Discursive patterns of anti-feminism and pro-feminism in Arabic newspapers of the KACST corpus

Abstract

This article presents the results of an analysis of the large-scale processed texts of Arabic newspapers in the King Abdulaziz City for Science and Technology (KACST) Arabic Corpus Project. I adopted methods modified from the Biber Connor Upton Approach to retrieve the expanded concordances of the lexical units almarᵓa (woman) and alnisāᵓ (women) from the corpus. The extracted text reveals the discursive patterns regarding a number of topics which are discussed in Arabic newspapers, namely, socio-culture (social practices) and eco-politics. The results of the study show how and why topics related to Arab women are discursively constructed, revealing negative patterns with regard to the topics discrimination, the veil, sports, male guardianship, driving, pro-divorce settlements and municipal elections, and positive patterns concerning the topics marital status, violence, education, travel, sexual harassment, the removal of male guardianship, employment, ministerial election and political participation. The specific case of the topic (un)veil is subsequently analysed in more detail and the results reveal that the topic is associated with negative patterns within a wider socio-cultural discourse.

Keywords

Anti-feminism Arabic newspapers Arab women corpus-based discourse analysis discourse studies discursive patterns eco-political practices pro-feminism social problems

Introduction

Discourse Analysis (DA) has been extensively used, focusing on language use and communication, language organization in terms of intralinguistic units (Stubbs, 1983), and the practice of spoken and written language itself (Burr, 1995: 47, 48; Fairclough, 1989; Fairclough and Wodak, 1997; Van Dijk, 1993, 1998, 2008). While it is important to consider the size of the texts analysed, it is more important to ensure that such corpus techniques adopt DA approaches (Baker, 2006; Flowerdew, 1997; Hardt-Mautner, 1995; O’Halloran and Coffin, 2004; Orpin, 2005; Stubbs, 1996; cf. McEnery and Hardie, 2012: 133–135). In other words, describing and interpreting the social practices embodied in written or spoken discourses using, for example, Sinclair’s approach (2004) is not sufficient by itself; DA and Critical Discourse Analysis (CDA) approaches are necessary as well (e.g. Stubbs, 1996, 2007; Teubert, 2004, 2005, 2007). It is essential to analyse texts in discourse studies, but there remains considerable flexibility in terms of discourse approaches (Fairclough, 1996; Titscher et al., 2000; Van Leeuwen, 1996; Weiss and Wodak, 2003; Wodak, 2006; Wodak et al., 1999).

To the best of my knowledge there has been no experimental analysis of Arab social and political practices and problems in relation to such topics for Arab women, and studies of anti-feminism and pro-feminism in Arab culture have to date concerned socio-cultural and eco-political practices regarding women’s issues. The issues facing Arab women are complicated and have been sporadically addressed in Arabic discourses, but with no systematic approach. This article documents an analysis of women-related discourses found in newspapers in the King Abdulaziz City for Science and Technology (KACST) Arabic Corpus Project. It also adopts corpus-assisted tools and techniques to analyse a large part of the previously constructed monitor corpus known as the KACST Arabic Corpus (Al-Thubaity, 2015). The study focuses on women and how social (and/or cultural), economic and political topics are addressed in the newspapers featured in this monitor corpus. As many aspects of these topics are regarded as either anti-feminist or pro-feminist, the present study will suggest a framework that will represent the social practices mentioned when discussing Arab women-related topics. The construction patterns of the discourses regarding each issue in question will be quantified. The article focuses on the construction of language when addressing Arab women-related topics, with a detailed interpretation of how and why anti-feminist and pro-feminist rhetoric are discursively constructed. The study goes on to qualitatively analyse a particular topic, namely, the (un)veiling of women’s faces. This topic was chosen because it occurs at low frequency in the texts and can therefore be manually analysed.

Anti-feminism and pro-feminism DA

There has been considerable analysis of discourse on the topics of anti-feminism and pro-feminism in languages other than Arabic. Recently, the concept of hegemonic masculinity has been contrasted to that of hegemonic femininity. Arguably, hegemonic femininity cannot be used in the context of women’s dominance in society, as Connell (1987, 1995) claimed, as any distinctive female identity is associated with that of powerful men (cf. McRobbie, 2009: 86–88). The dominant ideologies pursued by men and women in both writing and speech have received continuous research interest (e.g. Edley, 2001). Most studies on gender, particularly studies adopting a feminist perspective, have considered the macro-social and micro-social activities within which women take part, such as politics. These studies consider how women’s social activities are interpreted in the studied discourses, as well as women’s lifestyles and changes in physical appearance and attitudes – for example, the changes involved in the transition to womanhood (see e.g. the collection of papers edited by Holmes and Marra, 2010). Other studies have examined discursive socio-psychology (Charlebois, 2010).

In the Arab world, formal institutions and academic groups carrying out research on women produce more persuasive and influential research than tradition-oriented regional and local research, and furthermore, such reports are published in English rather than in Arabic. One example is a report made by Tohmé-Tabet (2001) to an international meeting managed by the United Nations Educational, Scientific and Cultural Organization (UNESCO), which presents a review of studies on Arab women and the social, cultural, economic and political obstacles they face, and is published in English. Additionally, a pioneering empirical study by the Federal Research Division of the Library of Congress (Offenhauer, 2005) concerning Muslim women in the Middle East, North Africa and other Islamic societies has investigated many topics regarding women, such as sex-role ideologies, feminine sacred texts, women’s rights, health, education, marriage, empowerment and elections. The data analysed in this study are derived from reports by a number of international organizations – UNESCO; the United Nations’ World’s Women: Trends and Statistics 2000; the World Bank; and the International Labour Organization (Offenhauer, 2005: 18–21). Surprisingly, the two main trends examined (positivity and negativity) proved to be fairly similar in size, and those who produced the texts being analysed tended to offer balanced views rather than being biased in either direction. Ottaway (2004) highlights the study’s claim of ‘positive results’ in some aspects of women-related issues in the Arab world, stating that ‘[t]he empirical evidence on the impact of women’s presence […] is extremely sketchy, anecdotal, and generally partisan, because the studies are often conducted by researchers who want to demonstrate positive results’ (p. 8).

In terms of women’s economic rights, notable works are the compendium and the gender overview issued by The World Bank (2009) and Livani (2007), respectively, which detail the status and progress of women in the Middle East and North Africa, with statistics implying a tendency to discourage women’s labour. There is a lack of data on discouragement in the Arab world, so no clear conclusion has been drawn regarding the employment of women or the opportunities they receive or do not receive. However, statistical data are presented in the compendium that show that men are more likely to benefit from economic rights than women, and that women’s employment inevitably causes unemployment for men (Livani, 2007: 8).

As for the political participation of women within the general socio-political and economic spheres, a case study by Sabbagh (2005) emphasizes the burden that women themselves must bear in order to build their own path, thus indicating that Arab women should not wait for privileges to be afforded to them from the hegemonic sphere of men. Despite this suggestion, Sabbagh interjects with the clause ‘if allowed’, thus referring to the hierarchal complexity of masculine distinction. However, Arab women’s leadership conditions are discussed in a study conducted by Skalli (2011), which reveals the gap between institutional mechanisms and the socio-cultural sphere where complementary knowledge and opinions do not easily converge to produce clear gender equality frames. The more formally institutionalized perspectives on the status of Arab women in the various aspects of their daily lives are more likely to represent positive views, while individual studies address the negative aspects (i.e. obstacles) that Arab women encounter. Furthermore, the shortage of data regarding the status of women progressing through varying paths in life is the main reason why women’s social activities have not been classified as either in accordance with or against feminine spheres.

