‘Future Talk’ in Newspaper Editorials: Predictions and Their Role in Argumentative Discourse

Newspaper editorials are known for taking a stance while fulfilling their goals of persuading the audience. In this regard, making future predictions is a crucial strategy in the argument structure of editorials. They are considered as risky acts since they are meant to outline future course of action as well as outcomes of such actions for their audience. This research is focused on the analysis of the speech acts of predictions among newspaper editorials of Pakistani, American and Malaysian newspapers. The analysis is focused on the exploration of forms, force and occurrence of these acts. The results indicate the preference of Pakistani and American newspapers in using strong predictions. The Malaysian newspaper meanwhile is found to be less explicit when predicting the future. This is indicated by less use of the strategy as well as adoption of implicit ways to express propositions related to the future. The results affirm the role of editorials as opinion leaders in their respective societies and the differences across cultures can be interpreted with respect to the extra linguistic and contextual factors that control editorial structures and strategies. The findings of the study are useful for future researchers to explore the relationship of language and its communicative purpose especially when fulfilling the goals of persuasion across cultures and contexts.


Introduction
The key feature of opinion discourses such as newspaper editorials is the ingenious use of opinions along with facts to carry out the goals of argumentation (van Dijk, 1998). Editorialists in this regard make use of a number of propositions to express their stance including suggestions, judgements and predictions (Huber & Snider, 2006;van Eemeren, 2002). In fact, the most crucial move in the structure of editorials is identified as taking a stance carried out through the strategies dominantly focused not only towards outlining the future course of action but also the outcomes of such measures for the target audience (Masroor, 2013;Masroor & Ahmed, 2017).
The aim of this research is to analyse this 'future talk' in newspaper editorial discourse that occurs in the form of predictions about the possible future outcomes of the actions taken or about to be taken in the present. This strategy, like the strategy of directing the readers (Masroor & Ahmed, 2017) is related to the future time frame and is a way of controlling the future actions of readers through making certain prophesies. Keeping in view the significance of stance taking in newspaper opinion genres and the role of language to fulfil these goals, this research is focused towards the analysis of linguistic choices as well as force and occurrence of these predictive statements in the argumentative structure of newspaper editorials across cultures. This research makes use of speech act theory to explain the concept of predictions and qualitatively analyses the strategy in the cross-cultural context. The study is significant as it carries out analysis of form, force and occurrence of predictions in a less explored area specifically in argument building in newspaper editorials. The rest of the paper, after delineating the relationship between argumentation and editorials, provides theoretical perspectives, methods and materials of the study, followed by analysis, discussion and conclusions. opinion discourse to present evaluations and assessments about future. Its usage in the journalistic genres, especially newspaper editorial, point to the fact that newspapers attempt to take hold of the future by most explicit ways such as talking about the future through future forecasts. By establishing that newspapers have greater knowledge than the readers reflected through the ability to foresee the future, newspapers attempt to exert their authority and supremacy over their readers. However, making future forecasts on events is a risky move. According to Smirnova (2009) making predictions is a risky job in journalism, and British newspapers prefer using reported speech to predict about future to avoid taking responsibility for such judgments. While analyzing business newspaper editorials, Katajamäki and Koskela (2006) observed that making predictions about the future on economic situations is a difficult task for editorialists and therefore newspapers prefer giving moral lessons rather than the much riskier act of giving predictions.

Editorials and Argumentation
Editorial language for the fulfilment of persuasive goals (Belmonte, 2008) has attracted the attention of many scholars. In this domain, some studies have been carried out in cross cultural settings. A few studies have focused on exploration of Asian cultures. Masroor (2013) and Masroor and Ahmad (2017) looked at editorial structure among Pakistani, American and Malaysian contexts. Persian and American editorials were analysed by Bonyadi (2010), while Malaysian and American editorials were analysed by Zarza and Tan (2016). The focus of these researches was on the organisational structure of the editorial genre. Some of the researches were focused on the exploration of linguistic features for the fulfilment of persuasive goals. In this regard, van Dijk explored the relationship between arguments and ideologies (1989). Alhudhaif carried out speech act analysis of Arabic and English editorials (2005) while Le focused on speech act and coherence analysis of French editorials (2009). The existing literature on the subject is focused less on linguistic perspective and more on the journalistic perspective (Ansari & Babaii, 2005), thus creating impetus for further researches on the role of this media for the fulfilment of argumentative goals through the speech act of predictions.

