# Introduction

any Asian countries have speeded up the process of English Language Education Policy (ELEP) programs, aiming to prepare their citizens for globalization and socioeconomic development. Their recent ELEP programs features in early introduction of English and English education for all the citizens (Kaplan et al. 2011). China is no exception. In 2001, Chinese Ministry of Education (MoE) issued Guidelines and Basic Requirements for English Teaching in Primary Schools (Draft) (2001, hereafter Guidelines), requesting English to be a compulsory subject taught from the third grade in primary schools nationwide (Jiang, 2003;Nunan, 2003;Pan, 2015).Curriculums and teacher guidebooks were compiled and distributed centrally by the MoE shortly after Guidelines.

It has been over a decade since the MoE's implementation of ELEP nationwide in 2001. The discussion on Guidelines, especially its implementation in various regions, does not lack the robustness (e.g. Hu, 2002Hu, , 2005;Perez-Milans, 2011; Pan, 2015). These studies have pointed out the limitations and deficiencies of the ELEP, including the neglect of local situation, limited teacher training, scarce of qualified teachers and lack of education resources. However, there is very limited research that touches upon deeper issues such as language ideology, power relations and education equality (with the exception of Perez-Milans, 2011; Guo, 2012;Pan 2015). Moving beyond a discussion of policy gaps and deficiencies, my inquiry seeks to uncover hidden assumptions and ideologies, and the power relations that shape MoE's ELEP and its actions towards local educators. More importantly, it aims to expose the way discursive practice maintain and reinforce the status quo.

Within this matter, critical discourse analysis (CDA) provides a solution to critically analysethe inherent language ideology and power relationships underneath the policy texts. CDA recognizes that power struggles often take the form of discursive practices. Policy texts are therefore an arena to study how relations of power are enacted in discourse and led to sorts of consequences (Johnson, 2011). It argues that the discourses of language policies can hegemonically normalize and legitimize what is acceptable and thinkable, while concomitantly delimiting other (Ricento, 2006).

In this paper, I will use the principles of CDA to critically analyse the 3000-word text Guidelines i M . My study is another serious attempt to find out the inherent ideology and power relationship in policy discourses. It will critically examine how policies normalize and legitimize what is acceptable and thinkable for commoners. In the next section, I will introduce my methodology. Following that, I shall start analysis of Guidelines, drawing on Fairclough's three-dimensional model. Discussion will be given on the way policy discourse maintain and reinforce the status quo. In the concluding part, suggestions are given to more successful implementation of ELEP in China and other developing nations.


# II. Critical Discourse Analysis: A Theory and Method of Analysis

The concept of power is a central notion in CDA, because discourse is socially consequential thus entwined in social power (Fairclough and Wodak, 1997). Power is signalled not only by grammatical forms within text, but also by a person or a group of people's control of a social occasion. As a result, discursive practices can produce and reproduce unequal power relations between races, classes, genders and other majorities and minorities. CDA analyses 'power' as central condition in social life that is manifested and challenged in discourse. As Wodak argues (2009, p. 11): 'language indexes power, expresses power, is involved where there is a contention over and a challenge to power. Power does not derive from language, but language can be used to challenge power, to subvert it, to alter distributions of power in the short and long term.' It is CDA's aim to point out the hidden power relations behind the discursive practices and challenge unequal power relations in making conventionalized and stable discursive practices.

There are various methods for CDA, but my analysis follows Norm Fairclough's (e.g. 2003, 2009; 2010) three dimensional model. His modal argues discourse, and any specific instance of discursive practices, should be seen as:

? A language text, spoken or written,

? Discourse practice (text production and text interpretation) ? Sociocultural practice. Each dimension requires different kind of analysis:

? The linguistic description of the language text, ? Interpretation of the relationship between the (productive and interpretative) discursive processes and the text, ? Explanation of the relationship between the discursive processes and the social processes. The three dimensions are interrelated: discourse practice is the link between text and sociocultural practices. Let me break down the process: In the first place, the discourse practice (how a text is produced or interpreted) depends on the sociocultural practice which the language text is a part of. Secondly, the discourse practice shapes the text and leaves 'traces' in surface features of the text (Fairclough, 2010, p. 132). As a result, discourse is both constituted and constitutive. It both shapes and is shaped by society: it is socially conditioned by the contexts in which it occurs (discourse practice) and it, at the same time, affects the social relationships and identities of people who are participated in these social events (sociocultural practices).

Let me explain the three dimension modal further by linking it with the policy text that I will analyse in this paper. A language policy, in particular Guidelines, can be understood and analysed as:

? The written text in terms of its linguistic features, such as syntax, modal words, argumentation, etc. ? The process of interpretation and production of this document, the discourse practices, among people or groups of people, e.g. a teachers' guidebook that explain the language policy document; a parents' meeting that interprets the document. ? The sociocultural practice or the situational, institutional or social context surrounds the discourse practices, e.g. the economic situation of the country; the language environment; the education system of the country, etc.

III. Exploring the 'Texture' of Guidelines


# a) Textual features of the Guidelines

In the beginning of the Guidelines, it is argued that ELEP is to cope with informationization and globalization. English is basic requirement for the citizens for it is important for the country's open up and international communication: Extract 1: Learning English is 21 st Century Citizens' Basic Requirement, direct translation of policy extract taken from Guidelines (MoE, 2001, preface).

1. In today's word, with information technology as main feature (,) technology 2. develops daily and monthly. Social life's informationization and economic 3. activity's globalization make foreign languages, especially English, increasing 4. daily become our country's open up and international communication's 5. important tools. Learning and grasping a foreign language is (,) to 21 century 6. citizens (,) basic requirement.

