Toward a genre-based characterization of the problem-solution textual pattern in English newspaper editorials and op-eds* ISABEL ALONSO BELMONTE Abstract The current study presents the results of a discourse analysis research on the problem solution pattern of textual organization (henceforth, the PS pattern) in a sample of English written newspaper editorials and op-eds. My main goal is to characterize the linguistic realization of the PS pattern in written opinion journalism and to describe its textual variations. The PS pattern is analyzed here as part of a wider framework, Tirkkonen-Condit's (1985) method of textual description, specifically designed to outline the global structure of argumentative texts. The application of Tirkkonen-Condit's method involves the dissection of texts into PS sequences and communicative acts. Both the PS sequences and the individualized communicative acts are assigned an illocutionary value and are analyzed in terms of the rhetorical relations holding between them. Findings reveal that the texts analyzed are complex networks of functionally arranged and interrelated PS sequences whose unmarked order can be purposely subverted with emphatic aims. In addition, an illocutionary and rhetorical analysis of each one of the PS textual components reveals different ways of addressing the reader in editorials and op-eds. Keywords: newspaper editorials and op-eds; problem solution pattern; textual organization; rhetorical roles; illocutionary value; communicative effectiveness. 1. Introduction The present study analyzes the linguistic realization of the problem-solution (PS) textual pattern and its textual variations in a sample of English written newspaper editorials and their corresponding op-eds.1 The PS pattern is a culturally bound rhetorical sequence articulated around certain fixed components which has been described as one of the 1860-7330/09/0029-0393 Online 1860-7349 © Walter de Gruyter Text & Talk 29-4 (2009), pp. 393-414 DOI 10.1515/TEXT.2009.021 394 Isabel Alonso Belmonte possible linguistic realizations of reader-writer interaction (Winter 1982, 1994; Hoey 1979, 1983, 2001; Thompson and Thetela 1995). It is pervasive in virtually all text types in English, although some scholars establish a strong link between the PS sequence and argumentation, which is conceived as a problem-solving process (Tirkkonen-Condit 1985; Teo 1995; Koester 2004; inter alia). Although the theory of problem-solution patterning is already well established for a wide range of genres, very few studies deal specifically with the discoursal characterization of the PS pattern in newspaper opinion texts. To fill this void, empirical evidence has been gathered from thirty editorials and thirty op-eds taken from the newspaper USA TOD A Y. This top-selling national newspaper offers a daily debate on issues of general interest with the publication of an editorial and an op-ed, that is, an article that expresses the opinion of a named writer, usually unaffiliated with the newspaper's editorial board. The comparative study of these text types expressing different views on the same subject enables the analyst to assess whether the linguistic realization of the PS textual pattern is determined by the variable of genre (Swales 1990) or by precise decisions taken by each writer in each communicative situation. To characterize the complex rhetorical organization of the sample texts, the problem-solution pattern is analyzed here as part of a wider framework, Tirkkonen-Condit's (1985) method of textual description, specially designed to portray the global structure of argumentative texts. The interest of Tirkonnen-Condit's method has been unanimously acknowledged by the scientific community in general. It has been successfully tested in discourse analysis for describing the textual structure of newspaper editorials (Teo 1995) and also used in contrastive rhetoric for the analysis of argumentative compositions by L2 students (Connor 1987; Connor and Lauer 1988). Findings obtained from the application of Tirkkonen-Condit's method to a wider sample of natural argumentative texts will be useful in identifying different top-level rhetorical arrangements made by professional writers in English newspaper editorials and op-eds. 2. The problem-solution pattern The problem-solution model of textual organization has been described as a culturally bound sequence articulated around certain fixed components that can combine in different ways (Jordan 1980, 1984, 2007; Crom-bie 1985; Hoey 1979, 1983, 1986, 2001). Each PS component is a textual chunk—whose size may range from the word to the clause level or higher —which is examinable as the writer's answer to an imaginary reader's The PS pattern in newspaper opinion pieces 395 question and contributes, with a specific function, to the overall coherence of the writing. The PS pattern is organized around the following textual components: a Situation, which describes facts and objective circumstances; a Problem, which introduces a problematic aspect of the situation and calls for a solution; a Response or Solution, which either explains how the problem has been resolved or puts forward suggestions and recommendations, and finally, an Evaluation and/or a Result,2 which positively evaluates the solution proposed. A negative Evaluation following a Solution component entails a recycling pattern