A Theoretical Account on the Study of Metaphor in Didactic Discourse

This article makes a literary review to the linguistic research in the use of metaphor in didactic discourse; especially the religious one. Acknowledging Conceptual Metaphor Theory as the primary theory in the field, the researcher embarks upon how metaphor is perceived and analysed in discourse in order to pertain its persuasive function. The article presents different approaches to metaphor analysis and their interconnection. The implications of these approaches are later deduced and interpreted within the scope of Islamic religious discourse as an example of didactic discourses.

In the 70s of the last century, several linguistic approaches emerged whose main interest was to come up with a comprehensive theory for the analysis of discourse and text.These approaches recognized the role of social factors in shaping the language and the discourse where they emerge.Discourse analysis can be recognised as approaching the relationship between the text and its context.It involves many tenets and methods which aim to scientifically dig the textual elements and hidden meanings in the discourse under investigation.Most of these approaches encompass large sets of designs and techniques acquired from mounting the perceptible bridges between the different disciplines of linguistics; namely socio-linguistics, semantics, pragmatics, and applied linguistics.However, the theory of discourse analysis proves how this intricate discipline is a challenge especially when it comes to simply conceptualize the definition of the word 'discourse'.

What is a discourse?
The foremost definitions of 'discourse' sound conflicting and overlapping because they are formulated from a range of different theoretical and disciplinary standpoints (Fairclough, 1992: 3).The simplest account of the topic argues that a discourse is "a continuous stretch of (especially spoken) language larger than a sentence, often constituting a coherent unit such as a sermon, argument, joke, or narrative" (Crystal, 1992:25).In the Foucauldian sense, a discourse involves "a group of statements which provide a language for talking about-a way of representing the knowledge about-a particular topic at a particular historical moment" (Hall, 1992: 291).This interpretation entails that the study of discourse, or discourse analysis, can be approached either synchronically or diachronically following both quantitative and quantitative methods of data collection and analysis in order to break in the essence of the discourse and uncover its textual, functional, and ideational contents.What is more, it tells us that discourse analysis should reveal how a discourse is constructed in reflection of the accumulative thoughts and ideologies of a society and its shared history.The attitude of such approaching to the text is the very radical assumption that most of meanings in a given text are hidden and not explicitly visible to the layman discourse receiver.Approaches and techniques of discourse analysis uncover any opaque discursive meanings and interpretations in the discourse of a text for the reader.
To elaborate, the linguistic study of discourse must tackle the historical, social, and political context where a discourse emerges in order to render the existing social relationships and structures; such as legitimization, power distribution and abuse, and ideologies.These issues are believed to be influential in shaping a discourse, and they are in turn shaped by it as well.Such issues are mostly approached and elaborated by the ethnographic study of different disciplined of social sciences; especially politics and religion.For instance, it is unsurprising that once political or religious groups reproduce a discourse, their argument mostly revolve around some constructed effects of an ideology.Such an ideology, as Teun van Dijk defines, makes shared social representations that have specific functions for social groups.And he accentuates that discursive practices between social actors are one site where ideologies are reproduced, transformed and challenged (van Dijk, 1998: 191).From this bold postulation, Critical Discourse Analysis emerged (See Wodak, 2001).

Critical Discourse Analysis
Fundamentally, Critical Discourse Analysis (CDA) can be seen as a means to arrange the ways in which difference of power emerges in conversation.In a series of different works which have been developed since 1989, Norman Flourishing Creativity & Literacy Fairclough draws most of the ideas of CDA which have influenced his contemporary following researchers.In this regard, CDA aims to critically associate language, ideology and social change in any given discourse by focusing on ideology and hegemonic politics, in order to challenge the 'value-free' assumptions and 'cause-effect' relationships which underlie most of the social science (Agger, 1991: 109).Fairclough accentuates that any piece of text (written or spoken language) can be simultaneously considered an instance of discourse that involves an instance of social practice and an instance of discursive practice (Fairclough, 1992;1995).Now, since discourse analysis is basically concerned with description of the textual (form) and ideational (meaning) meanings in the text discourse, CDA mainly takes the discursive production and interpretation of the text (discourse practice) as its point of departure as it operates at the level of broader social analysis, or the 'socio-cultural practice' (Fairclough, 1992: 4).
One may allegedly presume that CDA pays attention exclusively to the content of political-like genre of discourse by focusing on the ways in which different power relations and ideologies are introduced in such genres of discourses.In fact, the scope of CDA covers several genres of texts and discourses.For example, in conversation analysis, CDA draws attention to 'power' issues such as dominating a topic by an interlocutor(s) in a conversation; especially the manner of interaction and turn-taking.This can structurally involve showing how topics are introduced and changed by the dominant participant(s) in the conversation, showing how power-differences determine who speak(s) first, and about which topic, and for how long.Working on applying a consistent framework of conversation analysis in a given social context or domain of experience, CDA can even reveal the underlying (ideological) arguments and beliefs and prove how they are related to the different social order in a society (Pennycook, 2001:85-94).
CDA amplifies our critical awareness to the language and to the culture where discourse practices occur.And among the most salient discourse practices perceived are figurative language and metaphors.