Extract 1 states that learning English is for the benefit of China's open up and international communication (L4), especially since English is an important tool in informationization and globalization (L2-3). The conclusion is that learning and grasping a foreign language is a basic requirement to 21 st century citizen (L5-6). Overall, there is an invalid causal link presupposing that if one fact is true then the next is also true. That is to say, even English is important (L1-4), it does not necessarily mean learning English is basic English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China requirement for all the citizens (L5). For instance, another more valid link would be: English is important, and we should have more translators and interpreters expertise in English. Let me break down the linguistic description in details. In L1, it uses change of state verbs (develop), which presuppose the factuality of a previous state and positive evaluation. By saying technology develops, it presuppose that technology is moving ahead and it is good. For instance, a more neutral way to formulate the process would be: technology changes (though change is still a change of state verb, it is not evaluative in this case). In L2, the process of technology development is depicted as 'daily and monthly', which implies that technology is moving forward so fast that we should do something to not fall behind. L3 also describes that the importance of foreign language, especially English, 'increase daily'. L3 also uses 'increasingly' and 'become' to presuppose the factuality of a previous state, namely 'English is and has already been?important tool'. Another evaluative adjective 'important' is used in L4 to attach value to English. Overall, the usage of positive evaluation words in the Extract 1 has built up and produced a cumulative effect: English language is depicted as incontestable, indispensable, neutral and valuable for China's socioeconomic development and globalization.

L4 uses an inclusive pronoun 'our country', instead of 'your country' or 'China', to suggest that the readers are positioned as in-group members with the author and thus assuming shared perspectives. By contrast, L5 seems to use '21 century citizens', instead of 'our citizens' or 'citizens of our country', to refer to the readers. My interpretation is that the text is written in 2001, the conjunctions of two centuries (20 th and 21 st century), the usage of 21 st century citizen would make the initiative more timely and thus appear more acutely and acceptable. In other words, the text (written in 2001) urges that learning and grasping English is to cope with the approaching 21 st century.

I shall also point out that Extract 1 also features in a widespread elision of human agency. According to Fairclough, (2001), it is important to pay attention to both what is 'in' a text and what is 'left' out. In this sense, Extract 1leaves out human agency by using inanimate nouns and noun-phrases like 'technology', 'social life's informationization', 'economic activity's globalization' and 'learning and grasping a foreign language' as the agents of verbs. Fairclough (2002, p. 13) suggests that the elision of human agency is a common feature in discourses related to topics such as 'new global economy', 'neo-labourism', 'knowledge-driven economy', etc. It creates an effect that the responsibility for processes is in accounts of the inanimate subjects. For instance, in Extract 1, it seems to imply that the initiative for learning English nationwide is due to technology development, informationization, and globalization. It is not the initiative of any politicians (people) or government (groups of people). As a result, it is the inanimate subjects mentioned above that require that English to be a basic requirement, rather than any people or groups of people. This way, there is nobody in accounts of the initiative and consequently not responsible for any faults.

Overall, Guidelines uses an in valid causal link, change of state verbs, evaluative adjectives, inclusive pronoun, rhetorical expression and elisions of human agency to provide the rationales for the English education in primary schools nationwide. It tries to appeal to the common sense assumption that English is important in open up and international communication, which helps develop the country, otherwise the country will fall behind others in the new century.


# b) The use of common sense

Guidelines also uses the common-sense in a rather subtle yet coercive way. It urges that '21 st century's citizens' need to act immediately so as to catch up with the fast pace of informationization and globalization, which implies that readers (local agents in this case) are left behind, backward and incompetent. The usage of common-sense concurs with one of the central points of argument for CDA, namely the imbedded coercion in the discourses. I shall provide a brief discussion on the critiques of common-sense assumption in CDA field. According to Fairclough (2010), coercion exists in physical violence and coercive language. Coercion is also mostly notably exercised in consent. Common-sense assumption, according to Fairclough (2001), is a kind of ideology that serves for the purpose of coercive discourse. When people rely habitually on common-sense assumptions, the hidden power relations are produced, maintained and reinforced in these discourse.

In Guidelines, 'informationization' and 'globalization' is depicted as commonsensical fact. The importance of English to 'informationization' and 'globalization' is also presented as a 'fact', which is incontestable. As a result, the importance of English language education is naturalized consequently. The process of naturalizing and legitimizing the ELEP initiatives appears to present a commonsensical fact. In this way, ELEP has been discursively constructed as the indispensable, natural and technical tool for accessing advanced science and technology, which is beneficial for coping with 'informationization' and 'globalization'. In other words, when people accept the common-sense as factual, it is then difficult to find out the hidden power relations and coercion imbedded in the discourse.  c) Abstraction Guidelines also features in abstraction of highly complex series and sets of social events, past, present and predicted. Abstraction is a key point to look at for critical discourse analysts. It gauges the degree of abstraction/generalization from concrete events. The concept is firstly introduced by Bernstein (1990). According to him, political discourses always present a particular type of social event in different networks of social practices and genres. Fairclough (2003, p. 89) develops the concept and argues that 'in representing a social event, one is incorporating it within the context of another social event?This process affects how concretely or abstractly social events are represented, whether certain values are evaluated, explained, etc.'


# Volume XV Issue XI Version I

In Guidelines, there are abstraction over complex series and sets of social events, such as social life (e.g. instead of pointing out what is and what includes in social life), informationization, globalization, massage carrier, human life, education, pedagogy, economic construction, social development needs, time development's requirements. There are abstraction of past events (open up), present (English in every aspect of human life; English education's current situation), and predicted (the implication of role of English in informationization and globalization). There is also abstraction on the level of structural relations, such as the structural relation between social life's informationization and economy's globalization and importance of English. The problem of extensive use of abstraction here is that it hides away a lot of issues and concerns and naturalizes English as indispensable and natural medium for socioeconomic success. For instance, Guidelines abstracts complex series of social events as 'every aspect of human life'. English is described as being important in 'every aspect of human life'. Arguably, English is not important to many people who live in rural areas or who live in the lowest level of social ladder. The point here is that English might benefit those who already in a strong economic position and probably live in urban areas. English might not be important to 'every aspect of human life'.