and a return to the Problem. To these components, Hoey adds another one, called Plan, which can sometimes appear in the PS sequence, acting as an intermediate stage between the Problem and the Solution or Response. As Hoey (2001: 127) explains, "(it) either defines what might count as an adequate Response or makes a suggestion as to what Response to adopt." The PS textual pattern has been characterized as the main organizing principle of many kinds of written and spoken texts. From a genre-based approach, the PS pattern has been studied in academic writing associated with Swales's moves structure (Paltridge 2001) or has been subsumed within the structure of a particular move (Paltridge 1996; Flowerdew 2000). The PS pattern has also been characterized in corpus-based approaches to the study of report writing (Flowerdew 2003, 2008). As regards opinion journalistic texts, some studies have individualized the PS as the organizing principle in different samples of English written newspaper editorials, alone (Teo 1995; Scott 2002) and in intertwining with other existing textual patterns (Alonso Belmonte 2009). From the review of the existing literature, it seems that many researchers are aware of the existing textual variations of the PS pattern and the complexity of its linguistic realization in natural texts (Al Sharief 1998; Flowerdew 2003, 2008; Navas 2005; inter alia). However, most studies reviewed limit their task to the mere analysis of the textual realization of the PS sequence and do not venture to explain, for example, whether the different linguistic realizations of the PS textual model of organization in the texts analyzed respond to textual strategies on the part of writers to achieve a more effective communication with their readers, or to genre constraints. This issue will be specifically addressed in this paper. 3. Tirkkonen-Condit's method to describe the global structure of argumentative texts In 1985, Tirkkonen-Condit developed a method for the analysis of argumentative texts on the basis of the combination of different modes of 396 Isabel Alonso Belmonte analysis made mutually compatible: the already mentioned problem-solution analysis, the illocutionary analysis (the identification of illocu-tionary acts), and the interactional analysis (the identification of rhetorical relations).3 In order to solve some of the operational problems that the application of the Tirkkonen-Condit's method entailed when applied to the sample texts, it was updated with the more recent insights and empirical developments provided by some other scholars in the genre of written journalism (Teo 1995; Almeida 1992). A summary of its main characteristics is presented here. 3.1. The PS rhetorical pattern Tirkkonen-Condit proposes that any argumentative text can be divided into different text chunks that correspond to the different PS textual components: Situation, Problem, Solution, and Evaluation. Each textual component is made up of an indefinite number of communicative acts, which rhetorically develop the component meaning within the pattern (already explained in Section 2 for each one of the textual components) and whose grammatical correspondence is mainly the sentence and, occasionally, the clause.4 Readers usually make use of the question technique and observe the visible lexical signals present in the text to be able to trigger the recognition of any textual component as part of the PS pattern. Consider the following example: (1) {USA TODAY30/04/98 ED) creatine is a red-hot muscle candy. 2Thousands of athletes at all levels swear by it as a source of strength and endurance. 3Super Bowl quarterback John Elway of the Denver Broncos endorses it. 4So does the Baltimore Orioles fitness poster boy, Brady Anderson. 5But creatine can also hurt you, perhaps badly. 6It has caused diarrhea, dehydration, and muscle cramping and tearing. (...) 29What's needed is a step in the other direction, 30 one that protects against risky products without burdening those known to be safe. (...) 37In the absence of tougher consumer protection, take the counsel the FDA urges on users of creatine and every other diet supplement: 38 Talk to your doctor about potential side effects and interactions with other medicines. 39 After all, the idea of diet supplements is better health, not greater risk to health. The three first communicative acts belong to the Situation component, since they expose objective facts about the creatine substance. However, communicative act no. 5 signals the beginning of the Problem textual component, which explains the consequences of ingesting creatine on a regular basis. This textual component is triggered by the presence of cer- The PS pattern in newspaper opinion pieces 397 tain lexical signals—e.g., but, hurt, badly—and responds to the imaginary reader's question: what is the problematic aspect of the above-mentioned situation that needs to be solved? The writer continues exposing the different problems related to the ingestion of diet supplements in the text, until communicative act no. 29, in which the writer suggests an FDA new regulation of diet supplements and advises his/her audience to talk to their doctor before ingesting any substance on a regular basis. Directives and verbal phrases such as what is needed ... signal the beginning of this Solution component. Finally, the writer evaluates his/her proposal