KACST Arabic corpus and corpus-assisted discourse studies

The KACST Arabic corpus project was launched in 2012 in the King Abdulaziz City for Science and Technology by Al-Thubaity and his team. The corpus is more representative in terms of Arab discourse than other corpora as it contains multiple genres, domains and topics (Al-Thubaity, 2015). The provisional project website was not publicly available at the time this article was written, as the application was being updated to include additional cutting-edge language query functions to provide the user with sophisticated approaches to n-grams, collocation, expanded concordance lines and all corpus-linguistic statistical packages. The corpus encompasses a number of different genres from different domains, and covers about 14 centuries of the Arabic language. For example, the newspaper genre of the corpus comprises 241.5 million words, comprising 34% of the whole corpus (Al-Thubaity, 2015: 737–738) from 18 Arab countries (Saudi Arabia, Kuwait, Bahrain, Qatar, United Arab Emirates, Oman, Yemen, Jordan, Palestine, Syria, Iraq, Lebanon, Egypt, Libya, Tunisia, Algeria, Morocco and Mauritania).

This article focuses exclusively on the newspaper texts of the corpus. However, there have been no previous studies of the shape of social cognition from such a huge dataset of Arabic texts. In addition, when attempting to describe the principles of social cognition in terms of its discursive constructions and mental operations, such as interpretations, thinking, arguing and inferring (Van Dijk, 1993: 257), the women-related discursive constructions occurring in the newspaper genre of the KACST Arabic corpus are interpreted statistically and segmented for discursive analyses in the frame of the socio-cultural and economic and political spheres. The data, thus, are assembled using the sort-out and throw-out processes to retrieve the expanded concordances of the almarᵓa (woman) and alnisāᵓ (women) nodes. The segmentations of the intended texts which are regarded as the relevant expanded concordances of the Arab woman (and women) node from the newspaper genre are interpreted numerically.

Techniques for exploiting corpus-linguistic approaches to DA vary. Baker (2006) and Baker and McEnery (2005) outline the steps grounded in corpus linguistics (CL), which include collocations, keyness, concordances, dispersion and building corpora. The limitations of these steps have led to analyses of texts emphasizing the challenges in adopting DA approaches (e.g. Hoey, 1997; Louw, 2010; Mautner, 2007). However, when collocations, concordances and keyness are considered in DA, it might result in the loss of some prominent issues that occur in more lines of concordances (paragraphs) that should be tackled in DA. Discourse analysts are evidently free to choose between qualitative and quantitative methods (Stubbs, 1997), while the study of Magalhaes (2006) shows the usefulness of exploiting concordancing in a study of racism in Brazilian news discourses. Quantitative methods are evidently adopted in a partial or limited fashion depending upon the outcomes of the qualitative analysis, since the reliance on quantifying the discursive practices of individuals, institutions and organizations in discourses makes no hermeneutical sense (cf. Baker et al., 2008: 275).

In the work of Baker et al. (2008) and the methodological synergy that combines the advanced methods of CDA with the limited analyses of CL and natural language processing (NLP), the important question in corpus-based discourse studies (DA or CDA) is how to benefit from CL and NLP techniques in order to address the questions arising from the literature on DA and CDA. In contrast, it seems that the question of what CL and NLP techniques usually provide is seldom addressed. The tendency to fix the gap in the traditional corpus-based analyses is implied by taking the social, political, historical and cultural context of the data into consideration (Baker et al., 2008: 293). For example, studying the increase of technology in discourses in CDA (Fairclough, 1996), and how dominant social forces influence discursive practices, appears to require advanced CL and NLP techniques. The difficulties faced in employing any adopted CL or NLP technique to develop the frame of DA and CDA are not tackled in a thorough way. However, the Biber Connor Upton (BCU) Approach (Biber et al., 2007; Upton and Cohen, 2009) in the Move analysis provides a good way to segment the data from discourses in a top-down manner: from developing the analytical framework through the specification of the possible functional discourse types to segmenting each text into discourse units; from identifying functional types of each discourse unit to analysing lexical/grammatical characteristics; from describing the typical linguistic characteristics of each functional type based on sketching all discourse units; and from analysing complete texts shifting among the different functional types to delineating general patterns across the whole corpus. This is different from the bottom-up approach that focuses on particular words in the discourse, considering them as the starting point in combining corpus patterns with the unit types of the discourse.

In this article, adopting the top-down approach has helped in detecting the number of discourse units and text structures. The development I made in determining the analytical framework contributes to the bottom-up approach. The selection of women-related topics led to segmenting the text structures from their discourse units, and to analysing their expanded concordance lines in a 31n + grams span. The refining process was conducted using a multi-technique method. Multiple steps were considered in order to generate the text structures of the expanded concordance lines that relate to Arab women:

First step: refinement of 37,565 texts by extracting the textual lines of the almarᵓa (woman) and alnisāᵓ (women) nodes from the newspaper genre of the KACST Arabic corpus.

Second step: retrieval of the +15/−15 span of concordance lines of the intended node in comma-separated values (csv) format.

Third step: adoption of a coding protocol in csv format for segmenting discourse units into text structures (socio-cultural, economic and political); each text structure is inserted in a new csv format.

Fourth step: sorting of the intended nodal item and the expanded concordance lines per the key terms (keyness) that determine women-related topics via Notepad++ programming language.

Fifth step: detection of the previously determined women-related topics in each text structure, resulting in structuring 20 individual topics.

Sixth step: analysis of the fixed sequences of the topic-related span of the node in the expanded concordance lines (n = 31).

Seventh step: quantification of the discursive patterns in each topic to show positive, negative and residual aspects.

Eighth step: discussion of the arising patterns in text organization among all texts in the intended corpus.

Ninth step: the analysis of one selected case, namely, woman’s face veil.

The stances with- (or pro-), against- (or anti-) and neutral (or residual) are assigned to each mention of Arab women-related topics to indicate the macro-social aspects arising from the data. The sketches used to extract the intended genre from the whole KACST Arabic corpus, and to quantify the size of Arab women-related topics that this article embarks on, are illustrated in Tables 1 to 3. Notwithstanding the fact that the discourse units have many text structures that cannot necessarily be classified into one of the intended three groups (Tables 2 and 3), the social cognition and practices presenting such text structures and their selected topics (Table 4) cover several subjects in the socio-cultural text structure, such as education, nursing, sports and so on. Thus, the socio-cultural text structures are three times larger than the economic and political text structures.

Table 1.

KACST Arabic corpus and the size of woman/women’s data.

Corpus size	Genre size	Woman freq.	No. of texts	Women freq.	No. of texts
854,421 texts; 867,049,543 tokens; 7,996,617 types	241,500,000 words	65,633	20,591	32,668	16,974

Table 2.

The size of extracted data of the target discourse.

Discourse unit of the text structures	Total no. of texts	No. of concordances	Concordances of detected topics
News
Articles	37,565	98,337	12,154
Reportage

Table 3.

The size of women-related topics of the text structures.

Text structure	Topic size	Total no. of analysed concordances	n + grams span
Socio-culture	10,971
Economy	363	12,154	31
Politics	820

Table 4.

The analysed topics of the text structures.