In terms of presence, the only element of events consistently present is forms of activity(informationization, globalization, English education, our country's English scale, education and pedagogy), sometimes with abstraction of people (a lot of countries), or abstraction of objects (message carrier, every aspect of human life, economic construction and social development needs, profound achievement), more often without.

Moreover, the abstracted concepts (e.g. informationization, globalization, a lot of countries, social development, profound achievement) are not ordered or located in time and place, as if these events are indifference to time and place. In terms of time, English is set in a timeless frame of 'informationization, globalization' and universal place frame of 'a lot of countries, every aspect of human life'. In terms of space, since 'a lot of countries?all make English education as important part for quality education', 'our country's English' should do the same.

Overall, Guidelines shows a highly abstracted feature of complex series and sets of social events and the structural relationship between them. Fairclough (2003, p. 141) refers this sort of policy documents as one genre of governance. He points out that, in these policy documents, there will be a high degree of abstraction from and generalization across concrete events, and that causal and temporal relations will be specified between these abstraction. Such policy documents are important in linking scales-generalising over many local cases (and-a standard critique-thereby supressing difference) to make claims which have policy implications nationally or internationally.

The abstraction of complex series and sets of events and its relationship helps to make the proliferation of similar expressions in other policy documents. It helps to legitimate the importance of English in China nationwide. Arguably, globalization, informationization, English education are all very complicated matter. They should be examined within specific contexts, providing specific time, space, agents, etc. The initiative of English education in primary schools nationwide is arguably a much more complex matter that deserves more explanation, reasons and legitimacy than abstracted assumption of 'globalization and informationization or social development needs'. I will elaborate on this point further in this paper.


# d) The use of modal words

Apart from the use of change of state verbs, evaluative adjectives, inclusive pronoun, rhetorical expression and elisions of human agency, Guidelines also uses modal words to express the obligation, requirement and permission. According to Fairclough (2001Fairclough ( , 2003)), modal words say a lot about the power relations between the stakeholders. According to Halliday (1994), modality is the speaker's judgment of the obligations involved in what he/she says and it is the expression of the speaker's opinions. It is worth pointing out that judgment might not be explicit. However, even in cases where the judgment is only implied, the speakers' values can still be read and told through the analysis of the modal operators they use. Three values are attached to the speaker's judgment of the obligation to various extents, while there are different modal operators to express the judgment:

? High (required): must, need, have to, is to, ought to;

English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China

? Media (supposed): will, would, shall, should;

? Low (allowed): may, could, might, can There are also modal operators in Chinese language. According to Ross and Ma (2006), obligations can be strong or weak. Strong modal words are always used in pronouncements and in other formal spoken and written contexts. The modal verbs used to express weak obligations associated with social or moral responsibilities, and they can be used in both formal and informal contexts. There is a brief characterization of the force of the words used to express obligation:

? Strong obligation: must, have to (b ixu, bi dei, dei) ? Weak obligation: should, ought to (yinggai, ying dang, ying, hui, keneng, neng) Chinese modal verbs, like English, can express negative obligations or prohibitions, when modal words are used in negation form:

? Are not allowed to, should not (bukeyi)

? Cannot (buneng)

? Must not?not allowed (buxu) In Guidelines, different levels of modal operators are used throughout the texts to express the speakers' opinions on the degrees of the obligation and responsibility they require of the readers. Some examples of different degrees of modal operators are enlisted below (I have highlighted the modal operators in bold):

? Newly compiled textbooks ought to be evaluated and approved by our ministry (of education). ? Implementing English in primary schools ought to obey the rule of 'short sessions, high frequency'. Modal operators distinguish different levels or degree of commitment to truth on the one hand and obligation/necessity on the other. For instance, in the above examples 'Newly compiled textbooks ought to be evaluated and approved by our ministry (of education)' shows very strong commitment the writer makes, for example, in contrast to 'newly compiled textbooks might/probably/possibly be evaluated and approved by our ministry'. The point here is modal operators can represent different ways of doing of these, which make different commitments. The important question is: who has the socially ratified power of making strong commitments using the modal operators? Why? It is clear that the MoE has the socially ratified power of making strong commitments using the modal operators. The reason, I believe, using different levels of modal operators, MOE has declared its authoritative position clearly in a hierarchical way.

To prove my contention, I will first provide a brief analysis of modality and the manifested social relation. From different levels of commitment, strong or week, writers make commitment over their relationship with others. As Fairclough (2001, p. 166) argues, 'modality choices in texts can be seen as part of the process of texturing self-identity. But this goes on in the course of social process, so that the process of identification is inevitably inflected by the process of social relation.' Let us go back to the example I just mentioned. In saying 'newly compiled textbooks ought to be evaluated and approved by our ministry (of education)', the MoE does so as an authoritative and national institution giving authoritative information about ELEP to local policy agents who read the policy as a guideline for education reform. In this way, the texturing of identity is thoroughly imbedded in the texturing of social relations. By using excessive modal operators, the MoE has self-identified itself, in relation to its readers, as authoritative and prestigious. In this way, the top-down relationship between the MoE and local policy agents get reinforced and sustained through policy discourses. In this way, the MoE declares its authoritative power over local education bureaus, schools and teachers.

