concordance. Collocates tool used to create a list of 200 collocates associated with the words: /lad3iʔ/ ‘refugee’, /lad3iʔi:n/ ‘refugees’, /su:ri:/ ‘Syrian’, and /su:ryi:n/ ‘Syrians’. These collocates were organized into two thematic categories: ‘services and resources’ and ‘Jordanians and Syrians’. The study used a concordance tool to unveil the attitudes of newspapers towards the Syrian refugee crisis. The findings of the study showed that Jordanians see Syrians as “brothers” and “guests”. However, Jordanian newspapers overstated the negative effect of Syrian refugees on the Jordanian economy, education, healthcare, etc. Jordanians were frustrated because Syrians compete with them on their resources and governmental services.


Background
When the conflict started in Syria in 2011, Jordan has suffered from the negative impact of a massive influx of Syrian refugees. According to UNHCR, Jordan now hosts more than 1.6 million Syrians, of whom around 700,000 registered with the United Nations High Commission for Refugees. More than eighty percent of these Syrians live in Jordanian cities and villages, especially in the northern district, as opposed to refugee camps. UNHCR reveals that 29% of registered Syrian refugees live in the capital Amman, 25% in Mafraq (a city in north Jordan), and 21% in Ibid (a city in North Jordan). Nowadays, Syrian refugees constitute more than 10 percent of Jordan's population, which has influenced the social, political, and economic situation in Jordan. It is not a secret that Syrian refugees compete with Jordanians for their employment opportunities, healthcare, shelter, education, and scarce resources. and dominations.
The integration of CDA and CA resulted in the rise of corpus-assisted critical discourse studies (CACDAS) (Partington, 2004(Partington, , 2006. According to Baker (2006), CACDA has gained growing attention from scholars. They were concerned with research that addresses immigrants, social inequalities marginalization of some minorities, and political conflicts.
The current study uses the CACDA to explore the attitudes of Jordanian newspapers towards Syrian refugees. It aims at answering the following questions: • What do concordance analysis of the words "lad3iʔ 'refugee', lad3iʔin 'refugees', suri: 'Syrian', and suryi:n 'Syrians' in the corpus reveal?
• What is the attitude of Jordanians towards the Syrian refugee crisis?

Literature Review
Kandil (2009) combined Critical Discourse Analysis and corpus linguistics to investigate how media discourse represents the conflict between Israel and Palestine. The study focuses on the use of the lexical item 'terrorism' in three corpora of news reports collected from BBC, Al-Jazeera, and CNN news networks. The analysis of the data showed each network adopted different attitudes towards the Israeli-Palestinian conflict. CNN supports the Western/Israeli position and Al-Jazeera refuses that position. The BBC was less engaged than the other networks. Salama (2011) used the corpus method and critical discourse analysis to investigate how Wahhabi-Saudi Islam been ideologically recontextualized across post-9/11 opposing discourses via collocation. The study found out that that CDA corpus methods opened new horizons of the critical study of collocations at methodological and theoretical levels. Su and Xiao (2015) used critical discourse analysis and corpus linguistics and to explore the representation of the 'Chinese dream' among both Chinese citizens and officials based on a corpus collected from the Chinese press. The data revealed that the Chinese Dream had been successfully promoted as a slogan.
Aluthman (2018) used Corpus-assisted Critical Discourse Analysis to explore the depiction of immigration in the Brexit corpus. The corpus of the study consisted of 108,452,923 words collected from news, tweets, and blogs related to the Brexit context. The study used corpus-assisted statistical tools, such as collocation, keyword, and concordance, to analyze the data. The frequency findings revealed that immigration is one of the critical issues in the corpus. Concordance analysis of 'immigration' and 'immigrants' showed both negative and positive attitudes toward immigrants in the EU debate.
Tagnin and Teixeira (2012) used a corpus-based methodology to compile a bilingual, mono-directional English-Portuguese glossary of cooking terms. They gathered two sub-corpora: one in English and one in Brazilian Portuguese, consisting of cooking recipes collected from the Internet. After collecting the data, the researchers used WordSmith software to generate frequency and keyword lists.
Ensslin and Johnson (2006) investigated how newspapers represent themes relating to language and linguistics by using a corpus-based methodology. The corpus consisted of all articles from two British newspapers, The Guardian and The Times. The study employed the software program WordSmith Tools to identify those keywords that collocate with the words "language", "languages", "linguistic" and "linguistics". The findings of the study showed exciting ways in which language-related issues are represented in Newspapers.
Baker et al. (2013) employed a corpus-based methodology to investigate the representation of Muslims in British newspapers. The corpus used in the study is a 143 million words collected from British newspaper articles (1998−2009). The researchers used the Sketch Engine tool to analyze the noun collocates of the word "Muslim". The collocates were classified into different groups: conflict, characterizing/differentiating attributes, religion, culture, ethnic/national identity, and group/organizations. The findings of the study revealed that the 'conflict' category was lexically rich compared with other categories. The study also found out that the most frequent collocates with the word "Muslim" were "the Muslim world" and "the Muslim community". The analysis of these two pairs showed that Muslims were viewed as easily offended, isolated, and in war with non-Muslims.
Kim (2014)  The results of the study suggested that US news divided the world into allies or enemies. The analysis also showed that US news viewed North Korea as a 'rogue' country and as a threat to world peace.
Concerning the Syrian refugee crisis, the researcher did not find many corpus-based studies that explore the attitudes towards Syrian refugees. For example, Ulum (2016) used Critical Discourse Analysis to explore the ideologies constructed in English, German, and Turkish news headlines. The study found that the western media (English and German) represented the arrival of the Syrian refugees to Europe as a threat or a risk; the Turkish media viewed the issue from a humanitarian perspective.
To the best knowledge of the author, no study tackled the attitudes towards Syrian refugees in Jordan by using Corpus-assisted Critical Discourse Analysis (CACDA). Thus, this study attempts to bridge this gap by examining the attitudes of the Jordanian press towards the Syrian refugees using CACDA.

Data Collection
The corpus of the current study consists of 10140 articles (Word types: 103170 and Word tokens: 1956589) taken from the Petra news agency between 2016 and 2018.

Data Analysis
The study adopts the methodological synergy approach coined by Baker (2006). It makes use of discourse analysis methods and corpus-assisted statistical tools, including keyword, concordance analysis and collocation. Antconc Tools Version 3.4.4 was used to analyze the data. This software provides several functions by which the behavior of words can be studied, such as Wordlist, concordance, collocates, keywords and N-gram/cluster. This study uses collocates and Concordance function.
The researcher used the collocates tool to create a list of collocates associated with the words: lad3iʔ 'refugee', lad3iʔin 'refugees', /suri:/ 'Syrian', and /su:ryi:n/ 'Syrians'. However, before searching for these words and generating a list, it is essential to refine our search. The function words, including prepositions, articles, possessives, and conjunctions were removed from the list by using a stop list. It's crucial to remove these words from the list since they are ranked high on the list, and they are irrelevant to our study.

Results of Concordance Tool
After creating thematic categories, the researcher used concordance tool to investigate recurrent patterns and the use of these patterns in context. The thematic categories were discussed below.