Text structure	Women-related topics	No. detected
Socio-cultural	- Marital status: 1. marriage; 2. divorce; 3. khulᶜ; 4. nursing.	3265
	5. Violence	2870
	6. Education	1674
	7. Discrimination	1348
	8. Veil	874
	9. Sports	459
	10. Travelling	266
	11. Male guardianship	86
	12. Driving	84
	13. Sexual harassment	42
	14. Removal of guardianship	3
Economic	15. Employment	204
	16. Inheritance	93
	17. Pro-divorce settlements	66
Political	18. Ministerial election	494
	19. Municipal election	282
	20. Political participation	44

The words associated with the nodes almarᵓa (woman) and alnisāᵓ (women), as well as the frequency of their occurrence, initially indicate a more negative sense of the contextual patterns in a 2n + grams span. For instance, the most frequent collocates are rights ḥuqūq, against ḍidd, role dawr, issue qaḍiyya, and word ᶜamal. Processing the patterns in the expanded concordances is focal. In addition, the distributions of text structures reflecting the discourse units comprise the focal length.

The statistically distributed concordance lines extending in an n = 31 span for the identified women-related topics make up 12.4% out of the unrelated concordances that constitute the remaining 87.6% (Figure 1). The protocol code was used to extract the woman-/women-related topics using direct key words: marriage, divorce, dissolution of marriage by women, and so on (Table 4). The key words assist in re-structuring the relevant texts. The remaining (expanded) concordances, of greater quantity than those extracted, were checked using other synonymous key words, but few synonymous lines were detected. The large number of concordance lines from which no topics/key words came out were checked. Most of these remaining concordance lines are excluded from the analysis as they do not include the topics determined by the names of the 20 topics given in Table 4.

Figure 1.

Proportions of analysed and dismissed women-related (extended) concordance lines.

After checking the large quantity of concordance lines left after extracting the 12.4% of selected topics, and after calculating the collocates of the intended node in a 5n + grams span, the remaining topics are not specific, and most portray either poetic or empathetic language or irrelevant verbiage towards feminism. The calculation of the collocates in a 5n + grams span has been made using the standalone processing tool of the Arabic corpus: the computer software Arabic Corpus Processing Tools (Arabic name ‘khawas/ghawwas’, ‘diver’ in English) (see Al-Mujaiwel, 2016; Almujaiwel and Al-Thubaity, 2016; Al-Thubaity et al., 2014). These calculations assist in data analysis through emphasizing popular women-related topics and by cleaning up the majority of the text that does not include any of the topical key terms.

All concordance lines of the 20 topics excerpted according to the techniques given above are, in the next section, analysed for showing the frequencies of the discursive constructions of positive, negative and residual voices. Some salient examples are given as extracts. It will be a valuable addition if the concordance lines of the 20 topics can be interpreted according to the chronological and geographical sequences of the positive, negative and residual voices. This is still unfeasible, as the corpus used in the present study is still underdeveloped and the tools for searching and extracting the expanded concordance lines of the 20 women-related topics diachronically and geographically are still unavailable functions. However, the diachronic and geographical dimensions of the discursive construction patterns of positive, negative and residual voices are analysed manually for cases of veiling and unveiling.

Discursive construction patterns of positive, negative and residual voices

The discursive patterns detected reflect either positive or negative opinions of the women’s issue in question. The remaining residual patterns are found to occur in the context of general instructions that are neither positive nor negative, for instance what a woman should do to attain happiness in her personal or family life.

The frequencies of the positive, negative and residual construction patterns in the socio-cultural text structures are given in Tables 5 to 7, respectively. Their frequencies in the economic and political discursive patterns are given in Tables 8 and 9, respectively. The discussion of the positive, negative and residual discursive patterns in the 20 women-related topics of the text structures – socio-culture, economy and politics – addresses some issues with examples (extracts) given for clarity. The discursive patterns presented next are interpreted briefly from a quantitative standpoint for the 20 topics, except for the case study of the (un)veiling topic that will be interpreted qualitatively in the last section.

Table 5.

Positive frequencies, occurrences and patterns of socio-cultural structure topics.

Topic	Freq.	Positive occurrences	Discursive pattern
Marriage	2406	1408	Rights after divorce
Divorce	581	125	Open choice
Khulᶜ (dissolution of marriage)	143	129	With Khulᶜ
Nursing	135	6	Rights for nursing
Violence	2870	1131	Against violence
Education	1674	1611	Open education
Discrimination	1348	23	With CEDAW agreements
Veil
ḥijāb	661	7	Reasons for decorative veil
niqāb	213	Null	Null
Sports	459	3	No reasons for ban
Travelling	266	230	Acceptable in modern life
Maḥram (male guardianship)	86	10	Travelling without itNo need in publicNeed in case of insecurity
Driving (only in SA)	84	2	Women better at driving than men
Sexual harassment	42	33	Calling for law with no exceptions for prevention
Removal of guardianship	3	0	In case male guardianship prevents women from their choices in marriage

CEDAW: Committee on the Elimination of Discrimination Against Women.

Table 6.

Negative frequencies, occurrences and patterns of socio-cultural structure topics.

Topic	Freq.	Negative occurrences	Discursive pattern
- Marriage	2406	126	- Open-choice polygamy
- Divorce	581	104	- Not a woman’s choice
- Khulᶜ (dissolution of marriage)	143	14	- Underage marriage
- Nursing	135	11	- Against Khulᶜ
- Nursing	135	11	- No full rights
- Violence	2870	854	- With soft power
- Education	1674	63	- Limited education
- Discrimination	1348	1012	- Cases not tackled- Still not abiding by CEDAW- Banning CEDAW
- Veil			- Imposition
- ḥijāb	661		- Reducing men’s lust
- niqāb	213	867	- Fitna (causing sexual desire)- It is not an option
- Sports	459	16	- Risky for women’s bio-physiological functions
- Travelling	266	36	- No right to get passports without male permissionAccompanied by male guardians
- Maḥram (male guardianship)	86	76	- Travelling only with male guardians- Foreign private drivers’ exceptions- Breastfeeding adult male foreigners to make them Maḥram- Husband seeing his dead wife is forbidden- Segregation between male and female doctors in hospital is essential
- Driving (only in SA)	84	14	- Forbidden either being on own or with foreign drivers- Needs muscular abilities in case the vehicle breaks down
- Sexual harassment	42	9	- Veils causing harassment- Harassment is the responsibility of women- Harassment in private sectors- Imposing law against the encouragement of women adopting sexy appearance
- Removal of guardianship	3	3	- Disrupts women’s tribal lives

CEDAW: Committee on the Elimination of Discrimination Against Women.

Table 7.

Residual frequencies, occurrences and patterns of socio-cultural structure topics.

Topic	Freq.	Residual occurrences	The scenario
- Marriage	2406	872	- Pieces of advice
- Divorce	581	218	- Marrying non-Muslims
- Khulᶜ (dissolution of marriage)	143	133	- Neutral
- Nursing	135	118	- Neutral- Enumerated
- Violence	2870	885	- Enumerated
- Education	1674	Null	- Enumerated
- Discrimination	1348	313	- Enumerated
- Veil
- ḥijāb	661	Null	- Null
- niqāb	213	Null	Null
- Sports	459	440	- Irrelevant verbiage
- Travelling	266	Null	- Null
- Maḥram (male guardianship)	86	Null	- Null
- Driving (only in SA)	84	0	- Null
- Sexual harassment	42	Null	- Null
- Removal of guardianship	3	Null	- Null

Table 8.

Positive/negative/residual frequencies, occurrences and patterns of economic structure topics.