It is to note that, the MoE, in turn follows the instruction of the State Council, as manifested in the extract bellow: there are words directly express that the policies deriving from the State Council, the supreme political entity, must be implemented. In other words, there are very explicit textual items that depict the hierarchical power relationship between MoE and the local stakeholders: In Extract 2, the discursive connective 'in accordance with' (L1) followed by noun groups referring to the ideological guidelines of the State Council ('State Council's decisions on deepening Education Reform and Fully Promoting Quality Education' and "State Council's Decisions on reform and Development in Basic Education') represents the rest of the elements in the text as semantically subordinated. The predicative actions continue to emphasize the agent role of political authorities 'Ministry of Education' (L3). Furthermore, the semantic relationship, from State Council's guidelines to the emanating actions launched by MoE ('promoting' and 'adjusting' in L4), further strengthens the semantic authorization of these actions by placing them in a topdown hierarchical scheme. In this way, the top-down and hierarchical power relationship in Chinese education management is reinforced, sustained and maintained in the discourses (as presented in Figure 1). In this section, I have pointed out that the policy text has described English as incontestable, natural and requires for the socio-cultural development of the country. It also maintains and reinforces the absolute power of the MoE over the local agents. The hierarchical power relations in China's education system is maintained and reinforced through the discourse (Figure 1). In what follows, I shall frame the textual analysis in social analysis which can consider bodies of texts regards their effects on power relations. This concurs with Fairclough's three dimensional modal, in which the three processes, namely textual analysis, discourse analysis and sociocultural analysis, are interrelated. That is to say, I have attributed causal effects to linguistic forms above, now I will examine the effects through a careful account of meaning and context. In particular, I will find whether this sort of account of the 'English for everyone in globalization and informationization' is widespread in a particular type of text. I will also propose the level of the influence of such texts by looking at their wide national distribution and the extent to which they are 'intertextually' incorporated in other texts, particularly in media.


# IV. Text as 'Discursive Practice': Examining the Intertextuality in the Guideline

In this article, I have used CDA to analyse the imbedded messages in the policy documents on the legitimization of the ELEP nationwide. The assumption that 'globalization and informationization' is mediated through the network of texts, which is then used in different domains through genre chains: education, basic education, and English in primary schools. One important reason for the mediation at work is the abstraction of events and set of events. Guidelines was written in this background. For this reason, Guidelines and some other texts written for the education reform refer the initiative for English education nationwide as providing the important tool to meet the needs of social development in China for globalization and informationization.

For instance, State Council's Decisions on Reform and Development in Basic Education (1999, p. 1, hereafter Decisions) begins that 'today's word, scientific economy shows its first signs, and competitions of power among countries increase daily. ?which raises more urgent requirements for cultivating and making our country's 21st new generation'.

Likewise, according to the Programme for Curriculum Reform of Basic Education (MoE, 2000, p. 1, hereafter Programme), the overarching document of educational reform in China, the reason for the reform in basic education is that 'the overall standard of the basic education is not high enough, the current practice in basic education cannot meet the needs of the social development in China.' As a result, to meet the needs of the social and economic development in China, basic education should be reformed to develop a new generation or human resource. The young generation is thus educated to meet the needs of the social development in China.

As mentioned above, the account of coping with today's China with informationization and globalization is indeed a direct reporting of the policy documents written by more powerful people or institutions, such as Present Jiang and State Council. Reporting is the way we quote an important discursive feature is reporting, which refers to the way texts report, quote, claim and reproduced what was actually said or written in other texts. There are four types of reporting, i.e. direct reporting, indirect reporting, free indirect reporting and narrative report of speech act. Without further going into these types of reporting in details, I shall focus on the direct reporting here, for it is the linguistic feature in Guidelines. Direct reporting is defined by Fairclough (2003, p. 49) as the 'quotation, purportedly the actual words used, in quotation marks, with a reporting clause'. It is also a key indicator of the intertextual links among texts. Intertextuality refers that for any particular text or type of text, there is a set of other texts and a set of voices which are potentially relevant and potentially incorporated into the text. This intertextual links between texts is one of the key areas of study for Critical Discourse Analysis (CDA). Direct reporting of others in the text shows the relationship between authorial account and attributed speech. By directly quoting the account of 'informationization and globalization', the Guidelineshas indicated a close, affiliated and subordinated relationship of the MoE and State Council.

Let me explain further. Overall, the Extract 1 from Guidelines foregrounds legitimation through semantic relations of clauses. In the first clause, 'today's world' is the condition for the rest of the extract. Since it is a direct quote from the more powerful institutions and people, as I have mentioned above, the fact that it sees the direct quote as the condition for the rest of the policy initiatives shows the imbedded relationship of the MoE and its upper level policy makers. In other words, direct quotes from President Jiang: speech and the State Council's documents provide legitimation for the rest of the policy initiative in Extract 1. It is only with this condition, being backed-up and proved by President Jiang and the State Council, will the MoE and its policy work.

The legitimization, throughout the Extract 1, is not explicit and present. Rather, as I have analyzed above, it is foregrounded and laid out gradually. The extract indicates that if others accept that the national implementation of ELEP, they are more likely to accept it if they realize that the policy is initiated and supported by President Jiang and the State Council. In other words, being in line with the President and State Council, as indicated by the use of direct quotation from the national policy as the condition for the policy initiative, provides the legitimacy for the policy of implementing English education nationwide. To put the point in a commonsensical way, because the MoE has been supported by the President and the State council, its decision becomes legitimate and rationalized. The initiative of ELEP in this sense is not supported by detailed or careful empirical research. It is rather a political order from the State Council. As a result, the policy is centralised and is not consultative of the local schools or teachers. I will come back to this point shortly in this article.

In the above analysis, I have pointed out that the assumption of the 'learning English for its importance to nation-development' is widespread in a particular type of text. These include very influential texts produced by the State Council, MoE, the President's speech, and so forth. In a sense, they almost become a technology develops vigorously, knowledge-driven set of laws to obey by local agents. In what follows, I will also try and gauge the influence of such texts by looking at their wide national distribution and the extent to which they are 'intertextually' reinforced by genre chains and abstraction.


