DEONTIC MODALITY IN GHANAIAN PENTE-CHARISMATIC SERMONS

: The linguistic analysis of sermons has in recent times attracted some attention in linguistic research. Notwithstanding, studies on language use in sermons have predominantly examined the style of individual preachers but with limited attention to linguistic resources on deontic modality. To fill this gap, the present study examines the linguistic resources employed in expressing deontic modality in selected Ghanaian Pente-Charismatic sermons. It is theoretically underpinned by the Register Theory propounded by Halliday (1989) and Biber and Conrad (2009). The study is a qualitative case study that purposively sampled twenty Pente-Charismatic sermons delivered by three Ghanaian preachers namely Apostle General Sam Korankye-Ankrah; Bishop Charles Agyin-Asare and Bishop Nicholas Duncan-Williams. The findings reveal that modal verbs are used to make declarations and admonish listeners on various Christian principles, lexical verbs are used dominantly in the form of action verbs to construe preachers as powerful people who instruct congregants on their Christian obligations while auxiliary verbs are used to support the main verbs to emphasise various actions in the sermons. Phrasal verbs, adverbs, and clauses on the other hand are insignificantly used to enable the preachers to convey their messages in plain language that would be understood by their audience. This paper extends the scholarship on register analysis of sermonic discourse by focusing attention on Pente-Charismatic sermons, a context underexplored in the literature. It also illustrates the communicative implications that