Topic	Freq.	Positive occurrences	Discursive pattern	Negative occurrence	Discursive pattern	Residuals
- Employment	204	191	- Unrestricted jobs	13	- No physical works- Forbidden with men	- Null
- Inheritance	93	3	- Calling to apply Western law paradigms	19	- Culturally not having the right to it- Different between cities and villages- Totally prevented in Palestine	- How it works in Sharia law
- Pro-divorce settlements	66	4	- Imposing it on men legally	62	- Negligence- Still debating and not firm	Null

Table 9.

Positive/negative/residual frequencies, occurrences and patterns of political structure topics.

Topic	Freq.	Positive occurrence	Discursive pattern	Negative occurrence	Discursive pattern	Residuals
- Ministerial	494	461	- Not forbidden in Sharia law- Half should be women- Well-known nominated females	3	- Forbidden in Sharia- Fragile- Productive but not in politics	- Narratives about female ministers and- History of extraordinary female leaders
- Municipal	282	2	- Mandatory- Consequences of not having female members	269	- DismissedCondemned voices- On condition of Islamic instructions- Female names prevented	- Neutral- Enumerated
- Participation	44	39	- Urgent need to allow women	5	- Claim of inabilities	- Null

Positive voices in the socio-cultural text structures

The general themes of marital status and positive occurrences for marriage, divorce, dissolution of marriage by women and nursing exhibit affirmative aspects, indicating that women should possess the right to decide on their marital status with no exceptions. This has come up 1408, 125, 129 and 6 times, respectively. As for the dissolution of marriage (khulᶜ), a mixed representation in the corpus regarded this as a full right even if there is no reason (Extracts 1a and 1b). Another discursive pattern regarding the dissolution of marriage is the fatwa (the legal statement produced by Sheikhs as instructed by Islam) that welcomes this right for women, though surprisingly refuses it when promoted in the United Nations’ CEDAW agreements (Extract 2).

Extract 1.	a. If a woman gets to the point of hating her husband, Islam permits her to terminate the bonds of matrimony by filing for a ‘no-fault’ divorce (khulᶜ).
	b. Allah has granted the inherent right to initiate divorce proceedings by man. As a rule, it is a woman’s prerogative to file a ‘no-fault’ divorce (khulᶜ) in order to ensure that the covenant of marriage can be based on principles of equality and gender balance.
Extract 2.	We are at odds with the UN Committee on the Elimination of Discrimination Against Women (CEDAW) Convention over the prohibition of polygamy, which is incompatible with women’s rights to equality with men, given its emotional, material and other serious implications for women and their dependents.

The topic of the veil and the linguistic patterns and frequency of information arising in the positive form affirm that both the ḥijāb (the scarf folded across the hair and chin or around the whole head) and niqāb (a ruband covering the whole head and face except the eyes) are not compulsory, but are a tradition created in Islam. The decorated veil is a choice for women who must wear it to avoid the negative senses of the black ḥijāb, but some information arising in the data of the expanded concordances of the women-related topic of veil suggests that wearing the decorated ḥijāb may cause men to flirt/philander more.

Requiring the presence of maḥram (women’s kin with whom sexual intercourse is incestuous and marriage not allowed i.e. fathers, uncles or brothers) is not confirmed by Sheikhs, but positive patterns regarding maḥram in the discourse negate that. Interestingly, I found that some women prefer the presence of maḥram in situations in which they feel unsafe, such as in public places. In the context of religious activities, Sheikhs in Islam insist that women who want to do a Hajj (the annual pilgrimage to Mecca) must be accompanied by their maḥram, and only one concordance in the data was found that negates this (Extract 3). The removal of maḥram is suddenly affirmed once in the case where the existence of guardians results in a violation of women’s basic freedom (Extract 4).

Extract 3.	The Adviser to the Royal Court Sheikh Abdul Mohsen bin Nasser Al-Obeikan said that it is permissible for a woman to travel alone without being accompanied by a mahram (a relative with whom marriage is prohibited) if a woman can ensure her safety onboard, and provided that she will be received by a maḥram after arrival at the new destination.
Extract 4.	The competent judicial bodies are required to facilitate the proceedings to argue, hear these cases and eliminate inequities faced by women who have potentially been exposed to practices of al-ᶜaḍl (the act of a parent unjustifiably denying a daughter the right to marriage) and the removal of their male legal guardianship.

Negative voices in the socio-cultural text structures

In Table 6, the negative construction patterns indicate the extent to which the topics in question are structured in a masculine hegemonic way, being supported by loose Islamic laws.

It is an interesting observation, therefore, when, for example, polygamy (e.g. Extract 5) is an open choice for men, and divorce (e.g. Extract 6) is not the women’s choice. This is in addition to cases where underage marriage, for example, is not addressed seriously, and the negative consequences for the girls’ lives are not considered. In the case of nursing, I see negative patterns indicating mothers’ uncompromising denial of the nutritional and health benefits provided by nursing, which is a problem, and should not simply be accepted (e.g. Extract 7). In terms of violence, events of punishment for women are indicated by the concept of symbolic power, that is, soft power (Bourdieu, 1991), that belies gender dominance. The linguistic and lexical characteristics of the negative discursive patterns seen in the topic violence share similar lexical uses in discrimination whose negative discursive patterns include the soft power given to husbands to punish women who do not obey them (e.g. Extracts 8 and 9). With regard to education, 63 negative voices came up in the concordance lines towards unemploying women who must commute for their jobs by themselves, and preventing women from holding 24 industrial jobs (e.g. Extract 10). The dissolution of marriage by women was unwelcomed by male writers in this genre, as they believe that it might be impulsively considered by women and that it promotes women mocking their ex-husbands by publicly announcing divorce (e.g. Extract 11).

The negative discursive patterns for the topic discrimination (e.g. Extract 12) are of interest and should be discussed. Three expanded concordance lines convey masculine hegemony that is formidable and insurmountable. In total, 347 expanded concordance lines convey women being unable to prove cases of discrimination in court or unable to bring the cases to court, and the CEDAW agreement has existed for 30 years but has not seen a single act. Liberals who call for modernization, but not Westernization, are still shocked at fundamentalists and their viewpoints regarding women, that is, being considered a symbol of pre-Islamic barbarism. For the Arab countries, there were 147 mentions of Egypt, this being the first country to sign the CEDAW agreement, and Saudi Arabia, being the last; however, neither country has enforced the legislation in accordance with CEDAW. There were two mentions of low female representation in universities, showing the unequal proportions of admission of male and female students. The remaining texts convey a contradiction in terms with the articles of CEDAW that have already been passed in Sharia law and that always claim to protect and empower women. With regard to the topic of the veil, 29 expanded concordance lines convey the opinion that it is necessary to impose the veil. The others include different negative discursive patterns: criticizing women who have participated in the Olympic Games, emphasizing the difference between Arabs who must wear the veil and housemaids who have the freedom not to, and the lexical item fitna that is regarded as the main reason for imposing the veil and avoiding men’s lust (e.g. Extract 13). Some concordance lines display discursive patterns that are related to sports topics, for example whereby the suggestion is that partaking in sports causes women to lose their virginity (e.g. Extract 14). In the context of the topic of travelling, a social trend towards preventing women from being issued passports unless they obtain the permission of male guardians has been detected (e.g. Extract 15). This is combined with other texts where the topic of maḥram is constructed, whose 76 expanded concordance lines are negative in different forms: that there is limited mobilization when maḥram is not present; that it is acceptable for foreign drivers to be alone with women but not locals (only in Saudi Arabia); that the gathering together of men and women who are strangers, in public areas, is forbidden (only in Saudi Arabia); that making unrelated adults maḥram through breastfeeding should be permitted; and that male and female doctors in hospitals must be segregated (e.g. Extract 16). Two concordance lines relating to Saudi society are about two cases where female university students died because they did not immediately receive first aid from male paramedics due to the fact that they were not permitted to enter the female college. Sexual harassment is found in some concordance lines with unacceptable justification, such as men being excused from harassing women who are exaggerating their beauty in public places, that implementing/proposing laws to protect women from sexual harassment will encourage them to wear sexy clothes, and finally, that women in the governmental or private sector are sexually exploited by their managers, but they suffer it in silence in order to not lose their positions (e.g. Extract 17). The removal of male guardianship is a women-related issue concerning the right of women to be independent from their male guardians who otherwise force their female dependents to follow their demands and thus prevent them from choosing their own path in life, in particular with regard to marriage. This case occurred three times in the data in the negative sense, whereby it was stated that this right is given to women legally on condition that they return the dowry to the husband (e.g. Extracts 18a and 18b).