Theoretical Framework
The concept of language as an instrument to fulfil communicative intentions is found in speech act theory developed by Austin (1962) and Searle (1969), which propose that words perform certain actions. The intentions are expressed through the illocutionary point of an expression indicated by mostly performative verbs such as to inform, to apologize, to promise and so on. The illocutionary point can be determined with or without a performative verb and is dependent on other indicators as well including the position of verb and some non-verbal ques like gestures and intonation (Roth & Lawless, 2002). The five main types of speech acts based on the illocutionary intention include assertive, commissive, declarative, directive, and expressive acts (Austin, 1962;Searle, 1969).
The speech acts of assertive (also called representatives) comprise of the acts of reporting and predicting as indicated by performative verbs 'to inform' and 'to predict'. The difference between reporting and predicting pertains to something that has happened and that is about to happen (Searle, 1979). The predictive speech act is categorised as assertive due to its relation to the futuristic aspect. The English language has no explicit verbal category for expressing future and is carried out through modal verbs such as 'will' and 'shall'. The use of these modals is not reserved only for expressing the future but also modality (Huddleston, 1988). These speech acts may have varying degrees of force, depending upon how absolute or indefinite something is represented (Mey, 2005).
The process of argumentation involves the use of a number of propositions to convince adversary in favour of a particular standpoint (van Emeren & Grootendorst, 2004). It can be characterised as comprising of various speech acts. The argumentative structures have to be viewed in relation to their specific contexts. In this regard, a close attention can be paid to the participants, co-text and relationship between form and function as these propositions (and speech acts) can have varying degrees of force to convince within a particular context (van Emeren & Grootendorst, 2004).

Methods and Materials
For the purpose of analysis, 90 editorials (30 each) were selected from Pakistani, American and Malaysian newspapers. Since the nature of study is qualitative in cross cultural settings, and the purpose of analysis is based on exploration of forms as well as force and occurrence, the data size is kept to a minimum of 30 editorials for each cultural context. The selected cultures include one culture as a native while the rest are Asian cultures and can reveal interesting findings with respect to the functional use of the strategy in native and non-native contexts. The selected newspapers (Dawn, The New York Times, and News Strait Times) were influential and widely read newspapers of their countries. The collection period included 3 months (from January to March 2009) through systematic random sampling technique. This sampling technique involved the least chance of researchers' bias in the selection procedures. The emergent patterns resultantly can be generalized within the defined context. The time period of data selection was based on the ease of availability of the given corpus.
The analysis was kept to the level of sentence. A single assertion was treated as an instance of predictive statement, but compound sentences with more than one assertion were treated as separate predictive instances.

Future Predictions for Argument Building
This section is focused towards the analysis of the strategy of future predictions for stance taking in argument structure of editorials. Firstly, the total number of propositions used for making predictions is calculated (see Table 1), followed by the description of linguistic forms for their realization and their force as reflected through the forms and contextual clues.

Frequency of Predictions
All propositions expressing the strategy of making predictions were collected among the selected editorials. Table 1 shows the total instances of predictions in the whole corpus. The table above reflects the comparatively frequent use of the strategy by Dawn (35 instances) closely followed by NYT (32 instances). The least usage of the strategy is observed in NST (17 instances). The higher frequency in Dawn and NYT indicates newspapers' efforts to control future actions of its readers by making more predictions about future. NST's least frequent usage indicates its avoidance of controlling future actions of its audience.