Metaphorical Language in Discourse
As rule of thumb, a metaphor plays a key role in manifesting the creativity of any ideological discourse by adding more vividness to the ideational meanings of the discourse content and argument.Although metaphor has a significant and undeniable role in eliminating monotony and 'boring' uniformity from discourse, it is also has an 'ideological' significance which has to be 'ideologically' invested (Fairclough 1995: 74).Lakoff and Johnson demonstrated that metaphorical language holds a rather critical position in language: [M]etaphors […] highlight and make coherent certain aspects of our experience […] metaphors may create realities for us, especially social realities.A metaphor may thus be a guide for future action […] this will, in turn, reinforce the power of the metaphor to make experience coherent.In this sense metaphors can be self-fulfilling prophecies (Lakoff and Johnson 1980: 156).
As one of the most salient discourse practices, metaphors, in any language or culture, are not always subscribed to physical explanations; in fact they must be considered as another reflection of the socio-cultural practices of the culture.This is argued from the perspective that any discourse constructs its own context in accordance with a specific standpoint in which the same discourse is in turn constructed by this standpoint.Metaphors, from a cognitive linguistic viewpoint, are linguistic instantiations of discourse in texts where textual and ideational linguistic features and meanings noticeably emerge in connection to the overall cumulative cognitive representations of the individual and society.The cognitive power of metaphors and the way how it works allow the discourse producer to make his metaphors carry an explanatory and persuasive power that makes the strange and incomprehensible more familiar.Kittay (1987:39) maintained that "metaphor has cognitive value and that this stems not from providing new facts about the world but from a reconceptualisation of the information that is already available to us".Thus, metaphor can play on resolving the ambiguous and incomprehensible arguments by bringing into the surface what is considered more comprehensible aspects of the argument from different domains of experience.Kittay, added that "Metaphor actually gives us 'epistemic access' to fresh experience and, to the extent that we have no other linguistic resources to achieve this, metaphor is 'cognitively irreplaceable'" (Kittay, 1987:39).
Persuasive and ideological discourse exhaustively exploits metaphors.Taking political discourse as a (favourable) example, this type of discourse involves a great portion of language loaded with covert representations about abstract and 'hegemonic' ideological assumptions.And most of these assumptions, such as the existence of a 'wise' leadership, the common good of the people, the shared destiny, the actuality of internal and external threats, are quite inseparable from the accumulative cognitive memory of the given society.Such abstract representations demystify the inherent knowledge of politics to what is beyond its institutions.
A metaphor, then, can draw attention to intrinsic ideas about the actuality matter of the abstractness of politics, the state, and its structures and institutions.For example, history tells that leaders are constantly portrayed by their propaganda machine in terms of honorific metaphorical rather than literal titles.The title Pater Patriae (Father of the Nation) for instance is a metaphoric representation that used as an honorific title bestowed to a leader.This leader (unsurprisingly always a man) is considered the driving force behind the establishment of his country, state, or nation.The list of Pater Patriae involves, to mention but a few, Simón Bolívar, Sukarno, Jomo Kenyatta, Mohammad Ali Jinnah, Nelson Mandela, Mustafa Kemal Atatürk, and of course Gandhi.As the layman knowledge about leaders and leadership might naturally be deficient because of the lack to any experiential knowledge about them/it, the propaganda machine predominantly assigns them the metaphoric image of 'father' with all its entailments and implications.Herein, the image of the leader is drawn by directing the attention to images and traits of father on the basis of the conceptual metaphor 'THE LEADER IS A FATHER'.By combining the experiential knowledge about the real world to the hegemonic political beliefs, this metaphoric representation draws attention to particular 'symbolic' features of the leader such as his legitimacy, power, and compassion.Consequently, this metaphor, as a universal socio-cultural practice, facilitates the people's understanding of what the duty to their father is or should be, and even without questioning the real nature of this Pater Patriae or approaching his inaccessible characteristics.What is more, the 'THE LEADER IS A FATHER' metaphor may arouse the sense of weakness and limitation in relation to the leader's illimitable might.
Metaphors can take part in creating a persuasive impact on any hegemonic and political context making the followers of that political system or ideology willingly subdue to its statement of beliefs.As a result, we can say that metaphor plays an essential role in political discourse in portraying the necessity of the Pater Patriae as an imperative requirement for the people and the state.People, like children, are always in need for their father-figure to whom they turn in moments of despair; a figure that represents all sorts of power and authority in the eyes of his children regardless of his own character and behaviour.

Metaphors in Persuasion
A discourse is a purposeful activity, or a social practice.Its purposefulness significantly relies on how the discourse (re)producer recognises the different participants and factors involved in the context with the aim is to influence the discourse recipients.A purposeful didactic discourse then conveys a point of view carried in a message to be transmitted, and the outcome of this message can success in introducing new beliefs, or changing certain attitudes.A purposeful didactic discourse then that its ultimate function is to change attitudes and beliefs is what we call a 'persuasive' discourse.

Persuasion
Persuasion, according to Jowett and O'Donnel (1992:21) is an interactive communicative process in which a message sender aims to influence the beliefs, attitudes, and behaviour of the message receiver.The process of persuasion involves a process of interaction between two types interlocutor: the sender(s) and the receiver(s).The message carried in a persuasive discourse can be transmitted through different channels; visual channels such as images or texts shows; or auditory channels such as speeches and music.And it is very common to use the both channels such as in TV shows.
During the Middle Ages, persuasion was considered an art that must be mastered by any educated man who studied in the universities of Europe; from the days of imperial Rome through the Reformation, it was raised to a fine art by preachers who used the spoken word to inspire purposeful actions such as virtuous behaviour or religious pilgrimages; however, theorists emphasised the existence of some similarities between persuasion and education (Encyclopaedia Britannica, 2008: 'persuasion').They maintained that persuasion is strongly similar to the process of teaching new information through informative communication.Consequently, since repetition in communication modifies learning, they assume that it has persuasive impact as well; and that principles of verbal learning and conditioning are widely and profitably applied by persuaders (Encyclopaedia Britannica, 2008: 'persuasion').Figurative language, irony and sarcasm (allusions), recounting anecdotes and rhetorical questions are among the most rhetorical devices which can significantly be employed to (re)produce a persuasive discourse.These devises are the main ways of arousing audience interest and retaining the attention of the hearer (Charteris-Black, 2005:8).