Extract 5.	The call for polygamy has recently gained wide appeal among some women; they may be right against the backdrop of distorted social conditions.
Extract 6.	She said further that there are social obstacles highlighted as a key stumbling block for women as they exercise their right to divorce, because it is still perceived as morally repugnant, particularly in rural, depressed and densely populated areas.
Extract 7.	The current law states that a woman forfeits the right of custody of minor children (ḥaḍānah, ‘nursing’) if she gets married. The Muslim Jurists (fuqahāᵓ) observe that if the custodian is a female, she must not be married to a husband who is a stranger to the child, and the male custodian must not travel in a manner that would seriously harm the child in his custody.
Extract 8.	Outspoken advocates of Saudi women’s rights have left no room for ultraconservatives, who have supported the right of husbands to beat their spouse and created a web page for the purpose of promoting that idea.
Extract 9.	A significant number of women are beaten and humiliated in the presence of their children, but even more so in front of relatives and neighbors, and sometimes violence is carried out by the members of the husband’s household against the wife. In other words, the right to beat women is established by Iraqi law.
Extract 10.	Putting in place the online freelancing mechanism, feminization and saudization of industrial jobs and banning women from seeking work in 24 industries for jobs which are not suited to women.
Extract 11.	One of the drawbacks of the application of khulᶜ (divorce at the request of the wife in return for compensation to the husband) over the last ten years is that the divorced man by means of khulᶜ has been the subject of several jokes and mockery, and this is not acceptable at all.
Extract 12.	a. Religious Groups (Salafists) have called on Egypt to denounce the CEDAW Convention. The Arab States have not so far ratified the CEDAW, however the ratifying Arab States of the Convention have made some reservations.
Extract 13.	He has highlighted that the wearing of the ḥijāb by Muslim adult women, which means to hide or screen from view or to cover the intimate head from the eyes of adult men, is part and parcel of the norms of Islamic Law and its adopted provisions.
Extract 14.	A significant number of women lose their virginity automatically, and by that we mean when women have access to physical exercise.
Extract 15.	A woman is granted a passport only upon obtaining the permission of her guardian, and this implies that a significant number of achievements have been carried out by a person lacking full legal capacity.
Extract 16.	Gender-based segregation in schools and universities has taken place. Some attempts have been made as well to separate men and women in transport facilities but this broadly unpopular action has received considerable opposition.
Extract 17.	Many women and girls have experienced sexual harassment even in the workplace but they fear reporting the incident.
Extract 18.	a. Women are not required to return any other stuff. The law stipulates that women shall waive their financial rights guaranteed by the Sharia such as alimony and deferred dowry.b. The Personal Status Law No.1 of 2000 does not apply to this case, which provides that a woman has the right to file for divorce by means of khulᶜ if it is impossible for her to continue to live with her husband, provided that her husband is released and discharged from her dues.

Residual voices in the socio-cultural text structures

The naturally occurring residual connotations in the rest of the concordance lines of the topics appear as general information that does not have a positive or negative voice (Table 7). The case of Muslim men marrying non-Muslim women did not occur in the analysis as positive or negative. The general scenario of the information frequency of the residual expanded concordances of the node almarᵓa (woman) and alnisāᵓ (women) includes only the normality base, and with no positive or negative voice.

Positive, negative and residual voices in the eco-political text structures

In Table 8, it can be seen that most cases that express an opinion against women’s rights in employment, pro-divorce settlements (i.e. money paid to the woman) and inheritance are raised by individuals with formal institutionalized positions in society. Thirteen expanded concordances show the necessity of banning women from working in factories and of preventing any opportunity for women to be in workplaces side by side with men. Regarding pro-divorce settlements, there appears to be no case regarding taking serious steps to achieve this, and astonishingly, it is not supported in some tribes. In addition, the Sharia law that is followed in the courts does not consider modern financial situations and commitments; rather, it only considers that the divorced woman should receive enough to prevent starvation. The rules for inheritance whereby the woman, by Sharia law, receives half of her brother’s assets differ between cities and villages in Egypt, for example, and furthermore are totally dismissed in Palestine. Such cases appearing in the data raise a call for considering Western law, which would allow men and women to receive inheritance equally.

Despite the high number of positive voices that encourage the ministerial and political participation of women, the actual participation is still just a handful. This is due to continuing masculine hegemony and traditions that linger and override women’s rights. The most prominent point arising from the data, as in Table 9, is that the number of naturally occurring negative discursive patterns is higher than that of positive ones in the contexts of municipal elections.

Quantitative remarks on text structures’ voices

It was observed that the institutionalized discursive constructions arising from the data, despite a few concerning socio-cultural text structures, are pro-feminist in discourses of women-related topics, as opposed to the socio-cultural topics whose discursive patterns were more negative by social practices. The conclusion of this section is therefore that anti-feminism in the Arab world seems to be generated and exaggerated by individuals who are influenced by Arabic dictatorial social forums that maintain in people a sense of fear regarding the punishments that may be enforced upon women when decisions regarding their rights are released from the traditional norms. These forums reflect a movement towards the control of Arab societies by Islamic figures who hold positions that are respected by the public. The discursive patterns as a whole demonstrate that social practices are largely anti-feminist, and that eco-political courtesy is largely pro-feminist towards women-related topics.

The negative voices arising from the discourses of the data in this article are less common for the topics of society/culture and the economy. The dataset for the 12,154 expanded concordance lines shows that positive to negative and residual (neutral) patterns are significant in the Y- to X-axis (Figure 2). The positive patterns remain higher in the three text structures. However, the negative patterns are found more to be exhibited in the keyness of only discrimination, veil, sports, male guardianship, driving, pro-divorce settlements and municipal election.

Figure 2.

The distribution of positive (1), negative (2) and residual (3) (X-axis) in the three text structures (socio-culture/economy/politics) compared to the totality in the topics data.

The case of (un)veiling

The reason for choosing this case in particular was the result obtained from calculating the average number of whole concordance lines for the 20 topics detected (Table 4). The mean number was 714.9412. This number is closest to the number of the 31n + grams-expanded concordances of the women-related topic veil: 874.