Introduction
The practice of religion plays an important role in the life of people especially in Africa as affirmed by the Afro Barometer Report on religion in Africa in 2020 which indicates that "more than nine in 10 Africans (95%) identify with a religion" (Afrobarometer Report, 2019, p. 2). This picture is not different in the Ghanaian context as Ghana's most recent population census in 2022 indicates that in relation to religions among the population of Ghana, 71% of Ghanaians are Christian and 17% are Muslim (Ghana Population Census, 2021). This indicates that close to 90% of Ghanaians are affiliated with a particular religion. Christianity and Islam are the two main religions in Ghana. The practice of religion generally and these two religions specifically cannot be effective without the use of words of admonition from localized and national religious leaders. The practice of delivering religious messages of admonition is technically termed as a sermon. The delivery of sermons, therefore, permeates both the Christian and Islamic divides. With the importance attached to sermons, linguistic studies have been carried out on both Christian and Islamic sermons. Most of the studies on Islamic sermons are centered on their Friday Sermons (Underwood and Kamhawi, 2015;Abdel and Eldin, 2014;Olaniyan and Oyekola, 2006;Kareem, 2018). Relatively, Christian sermons have received more linguistic research attention (Kareem, 2018). The preaching of Christian sermons seeks to work on the salvation of worshippers and deliver them from emotional, spiritual, social and political challenges (Idowu, 2007). Idowu, Aliu, and Ikuelogbon (2021) aver that language is dynamically used in the delivery of sermons to persuade Christian believers to lead a life in sync with their religious beliefs. From the Christian context, sermons are a cardinal form of discourse with evangelical and non-evangelical purposes. They are delivered weekly, sometimes more frequently depending on the particular Christian denomination. Rose (1997) identifies four types of contemporary Christian sermons namely traditional sermons, kerygmatic sermons, transformational sermons and conversational sermons.
The sermon as a genre has its own conventions depending on what a preacher intends to achieve. In sermons, various linguistic strategies are adopted in as much as an atmosphere of spirituality is maintained (Berghegger, 2009). As required, listeners are able to discern in order to tow a particular line of life as prescribed by their belief. Also, with the current situation where churches have sprung up all over, and sermons preached everywhere, especially in public transports, on TV and radio stations, preachers also strive to keep up with current trends of preaching in order to secure their loyal members in their churches. This phenomenon however has drawn the attention of scholars to the study of the nature of religious discourse. In the past few decades, various linguists such as (Garner, 2007;Berghegger, 2009;Dzameshie, 1995) have focused on the study of sermons. Findings from these studies also awakened the need to look further into the linguistic and rhetorical features of sermons. With this, scholars such as Ladzekpo, (2012) also saw the need to study sermons as one of the genres that are worth exploring. The various areas explored by earlier researchers in sermons include rhetoric, metadiscourse, and communicative events of sermons among many others and they have revealed interesting findings about the study of religious discourse. Some of these earlier studies highlighted the important elements of sermons and these also have provided certain evidence that have served as a guide for future studies such as this. The study of sermons in the early days was based on theology but gradually, attention shifted to phonological activities in the delivery of sermons (Wharry, 2003), historical linguistics of sermons (Boggel, 2009), and now the study has moved beyond all these levels to the investigation of how language is used to convey intended messages to the listeners (Koncar, 2008;Adams, 1995;Uzuako, 2013;Uzuako 2014;Ekeoha, 2015;Apata, 2016;Idowu, et al., 2021). Apata (2016) observed an increase in Pentecostalism in different parts of the world especially in Africa. This increase in Pentecostalism has come with commensurate attention in terms of linguistic research.
Literature on sermons has focused on the nature of sermonic discourse (Hills, 1989;Dzameshie, 1995). Other studies (Wharry, 2003;Berghegger, 2009) also analysed prosodic features in sermons. Studies by Koncar (2008) and Adams (1995) also investigated the rhetorical design of sermons and the accountability of sermons respectively. Uzuako (2013) and Uzuako (2014) explored the lexical collocations and lexical choices of sermons while (Rajtar, 2012;Kim, 2016) examined the portrayal of women in sermons. Recent studies have examined the use of pragmatic strategies in Pentecostal Church Sermons, focusing on the Nigerian context (Ekeoha, 2015;Apata, 2016;Willy and Mbakop, 2018;Akinwotu, 2021). Other studies have also examined the use of appraisal resources in Christian sermons (Idowu, et al., 2021). There is a gradual paradigm shift from Pentecostalism. Pente-Charismatism has become the new trend in Christianity today and is the preference for most Christians especially those in their middle ages as the orthodox associations are now related to loyalty or a family's long-standing association with the said church. On what constitutes the uniqueness of a Pente-Charismatic church, Yong (2004, p. 282) is of the view that it is "not just the utterance of strange tongues but also the gift of interpretation of tongues which preserves the distinctiveness of the communicative and receiving languages". By virtue of how these churches conduct their services, Pente-Charismatic churches have become attractive to most Christians. The youth are attracted to these sermons because they are characterized by signs and wonders which are rooted in the scripture. Chapman (2004, p. 230) also affirms this assertion that "the pages of confidence are full of examples" which cause listeners to believe whatever these preachers preach. These preachers believe that the level of confidence they exercise in the process of preaching is from the "Holy Spirit" (Klaus, 2006). The present paper contributes to Christian sermon discourse, focusing on a key yet neglected text type, Pente-Charismatic sermons. Specifically, we analyse Pente-Charismatic sermons delivered by three Ghanaian preachers namely Apostle General Sam Korankye-Ankrah; Bishop Agyin-Asare and Bishop Nicholas Duncan-Williams, focusing on the deontic modality in relation to the resources of the register provided by the register theory (Halliday, 1989;Biber and Conrad, 2009). These sermons are symbolising the current trends of Christianity with its history hinged on both charismatism and Pentecostalism. Their investigation will contribute towards a fuller understanding of the amalgamation of charismatism and Pentecostalism into a whole and increase our knowledge of how the use of register by preachers within this new trend differs from that of the former trends of Christianity. The present study, therefore, goes beyond previous studies to examine deontic modality in Ghanaian Pente-Charismatic sermons. We focus specifically on two objectives that seek to identify the linguistic resources used to express deontic modality in Pente-Charismatic sermons and examine the discourse functions that underpin the use of deontic modality patterns in Pente-Charismatic sermons. The rest of the paper reviews the theoretical lens that frames the study, the methodology, analysis and discussion of results and conclusion of the study.