with communicative act no. 39; lexical evaluative signals such as better or greater indicate so. Not all textual components are obligatory. Tirkkonen-Condit identifies the Problem component as the only nucleus of the PS sequence and claims that the presence of other textual components is optional. Other more recent studies, however, consider the Evaluation component as another nucleus of the sequence (Teo 1995). Indeed, Hoey (2001) acknowledges that the Situation component only belongs to the PS pattern retrospectively. Thus, for the present study, I assume that the Problem and the Evaluation textual components are the nuclei of the textual pattern and that therefore their presence is obligatory to identify the pattern; whereas the presence of the remaining ones is communicatively determined. In other words, it is a writer's decision to make them explicit or not. 3.2. Interactional analysis Each textual component and each communicative act is analyzed in terms of their superordinate or subordinate nature to each other, and if subordinated, in terms of the rhetorical nature of the connection with the superordinate textual pattern or the subsequent communicative act. To be able to characterize the possible hypotactic and paratactic rhetorical relations among communicative acts and text components in this study, a list of subordinated and coordinated rhetorical roles was elaborated on the basis of previous research on the subject (cf. Grimes's [1975] predicates, Tirkkonen-Condit's [1985] and Teo's [1995] interactional roles, Hoey's [2001] PS components). As Table 1 shows, some rhetorical roles match their names and functions with their equivalents in the PS pattern— situation, (positive, negative and neutral), evaluation, solution— while others further work on the information presented in the previous roles: justification, explanation, conclusion, enlargement, elaboration, exemplification, metastatement, reformulation, and addition. All respond to readers' imaginary questions to writers and can be characterized as shown in Table 1. 398 Isabel Alonso Belmonte Table 1. Some common rhetorical roles in written argumentation Interactional roles Hierarchical relationships to preceding acts Reader's questions they elicit Lexical signals and illustrating sentences Situation Negative evaluation Positive evaluation Neutral or analitical evaluation (Teo 1995) Solution Justification Explication Conclusion Elaboration Exemplification Enlargement Coordinated Coordinated Coordinated Coordinated Coordinated Subordinated Subordinated Superordinate Subordinated Subordinated Superodinate What is the situation? What is your response to the situation you have just described? What is your response to the situation you have just described? What is your response to the situation you have just described? What should be done about the problem? On what grounds are you asserting this? What is the cause of this? What conclusion can you draw from this? Can you give some details of this? Can you give an example of this? What generalization can be made on the basis of what you have just said? But, However, Yet, Nevertheless, Bad, i.e., "The CIA has had enough 'scrubbings'." Important, Good, i.e., "But more than shareholders will benefit" i.e., '' Internationalization is essential to success"; "(...) The episode has reignited a simmering debate"; "Bush needs time." Therefore, So, Must Because Because Thus, In conclusion, To sum up, Therefore, As a result, In a word, All in all, To summarize, In summary In particular For instance, For example Thus, All in all The PS pattern in newspaper opinion pieces 399 Table 1 {Continued) Interactional roles Hierarchical relationships to preceding acts Reader's questions they elicit Lexical signals and illustrating sentences Addition Coordinate Reformulation Metastatement Subordinated Subordinated Do you have any other examples (elaborations, conclusions, justifications, etc.) that you would like to add? Can you express this in other words? (A question explaining the relationship between preceding and subsequent acts) In addition, Moreover, Also, Likewise In other words Indeed, In short "This is exemplified by the following"; "I will give an example"; "Let us take an example"; "To understand why (■■■)"; "I will discuss ... "; "... has expressed the following idea" 3.3. Illocutionary analysis Regarding the illocutionary analysis, Tirkkonen-Condit designed a typology of illocutionary acts based on the contributions of speech act theory (Searle 1979) and, more specifically, on work by Aston (1977) on representative speech acts and by Edmonson (1981) on directives. Within the class of representatives, Tirkkonen-Condit distinguishes the following categories: statements, assertions, reported assertions, and shared-knowledge assertions. The controversial dichotomy between facts or objectively true statements and opinions or evaluative expressions underlies Tirkkonen-Condit's and Teo's discrimination between statements and assertions. However, since this criterion was not proved to be operational either in Tirkkonen-Condit's method or in Teo's study on the textual structure of newspaper editorials, it was decided here to adopt Almeida's (1992) distinction between factual and nonfactual statements, which is based on the definition of factuality in news discourse. Thus, a fact is considered here to be understood by the