To give details of the diachrony and geographical areas of the intended concordances of veil, Table 10 shows the chronological sequences of the strategies of constructing the reasons for (un)veiling according to their frequencies in percentage as given in Table 11. In addition, Figure 3 shows the ratios of the 31n-grams concordances span of veil + woman/women among all 18 Arabic countries, except Qatar and Mauritania where the retrieval of data has failed to find archived newspaper texts.

Table 10.

The chronological dimensions of the strategies toward (un)veiling.

Freq. in %	Time periods			Freq. in %	Time periods
Freq. in %	1990–2000	2000–2010	2011–2016	Freq. in %	1990–2000	2000–2010	2011–2016
0.34	–	–	0.34	0.45	–	–	0.45
1.37	–	0.4	0.97	1.83	–	–	1.83
0.68	–	–	0.68	1.71	–	–	1.71
4.46	0.4	1.1	2.96	10.18	–	0.8	9.38
6.2	–	–	6.2	1.37	–	–	1.37
3.77	–	–	3.77	9.95	–	–	9.95
25.52	–	1.6	23.92	9.26	–	0.6	8.66
7.89	–	–	7.89	0.68	–	–	0.68
0.68	–	–	0.68	0.34	–	–	0.34
1.02	–	–	1.02	1.71	–	–	1.71
1.37	–	–	1.37	0.34	–	–	0.34
0.68	–	–	0.68	0.22	–	–	0.22
0.45	–	–	0.45	1.83	–	–	1.83
1.83	–	–	1.83	1.25	–	–	1.25
1.6	–	0.4	1.2	1.02	–	–	1.02
				100.00

Table 11.

Contexts, actors, reasons and strategies in the concordances of veil.

Context	Actors	Reason to veil	Reason to unveil	Strategy	Freq. in %
- Actresses and news presenters	Islamic elites		- Compulsory for job	- Flexible only in the given context	0.34
- Society		- Morality		- Showing faces combined with wearing jeans results in seduction	1.37
				- Faces are regarded as genitals	0.68
				- Sinfulness	4.46
- Olympics		- It is Allah’s choice, not a tradition		- Women, thus, have no right to argue	6.20
			- It is not the core of Islam	- The need is for decency not for veiling	3.77
				- Where are the human rights if Arab women want to?	25.51
				- Gives her a social privilege
- Constitution		- Women elected or voting should oblige with the veil			7.89
- Constitution		- Women’s choice		- Women themselves obey the Islamic instructions	0.68
- Instability	Shia referents	- Western life/night clubs cause instability in Iraq		- Harmful consequences of liberating women	1.02
		- Mandatory symbol		- Shia women without the veil are at risk	1.37
- Constitution		- Shia MPs declaration		- The Shia woman is a modal example	0.68
- Society	Extremists	- The spread of vices		- Terrorizing women by any means in order to veil	0.45
		- Sitting on chairs in public tightens clothes and shows women’s body edges		- Claiming such a scene	1.83
		- Make young girls accustomed to veil		- Educating young girls about the consequences of vices	1.60
- History	Liberals		- Woman’s face is as normal to look at as her hands	- Face is not a genital; this has arisen from Ibn Taymiyya (Salafist school)	0.45
			- Examples of women not veiling in front of Prophet Mohammad	- Narrating how it was without referencing resources	1.83
- Constitution			- No direct evidence in Islamic-Quranic instructions	- How is it? Different sayings with different arguments about unclear exegetic interpretations of Quran	1.71
			- Distortion of women’s reputation	- Against global normality	10.18
			- Identity should be totally released	- Elected women in politics must show their faces	1.37
- Society			- Recalling equality	- Veils have nothing to do with respect or religion	9.95
				- Women’s choices only	9.26
			- To walk and breathe comfortably	- Men should wear it once to feel how annoying it is	0.68
			- It is not imposed during praying or in Makkah holy mosques	- Why is it not imposed during worships but it is so in normal life?	0.34
- Mask	Reporters	- Obliged by themselves in western countries	- Against the law in some western countries	- Masks are not banned as are veils	1.71
- Black veil and abaya		- Seducing religious men		- A semi-naked woman is not as girly and charming as an niqāb-clad one	0.34
- Black veil and abaya		- Living in their country	- Travelling to liberal countries	- Depends on common sense	0.22
- Society	Women	- No reason except what we are told from Quran	- Difficulty in achieving success in life	- Comparing women with no veils in private sectors with those wearing veils	1.83
				- Getting rid of the veil is helpful	1.25
		- Exploiting the veil to hide the face for the purpose of thieving	- Cannot get married to whom I choose	- Black veil and abaya are the best means for thieves	1.02

MP: member of parliament.

Figure 3.

The geographical dimensions of the strategies toward (un)veiling.

To qualitatively analyse these 874 expanded concordance lines, three most important aspects will be considered: transparency, consistency and reflexivity (Lombardo, 2009: 24). The discursive discourse of the subjects is interpreted according to their backgrounds and current positions, and is performed in four stages: interpretation of the contexts, the actors, reasons behind the stances taken regarding veiling and unveiling, and the strategy in the captured concordances that can be embedded in the extended texts of the concordances, as shown with their frequencies in Table 11, respectively.

The salient remarks arising from Table 11 are as follows:

First, the inclusion of ḥijāb or niqāb associated with the nodes almarᵓa (woman) and alnisāᵓ (women) draws the most attention regarding negative ideologies on the one hand, and regarding ignoring other voices that are supposed to only be from women on the other. The actors are Islamic elites, Shiite referents, extremists, liberals, report writers and women. There are major differences between the contexts of each actor regarding the rhetorical strategy employed for reasoning about either veiling or unveiling. Each actor constitutes a reason for a context, but none gives details about how to veil and what to veil.

Second, a large proportion of the opinions of the Islamic elites were contradictory. Islamic elites speak on behalf of women and call for respect and freedom for women to wear ḥijāb or niqāb, but they curtail women’s freedom to choose how to wear them in general. This opinion increased to 25.51% in the data between 2010 and 2015. In particular, there was a very low occurrence (0.34%) of the trend towards facilitating the jobs of women as actresses or news presenters where the veil is not allowed, but where they may experience hardships as a result. However, if such hardships are the reason for not veiling, it could be argued that women may face many hardships outside of this context, for instance in the private sector where women are not allowed to wear the veil. The strategy used by Islamic elites to impose the veil on women in society is reflected in the beliefs of the Salafist school whereby Ibn Taymiyya regards the woman’s face as a genital (0.68% occurrence). A similar low frequency is found in the topic of constitution whereby Islamic elites misleadingly give the reason for veiling as women’s choice.

Third, the most unwelcome trends, as seen in Table 10, have emerged with actors from the extremist Sunni/Shiite groups, where cases of violence against and intimidation of women in Arab society are implied in order to forcibly make women wear the full veil.

Fourth, opinions of liberals make up most of the strategies for not wearing the veil. The veil is argued to be inconsistent with the common sense of global human character; this opinion is held by 10.18%. A similar percentage (9.95%) believes that veiling does not correlate with respect or religion itself. Only slightly fewer (9.26%) believe that women are free to choose what they want, and many also believe that it is self-conceit to do so due to the failure to admit that women will only have freedom to wear it if women, and only women, want to.

Fifth, with regard to report writers, some outlandish ideas have arisen in the discourse, such as a call for replacing the veil that is banned in European countries with face masks that are usually used to prevent harmful smells, and another stating that niqāb and ḥijāb seduce men more than semi-naked women. The third stance concerns the women who wear ḥijāb in their country but choose not to when in a foreign country. This can be explained in two ways: first, they may feel shame wearing the ḥijāb in advanced countries, and second, they may fear the negative consequences of not wearing it in their own country. In addition, there is more safety, according to what most women said, associated with the word ḥijāb in the data, in developed countries than their own Arab countries.