Theoretical Framework
The study is grounded in Halliday's (1989) and Biber and Conrad's (2009) theory of register. This theory is an essential part of the Systemic Functional Linguistics (SFL) proposed by Halliday which basically highlights language use and its relationship with the culture associated with a particular situation. Register generally relates to the use of language depending on the situation and this is reflected through the appropriate choice of words relating to the particular field. Harold (2007) avers that register refers to the language choices users make to appropriately convey their messages in an acceptable or appropriate manner. This implies that the acceptability of a message is largely dependent on the choice of language by the user based on the context of the situation. Certain events are highly expected to take place in connection with certain situations, and there are words relating to such events. The ability of the user of a language to appropriately use these particular words demonstrates his linguistic competence. Register relates to language variety which is connected to the situation of use and its pervasive linguistic features that perform significant functions within the situation of use (Biber and Conrad, 2009). In other words, the register could be looked at as the use of language that authenticates the fact that a genre belongs to a particular field. There are numerous options available to language users in expressing the same thing but the user has to select the one that is most appropriate in the particular situation. Therefore, the ability of the user of a language to make the appropriate linguistic choices for the right situation makes him linguistically competent. Yet all these are made possible by adhering to the three configurations of register namely the mode, field and tenor. In the creation of such linguistic distinctions, Halliday's three configurations of register-field, tenor and mode, are relevant. This configuration is in line with Hassan's (1989, p. 56) argument that "the total features, from field, tenor and mode need to be interpreted as one configuration". This implies that these three must be identified in every discourse; whether spoken or written. Guazzieri (1998) defines the field as 'the subject matter' and it may be similar to certain uses of the term domain in computational linguistics to mean; when, where and why something is happening. He notes tenor as "the social relation existing between the interactions in a speech situation'. He also describes mode as 'the way language is used in the speech interaction, including the medium as well as the rhetorical mode".
The Hallidayan theory of register as situated in the systemic functional tradition construes register as how language is used in a particular context based on the variables of field, mode and tenor. Biber and Conrad (2009) on the other hand construe register in relation to the situational background underpinning various linguistic choices based on seven factors namely the background of participants involved, the relationship between participants, the channel of communication, the setting, the production and comprehension of circumstances, the communicative purpose, and the topic. Despite the difference in the number of variables (three and seven respectively) in these two theories of register, there exists a strong functional complementary nexus between them. Biber and Conrad's model presents relevant elucidations to the variables in Halliday's model. Frimpong (2015) details the complementary relationships that exist between various aspects of the two theories. He states that the "field is roughly relatable to topic, communicative purpose and setting of Biber and Conrad's situational variables. Mode relates with the channel of communication and production and comprehension circumstances, while, finally, tenor corresponds roughly to participants and relations among participants." (Frimpong, 2015, p. 75). In the study, the field, mode and tenor as the main variables in Halliday's model are used as the major points of discussion which the seven situational variables in Biber and Conrad's model are respectively aligned to.

Field
The field is the dimension of the register that concerns what is being talked about and the role of language in the activity that is going on (Thompson, 2014). It also refers to the subject matter which may concern what is happening, where and when, and why it is happening (Guazzieri, 1998). A text or discourse always belongs to a particular domain, and this domain in register studies is regarded as the field of a text or discourse. The field is the content of an interaction or text. It is the total event in which the text is functioning, together with the purposive activity of the speaker or writer; it thus includes the subject-matter as one element in it (Halliday, 1994). Martin construes the field in two perspectives: "what people are doing and what they are doing it to" (Martin, 2010, p. 20). The field is the social context in which a text is typically produced, the constraints the particular setting imposes on the speakers and the communicative purpose of the text. The delivery of sermons represents a particular human activity. In it, preachers engage their congregants with the main intention of presenting the word of God to them. To do this, they make use of language to influence their audience. These factors have a tendency of shaping sermon texts linguistically.

Mode
Mode refers to the conventions expected of a discourse or text belonging to a particular domain. It is the shared knowledge required by speakers concerning a selected genre. According to Guazieri (1998), mode describes the way language is used in interaction as well as the rhetorical mode. Martin (1992) defines mode as the role language plays and what participants expect to achieve in the situation. It is what is expected by the participants of the linguistic exchange through their use of language. Thompson (2014) also views mode as the way language functions in the interaction with regard to whether an interaction is spoken or written. The channel of communication also plays a cardinal role in the field. A written text may therefore differ from a spoken text in that, a written text may be well thought-through and edited but a spoken text may be spoken spontaneously. This may create linguistic complexity in written texts and relatively less complexity in spoken texts. The mode is as relevant to our understanding of the delivery of sermons as the communicative function. In fact, it is one of the most important physical situational contexts. In the first place, sermons, like any other spoken genre are composed under situational conditions that make them linguistically less complex. Though some sermons may be scripted, their delivery may be marked by some occasional off-script comments.