What Makes a Metaphor Persuasive?
Many theories highlighted how the comprehension of metaphor may attain its persuasive force in discourse such as the 'literal-primacy' view, the 'salience-imbalance' theory, and 'structure-mapping' theory.From these theories and views six perspectives on the relationship between metaphor and persuasion were reviewed (Sopory and Dillard, 2002: 383-391): (a) pleasure or relief, (b) communicator credibility, (c) reduced counterarguments, (d) resource-matching, (e) stimulated elaboration, and (f) superior organization.
First, any instance of metaphor may have a persuasive power from its 'pleasure' or 'relief'.According to the 'literalprimacy' view metaphor is an exceptional language, but it is literally false and involves 'semantic anomalies'.Both 'pleasure' and 'relief' perspectives indicate that a metaphor involves a 'pleasure' aspect.However, a metaphorical expression causes a negative tension because of its semantic anomaly, and to reveal this aspect it is important to solve the semantic tension.Then, solving this tension can only be done by making the metaphorical meaning becomes comprehensible (Reinsch 1973;Tudman 1971).In 'pleasure', it is important to determine the metaphorical meaning and uncover the "unexpected similarities" between the components of the metaphor to make 'pleasurable'.In 'relief', it is important to find how metaphorical meaning drives way the negative tension to lead to 'relief'.Consequently, since literal language does not cause any semantic tension to be resolved, neither 'pleasure' nor 'relief' will be resulted from its understanding.On the other hand, in metaphorical expressions, the outcome of the states of 'pleasure' or 'relief' will cause the reinforcement and evaluation of the metaphorical meaning, thereby increasing persuasion.
From the 'communicator credibility' perspective, any communicator, a discourse maker, will be judged more credible if he uses metaphors more than one who uses literal language (Osborn and Ehninger, 1962;Reinsch, 1970).The credibility of the metaphor users are attributed to two reasons: the first is the Aristotelian position that those (speakers or writers) who employ metaphors should be judged quite positively; the second is that metaphors can draw attention to some unfamiliar similarities existed between different entities.For the first argument, it is acknowledged that Aristotle once stated: "The greatest thing by far is to be a master of metaphor.It is the one thing that cannot be learnt from others; and it is also a sign of genius" (Aristotle, 1952: 255; cited in Sopory and Dillard, 2002: 385).For the second argument, the detection of similarities existed between different entities may be highly appreciated by the discourse receiver.In addition they may become a source of interest and pleasures to the discourse receiver who in return will be impressed by the discourse maker and favourably judge his (the discourse producer's) credibility according to the 'literal-primacy' (Bowers and Osborn, 1966;Osborn and Ehninger, 1962).
The third perspective, according to Guthrie (1972) suggested that the process of metaphor comprehending produces a lot of associations that may lead to "an overload in the receiver's mental circuitry" (Guthrie, 1972: 4).The key argument of this 'reduced counterarguments' view is that the processing of any metaphor in discourse require more cognitive resources than processing literal language.Therefore, a metaphor receiver will need a high proportion of the cognitive resources to 'decipher' the metaphorical persuasive message.Consequently, smaller amount of resources are left to "derogate or exclude the message content or the source" (Guthrie, 1972: 4).In addition, the reduced counterarguments view assumes that all message recipients are inclined to 'counterargue' a message apart from of what it is advocating.Consequently, a result of this counterargument caused by the existence of metaphor will be a greater agreement with the ideas advocated in the message of the discourse.And this is apparently what causes a disruption between an argument and its counterargument which leads to the increase of the persuasive power of metaphor.
The fourth perspective assumed that the relationship between metaphor and persuasion can be pointed to from a more sophisticated use of the cognitive resources.This idea is suggested by the 'resource matching' view (Jaffe 1988, cited in Sopory andDillard, 2002).This view proposed that the meaning of a metaphorical expression can be derived by calling some cognitive elaboration to guarantee a better 'integration' in memory of the discourse message and its arguments, and that leads ultimately to a greater persuasion effect in relation to the literal message (Ortony, 1979).As elaboration calls for more efforts and more 'mobilization' of the cognitive resources in the human brain, the greatest portions of elaboration and comprehension of metaphors take place when there is a mutual conformity between the available resources required to realize the metaphorical message from one hand and the other resources available to the discourse interpreter, or receiver.Consequently, persuasion is guaranteed if there are disconformities between the two cognitive resources because of the paucity of complete comprehension.Equally, the overload in available resources in the cognition of the discourse receiver may lead to the diminution of the persuasive impact of the metaphor in the message and its content.
The fifth 'stimulated elaboration' perspective perceives that the understanding of metaphors motivates our thought by drawing attention to the similarities of the relational structures between the metaphor components rather than simple linguistic and rhetorical features.This process leads to bringing to mind more affluent associations in the discourse receiver's semantic memory when compared to what literal language does (Whaley, 1991).Hence, the enhancement of semantic connections can produce a greater portion of elaboration in the message content that leads to an increase in the impact of persuasion.Since certain types of metaphors appear to incite more arguments; then their processing and comprehension may produce more elaboration than that of messages transmitted in literal language (Whaley, 1991).Therefore, if the message, or the discourse, recipient has high motivation and ability to process the metaphor, and if the content of the message is convincing, then the outcome will be a greater portion of thoughts which agree with the message advocacy; and thereby to a greater persuasion (Chaiken et al., 1989).
Finally, the 'superior organization' view proposes that a metaphor facilitates structuring and organizing the message arguments better than literal language (Mio, 1996).As a metaphor brings to mind a greater portion of semantic associations, and if these associations prove its consistency with the metaphor, then the diverse raised arguments will connect themselves more coherently by using the available semantic associations.What is more, the links between the semantic associations and metaphor will show up the hidden arguments in the discourse and make them more prominent, and this makes them easier to be processed by the discourse receiver and improves the understanding of message arguments and persuasion (McGuire, 1972;1985).Sopory and Dillard (2002) argued that in spite of the fact that each one of these perspectives stands in contradiction to one another, a variety of hypotheses and assumptions can be derived from them.For example, they argued that each of the six perspectives predicts that metaphors should show 'suasory' (tendency to persuade) superiority over literal language and those messages with metaphors are more persuasive than the literal ones.This leads to the presumption that the number, or density, of metaphors in the message and their locations in the discourse text can be of positive function in persuasion.

Studying Metaphor in Language, Text, and Discourse
In this section, I present the most common approaches in studying metaphor in text and discourse.Most of my presentation is oriented toward the theoretical cognitive account of contemporary metaphor theory and the technical procedural corpus-based approaches in studying metaphor.I point here to the contributions of major metaphor scholars who have combined corpus-linguistics with the traditional cognitive approach of metaphor research such as Lakoff and Johnson, Zoltan Kovecses, Fauconnier and Turner, Alice Deignan, Andrew Goatly, and Jonathan Charteris Black.