Finally, the last stance concerns what women say about veiling and unveiling, which has the same meaning but is considered more formal. One strategy employed is that there is no clarity about the ḥijāb or what it was like during the inception of Islam. In addition, neither the ḥijāb nor the niqāb help women to succeed in life, particularly those who are working in the private sector. Moreover, the black abaya does not help women to choose their partner and get married as they wish, as they are unable to reveal their face and identity when meeting people, which is essential when making acquaintances. The last stance regarding the reasons for unveiling is that the abaya can be exploited for stealing or carrying out acts involving violation of privacy, to the extent of also being exploited by men through disguising themselves and begging.

The greatest discrepancy observed in the rate of employment of different strategies exists between Islamic elites and liberals. Both possess extremist ideologies, but the Islamic elites are more negative than the liberals (cf. Read and Bartkowski, 2000). This conclusion might also be drawn through a detailed analysis of any women-related topic in terms of positivity and negativity in the future.

Concluding remarks

The social practices embedded in the women-related topics discussed previously exhibit a range of problems. These problems are due to ideology and hegemony (Fairclough, 2009), according to my analysis of the newspapers of the KACST Arabic corpus with regard to the aforementioned 20 Arab women-related topics. The hegemonic reproductions are largely embedded in the discourses of Arab women-related topics rather than in the (actual) transformations that help to result in more sustainable developments.

The social power featuring in the three text structures – socio-cultural, economic and political – was established through the newspaper genre of the KACST Arabic corpus and showed that the socio-cultural textual structure is larger in quantity but less negative compared to eco-political hegemony. The topics vary in terms of the reinforcement of positives and negatives regarding Arab women. The Arabic discourses are loud and verbose, and need to be interpreted carefully in a separate study for each topic. The power element that controls Arab women’s issues and lives was found to be resistant negativity, although it was not as coercive as the social powers, for which the constructive patterns in Arabic women’s topics were positive. The social power embedded in the data for each individual topic related to Arab women only maintains the status quo, whether negative or positive.

However, the Biber Connor Upton approach, which was partially applied to the topics, was implemented for the data that were classified by the sequences of expanded concordance lines according to the almarᵓa (woman) and alnisāᵓ (women) nodes. Furthermore, the techniques employed in the quantitative and qualitative analyses of women-related topics in newspapers of the KACST Arabic corpus revealed the formation of discourses as positive, negative or residual. The positive voices offered some sort of relief without actual consequences in reality, while the negative voices tended to be based on a number of firm ideologies.

The analysis of the case of (un)veiling apparently indicates that no actors, except reporters and women themselves, considered giving complete freedom of choice to women. This, of course, excludes the stances of extremists and fundamentalists whose strategies are terrifying. In addition, it seems on the whole that Islamic elites are being stubborn when reiterating the mandate to wear the veil without giving a detailed explanation for the need; the concept of covering the whole head is simply highly embedded in most of their discourses analysed here.

To sum up, this article has represented, in terms of frequency, the elements of how Arab women-related topics are discursively constructed in more negative than positive ways regarding discrimination, the veil, sports, pro-divorce settlements and municipal election. The question of why there is a higher frequency of negativity for these topics needs further detailed discursive analysis through the contexts and actors introduced in the texts. However, positive discursive patterns, on the whole, remained higher. In the data analysed, the anti-feminist and pro-feminist stances do not indicate a great enough tendency for either direction. The number of occurrences of women taking stances about veiling and unveiling is small. Further discourse analyses are urgently needed for each stance and each topic, in particular contexts or between different discourses, genres, domains and topics, to fill the huge knowledge gap regarding the reality of the construction of Arab social, economic and political practices and problems towards women-related topics.

Footnotes

Acknowledgements

I would like to thank the anonymous referee(s) for their suggestions that helped clarify a number of points in the earlier version of this article. My thanks are also extended to Dr Abdulmohsen Al-Thubaity for showing some technical considerations, as a head of the KACST project.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author extends his appreciation to the Deanship of Scientific Research at King Saud University for funding this work through the Research Project No: NFG-14-01-02.

Author biography

Sultan Almujaiwel is an Assistant Professor of corpus linguistics in the Arabic Department at King Saud University, Riyadh, Saudi Arabia. He received his PhD in corpus/applied linguistics from the University of Exeter, UK, in September 2013. He teaches and researches in various aspects of corpus-based discourse analysis and corpus-linguistic approaches, with an interest in how large/small-scale data of natural language are used in discourse studies, linguistics and social sciences.

References

Al-Mujaiwel

(2016) Free/Open KACSTAC and its processing tools: Lexical resources for Arabic lexicogrammatical microstructures based on collocational indicators. In: Alonso Almeida

Ortega

Quintana

et al . (eds) Input a Word, Analyze the World: Selected Approaches to Corpus Linguistics. Newcastle upon Tyne: Cambridge Scholars Publishing, pp. 153–170.

Almujaiwel

Al-Thubaity

(2016) Arabic corpus processing tools for corpus linguistics and language teaching. In: Proceedings of the 2nd International Conference on the Globalization of Second Language Acquisition and Teacher Education (G-SLATE), Fukuoka, Japan, 4–6 August. Gifu City, Japan: INTESDA, pp. 103–108.

Al-Thubaity

(2015) A 700M + Arabic corpus: KACST Arabic corpus design and construction. Language Resources and Evaluation 49(3): 721–751.

Al-Thubaity

Khan

Al-Mazrua

et al . (2014) KACST Arabic corpora processing tool ‘Khawas’ (computer software). Available at: https://sourceforge.net/projects/ghawwasv4/ (accessed 2 July 2016).

Baker

(2006) Using Corpora in Discourse Analysis. London: Continuum.

Baker

McEnery

(2005) A corpus-based approach to discourses of refugees and asylum seekers in UN and newspaper texts. In: Ballington

Karam

(eds) Journal of Language and Politics, pp. 197–226.

Baker

Gabrielatos

Khosravinik

et al . (2008) A useful methodological synergy? Combining critical discourse analysis and corpus linguistics to examine discourses of refugees and asylum seekers in the UK press. Discourse & Society 19(3): 273–306.

Biber

Connor

Upton

(2007) Discourse on the Move: Using Corpus Analysis to Describe Discourse Structure. Amsterdam: John Benjamins.

Bourdieu

(1991) Language and Symbolic Power (trans. G Raymond and M Adamson ). Oxford: Basil Blackwell Ltd.

10.

Burr

(1995) An Introduction to Social Constructionism. London: Routledge.

11.

Charlebois

(2010) The discursive construction of hegemonic and pariah femininities in the spoken accounts of a group of Japanese women. PhD Thesis, Lancaster University, Lancaster.

12.

Connell

(1987) Gender and Power. Stanford, CA: Stanford University Press.

13.

Connell

(1995) Masculinities. Oxford: Polity.

14.

Edley

(2001) Analysing masculinity: Interpretative repertoires, ideological dilemmas and subject positions. In: Wetherell

Taylor

Yates

(eds) Discourse as Data: A Guide for Analysis. London: SAGE, pp. 189–228.

15.

Fairclough

(1989) Language and Power. London: Longman.

16.