Tenor
Tenor refers to the participants in a discourse and the role and relationships the participants of the speech share (Martin, 1992). Some scholars such as (Esser, 2009) consider tenor as the people involved in an interaction and the status; they hold in the communication process. He compares tenor to style of discourse and concludes that tenor describes the people that participate in an event as well as the relationships they hold, and their status. Tenor as an essential part of the register is synonymous to style or what is otherwise known as the formality of language. Guazzieri (1998) emphasises the social relation existing between the interactants in a speech situation. He considers the level of formality, power relation and the effect of these on the speech situation. Gledhill and Diderot (2013) added that tenor involves the roles played by the participants in a particular genre. Frimpong (2015) posits that Biber and Conrad's situational features of participants and relations among participants perfectly align with tenor. His assertion is substantiated by the fact that the background of participants of a discourse can influence their linguistic choices which may unveil the status of participants at both ends of the communication. In a sermon text, the address or is the preacher. The audience is the congregants and sometimes absent audiences who receive the message via other mediums made possible by technology. The recipients of the message usually constitute people from varied backgrounds.

Methodology
The study is a qualitative study that investigates how preachers of Pente-Charismatic sermons convey their messages concerning the Christian faith across to their congregation using deontic modality. It is theoretically framed by the register theory based on Halliday's (1989) and Biber and Conrad's (2009) models. These models are applied as complementary models to boost the theoretical rigor of the study. Convenient and purposive sampling techniques were used by the researchers. Convenience sampling became necessary because it was the most suitable option available in determining the Pente-Charismatic preachers whose sermons would be used for the study. With this, the researchers made sure that all preachers whose sermons were sampled had enough CDs readily available and accessible. It also helped to gather enough sermons from the selected preachers. These were sermons that were preached between September, 2016 and April, 2017. The study employed a dual data collection approach. CDs of sermons from the selected preachers were obtained from various church stores in addition to what the researchers had recorded on tape recorders. A period of seven months (Sundays only) was used to collect data from the three preachers and this was done on a rotational basis. The period for the data collection was determined by the fact that some of the church stores contained a large quantity of the same sermons hence the inability of the researchers to easily get a variety of sermons. The researchers again visited some of the churches personally to record some of the sermons to complement the ones on the CDs. The researchers listened to sermons and sampled them for the study. A total of twenty (20) sermons were selected from the three Pente-Charismatic preachers respectively from September 2016 to April 2017. Sampled sermons were transcribed from the recorded form into textual forms ready for analysis. Pente-Charismatic preachers such as Bishop Agyin Asare, Arch-Bishop Duncan-Williams and Apostle General Sam Korankye-Ankrah were conveniently sampled because their recorded sermons were readily available and accessible. The purposive sampling technique was used to select a varied number of sermons from each preacher based on the variations in the length of the sermons and the presence of linguistic items denoting deontic modality. Five (5) sermons each were selected from Bishop Agyin Asare and Arch-Bishop Duncan-Williams respectively and ten (10) from Apostle General Sam Korankye-Ankrah. A total of twenty sermons were sampled from all three preachers and the selected number was predicated on Biber and Conrad's (2009) position on register analysis which suggests that the number of data must be on a representative sample of texts. Register analysis "must be based on analysis of a sample of texts selected to represent the register as fully as possible" (Biber and Conrad, 2009, p.10). We, therefore, considered this as a vital aspect in the selection of the number of texts for the analysis in this study. The researchers listened carefully to the sampled sermons on CDs and concurrently transcribed them from their initial recorded forms into texts. This instrument was necessary because it gave the researchers access to sermons in textual forms to be able to study them critically in order to achieve the objectives of the study. After the individual transcription, the researchers cross-checked their transcripts to reconcile some variations which boosted the reliability of the transcripts. To reduce subjectivity and increase objectivity, the consensual coding strategy was used where each of the three researchers carried out a separate coding of the instances of deontic modality in the texts after which, the identified instances of the use of deontic modality were compared in order to minimize the margin of error. If there were a disagreement on a particular instance of deontic modality, the researchers discussed the issue to come to a mutual agreement.