Metaphor and Thought
The study of metaphor has flourished as an academic discipline since the emergence of major works such as Ortony's collection Metaphor andThought (1993[1979]) and Lakoff and Johnson's Metaphors We Live By (1980).These leading works deserve recognition for the renaissance of scholarly interest in the queen of tropes, metaphor.However, other important works contribute in highlighting different aspects and characteristics of metaphor from different viewpoints.
In his work Metaphor: A Practical Introduction (2002) Zoltan Kovecses introduces different aspects of the contemporary theory of metaphor that are based on Lakoff and Johnson's Conceptual Metaphor Theory (CMT).This work provides rich linguistic examples of metaphors from different languages and cultures; emphasizing the linguistic, cognitive, psychological and cultural aspects of metaphor.However, although it is considered the most conventional theory of metaphor, it is crucial to note that CMT has also some limitation in that Lakoff and Johnson, and their followers, present frequent examples of verbal manifestations of metaphors from different domains, but their examples seem to be invented and artificial, rather than citations of everyday language and public discourse.Most of the examples in these works are quoted out of their context without raising any question about who is using the metaphor and to whom, and in what textual, or discursive context, considering the discursive functions that these metaphors achieve in the given domain.It is imperative in linguistic study and research that the readers, or the scholar, be given more details about the examples of metaphors and their context rather than giving them the option to imagine a context from themselves.Furthermore, the 'traditional' works of Lakoff and Johnson, Kovecses, and even Gillis Fauconnier and Mark Turner's works on Integrated Blending Theory mostly give unsatisfactory consideration to the different textual forms of the metaphors themselves.This theory appeared in 1994 shortly after Lakoff and Johnson's CMT.This theory emphasises that the study of conceptual metaphors must not be restricted to the study of conventional metaphors in spite of their importance.A conventional metaphor will not become 'conventional' if it has not been once a 'novel' metaphor.For this reason Lakoff's emphasis on studying conventional metaphors was not highly appreciated by other scholars.The (Integrated) Blending Theory (BT) suggests that metaphors are the product of many general processes which take place at the human cognitive system.These cognitive processes involve a cross-domain series of combinations between different types of information which already exist in particular domains, or 'mental spaces', in the human cognitive system, and this process is called 'conceptual integration' or 'blending' (Grady, 2007:198).
To a degree, the Blending Theory (BT) shares the same cross-domain model with Lakoff and Johnson's CMT.The two theories suggest models which construct conceptual metaphors by means of 'structural' conceptual domains; the source and the target domains, and both of them involve a process of mapping process between the two domains (spaces in BT); source and target.Nevertheless, the BT considers conceptual metaphors differently where conceptual metaphors are resulted from more complex relationships between the elements in separate conceptual domains, or 'spaces'.The conceptual metaphor occurs when these relationships are highlighted in different scenarios of mappings where they are 'blended' in a third domain, or 'space', in the cognitive system, and this process is called the 'Conceptual Blending'.So, the BT extends the traditional two-domain model into a new 'four-space' one.Each 'space' in the BT model is understood as a mental space in the sense of a coherent bundle of information activated in the mind at a particular time, representing an understanding of a real or imaginary scenarios (Grady, 2007:199).
Fauconnier and Turner (1998) presented a schematic framework for blends in which conceptual metaphors are characterized by the existence of some degree of irregularity in the mapping process.In order to outline the structure of a blend we need to different types of information from the two input spaces before drawing the mapping in the conceptual metaphor.In other words, to make a conceptual metaphor we 'blend' some information that we already have with new acquired experiential knowledge about the object, or person, that we are describing metaphorically.To illustrate, the following figure exemplifies the different mapping processes between the conceptual spaces according to the Blending Theory: To make it in a more illustrative example, Grady et al. pointed out that in the expression "this surgeon is a butcher" the speaker, or the writer, aims to convey an idea that the surgeon is incompetent (Grady et al. 1999).The BT model suggests that in such example the speaker is 'blending' some information from his knowledge about the domain of surgery in all its features with his experiential knowledge from the domain of butchery.Accordingly, the BT sees that when perceiving such metaphorical expression a series of conceptual mappings occur between the four conceptual spaces in the cognitive system.First, our mental representations will structure two 'input spaces', where the first involves the real experience and knowledge about the domain of surgery.The second involves what we already know about the domain of butchery.Then, on base of the common features of the two spaces and which are already exist in the 'generic space', the two input spaces assign a conceptual mapping into a fourth space, the 'blend space'.This space carries the new metaphorical structure resulted from the mapping of selected conceptual materials in the two input spaces.To illustrate is the following figure: We see from Fig. 2 above that a mapping occurs between the distinctive characteristics of each input spaces on base of the common feature which they already have in the generic space.Then, the different elements from the two input spaces project into the blended space to establish a new idea with a new relationship which is described by the conceptual metaphor.Croft and Cruse explained the example "this surgeon is a butcher" according to Grady et al.By saying that: "[T]he inference of incompetence arises through an elaboration of the basic elements of the blended space, that is, we imaginatively reconstruct a scene in which a butcher is in charge of an operation, and uses his normal butcher's techniques on the patient: there is a basic incompatibility between the goal and the means, which leads to the inference of X's [the surgeon's] incompetence".(Croft and Cruse 2004:209) The previous example draws our attention to the existence of three successive stages which must be performed in order to construct the blend; 'composition', 'completion', and 'elaboration' (Fauconnier and Turner 1998).By 'composition' Fauconnier and Turner argued that any blending involves certain kinds of attribution between the different elements from the input spaces which leads to providing relations that do not exist in the separate inputs.In the 'completion' stage we have a situation where the structure in the blend agrees with the information already stored in our long-term memory.Finally, in the 'elaboration' stage certain processes which involve imaginative cognitive simulations are developed in the blend.After these successive stages a new structure emerges in the blend that is not entirely copied from the inputs.
On the other hand, Fauconnier and Turner raised five standards for any given conceptual blending metaphor in order to work effectively, and they call these standards the 'optimality principles (Fauconnier and Turner, 1998).The first of these principles is the 'integration' principle where the scenario in the blended space must be in complete integration with its component and form a perfect scene for the metaphor.The second principle is called the 'web' principle where we must have some sort of fixed and sustained connections between the blend and the inputs in a way that any event occurs in one of the input spaces will entail an immediate change in the matching event in the blend.Thirdly, we have the 'unpacking' principle that says structure of connecting networks of the two input spaces must be simple and straightforward for interpretation to give the appropriate blend space.Fourthly, there is the 'topology' principle which suggests that all the elements in the blend space must participate in similar sorts of relations as their counterparts in the inputs.Finally, we have the 'good reason' principle that says any element appears in the blend must be meaningful.
The BT did not come as a reaction against the CMT or to disprove it.A meticulous understanding to both theories reveals that they complement each other.While the CMT pays much more attention to conventional metaphors, the BT seems to be more feasible and practical in studying novel and creative metaphors.The four-space model of BT greatly elaborates in detail what are the particular elements exist behind the linguistic metaphors such as their context sensitivity.This last postulation gives the BT the capability to deal with the instances of innovative and nonconventional metaphors in different kinds of discourse such as the political and religious discourses.The BT gives more attention to the elements which each mental space has and their connections with the other spaces of experience and knowledge which greatly participate in construct the conceptual metaphor.