Lexical Forms for Making Future Predictions
The ways of making forecasts differed remarkably in the corpus depending upon the newspaper's level of conviction of the forecast. This is mainly expressed through the use of modal verbs which are among the most common ways of expressing stance (Biber et al., 2002). It has been observed that in most of the cases, predictive statements are lexically realized through modal auxiliary verbs that can refer to the future time such as will, would, can, could, may and might. It is to be noted that modal verbs when used for making predictions are used in the sense of logical (extrinsic) modal meaning referring to the "levels of certainty, likelihood, or logical necessity" (Biber et al., 2002). It would be useful to take a look at the usage of the forms in the data to express predictions.

Will/Would
The modal forms "will" and "would" are used frequently in the corpus to express future prediction. According to Biber et al. (2002), modal verbs can have either personal or logical meanings, and in that sense the logical meanings of "will" and "would" mostly express the likelihood or probability of a prediction about certain future events. The most frequent form used to make predictions in the corpus is "will" used extensively by all newspapers.

Example 1
Vigorous trade will help the world recover. (NYT24) As the Example 1 above illustrates, newspaper's assessment about future through the use of 'will' and 'would' in the corpus usually indicates the definiteness of future prediction and newspaper's conviction about the proposed judgment about future occurrence of events. However, according to Biber et al. (2002) the prediction expressed by the use of 'would' has most of the times hypothetical meanings as the Example 2 below from NST indicates.
Example 2 It would be damaging not only to MCA but also to its partners in Barisan Nasional for its top leaders to continue to bash each other. (NST18) It has been found in the data that sometimes the usage of 'will' and 'would' is accompanied by the conditional 'if' which means that the future occurrence of an event depends upon the fulfilment of a certain specified condition If Section 5 is struck down, states and localities would have far more freedom to erect barriers for minority voters …. (NYT9) The above examples from Dawn and NYT show the use of conditional clause when making predictions about future.

Can/Could
The modals of 'can' and 'could' in predictive sense in the corpus express logical possibility. The use of modal 'can' in the predictive sense can be distinguished sometimes easily through the context. Example 5 below expresses the future possibility through an explicit reference to the future time frame by the use of time adverbial 'tomorrow'.
Example 5 Gen Kayani has been applauded for his role in the restoration of Chief Justice Chaudhry, but a good choice yesterday can become a bad one tomorrow. (D27) The presence of possibility modal 'can' shows tentativeness but the negative of possibility expressed through 'cannot' reflects impossibility and therefore is considered as more decisive than when used in the positive sense through 'can'. Example 6 below illustrates the use of 'cannot': Example 6 On the one hand it claims that Pakistan "uses terrorism as an instrument of state policy." At the same time India wants the state of Pakistan to take action against those who masterminded the Mumbai attacks. This makes no sense and cannot serve as the basis for a working relationship. (D3) In the example above, the use of 'cannot' makes the prediction much decisive in comparison to the use of 'can' in the previous example which was tentative. The use of 'could' usually adds unreality to the meaning. According to Biber et al. (2002), the use of modals conveying past time like 'could' is usually associated with hypothetical meanings and conveys greater degree of doubt. The use of 'could' is provided in Example 7 below: Example 7 We don't know that his industry ties would influence his judgments on health issues, but they could potentially throw a cloud over health care reform. (NYT12) The hypothetical situation as expressed through the modal 'could' above refers to the future time frame as evident through the use of adverbial 'potentially'.

May/Might
The modals of 'may' and 'might' in the predictive sense are used to express logical possibility. The following example from NST provides the usage of the modal 'may' to express tentative possibility.

Example 8
It is gratifying that the police have not stood in the way of a second autopsy on Kugan's remains, though there may be dire consequences on the medical officer who signed the first report, should the second refute the innocuous initial conclusion of death due to "fluid in the lungs". (NST8) In the corpus, 'may' is also found in the conditional sense through the use of 'if' clause.
Example 9 There are disturbing reports that the rocket launchers and explosives used in the attack in Lahore on the Sri Lankan cricket team were standard issue weapons used by Indian forces. If true, these findings suggest that genuine regional cooperation in the fight against militancy and terrorism may still be a distant dream. (D28) The past form of 'may' i.e., 'might' like 'could' is used to convey greater degree of uncertainty and tentativeness (Biber et al., 2002). The following Example 10 from NYT shows the use of 'might' for making predictions.
Example 10 ijel.ccsenet.org International Journal of English Linguistics Vol. 10, No. 3;2020 A real investigation might serve as a channel for the public anger now used by politicians to score quick populist points on television without tackling the real issues. (NYT30)

Primary Verbs (be and have)
Besides the use of modals, the data from the study also makes use of primary verbs such as 'be' and 'have to' make predictions about future. According to Biber et al. (2002), a reference to future can be achieved through the present tense accompanied by either a time adverbial or an adverbial clause referring to the future time frame.
The following Example 11 illustrates this use.