Results and Discussions
This section of the study presents the results and discussion of the findings of the study. It is intended to show the distinctive feature of the register of Pente-Charismatic sermons which are explored by analyzing the preachers' use of deontic modality. As indicated by Biber (1994) that registers are distinguished by differing exploitations of core linguistic features, the distinctiveness of Pente-Charismatic sermons is explored by studying such dominant linguistic features that set the register of sermons apart. Various language resources were found performing the functions of deontic modality in these sermons. Table 1 below shows the various linguistic items used in this regard, their frequency of usage and percentages. From the distribution, it is evident that in the sermonic discourse, 179 modals were used representing 43.0%. The most common resource used as represented in Table 1 was the lexical verbs with a frequency of 187 which represents 45.0% of the total distribution. The third most common form in the sermons is the primary auxiliary verbs with a frequency of 45 and a percentage of 10.8. Next are a pair of items; the phrasal verbs and clauses with the same frequency of 2 representing 0.5% each. The least used of all the linguistic resources in Pente-Charismatic sermons is the adverb with a frequency of 1 which represents a percentage of 0.2. The preponderant use of lexical verbs reflects the imperative mood in the use of language to pronounce blessings and oppose the workings of satanic forces in the sermons. A further and detailed discussion on the use of these language resources and their communicative implications in the sermons are presented below.

Modals
Among the group of linguistic resources expressing deontic modality is the modal category. To a large extent, modals were used in Pente-Charismatic sermons to express the preachers' authority to give hope to the listeners. Below are some instances in the sermons demonstrating such authority through the use of deontic modals. The register analysis of Pente-Charismatic sermons indicates that preachers have the authority to make declarations in the form of promises. This is a confirmation of Ronan's (2013) and Benicot et al.'s (1993) view that speakers show their commitment to some future behaviour or activity by making a promise and this type of modality usually manifests linguistically as declarative sentences and this is done by agents that have the ability to do so. Examples 1 to 5 demonstrate that in preaching Pente-Charismatic sermons, preachers give hope to the listeners of their sermons with a certain kind of authority because by virtue of their standing as spiritual leaders of the church, they are considered as agents that have the power to do so. Their source of power emanates from the fact that they believe they have been divinely reposed with such authority and so whenever they stand in front of the congregation at that moment, whatever they say is regarded as a message from God. This authority to preach and give hope is a significant characteristic of charismatism which is synonymous to the infilling of the Holy Spirit and prophesy (Baumert, 2004). This signals that under the influence of the Holy Spirit, preachers can make promises that can give hope to their listeners. The use of modals such as 'would' and 'will' as in examples 1, 2, and 5 show that the Pente-Charismatic preachers make declarations in the form of promises to their listeners in the process of preaching. The use of 'must' on the other hand as in example 3, shows that sometimes declarations and promises made by preachers Pente-Charismatic sermons become definite and compelling such that the listener is bound to receive them. The analysis also shows that preachers' declarations and hopes are given in the form of appealing to God to grant the listeners their desires. This shows that in Pente-Charismatic sermons, certain declarations are made to the congregation that is tied to certain conditions. Thus, the system in this regard is a bargain that allows the listener to benefit from the expected future occurrence if only they also fulfilled their part of the bargain. In example 6, 'when' has also been used to express the same purpose of the condition that the listener should expect the benefit in the main clause by fulfilling the expected role in the subordinate clause. The implication of the use of this resource shows that the listener should expect the outcome either good or bad for their actions and inactions.
From the analysis, modals have also been proven to have been used in Pente-Charismatic sermons to give godly instructions to the congregation. These instructions pertain to what a Christian is expected to do either by God or by the preacher. Examples of such usage are evident in the following lines in the sermons. Evidently, two modal categories 'should' and 'must' have been proven to have been playing the role of giving Godly instructions in Pente-Charismatic sermons. The use of 'should' as in examples 11, 12, and 13 shows obligations that are expected to be carried out by the listeners. It further shows that the listeners should make conscious efforts to perform such obligations. These manifestations in sermonic discourse highlight Durst-Anderson's (1995) position that deontic modality can manifest as advice, admonition, permission, command, or request. On the other hand, 'must' has been used to show compelling obligations that the listener is bound to perform and has no choice but to do so. This may be seen largely as Godly advice that may be considered as commands and it highlights Allan's (2013) view that deontic modality pertaining to obligations concerning someone doing something is expressed by modal verbs of obligation.