# b) Reinforcement of the assumption in genre chains

In Extract 1 and 2, the values and functions of English is described as 'the most important carriers of information', 'the most widely used language', 'an important tool in opening up and international communication for our nation'. The initiate for learning English are that 'the informationization of social life and globalization of the economy has strengthen the importance of English.' From the speech in the 15th National Congress, to the State Council Decision (1999), to MoE's Programme (2001) and Guidelines (2001), the initiative (a kind of discourse) change the genres from speech, to political documents, to curriculum (more academically based). Fairclough (2003) refers to the transformation of genres as genre chains. According to him (Fairclough, 2001, p. 31): 'These are different genres which are regularly linked together, involving systematic transformation from genre to genre. Genre chains contribute to the possibility of actions which transcend differences in space and time, linking together social events in different social practices, different countries, and different times, facilitating capacity for 'action at a distance' which has been taken to be a defining feature of contemporary 'globalization' and therefore facilitating the exercise of power.'

The 15th National Congress of CPC (1997, 12 Sep-18 Sep) was limited to little more than 2500 audiences, consists of CPC members, members of other parties, representatives of religions, etc. President Jiang made the opening and closing speeches, in which he said the country need education reform and technological innovation given the globalization and informationization in the 21st century. Recordings were not allowed during the meeting. Jiang's transcript was later edited and released as a report by Xinhua agency, the party-sponsored state media in China. Shortly after Xinhua's release of the report, other mass media, government owned and private, soon distributed the report through print, radio, television, and the Internet. There were 'learning meetings' that pass the spirit within CPC department in each institute, television programs and newspapers that interpret the report. In a lot of domains, such as meetings and media, the expression is used to describe the need for innovation and reform. For instance, shortly after the 15th National Congress, a Scientific Innovation Meeting was held in Beijing in 1999, which uses the same expression (Today's world, technology develops daily and monthly, with hitechnology as main features). In other words, Jiang's speech is of local scale, with limited audiences discussing on limited issues. Yet, reports of this sort, given by powerful gatekeeper who can exert influence through mass media, can circulate regionally and nationally. In other words, through the mediation of mass media, it manages to become more powerful discourse. In this process, the follow-up is of reinforcing the gatekeepers' ideas rather than discussion.

I shall briefly introduce the concept of mediation for a moment. Mediation, according to Fairclough (2002, p. 30), 'involves the ''movement of meaning'' from one social practice to another, from one event to another, from one text to another'. This implies that mediation happen within networks of texts or chains of genres (as I have mentioned above). Fairclough (2003, p. 219) further points out that mediation often makes use of copying technologies, such as print, broadcasting, internet, which disseminate communication and preclude real interaction between 'sender' and 'receiver'. Modern world depends largely on mediation which involves the expanded capacity for groups of people to act upon and shape the actions of others over considerable distances of space and time.

As argued above, the genre chains manage to infuse ideas (globalization, technology development, informationization), inculcate ideology (through mass media) and enact change (e.g. 'learning meetings', ELEP nationwide, etc.). It does so by managing to control various (local, regional, national, global) contexts of discourse use. Contexts, according to Van Dijk (2007), are crucial in discourse use and in relation of discourse access, control and power. He defines contexts as the mentally represented structure of those properties of the social situation that are relevant for the production or comprehension of discourse. In another occasion (Van Dijk, 2001, p. 357), he points out that context 'consists of such categories as the overall definition of the situation, setting (time and place), ongoing actions (including discourses and discourse genres), participants in values communicative, social, or institutional roles, as well as their mental representations: goals, knowledge, opinions, attitudes, and ideologies.' According to him, controlling context involves control over one or more of these categories. For example, the powerful might control over or have access to contexts by defining the goals of the social actions, the time and space of the communicative event or the participants who may or must be present.

The discourse of 'globalization and informationization' and its justification of ELEP nationwide is reinforced again and again through its use in different contexts, until the point that it appears as a 'common-sense assumption' or a reality that would not be argued against. To start, globalization and English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China informationizationare very complicated and contentious issues. For instance, Fiarclough (2002, p. 47) points out that globalization is always taken for granted yet 'there is a need to redress imbalances of power in the way in which international trade in increasing', especially for some third world countries. I will not go through the detailed discussion on (assumed) globalization, since it is not my focus. What I want to point out (also secondly), globalization and informationization (even it may be a prevalent phenomenon) do not necessarily mean English should be taught nationwide (my emphasis). In other words, the discourse and its genre chains manage to presuppose/assume an invalid causal effect (Globalization and informationization are going on, so English should taught nationwide) through reinforcing the discourse again and again and through its use in different contexts.


# c) Reinforcement of the assumption through recontextualization

I have analysed above that the assumption is partly reinforced through genre chains and its use of discourse in various contexts (local, regional and national). I have also pointed out above that the assumption is indeed based on an invalid causal effect that globalization and informationization are going on, so English should taught nationwide. Here, I will analyse the reinforcement of the assumption through recontextualization Let me have a brief discussion on recontextualization for a moment. The analysis of recontextualization is very important in CDA, for it shows how certain ideas, ideology, suppression and hegemony maintained and reinforced through the discourses (Bernstein, 1990;Wodak, 2009, Wodak andFairclough 2010). According to Linnell (1998, in Lin, 2013, p. 5), recontextualization is 'the dynamic transferand-transformation of something from one discourse/text-in-context (the context being in reality a matrix or field of contexts) to another. Recontextualization involves the extrication of some part or aspect from a text or discourse, or from a genre of texts or discourses, and the fitting of this part or aspect into another context, i.e., another text or discourse (or discourse genre) and its use and environment.' Linnel's definition seems to focus on recontextualization happens within discourses, or recontextualization of discourse/text-in-context from text(s) to text(s). Take the concept 'globalization' for instance. The recontextualization of globalization can mean the transformation of the concept through different discourses (texts) or the representation of the concept in differ networks of social practices. In Jiang's report (1997), English education was not mentioned. Jiang said: 'In today's word? knowledge-driven economy shows its first signs which foresees that human beings economic and social life will have new and magnificent change..' Then he went on to point out ten aspects that need to be reformed: market, education, technology, political science, etc. The SC's Decisions (1999) then recontextualizes the discourse (globalization, informationization in 21st China) to all education sectors, ranging from primary education, secondary school, higher education, special education, adult education, etc. Programme (MoE, 2000) then recontextualizes the discourse to basic education reform, which involves pre-school, nine-year compulsory education from elementary to junior high school, standard senior high school education, special education for disabled children, and education for illiterate people. Guidelines (MoE, 2001)   ?English in Primary Schools