Studying a Metaphor in its (Linguistic) Context
Despite the genuine technicality of cognitive theories of metaphors, they have not been principally immune to criticism though.The most critical argument against these approaches was in regard of the 'artificiality' of their illustrative examples.Metaphors do not always occur in ordinary language in the prototypical form (X is Y, or X is like Y); in fact, they predominantly take a variety of structural forms especially when semantic factors and ideational interpretations are fore-grounded.For example, both "a budding love" and "he cultivated his love relationship with her" could be considered amenable to the conceptual metaphor LOVE IS A PLANT, but the two metaphors are not completely interchangeable in terms of their impact upon discourse and their interpretation; as the first metaphor emphasizes the nature of the love relationship, and the latter pay more attention to the course of love, implying impediments and problems that face this relationship.In addition, it is common that a metaphor comes in the company with other rhetorical devices, or 'schemes' such as end-rhyme, alliteration, and assonance, or by making another a combination with other tropes such as metonymy, irony, and hyperbole (Leech, 1969).These devices may affect the form of metaphor and change its prototypical structure in order to serve some intended ideas in a discourse.
In one of his speeches the former American president Bill Clinton said "...and so today we pledge an end to the era of deadlock and drift, and a new season of American renewal has begun (20 January 1993;cited in Charteris-Black, 2005:129).Here, the combination between the alliteration and metaphors form serves the rhetorical purpose of the discourse.Hence, any change that occurs in the metaphor form may affect the attraction of the alliteration and viceversa.Nevertheless, the study of metaphor had to overtake a remarkable shift by moving from cognitive-based approaches to the advantage of corpus-based approaches which pay more attention to the discourse details and its immediate linguistic context.Jacques Derrida was among the first prominent scholars who emphasized the necessity of studying language in its context.Derrida maintained that meaning can never be in possession of language speakers exclusively, and any language speaker has a false impression that he has a control over meaning while he speaks; ignoring that the meaning of utterances and statements is determined by the place they hold in a discursive system (Derrida, 1984).
Many linguists emphasize that there must be a return to language itself when studying metaphorical language, yet without denying its conceptual nature; thus, it is essential to pay more attention to the different details of the conceptual nature of metaphorical expressions in addition to the qualities of metaphors which they involve (Crisp, 2003;Semino and Culpeper, 2002;Müller, 2004).The most fruitful approach then was to extensively study the immediate linguistic context in which a particular instant of metaphor in different types of discourse.Recently, this approach of handling metaphors and their linguistic textual context becomes more methodological and uncomplicated by means of computer software.Such software paved the way to corpus-approaches of metaphor analysis.

Corpus-Based Research of Metaphors
Corpus-based research of metaphors is based on studying the different linguistic realizations of a metaphor and all its forms in a corpus developed by the researcher (that contains text under study).Alice Deignan (2005a: 75) states that the term "corpus" refers to "collections of citations of various kinds, such as dictionary entries, or pre-selected collection of utterances containing linguistic feature of interest, such as metaphor".Furthermore, she emphasizes that this "collection" must comprise naturally-occurring texts which have been stored in machine-readable form that can be studied using various computer programs (Deignan, 2005a: 76).In fact, most corpus-based linguistic studies and theories are built around the orthographic nature of an individual word or a group of words and their occurrence in any given text.Thus, corpora as large collections of orthographically represented texts can only be accessed by searching for individual entry (a word) or any other groups of words by the means of computer software.
Early empirical studies of naturally occurring metaphors in both spoken and written corpus-based forms have focused on providing the specific contextual data needed in order to predict the appearance of metaphors in certain kinds of discourse (Martin, 1994: 211).For instance, Pollio et al. were amongst the earliest to comprehensively analyze the frequency of metaphor and metonymy in a wide variety of texts (Pollio et al., 1990).The results of their analyses reveal that both metaphors and metonymies emerge in any text relatively frequently, and they estimate the emergence of metaphor in a text on average around 5 times per 100 words of any given text, regardless of its nature.However, it is found that one of the major obstacles that face corpus-based analysis of metaphors is the fact that metaphorical mappings are not associated with specific lexical items.Still, early corpus-based studies of metaphors prove that corpus linguistics can be a helpful instrument for establishing many empirical bases for studying other aspects of metaphorical language structure such as their lexical semantics and grammatical forms.To solve this problem, it is very important to deal with metaphors in any given corpora by looking at the metaphorical markers.
Metaphorical markers are those linguistic expressions that can be understood as indicating a particular metaphor, and once they are identified within a discourse it becomes easier to find them and study their recurrence within the corpus.Goatly (1997) refers to metaphorical markers by elucidating them as the words and phrases occurring in the environment of a metaphor's vehicle term (or the 'Source' domain in CMT).Furthermore, a metaphorical marker can be any unit of discourse that unconventionally refers to or colligates with the topic of a metaphor on the basis of similarity, matching or analogy (Goatly, 1997: 8).In view of that, Sznajder and Angordans (2005)