"You
Of the individual modals identified to have played the role of deontic modality, there were some that were more pronounced than the others and the analysis of these and the possible reasons for their excessive use would be further reported below. The analysis presented in Table 2 projected that the modal 'would' was predominantly used in the selected sermons. This modal element was used 67 times representing 37% which shows that the modal 'would' is indispensable in Pente-Charismatic sermons. In other words, the number of occurrences of this modal element also suggests that 'would' is obligatory in Pente-Charismatic sermons. As an essential element of register analysis indicated by Bhatia (1993;5) that 'register analysis focuses mainly on the identification of statistically significant lexicogrammatical features of linguistic variety', the modal 'would' therefore have a significant role to play in the register of Pente-Charismatic sermons. The modals 'may' and 'will' are invariably the second most used modals in the discourse involving Pente-Charismatic sermons. The optional notion inherent in the use of the modal 'may' suggests that in as much as certain aspects of sermons are compelling and obligatory, there are also aspects that are obligatory and leaves the listeners to make their own choices concerning certain pronouncements and declarations. 'Will' on the other hand was also used with the same frequency as that of 'may'. This frequency balance of usage could be interpreted as referring to a balance in formality. Whereas the former gives the listeners the option to choose or not, the latter suggests finality in choice or decision. Also, the fact that one of these is formal and the other informal yet they both have the same frequency of occurrence means that there is a balanced level of rigidity in Pente-Charismatic sermons.
The occurrence of modals such as 'must' and 'cannot' having the same frequency can be described in terms of aspects of the sermon that are made compelling to the listeners. Both 'must' and 'cannot' display a typical indication of power and authority exerted in sermonic discourse. This does not only reveal the preacher's authority to compel listeners to perform or not to perform an action but also depicts the listeners' ability to listen and obey whatever is preached in sermons. It is surprising however that the modal 'must' performed poorly in occurrence considering the fact that sermons cause listeners to follow a certain path of life. This also confirms the use of the two modal verbs 'must' and 'cannot' as a tool of infiniteness that is used to reflect the voice of the Supreme God.
The modal verbs 'should' and 'shall' were less frequently used than 'may' and 'cannot'. The two were used 14 times representing 8% contributing to the instructive nature of Pente-Charismatic sermons. The modal verb 'should' for instance was used to indicate a strong authoritarian position of the preachers in the sermons.
As regards the modal 'can', it was used only 11 times as indicated in example 6 to resonate admonishing and warning to listeners of things that are possible to happen if they do or not do certain things. Following Bhatia's (1993) model of register analysis, the position of the modal verbs in Figure 2 above, the frequency of occurrence of the modal verb 'can' and its position on the table renders it not so significant an element in Pente-Charismatic sermons but yet cannot be disposed of the sermonic discourse.
The distribution of the modals in Table 1 also reveals another interesting thing about Pente-Charismatic sermons. As a component of Halliday's tenor of discourse, the sermons were full of modals that suggest certain levels of formality and politeness. The modals 'would' and 'may' were possible choices over their less formal and less polite counterparts 'will' and 'must'. This suggests that to some extent, Pent-Charismatic sermons are delivered in a polite manner and are generally formal.