In the last three sections, I have analysed the way the government legitimizes their decisions through linguistic expressions, genre chains and recontextualization. More specifically, the initiative of 'implementing ELEP nationwide' is based on the abstraction and assumption of 'meeting the demands for globalization and informationization'. Although the analysis above is limited to the policy discourses in question, the legitimization of reforms based on the argument of 'globalization', 'knowledge-driven economy' or 'informationization' is not limited to the ELEP in China. Fariclough (2002) provides several examples in which politicians (e.g. Tony Blair), experts (e.g. Rosabeth Kanter from Harvard Business School) and organizations (e.g. World Economic Forum; BBC; European Council) would legitimize their initiatives and ideas based on the abstraction of 'globalization' and 'neo-capitalism'.

At this point, one might ask: in what way does the assumption and abstraction of 'meeting the needs for globalization and informationization' matter in reality? Or, what impact could the legitimization (largely based on assumption and abstraction) bring or potentially bring? The next section will evaluate the influence and cost of Guidelines as a socio-political discourse.


# V. Sociopolitical Discourse Analysis

The implementation of ELEP nationwide by 2001 is, without doubt, a very time and money consuming business. Given the lack of data from the government (there is not open-accessed data from the government), I cannot give a specific number for the budget of this national decision (one might argue, this indicates the lack of negotiation and participation from the people as citizens, since this information should be entitled to people as tax-payers). As a result, I will use several figures to roughly give a feel of the scale and cost of such decisions.

From 2012 to 2013, the annual national education budget is 3.88 trillion, counting 4% of the gross domestic product. By 2011, there are 254,000 primary schools in China (MoE, 2014). Each school needs to provide educational resources for English teaching, including teachers, technic support, textbooks, student' workbooks, etc ii Without doubt, implementing ELEP nationwide is a very expensive matter for a big country like China. One would argue that the decision should be based on discrete empirical analysis that considers the needs, the possibility, the feasibility, the cost, the effect, the necessity of implementing ELEP nationwide. By and large, it should be a careful decision based on valid analysis of contexts and local regions. However, numerous studies have shown that ELEP nationwide in 2001 was a hasty decision that does not take full consideration of local needs and abilities. Nunan(2003) comments that the state policy is top-down, assumption-based and hasty, since there were no . Needlessly to say, the decision of implementing English education nationwide is very expensive to the government. Meanwhile, since English is implemented nationwide, parents tend to provide financial support for their children by hiring tutors and sending children to private institutions. learners can do better, because of their greater cognitive development. For most Chinese children in China, the basis for their advantage over adult learners (a long period extensive exposure in L2 context) is not likely to be the case. Since most of the children in China can only get English classes from 1-4 hours per week (very limited exposure), the effect of early introduction of English education nationwide is highly unlike to be successful. Ellis (2010), a decade after the implementation of ELEP China nationwide, concludes that the implement of ELEP in China is extremely patchy and unlikely to be successful given the limited exposure of English language for most Chinese students.


# VI.


# Conclusion

In this paper. I use principles of critical discourse analysis (CDA) to examine the English Language Education Policies (ELEP), such as the analysis of recontextualization, abstraction, assumption, common sense and modal operators, to analyse Guidelines, in terms of the textual features, discourse practice and socio-political discourses.

Following the three dimension modal (Fairclough, 1989(Fairclough, , 2001) and other CDA, in each theme, my analysis is carried out through three interrelated processes: a. the linguistic description of the features of the policy texts; b. the interpretation of the discourse practice, or how people interpret and produce the discourse; c. the way policy practice and texts are shaped by the sociocultural contexts. a. the policy texts (written, oral or multimodal); b. the discourse practice (how policy texts are interpreted and produced by people or groups of people); c. the sociocultural practice, or the situational, institutional, societal contexts, which shapes the discourse practice and in which the policy texts is a part of.

My analysis fits with the central argument of CDA. To use the words of Wodak, (1996, p. 15), Describing discourse as social practice implies a dialectical relationship between a particular discursive event and situation(s), institution(s) and social structure(s) which frame it: the discursive event is shaped by them, but it also shapes them. That is, English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China discourse is socially constituted, as well as socially conditioned-it constitutes situations, objects of knowledge and the social identities of and relationships between people and groups of people. It is constitutive both in the sense that it helps sustain and reproduce the social status quo, and in the sense that it contributes to transforming it. In my analysis, I have identified how ideology and power is maintained, sustained and reinforced in the discourse, while discourse, in turn, sustain and maintain the socio-political contexts. In particular, I have analysed through this paper that the policy texts have depicted English as a natural, incontestable, neutral an indispensable tool that is linked to China's economic development and personal benefits. This depiction of English is interlocks with the argument that English is timely needed to cope with informationization and globalization. The interlocking connections between ELEP and globalization make the competition on the terrain of English being naturalized. The importance of English and the competition in the terrain of English further legitimize and reinforce the MoE as a national guidance/leader/decision maker in controlling the local policy agents, who are in caparison with the authoritative and prestigious status of the MoE, become incompetent and inferior.