As it were
However, Sznajder and Angordans (2005) see that although it is not fully described in his studies, Goatly's classification of metaphorical markers would allow for relatively direct access to potential metaphorical material in any linguistic corpora that are stored electronically.This method enables the researchers to highlight the phrases used as markers of metaphors in context and find it easily by using suitable concordancing computer software.
Among the most significant works to study metaphors in real contexts are the works of Alice Deignan (Deignan, 2005a;2005b;2003;1999a;1999b;1999c;Cameron and Deignan, 2003).In her work Metaphor and Corpus Linguistics, Alice Deignan (2005a) recapitulate her long experience in the field of corpus-based studies of metaphors.This work is a practical survey of different approaches to the analysis of metaphor in discourse.Deignan is one of the exceptional metaphor scholars who have combined corpus linguistics and metaphor research.Her work mainly involves the investigation and analysis of metaphors in different electronic texts or transcripts of spoken discourse from different domains.Deignan's investigations, based on CMT, examine how metaphorical language is used for meaning-making in different types of spoken and written language.Deignan's investigations also aim to answer her primary questions about the extents to which a theory of metaphor-as-thought can account for patterns found in natural language.To pursue the answer, she relied on the analysis of a cross-section investigation of some corpora of the Bank of English Corpus.According to Deignan, evidence from corpus analysis of metaphors, such as their frequency and context of use, can draw sensible conclusions about distinctions between the different categories of metaphor which could be classified in terms of their 'metaphoricity' as innovative, conventionalized, dead, or historical metaphors (Deignan, 2005a: 36-46).For example, she states that "any sense of a [metaphorical] word that is found less than once in every thousand citations of the word can be considered either an innovative or a rare [metaphor]" (Deignan, 2005a: 40).On the other hand, she sees that an historical metaphor is one which is found to only show the metaphorical meaning in the citations of an expression in the corpus if the literal (not metaphorical) sense has since dropped out of use (Deignan, 2005a: 40).
The most significant aspect of Deignan's work (2005a) is her introduction of many of the basic concepts and techniques of studying metaphors by relying on corpus linguistic work.In fact, corpus linguistics raises new questions for research under CMT.If we take as examples the different tense forms of the same verb, or the singular versus the plural form of a given noun, we notice that they often have different meanings.For example, Deignan analyzed the singular noun rock, which metaphorically signifies the feature of stability, while the plural form rocks indicate danger (Deignan, 2005a: 87).
Herein, the two different uses of the words rock and rocks participate in defining a situation in which the two grammatically different forms of the same word rock can correspond to different target domains.
In addition, Deignan pays attention to the cognitive and psycholinguistic approaches to metaphor research.In this respect she draws attention to some problematic issues of the study of metaphors in corpora; their ambiguity, frequency, context and authenticity, for example, in terms of the frequency of certain words and the assessment of their most frequent use.In contrast to the expected intuitions of most native speakers of English, Deignan discovered that less than 10% of the corpus citations of the verb soar (and its inflections) in her sample are literal (Deignan, 2005a:118).She states that the norm for most of her examples of the verb soar in expressions like "...first time buyers, driven out of Dublin by soaring housing prices" involves a metaphorical rather than literal (or physical) rising of house prices.On the other hand, Deignan emphasizes the problem of context and its significance to corpus-based studies of metaphor.She points out that some experiments have attempted to provide the context of any expression that involves metaphors.However, these texts which aim to explain the context are not part of the real context, but they are imposed by the researcher himself or herself.This fact, according to Deignan, brings many problems such as the inadequacy of these texts (Deignan, 2005a: 118) and their contribution to the understanding of the metaphorical or literal meaning because of the need to attain some shared knowledge between the speaker or the writer from one side and the listener or speaker from the other.Furthermore, Deignan sees that in any invented text used to describe a context in a corpus "indications of the people involved, their relationship and the channel of communication are often absent" (Deignan, 2005a: 118).So, even though these invented texts could be of significant importance in explaining the surrounding environment of the real text in a corpus, they may also raise other problematic issues which can shift the attention from the formal context of the speech to other informal and pointless content.
To sum up, Deignan (2005a) argues effectively in favour of employing different types of texts corpora from different domains in metaphors research.She also sees that even large corpora, such as the Bank of English Corpus, are at the heart of corpus linguistic studies, small and special invented corpora of texts from different domains such as politics, economy, and advertising can better suit the researcher's own purposes when studying metaphors in different domains.Furthermore, such small corpora may have the advantage of being accessible and searchable by using simpler software than those highly sophisticated ones which are normally used with the larger corpora databases.
Deignan's contribution to the fields of corpus linguistics and metaphor research has paved the way for other researches of metaphor using small or large corpora in studying metaphors in genuine texts.These corpus-based studies of metaphor involve texts from different domains such as politics and ideology (Goatly, 2006;Charteris-Black, 2004), advertising (Lundmark, 2005), educational contexts (Cameron, 2003), and politics, race, and economics (Skorczynska and Deignan, 2006).