Fairclough

(1996) Technologisation of discourse. In: Caldas-Coulthard

Coulthard

(eds) Texts and Practices: Readings in Critical Discourse Analysis. London: Routledge, pp. 71–83.

17.

Fairclough

(2009) Critical discourse analysis as a method in social scientific research. In: Wodak

Meyer

(eds) Methods of Critical Discourse Analysis. London: SAGE, pp. 122–138.

18.

Fairclough

Wodak

(1997) Critical discourse analysis. In: Van Dijk

(ed.) Discourse as Social Interaction. London: SAGE, pp. 258–284.

19.

Flowerdew

(1997) Interpersonal strategies: Investigating inlerlanguage corpora. RELC Journal 28(1): 72–88.

20.

Hardt-Mautner

(1995) Only connect: Critical discourse analysis and corpus linguistics. UCREL technical paper no. 6. Lancaster: Lancaster University. Available at: http://ucrel.lancs.ac.uk/papers/techpaper/vol6.pdf (accessed 14 July 2016).

21.

Hoey

(1997) From concordance to text structure: New uses for computer corpora. In: Lewandowska

Melia

(eds) Proceedings of the Practical Applications of Language Corpora Conference. Łódź: Łódź University Press, pp. 2–23.

22.

Holmes

Marra

(eds) (2010) Femininity, Feminism and Gendered Discourse: A Selected and Edited Collection of Papers From the Fifth International Gender and Language Association conference (IGALA5). Newcastle upon Tyne: Cambridge Scholars Publishing.

23.

Livani

(2007) Middle East and North Africa: Gender Overview. Washington, DC: The International Bank for Reconstruction and Development, pp. 1–43. Available at: http://siteresources.worldbank.org/INTMENA/Resources/MENA_Gender_Overview_2007.pdf (accessed 14 July 2016).

24.

Lombardo

(2009) Introduction: Establishing guidelines for the use of corpora as resources for learners (and their teachers). In: Lombardo

(ed.) Using Corpora to Learn about Language and Discourse. Bern: Peter Lang, pp. 7–38.

25.

Louw

(2010) Collocation as instrumentation for meaning: A scientific fact. In: Van Peer

Zyngier

Viana

(eds) Literary Education and Digital Learning: Methods and Technologies. Hershey, PA: IGI Global, pp. 79–101.

26.

McEnery

Hardie

(2012) Corpus Linguistics. Cambridge: Cambridge University Press.

27.

McRobbie

(2009) The Aftermath of Feminism: Gender, Culture and Social Change. London: SAGE.

28.

Magalhaes

(2006) A critical discourse analysis approach to news discourses and social practices on race in Brazil. DELTA 22(2): 275–301.

29.

Mautner

(2007) Mining large corpora for social information: The case of elderly. Language in Society 36(1): 51–72.

30.

Offenhauer

(2005) Women in Islamic societies: A selected review of social scientific literature. A report prepared by the Federal Research Division in the Library of Congress under an interagency agreement with the Office of the Director of National Intelligence/National Intelligence Council (ODNI/ADDNIA/NIC) and Central Intelligence Agency/Directorate of Science and Technology. Available at: https://www.loc.gov/rr/frd/pdf-files/Women_Islamic_Societies.pdf (accessed 14 July 2016).

31.

O’Halloran

Coffin

(2004) Checking overinterpretation and underinterpretation: Help from corpora in critical linguistics. In: Coffin

Hewings

O’Halloran

(eds) Applying English Grammar: Corpus and Functional Approaches. London: Arnold, pp. 275–297.

32.

Orpin

(2005) Corpus linguistics and critical discourse analysis: Examining the ideology of sleaze. International Journal of Corpus Linguistics 10(1): 37–61.

33.

Ottaway

(2004) Women’s Rights and Democracy in the Arab World. Washington, DC: Carnegie Endowment for International Peace.

34.

Read

Bartkowski

(2000) To veil or not to veil? A case study of identity negotiation among Muslim women in Austin, Texas. Gender & Society 14(3): 395–417.

35.

Sabbagh

(2005) The Arab states: Enhancing women’s political participation. In: Women in Parliament: Beyond Numbers, rev. edn. Stockholm: International Institute for Democracy and Electoral Assistance (IDEA), pp. 52–71.

36.

Sinclair

(2004) Trust the Text: Language, Corpus and Discourse. London: Routledge.

37.

Skalli

(2011) Constructing Arab female leadership lessons from the Moroccan media. Gender & Society 26(3): 382–405.

38.

Stubbs

(1983) Discourse Analysis: The Sociolinguistic Analysis of Natural Language. Chicago, IL: University of Chicago Press.

39.

Stubbs

(1996) Text and Corpus Analysis: Computer Assisted Studies of Language and Culture. Oxford: Blackwell.

40.

Stubbs

(1997) Review of T. McEnery and A. Wilson (1996) Corpus Linguistics. International Journal of Corpus Linguistics 2(2): 296–302.

41.

Stubbs

(2007) On texts, corpora and models of language. In: Hoey

Mahlberg

Stubbs

et al . (eds) Text, Discourse and Corpora: Theory and Analysis. London: Continuum, pp. 127–162.

42.

Teubert

(2004) Language and corpus linguistics. In: Halliday

Teubert

Yallop

et al . (eds) Lexicology and Corpus Linguistics: An Introduction. London: Continuum, pp. 73–112.

43.

Teubert

(2005) My version of corpus linguistics. International Journal of Corpus Linguistics 10(1): 1–13.

44.

Teubert

(2007) Parole-linguistics and the diachronic dimension of the discourse. In: Hoey

Mahlberg

Stubbs

et al . (eds) Text, Discourse and Corpora: Theory and Analysis. London: Continuum, pp. 57–88.

45.

The World Bank (2009) The Status and Progress of Women in the Middle East and North Africa. Washington, DC: The Social and Economic Development Group and the Office of the Chief Economist of the Middle East and North Africa. Available at: http://siteresources.worldbank.org/INTMENA/Resources/MENA_Gender_Compendium-2009-1.pdf (accessed 14 July 2016).

46.

Titscher

Meyer

Wodak

et al . (2000) Methods of Text and Discourse Analysis. London: SAGE.

47.

Tohmé-Tabet

(2001) Women, intangible heritage and development in the Arab world. In: International Editorial Meeting and Future Activities in the Domain Iran-National Commission for UNESCO, Tehran, Iran, June. Available at: http://www.unesco.org/culture/ich/doc/src/00161-EN.pdf (accessed 14 July 2016).

48.

Upton

Cohen

(2009) An approach to corpus-based discourse analysis: The move analysis as example. Discourse Studies 11(5): 585–605.

49.

Van Dijk

(1993) Principles of critical discourse analysis. Discourse & Society 4(2): 249–283.

50.

Van Dijk

(1998) Ideology. London: SAGE.

51.

Van Dijk

(2008) Discourse Reader. London: SAGE.

52.

Van Leeuwen

(1996) The representation of social actors. In: Caldas-Coulthard

Coulthard

(eds) Texts and Practices: Readings in Critical Discourse Analysis. London: Routledge, pp. 32–70.

53.

Weiss

Wodak

(eds) (2003) Critical Discourse Analysis: Theory and Interdisciplinarity. Basingstoke: Palgrave Macmillan.

54.

Wodak

(2006) Review article: Boundaries in discourse analysis. Language in Society 35(4): 595–611.

55.

Wodak

De Cillia

Reisigl

et al . (1999) The Discursive Construction of National Identity. Edinburgh: Edinburgh University Press.