Policies define how we are to act and by what rules we must abide (Fairclough, 2006). Through policies, we come to be socialized in many ways into what is thinkable and unthinkable. Policy and political discourse represents the authoritative allocations of values and goals and socially situated representation of the world. In the analysis of the policy discourses in this article, I have identified the way discourse maintain, sustain and reinforce its ideological control and message. It examines how political power constructs and is constructed by larger social practices.

Drawing on Fairclough's dialectical-relational approach, I have used text analysis, processing analysis and social analysis to analyse the hierarchical structure manifested in the policy discourses. The analysis shows that MOE sees itself as an authoritative government branch that gives out orders for the lower branches to follow. The hierarchical structure in Chinese educational policy process is clearly identified (Figure 1). I also point out that the policy maker in upper level is presented as a face-less organization, which makes it authoritative, untouchable and unquestionable. As a result, the mysterious organization becomes more authoritative and unquestionable.

It is to note that the absolute power of the national educational government is not unique to China. This is why my article has implication larger that the context of Chinese primary English education. Kaplan et al. (2011) discuss the reasons for the failure of primary second/foreign language programmes in Asian. One of the reasons they find is that the language planning always features in the dictate and absolute power of the educational government. It is not consultative of the local schools and teachers, who are always minimised as mere implementers. As a result, the language policy and planning desired and designed by the national government always does not fit with reality. As in the case of the Chinese ELEP, result is that the implementation efforts are haphazard and do not lead to the expected increase in proficiency level. In this sense, the national government should listen to the voices of the local schools and teachers. It is only among schools and teachers will the policy make a difference in reality. Teachers always have some expertise in the LPP matter. Secondly, the national government should take responsibility in addressing related policy issues on the system levels, such as national examine system, university enrolment policy and education resource allocation system. It is rather unfair for the local educators and teachers to take the blame of the lack of success in Chinese ELEP.

Thirdly, successful implementation of the policy relies on the careful and cautious planning of the language policy in the first place. As I have mentioned, Chinese ELEP is not well researched before its implementation. It is largely a political order that is not consultative of the teachers nor does it take full account of the education reality in the diverse and vast territory of China. The amount of time allocated in school curriculum, in a lot of Asian countries, is often grossly inadequate to achieve the desired language fluency (Kaplan, et al. 2011a). The early introduction to English is not panacea for success English language education. The introduction of English education nationwide is not panacea for the socio-development of China in globalization. As a result, the hasty decision in language policy might lead to a waste of resources.
![English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China](image-2.png "")
![Extract 2: 'In accordance with the 'State Council's decisions', direct translation of policy text from Guidelines (MoE, 2001, p.1) 1. In accordance with the 'State Council's decisions on deepening Education Reform and Fully 2. Promoting Quality Education' and "State Council's Decisions on reform and Development in 3. Basic Education', the Ministry of Education has decided to make a vigorous effort in promoting 4. the basic education reform, adjusting and innovating in basic education the curriculum system, 5. structure and contents to meet the requirements of quality education. English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China](image-3.png "")
1![Figure 1 : Devolution of educational management from the central to the local level](image-4.png "Figure 1 :")
![a) Direct quotesIn 1999, President Jiang Zemin, the then top leader inChina (1989China ( -2002)), made a speech in The 15th National Congress of the Communist Party of China (CPC). He says: 'In today's word, with information technology as main feature, technology develops daily and monthly. .. knowledge-driven economy shows its first signs which foresees that human beings' economic and social life will have new and magnificent English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China change?facing this trend, we need to follow the tide and move forward.' The 15th National Congress puts forward the Revitalizing Action Plan for the 21 st Century. The Action Plan argues that to rejuvenate the country in the 21st century, the country should promote educational reform and development and improve the nation's quality and innovation ability. The key to the rejuvenation is the technological development. Based on the Action Plan and the spirit of the 15th National Congress of CPC, an education reform was launched to meet the needs of technologic and scientific development of the country in the 21st century.](image-5.png "")
2![Figure 2 : Sequence of recontextualization](image-6.png "Figure 2 :")

furtherrecontextualizes the discourse in the context of Englisheducation in primary schools nationwide. We canschematicallypresentthissequenceofrecontextualization:
		
		
			

enough qualified teachers available when the policy was implemented. Li (2009) also points out that the language competence of English teachers, especially in suburban and rural areas, shows that they are not ready for the policy to be implemented nationwide. Hu (2008) goes to four schools in Wenzhou area, two of the schools are wealthy and two of them are less privileged. Hu considered implementation in terms of 1. Provision for English classes; 2. Time allocated to English classes; 3. Availability of trained English teachers; 4.Environmental support for English within the school; and 5.Parental support. Hu concludes that the extent to which policy was being successfully implemented in China was very varied. Hu (2008, p. 533) notes that there are a serious teacher shortage in some schools, and there are a considerable disparity among schools in the implementation of policy and that this could reinforce and perpetuate social stratification. His final verdict was 'it can be argued that the policy was implemented prematurely'. Guo (2012) also show similar findings in Chinese more developed coastal areas as well, which suggests that issue is not restricted to developing areas but prevails in a larger scale. Primary schools have to offer English teaching in primary schools according to the national policy requirements, however, there is no consequential and sufficient teacher training to allow teachers manage the teaching in reality. The language competence and skills of the teachers are even less sufficient in inland cities and rural areas, especially given their limited government funds.

By and large, ELEP in China in 2001 was a hasty decision, without the support of valid analysis and empirical research. A part from the argument of 'to prepare China for globalization and informationization', another initiative for the ELEP is that early introduction of English is thought to bring better learning results. I shall briefly discuss the fallacy behind this initiative.

According to Krashen et al. (1979( , p.161, in Ellis, 2010, p. 11), p. 11), the key opponents of English immersion programmes, there are three conclusions on language acquisition:

? Adults proceed through the early stages of syntactic and morphological development faster than children (where time and exposure are held constant). ? Older children acquire more quickly than younger children (again, in the early stages of syntactic and morphological development where time and exposure are held constant). ? Acquirers who begin natural exposure to a second language during childhood achieve higher secondlanguage proficiency than those beginning as adults.