Critical Metaphor Analysis
Another noteworthy approach of metaphor analysis is Jonathan Charteris-Black's Critical Metaphor Analysis (CMA).This approach combines the techniques of corpus linguistics, cognitive semantics and Fairclough's Critical Discourse Analysis.In his monograph Corpus Approaches to Critical Metaphor Analysis, Charteris-Black examines metaphor in different types of discourse: political discourse, press reporting (including sport and financial reporting), and religious discourse.Furthermore he emphasizes the idea that metaphors should not be studied in isolation from the context in which they are used (Charteris-Black, 2004: 10).CMA approach presupposes a process of identifying and analyzing metaphors through close reading of the texts for identification, interpretation and explanation of the metaphors (Charteris-Black, 2004: 35-39).
In his analysis of the different corpora that he compiled, Charteris-Black makes the first move in metaphor analysis by looking for the different source domains of metaphors in each text such as journey metaphors, building metaphors, plant metaphors conflict metaphors, and religious metaphors.Then he investigates their representation within different small samples of texts for their occurrences, frequencies and values in the different parts of the corpora, or 'registers'.This methodology of metaphor analysis in different kinds of discourse depends on looking for "the presence of incongruity or semantic tension -either at linguistic, pragmatic or cognitive levels -resulting from a shift in domain use" (Charteris-Black, 2004: 35).This incongruity is considered the key factor for identifying a metaphor.Then, Charteris-Black suggests classifying metaphorical expressions which are commonly used with a metaphoric sense as 'metaphor keywords'.The analysis of these metaphor keywords involves a quantitative search for such keywords in a text corpus.By this quantitative analysis, Charteris-Black aims to assess the degree of conventionality of each metaphor keyword, and whether it is used metaphorically or not.So, wherever a metaphor keyword is found to be of high frequency, it is a conventional metaphor, which is, according to Charteris-Black, more important than creative ones, because they give us more indications about the different rhetorical or ideological strategies employed by the discourse maker and these may reflect hidden aspects of his character.
One the most elaborate examples that Charteris-Black discusses is his discussion of the word 'crusade' in the phrase 'crusade against terror' which was uttered by the American President Bush after the attacks on the two towers of the New York Word Trade Centre in 2001 (See Charteris-Black, 2005).Charteris-Black sees that this example as topical but controversial since it is argued that President Bush intended crusade metaphorically whereas many Moslems interpreted it literally.Charteris-Black points out the word crusade is particularly painful to Muslims since it takes them back to its literal and historical meaning ('any of the mediaeval Christian military expeditions to win the Holy Land from the Muslims') which implies that this war against terrorism is, in fact, against Islam.However, Charteris-Black claims that he looked for the keyword expression 'crusade against' in a corpus of different academic and newspaper texts, and found that its metaphoric sense often collocates with 'corruption', 'slavery', 'communism', 'cancer' and 'crime'; thus, it is obvious that all of these collocations carry strongly negative evaluations (Charteris-Black, 2005: 36).
Charteris-Black argues that Bush's rhetoric draws on the conceptual metaphor POLITICS IS RELIGION whereas his opponent in ideology, the Islamist terrorist Osama Bin Laden, draws on the metaphor CONFLICT IS RELIGION.
Furthermore, he argues that the constant use of these metaphors from the domain of religion has created a problematic link between the domains of 'politics', 'conflict', and 'religion', and each of them participates in creating different conceptualizations of terrorism in the audience (Charteris-Black, 2005: 41).
Charter-Black has also studied metaphors in religious discourse from the Bible and the Koran, for example, in his study of the semantic field of 'Light' and the way it is used in the Holy Koran as an example of metaphor in religious discourse.He argues that both 'Light' and 'Dark' metaphors are a common collocation in the Koran, with sixteen occurrences in the same verse.One example from the Holy Koran states: With Allah guides him who will follow His pleasure into the ways of safety and brings them out of utter darkness into light by His will and guides them to the right path.(The Dinner Table,5:16) Charteris-Black sees that this systematic occurrence aims to highlight the conceptual metaphor SPIRITUAL KNOWLEDGE IS LIGHT and SPIRITUAL IGNORANCE IS DARKNESS (Charteris-Black, 2004:231).This argument within religious discourse in general and the Koran in particular makes metaphors "carrying a strong role in evaluation of behaviour and its relationship to spiritual welfare predominate" (Charteris-Black, 2004: 230-231).
In fact Charteris-Black emphasises in his corpus-based analysis of different kinds of metaphors in the domains which carry an ideology that has to be transmitted to other people.He argues that cognitive semantics conceals the fact that "metaphor selection in particular types of discourse is governed by the rhetorical aim of persuasion" (Charteris-Black, 2004: 247), and according to him his corpus-based approach fills this gap in metaphor research.Another case in point of Charteris-Black's corpus-based analysis of metaphor is his focus on the link between the frequency of certain metaphors and their values within the discourse.For example, he notes a high frequency of body part metaphors in American presidential speeches.Nations, as well as cities and all human and political communities, are often metaphorically conceptualized as persons or bodies: "These [body part metaphors] are quite high frequency in the corpus and are perhaps best considered as blends of metaphor and metonymy based on some familiar relations of correspondence of particular parts of the body with particular actions.The hand is metonymically associated with all types of physical action, the heart with feeling, the head with thinking and the eyes with seeing (and metaphorically with understanding)."(Charteris-Black, 2004: 105).
As a result, Charteris-Black argues that CMA approach complements the traditional cognitive theory of metaphor.Furthermore, he insists that cognitive semantics can present a good model of the interpretation of metaphors, but it is not adequate to account for why certain metaphors are chosen by certain discourse makers and in specific discourse contexts.In fact, Charteris-Black's main argument is that social, ideological, and societal factors contribute greatly to the choice of conceptual metaphor, which in turn implies the inevitability of studying context in any research that deals with metaphors.Charteris-Black's work demonstrates that after the remarkable shift to the study of metaphor from a cognitive to a corpus-based point of view, metaphor researchers are right to redress the balance and pay more attention to the study of the context.