Example 11
He believes he can but whatever the ultimate truth, this much is certain: Afghanistan is set for a long, difficult summer, politically and militarily. (D15) The future prediction in the example above is made in the present tense through be verb 'is', while 'is set for' refers to the future time frame. Similarly, Example 12 below makes use of 'be' verb with reference to future time through the use of 'ahead'.

Example 12
The challenge ahead is daunting in the extreme. (D28) Predictions with present tense are also carried out by using conditional clauses. Example 13 below shows the conditional prediction through the primary verb 'are'. The use of conditional 'unless' shows events that might occur in future upon the fulfilment of the specified conditions.

Example 13
It sold one lease, now under court challenge, covering 2.3 million acres in 2008. Unless the Obama administration changes course, other leases within these areas are sure to be offered. (NYT28) Besides the use of 'be', it has been found that primary verb 'have' is also found in the corpus for making predictions about future. In the Example 14 below, 'has dimmed' refers to the present state of events as a result of some actions already taken in the present which have a specified impact on the future.

Example 14
By invading Gaza, the Jewish state has dimmed what little hope there was for a settlement of the Palestinian-Israeli conflict. (NST2) The forms used for making predictions can have varying degree of strength which is explored in the next section.

Strength of Predictions
The strength of prediction was determined based on newspaper editorial's level of conviction and certainty while predicting. The predictions can be made tentatively or with a full conviction depending upon newspapers' confidence on the prediction being made. Assessments and warnings about future differ in their force depending upon the type of form used to express them. Following Alhudhaif (2005), the predictions in the data were categorized into either strong predictions or weak predictions.
The strength of a prediction is mainly expressed through lexical choices made by newspaper editorial to predict future. Mostly, the predictive statements in the data are lexically realized through modals. Prediction and possibility modals together are used to predict the future as they inform about the likelihood of an event's occurrence. The predictive modals such as will and would in the study were mostly categorized as having stronger force than the weaker force of the possibility modals can, could, may and might.
The additional force is determined by the co-text and contextual meanings which may help strengthen or weaken the force of a prediction. The use of lexical devices such as boosters may increase the conviction of a prediction. In the Example 15 below from Dawn, the use of 'naturally' increases the conviction level of the modal 'will' and therefore the proposition is considered as having a strong predictive force.

Example 15
As the new year dawns, the primary talking point will naturally be the ongoing tension in relations with India. (D1) Just as boosters could help strengthen the force of a prediction, the use of vague expressions such as likely, unlikely, possible, slight chance, probably, seem, and so on, help mitigate the force of predictions. According to Toulmin (1964), when the expressions such as 'probably' are used in making predictions, it makes it a guarded ijel.ccsenet.org International Journal of English Linguistics Vol. 10, No. 3;2020 prediction which shows the writers' tentativeness in fully committing themselves about the occurrence of a predicted event. In the excerpt below, Toulmin (1964, p. 49) explains the difference between a commitment and reservation.
When I say "I know", I give others my word: I give others my authority for saying that "S is P".' So also, saying 'S is probably P' is not saying 'I am fairly confident, but less than certain, that S is P', for 'probably' does not belong in this series of words either. When I say 'S is probably P', I commit myself guardedly, tentatively or with reservations to the view that S is P, and (likewise guardedly) lend my authority to that view. (Toulmin, 1964, p. 49).
The Example 16 below, regardless of the use of modal 'will' was categorized as having weak strength due to the use of probability marker 'seem':

Example 16
Unfortunately, chances that a bank would return to health after years on government life support do not seem especially good. (NYT21) Similarly, in the following Example 17 from NYT, the presence of probabilistic expression 'perhaps' shows uncertainty:

Example 17
Perhaps the toughest part of that fight will be pushing for Medicaid reforms that would curb reimbursements for most hospitals, nursing homes and home care and encourage more cost-effective care.