Chinese national ELEP is largely based on the third composition, which is always simplified as 'the earlier to start English education, the better outcome'. However, the real difference lies in naturalist learners and in school-based learners. As Ellis (2010, p. 11) rightly points out, the advantage that children have over adult learners only becomes evident in contexts where the learners have extensive exposure to the L2 over a long period of time. For the implicit language knowledge, such as pronunciation and oral English, the young learners might acquire English faster after a long period of extensive exposure in L2 context. For the explicit language knowledge, such as grammar text, old
			
			

				


* 
	
		The structuring of pedagogic discourse
		
			BBernstein
		
		
			1990
			London: Routledge
		
	




* 
	
		English in elementary schools: limitations and possibilities
		
			REllis
		
	
	
		Journal of Foreign languages
		
			33
			1
			
			2010
		
	




* 
	
		Critical discourse analysis
		
			NFairclough
		
		
			RWodak
		
	
	
		Discourse as social interaction
				
			TVan Dijk
		
		London: Sage
		
			1997
			
		
	




* 
	
		Language and power
		
			NFairclough
		
		
			2001
			Longman
			London
		
	
	second edition




* 
	
		Language in New Capitalism
		
			NFairclough
		
	
	
		Discourse & Society
		
			13
			2
			
			2002
		
	




* 
	
		Analyzingdiscourse: Textual analysis for social research
		
			NFairclough
		
		
			2003
			Routledge
			London
		
	




* 
	
		English as a Basic Requirement for 21 st century Citizens': A Critical Discourse Analysis of English Language Education Policy in China
		
	




* 
	
		Language and globalization
		
			NFairclough
		
		
			2006
			London: Routledge
		
	




* 
	
		Critical discourse analysis as a method in social scientific research
		
			NFairclough
		
	
	
		Methods of Critical Discourse Analysis
				
			Ruth&Wodak
			MichaelMeyer
		
		London
		
			2009
			
		
	




* 
	
		
			NFairclough
		
		Critical discourse analysis
				London
		
			Longman
			2010
		
	
	second edition




* 
	
		Teaching English for economic competiveness: Emerging issues and challenges in English education in China
		
			YGuo
		
	
	
		Canadian and international education
		
			41
			2
			
			2012
		
	




* 
	
		An introduction to functional grammar
		
			MHalliday
		
		
			1994
			Edward Arnold
			London
		
	
	second edition




* 
	
		Recent important developments in secondary English-language teaching in the People's Republic of China. Language, Culture and Curriculum 15
		
			GWHu
		
		
			2002
			
		
	




* 
	
		Contextual influences on instructional practices: a Chinese case for an ecological approach to ELT
		
			GWHu
		
	
	
		TESOL Quarterly
		
			39
			
			2005b
		
	




* 
	
		China's English language policy for primary schools
		
			YHu
		
	
	
		World Englishes
		
			27
			
			2008
		
	




* 
	
		English as a Chinese language
		
			YJiang
		
	
	
		English Today
		
			19
			2
			
			2003
		
	




* 
	
		Critical discourse analysis and the ethnography of language policy
		
			DCJohnson
		
	
	
		Critical discourse studies
		
			8
			4
			
			2011
		
	




* 
	
		Why educational language plans sometimes fail
		
			Kaplan
		
	
	
		Current issues in language planning
		
			12
			2
			
			2011
		
	




* 
	
		China and English: Globalization and the dilemmas of education
		
			JLi
		
	
	
		Bristol: Multilingual Matters
		J. L Bianco et al.
		
			
			2009
		
	
	Motivational force and imagined community in 'Crazy English




* 
	
		Critical discourse analysis: overview
		
			ALin
		
	
	
		The encyclopedia of applied linguistics
				
			CAChapelle
		
		
			London Blackwell publishing
			2013
		
	




* 
	
		Programme for Curriculum Reform of Basic Education (Draft)
		
			Moe
		
		
			2000
		
	
	Beijing: Ministry of Education




* 
	
		Guidelines and basic requirements for the teaching of English courses in elementary. (written in Chinese)
		
			Moe
		
		
			2001
			Ministry of Education
			Beijing
		
	




* 
	
		The impact of English as a global language on educational policies and practices in the Asia-Pacific Region
		
			DNunan
		
	
	
		TESO Quarterly
		
			37
			4
			
			2003
		
	




* 
	
		English as a global language in China: deconstructing the ideological discourses of English in language education
		
			LPan
		
		
			2015
			Springer
			New York
		
	




* 
	
		Caught in a 'West/China' Dichotomy: Doing a Critical Sociolinguistic Ethnography in Zhejiang Schools
		
			MPérez-Milans
		
	
	
		Identity & Education
		
			10
			
			2011
		
	
	Journal of Language




* 
	
		An introduction to language policy: Theory and method
		
			TRicento
		
		
			2006a
			Blackwell publishing
			Malden, MA
		
	




* 
	
		Modern mandarin Chinese grammar: A practical guide (Modern Grammars)
		
			CRoss
		
		
			JMa
		
		
			2006
			Routledge
			London and New York
		
	




* 
	
		State Council's decisions on deepening Education Reform and Fully Promoting Quality Education
		
			StateCouncil
		
		
			1998
			State Council
			Beijing
		
	




* 
	
		Critical discourse analysis
		
			TVan Dijk
		
	
	
		Thehandbook of discourse analysis
				
			.DSchiffrin
		
		Oxford
		
			Blackwell
			2001
			
		
	




* 
	
		Discourse and Manipulation
		
			DijkVan
		
		
			T
		
	
	
		Discourse & Society
		
			17
			2
			
			2007