Lexical Verbs
Having discussed the use of modal verbs as a linguistic resource that is used to express deontic modality, other elements identified to be performing the same function also need to be discussed. In the register of Pente-Charismatic sermons, lexical verbs were found to have the highest occurrence as elements expressing deontic modality. This function of lexical verbs is realized in the context due to the imperative mode of the verbs although traditionally, some may not be considered deontic. Throughout the Pente-Charismatic sermon texts, lexical verbs were used to express actions and things that the preachers expect the congregation to do. Interestingly, most of these verbs are action verbs that instruct listeners on what to do as instructed in the sermons. Below are some instances of the use of lexical verbs in the sermons. As a key element of charismatic sermons, the preacher is baptized and filled by the Holy Spirit hence, speaks under the unction of the Holy Spirit (Baumert; and the congregation readily performs whichever tasks they are asked to perform. Evidence from the examples above vividly explains Bhatia's (1993) definition of register as the language choices usually involving vocabulary and syntax that speakers of a language make in conveying the appropriate message in a given context. In the examples above, example 16 for instance demonstrates finality in the preacher's speech as someone in whom divine power has been reposed. This was achieved by the use of the action verb 'stop' as a means of commanding the listeners to refrain from performing an act as in example 16. As an element expressing deontic modality, it is a confirmation of Smart's (1984) assertion that imperative elements are inherent in deontic statements and this is evident in Pente-Charismatic sermons.
Results from the analysis also show that in Pente-Charismatic sermons, preachers have the characteristic feature of exercising authority over things that are not seen. When on the pulpit, these preachers feel so empowered that they feel everything is possible and everyone listening to them is subject to their authority. The examples below further illustrate this assertion. Analysis of the above categories reveals that preachers of Pente-Charismatic sermons demonstrate that they have been empowered by their belief just as Jesus Christ to evoke and command things in the unseen realm. The lexical verb 'bind', as used in example 19 for instance, indicates that these preachers directly deal with the unseen spirits by stopping them from operating. Also, the use of the verb 'command' in examples 21 and 22 portrays the authority of Pente-Charismatic preachers to command seen and unseen forces. Demonstration of power over the unseen is also exhibited in example 20 where the preacher used the verb 'release' to evoke the blood of Jesus. The use of these lexical verbs above presents the preachers as powerful persons whose source of power is drawn from Jesus Christ.

"I bind
Again, the use of lexical verbs in Pente-Charismatic sermons signals that preachers have the power to grant or give to the listeners whatever their desires are. They make declarations that put the listener in an instant mood of reception. Examples 23-25 further explain this. Evidence from examples 23, 24, and 25 suggests that preachers have the power to give divinely endowed gifts to the listeners. This element of power is illustrated in their use of verbs suggesting that a person who possesses something and is willing to give out to others. Verbs such as 'receive' and 'increase' in examples 23 and 24 respectively show that the preachers possess such powers and can also give out anything they feel the listeners lack or need. These verbs were used without subjects and this again suggests that the speaker has dominion or sovereign authority to make such proclamations. In example 25, the verb 'confer' has been preceded by the subject 'I' which also reveals that preachers of Pente-Charismatic sermons consider themselves as repository bodies that can then give out to the listeners from the store of power and authority they have from God [PCS05.dm38]. This thus gives the justification that Pente-Charismatic preachers sometimes do not always ask God to bless the believers but do it on God's behalf because they have the power to do so. Lexical verbs are indispensable in the discussion of linguistic resources that express deontic modality, especially considering the quantity used. In other words, it could also be described as the most important item that can vividly convey deontic sense in Pente-Charismatic sermons.

Auxiliary Verbs
One of the language resources identified in sermonic discourse is the use of auxiliary verbs. Evidence from the register analysis of Pente-Charismatic sermons shows the extensive use of auxiliary verbs basically to express deontic modality emphasis. Auxiliary verbs are largely dependent on main verbs in clauses and this is also true in the case of the language of Pente-Charismatic sermons.
In several respects, auxiliary verbs have been used in sermons in a plethora of ways. They help to emphasise whatever action the speaker needs to express in his speech. The examples below feature three auxiliary verbs in Pente-Charismatic sermons which would be subsequently discussed.
28. "You also need to persevere." [PCS20.dm 21] 29. "You have to go into marriage with a mind of permanence." [PCS19. dm03] 30. "You do not treat your wife like a maid." [PCS20.dm09] The auxiliaries used in the examples above emphasized the actions that were meant to be performed by the main verbs 'persevere' 'go' and 'treat'. In example 28 for instance, the action the preacher wants the listeners to perform is to persevere but the auxiliary 'need to' adds a sense of obligation to the action meant to be performed. Example 30 also features the auxiliary verb 'do not' preceding the main verb 'treat' to show a prohibition of an action the preacher feels the listeners are likely to perform. This finding in the Pente-Charismatic sermon is not surprising considering the fact that in deontic modality, the speaker could prohibit the listener from performing an action.