(NYT18)
Besides the lexical choices, the contextual meanings were also considered as influential in determining the force of prediction. The following Example 18 from Dawn is considered as having a weak predictive force despite of the use of strong modal 'will'. This is due to the indefinite prediction made on the part of newspaper regarding the future course of action of President Obama's administration.

Example 18
Choices exist; the next few months will make it clear if the Obama administration prefers the stick or the carrot when it comes to Pakistan. (D4) The primary verbs (such as be and have) when used for making a prediction were categorized as having strong force mostly depending on the presence of certainty markers such as 'this much is certain' in Example 19 below from Dawn.

Example 19
He believes he can but whatever the ultimate truth, this much is certain: Afghanistan is set for a long, difficult summer, politically and militarily. (D15) Although the possibility modals such as 'can' and 'could' were categorized as having weaker predictive force, however, the negative form of 'can' was considered as having stronger predictive force as is clear from the example below from Dawn. The use of 'cannot', as in the Example 20 below, makes the possibility of 'can' as impossibility and thus increases the level of conviction.

Example 20
On the one hand it claims that Pakistan "uses terrorism as an instrument of state policy." At the same time India wants the state of Pakistan to take action against those who masterminded the Mumbai attacks. This makes no sense and cannot serve as the basis for a working relationship. (D3) Therefore, after describing the criteria for determining the force of predictive statements, all the instances were collected and categorized based on the type of linguistic forms used for predicting. Based on the total number of forms used by each newspaper reflecting the strength of predictions, the results were compiled. The following Table 2 shows the results about strength of predictions used across the corpus.  As the table above reflects, all newspapers show a preference over using stronger predictions than the weak ones. Similar to results of frequency analysis, Dawn surpasses others by again making greater use of strong predictions when compared with the other newspapers i.e., 63 percent, closely followed by NYT (62.5 percent) and the least frequent usage is observed by NST (i.e., 59 percent). The difference is more striking when we compare the number of instances used by each newspaper to make stronger predictions i.e., 22 instances by Dawn, 20 instances by NYT and 10 instances by NST. It has to be noted that NST follows a consistent pattern when taking a stance as its percentage of making strong predictions is the lowest and making weaker predictions is the highest among all newspapers.

Discussion
Predictions act as a measure of stance in opinion discourse such as newspaper editorials (Masroor, 2013;Masroor & Ahmad, 2017). Among the chosen newspaper contexts, Dawn is observed to make greater use of the strategy among the other newspapers with respect to the occurrence of the strategy. This is followed closely by NYT, while NST is observed to make the least usage of the strategy.
It has been observed that all newspapers show a preference over using stronger predictions than the weak ones. The strong predictions are a means of strengthening the stance and thus enhancing the persuasive impact in the argumentative structure of editorial genre. The trend therefore points out towards the function of editorial genre as an opinion discourse, i.e., as a place for the explicit articulation of newspaper's viewpoints. Being an opinion discourse, newspaper editorials have to take risks while taking a stance for or against an issue. It is to be noted that vagueness is not always meant to avoid confrontation. At times, it might be a prerequisite for making an honest judgment. However, it is a preferred act especially in journalistic writing where taking strong and clear positions is highly desirable (Simurda, 1997). It is clear through analysis that Dawn surpasses other newspapers by making greater use of strong predictions i.e., 63 percent, closely followed by NYT (62.5 percent) and least frequent usage in NST (i.e., 59 percent). The difference is visible by looking at the number of occurrences per newspaper to make stronger predictions i.e., 22 instances by Dawn, 20 instances by NYT and 10 instances by NST. It has to be noted that NST follows a consistent pattern when taking a stance. Its percentage of making strong predictions is the lowest and making weaker predictions is the highest among all newspapers.
The difference in the frequencies is an indicator of different contextual circumstances prevalent in the countries of each newspaper. Dawn's greater usage of the strategy could be explained in the light of the critical socio-political circumstances in Pakistan during the times of data collection and an attempt by the newspaper to take hold of the future by making its audience aware of the outcomes of the actions taken in the present to avoid repeating follies of the past. NYT's strategy of taking greater risk through predictions is also a result of increasing threat of economic instability and a more authoritative role that the newspaper takes upon itself in the society. Since making predictions is a far riskier act (Smirnova, 2009;Katajamäki & Koskela, 2006), NST's avoidance of confrontation through the less usage of the strategy shows newspaper's lack of relative authority and control over the future actions of the readers and therefore exhibits a less persuasive stance. The tentativeness could be the resultant of a desire to project deference towards its audience. The multi-ethnic fabric of the country's society and the newspaper's target audience, which includes more elites than the common public, and its relationship with audience, could be the critical factors in developing this stance.