Implications
An oral or written discourse represents an organized form of human experience, and this representation is established through processes of recitation and interpretation of different events and in different contexts within many domains of experience such as literature, trade, politics, or religion.The presentations above accentuate the significance of the scientific contextual study of metaphorical language because a discourse does not constantly constitute a continuous set of literal meanings.The nature and function of any type of discourse in addition to its context can affect its language.Consequently, I conclude this theoretical account by commenting on the use of metaphorical language in didactic discourse taking the Islamic religious discourse as my case study.
The term 'religion' covers ancestral or cultural traditions, writings, history, and mythology, as well as personal faith and experience.And any religion involves a set of tenets and practices that presume the existence of a supernatural power that controls life.This power can be the 'best' and most accepted explanations for many incomprehensible events and phenomena such as the creation of earth and universe, human nature, his birth and death.In general, the religious traditions come in a form of verbal and 'linguistic' activities such as prayers, rituals, and ethical principles and instructions.And in many religions, these traditions are codified to encompass a great portion of the componential burden of rites, principles, and laws which codified within the religious scripts.However, most of these codified forms where written in a distinguished style that provides the necessary magnitude and dignity of the religion.The rhetorical and stylistic components exist in the linguistic repertoire of the community where a religion emerges can play a significant role in arousing the awareness of dignity and reverence for the religious text.In addition, the narrative interpretations of principles, instructions, ideas and beliefs of the religious experiences of any group of people who belongs to a religion can make the religious language of this group.When a religion emerges in a community, its language (the language of the community) will be primarily used as an instrument for worshipping.Sometimes, and after a quite long period of time, the early forms of the language may 'suffer' from some changes in its linguistic system.Nevertheless, the new religion followers often assign a sort of virtuous and divine nature to the early forms of this language in order to protect it from change, because it was the first language used to carry the message of their religion.Consequently, solemnity and dignity will be associated to the early forms of the language to promote its place over other existing new varieties of the language.Furthermore, the religion followers will start believing that it is important to learn the 'correct' language and its words to use them in their worshipping.
The religious language of the expressions used in sermons such as prayers and rituals has an undeniable role in arousing the feelings of ecstasy that accompany the religious rituals.The words of these expressions carry religious or other emotional meanings which can convey the speaker's religious emotions, and they can arouse similar emotional responses from the listeners.Religious language has been studied at first by philosophers who focused primarily on the epistemological nature of the religious beliefs of certain communities.In general, most of their interests have been oriented to the connotation and denotation of the religious words and their roles in arousing certain kinds of 'necessary' religious emotions.However, religious language can serve other purposes beyond stimulating emotions.From this point, many scholars highlighted the significance of investigating the ideological and intellectual functions and beliefs impeded in religious discourse.
Metaphorical language was appreciated in Arabic rhetoric but did not enjoy similar appreciation by philologists.In the past, it has been considered as supportive and ornamental instruments in the language that involves arguments and debates to attract the reader or the hearer.However, metaphorical language enjoyed a special status in didactic discourse because of its capability on revealing deep insight into a few impeded relationships between different things.Although a metaphor is essentially used as an ornament for a discourse, its importance stems from its capability to influence the readers', or hearers', feelings.This emotional appeal is attributed to the admiration it acquires by the reader for its being strange and extraordinary.Now, in a didactic discourse, like the religious one, a metaphor can be used to explain an abstract metaphysical spiritual message for the purpose of emphasizing its meaning and overstating its properties.A metaphor is capable to refer to a meaning rhetorically in a few words or clarifying the intended message and is meanings in a way that could not be achieved entirely by using everyday literal language.This postulation is imperative within the Arabic religious context as Islamic religious discourse has been largely influenced by the necessity to interpret the Holy Quran and the Prophetic Tradition, the Hadith.The development in the science of inference and explanation (Tafsir) of the divine word in the scriptures of the Holy Quran and the Hadith relies mainly on the full mastery of the sciences of the Arabic language including its rhetoric.Accordingly, it was necessity for studying metaphors in order to correctly deduce the principles and orders of Islam from the sacred texts.In addition, the study of metaphorical language methodically questioned the nature of metaphors in Islamic religious discourse, especially in the Holy Quran.As some philosophers and theologians believed that the word 'metaphor' is a synonymous to 'untrue' and 'false', they refused the idea of the existence of metaphors in the Holy Quran and the Hadith.On the other hand, some other religious "Sufi" (puritans) groups (such as the Esoteric (Batinites)) maintained that almost all the words in the Holy Quran, and many other kinds of religious texts, are metaphorical, and that they must be interpreted metaphorically.
On the basis of the long tradition of studying (cognitive) metaphors in language and discourse, one may sustains the argument of the inevitability of metaphors in any religious discourse.The incomparable nature of divine communication entails the existence of metaphors which can transmit the divine message content into the human language.Since religion does not have a special language of its own; it must resort to ordinary language in accordance with society's convention in which the given language operates as a means of conversation.Accordingly, messengers (of faith) easily expressed the distinguishing qualities of the divine language to the common people through similitude; taking into consideration that messengers were always sent speaking the language of their societies.
Still, the copious existence of metaphors in religious language does not necessarily entail that a metaphorical interpretation can be always taken for granted.In Islam, a metaphoric interpretation can be deemed unsound and disregarded if it goes, explicitly or implicitly, against the well-established principles of Islamic faith and creed explicitly mentioned in the Holy Quran or the Prophetic Tradition.Metaphorical expressions in any given religious discourse require some sort of metaphorical interpretation (ta'wil) related to the inference or exegesis (tafsir) of the religious text.
A sound metaphorical interpretation, then, must be subject to the norms and rules.First of all, a professional mastery of the Arabic language and its conventional use is indispensible.Secondly, we must always look for a 'religious evidence' (dalil naqli) that can support our 'rational evidence' (dalil aqli) to maintain the metaphorical interpretation that we have.Thirdly, and possibly most importantly, we can never accept any metaphorical interpretation if it goes against the known principles of religion.And we can add that it is vital to look at the society in which the metaphors were delivered taking into consideration that words meanings evolve over time.So, some metaphors may have certain implications when the Prophet and his Companions used them that have not the same effect or meaning in our days.
We may willingly admit that symbolic language, in general, and liberal use of metaphors in particular may have a deep and continuing impression on the heart, and give the religious texts more esteemed and 'divine' status.Hence, the immense weight of metaphors in didactic religious discourse cannot be ignored since no language, especially a didactic religious one, means always literally what it appears to mean.So, no one can deny that metaphorical language is a normal, perhaps even necessary, phenomenon in religious language (Ayden, 1997: 2).

Figure 2 .
Figure 2. The conceptual metaphor "A SEGEON IS A BUTCHER" according to the BT (See Fauconnier and Turner 1998; Croft and Cruse 2004:208); source: my drawing.