Phrasal Verbs
Findings from the analysis revealed that there is minimal usage of phrasal verbs in the Pente-Charismatic sermons. Sinclair (1992)  As a significant feature of phrasal verbs, evidence clearly shows that in the culture of Pente-Charismatic sermons, preachers advise listeners to do certain things sometimes not considering the individual meanings of the words they use but the idiomatic sense inherent in them. For instance, the phrasal verb 'put up' as illustrated in 31, the preacher seeks to inform the listeners to cultivate an attitude of humility but not the literal sense which may mean to raise something up. The minimal use of this language resource in Pente-Charismatic sermons suggests that preachers prefer to convey their messages in plain language that would be understood by every listener.

Adverbs
The analysis did not show any significant use of adverbs in the data. The adverb 'no more' was the only adverb found expressing deontic modality in Pente-Charismatic sermons. It was realized that 'no more' was used by preachers to forbid any force oppressing the listeners. The following exemplify the use of adverbs in sermonic discourse.
Another characteristic feature realized about the usage of the adverb in sermons is that it exhibits the preacher's commitment to ensuring the liberty of the listeners. This is revealed through the finality expressed in the use of this adverb. However, since this linguistic resource does not re-occur in the sermons, it cannot be considered as a valuable constituent of the register of Pente-Charismatic sermons according to the views of Firth (1935).

Clauses
Following the analysis of Pente-Charismatic sermons, it was realized that there is very minimal usage of clauses in expressing deontic modality. It became evident that sometimes, a full clause may be found expressing deontic modality in sermons. With this, no particular lexical item was found expressing deontic modality but the intention is drawn from the whole clause in expressing such. The following examples provide evidence for this finding. In examples 34 and 35, it is evident in Example 34 that the intention of the preacher is to assure the listener through a declaration about a pleasant thing that he would be experiencing something in the future. Example 35 also expresses a declaration of the possibility of a happening in the life of the listener. By applying the register studies to sermons, it was possible to find answers to the research questions posed in 1.06 above. The analysis of Pente-Charismatic sermons revealed that the register of sermons regarding the use of deontic patterns depends largely on the particular function they are meant to perform.

Conclusion
The study is guided by two fundamental objectives. The first objective has been to identify the linguistic resources that are used to express deontic modality in Pente-Charismatic sermons while the second objective examines how forms of deontic modality patterns relate to the discourse functions of Pente-Charismatic sermons. The study identified linguistic resources such as modals, auxiliary verbs, lexical verbs, phrasal verbs and clauses as elements of deontic modality in the data. The analysis of modals reveals that in Pente-Charismatic sermons, preachers exert a certain level of authority in giving hope to listeners through declarations in the form of promises, admonishing them on what they should do and the results of their actions and inactions. The use of lexical verbs signals that preachers have the power from God to grant the desires of their congregants. The other elements of deontic modality such as phrasal verbs, adverbs, and clauses are used to enable the preachers to convey their messages in plain language. All the findings above concerning linguistic resources that express deontic modality are not new to earlier findings. They are confirmations only to the views of certain renowned linguists such as Palmer (2001) and Finnegan (2004) that apart from modal verbs such as order, assume, allow and permit; auxiliary verbs such as may, should, must, ought and will; and adverbials such as possibly, certainly, virtually; other constructions also perform the same functions. Hence, the finding that phrasal verbs and clauses perform deontic functions in Pente-Charismatic sermons may not be entirely new but rather confirm the 'other constructions' pointed out by Palmer (2001) and Finnegan (2004). Future studies could analyse other linguistic elements in sermons under the remits of Critical Discourse Analysis to examine how power relations come to play in Pente-Charismatic sermons as well as examine the semantic implications of sermonic discourse and sermons of other religions other than the Christian religion from various linguistic perspectives. It is by realising these two objectives that this paper makes a contribution to knowledge on Pente-Charismatic sermons in general and how Ghanaian preachers use deontic modality in preaching their sermons. These findings will conscientise preachers of various religious messages in their preparation of sermons in order to present exactly what they intend to convey to their listeners.
Grace Asante-Anyimadu is an English Language Tutor at the Department of Languages, Methodist College of Education, Akim Oda, Ghana. Her research interests include Systemic Functional Grammar and Stylistics Analysis.