Conclusion
This study has revealed the relationship of language with the communicative purpose of an argumentative genre.
Since the goal of an argumentative text is to make use of a number of propositions and strategies to convince and change the minds of the addressee (van Eemeren et al., 2002), editorial genre makes use of prediction statements to carry out its goals of persuasion. The results have highlighted the role of future assessments and evaluations in the expression of opinions in editorial structure. In the light of the posed research objectives, the research has highlighted the use of various forms for making future predictions in the selected corpus. The forms uncovered are helpful for marking and categorising this particular speech act of future predictions. The forms are explored in the context of their usage by exploring the role of context and co-text in determining the force of prediction.
The findings of such studies which highlight the role of linguistic features for argument building are equally helpful for the students of language (Masroor, 2016) and media studies.
Its usage in the journalistic genres, especially newspaper editorials points to the fact that newspapers attempt to take hold of the future by most explicit ways such as talking about the future through future forecasts. By establishing that newspapers have greater knowledge than the readers reflected through the ability to foresee the future, newspapers attempt to exert their authority and supremacy over their readers.
The strategies adopted by newspapers to fulfil their goals can be varying in differing contexts. The use of more direct and clear language while making predictions can help enhance the persuasive effect of a text, whereas a less confrontational language fails to address the doubts existing in the minds of the readers. The direct interaction with the readers can be viewed in terms of considering argumentation process as a form of dialogue. Thus, the use of more involvement strategies in the form of a direct address can lead to an increase in the persuasive impact of a genre.
The process of argumentation can be considered as a dialogue where writers can choose the ways of addressing the audience and establishing their relationship with them (Hyland, 2002). The stance of editorialists is depended upon various contextual factors and institutional constraints. In this regard, the role of a newspaper in a particular society and more importantly, socio-political circumstances of the country cannot be ignored when analysing argument structures (Ahmed & Masroor, 2018;Masroor, 2013;Masroor & Ahmad, 2017). The critical times in Pakistan are reflected in Dawn's need to voice controversies more explicitly and therefore resulting in the adoption of a comparatively more aggressive and engaging stance. Dawn's predictions in comparison to the other two newspapers reflect bolder, alarming and authoritative tone. NYT's stance is also authoritative but far from aggressive. Its role as an international newspaper requires the newspaper to reflect authority in a less confrontational and more polite manner. NST's stance also reflects the role of circumstantial factors when carrying out argumentation. The existence of multiple races in the country requires a careful stance when making future forecast which are considered as risky acts (Smirnova, 2009;Katajamäki & Koskela, 2006). NST's avoidance of conflict while expressing its authority is reflected through its implicit use of the strategy while carrying out the goal of persuading the audience.
This study has revealed valuable findings with respect to the types of linguistic choices preferred by newspapers across cultures when arguing through predictions and the role of extra-linguistic factors that can influence the lexical choices. The study also highlights the role of co-texts and devices for mitigation and imposition in a proposition. The findings of the study are useful for research across cultures and contexts. It encourages future researches in this regard to explore further the argumentation strategies and structures in opinion discourses like editorials.