reader as describing an actual situation— an event or state, present or past—based on the writer's knowledge of the world and for which he/she accepts all responsibility: i.e., "On Satur- 400 Isabel Alonso Belmonte day, for instance, the Mercosur trading block—Brazil, Argentina, Uruguay and Paraguay—announced plans to strike a trade deal with Central American countries, which plan to form a free trade bloc." On the other hand, assertions are taken here as nonverifiable evaluative propositions, speculations, predictions, etc.: i.e., "Enacting unnecessary barriers—such as requiring a written authorization every time information is used— would harm patients." Regarding reported assertions, they are defined as the "assertions of others," that is, embedded nonfactual propositions which can include a specific reference to the source or not: i.e., "(...) President Bush said the public's decision to re-elect him was a ratification of his approach toward Iraq and that there was no reason to hold any administration officials accountable for mistakes or misjudgments in prewar planning or managing the violent aftermath." Finally, shared-knowledge assertions are considered to be evaluative propositions that appeal to the shared knowledge between writers and readers: i.e., "(...) Blood shortages are an everyday concern for every American." Tirkkonen-Con-dit also identifies requests, suggestions, proposals, recommendations, and warnings as subtypes of directive speech acts, although they occur infrequently in her study. 3.4. Summary In sum, the application of Tirkkonen-Condit's method involves the dissection of texts into PS sequences and communicative acts. Both the PS sequences and the individualized communicative acts are analyzed in terms of their cotext (by unraveling rhetorical relations holding between these units) and in their context (by assigning them an illocutionary value). On the one hand, the analysis of the problem-solution pattern and the most common rhetorical roles present in each one of the PS components reflects the ways in which writers take the imagined reader's expectations, knowledge, and interests into account when constructing their texts. On the other hand, the analysis of the dominant illocutionary acts present in the texts analyzed shows evidence of how writers conduct interaction with their readers, particularly by assuming for themselves and assigning to the reader roles in the interaction and by intruding in the message to comment on and evaluate it. 4. Sample and methodology of analysis The sample consists of thirty newspaper editorials (in total, 14,554 words) and their corresponding op-eds (9,945 words) taken from the online ver- The PS pattern in newspaper opinion pieces 401 sion of USA TODAY (www.usatoday.com). Editorials and op-eds are public, mass-communicated types of discourse that play a definitive role in the formation and altering of public opinion, promote social interaction among journalists, readers, and the rest of participants in the language event, and influence decision making and other forms of social and political action. Both newspaper editorials and op-eds have similar functions: to comment on recent events and attempt to persuade the reader of the newspaper's point of view, or, of the commentator's point in the case of op-eds. All texts analyzed cover different economical, social, and political topics and were selected with the help of a random number table from a bigger sample of 472 texts (236 editorials and 236 op-eds) in total, gathered weekly during 12 months. Each sample text was saved as a Microsoft Word document and classified by date and group: the code ED refers to newspaper editorials and the code OP.ED to the corresponding opinion articles: i.e., USA TODAY 19/04/98 OP.ED. Due to the intricate nature of the method applied, texts were analyzed manually by identifying: (a) the presence of PS sequences, and their position within the text (i.e., initial, middle, final position); (b) the most common rhetorical roles in each one of the PS components; (c) the most common illocu-tionary acts in each PS component. As a result of this type of analysis, 201 PS sequences (103 in newspaper editorials and 98 in op-eds), 1402 rhetorical roles, and 1440 communicative acts were identified and characterized. Within the scope of this paper and to facilitate reading, the different communicative acts presented here will be signalled with small correlative numbers (123' etc.), the name of the different textual components will be presented in regular typeface, and the rhetorical roles adopted by the subordinated PS sequences, if any, in small capitals. The entire corpus was coded by the author. To verify the stability of the coding, thirty randomly chosen texts (50% of the total sample) were re-coded independently by a graduate student. After comparison of the results, 1.5% of the codes for these thirty sample texts were revised. Then the coding of the entire corpus was checked for any discrepancy between the original and the revised coding. The resulting coding of the corpus is considered stable enough for reliable results. Once the analysis was undertaken, results were systematized in two Access databases; one for the newspaper editorials; the other one for the op-eds. Each database allows the analyst to extract information about each one of the sentences or clauses analyzed in terms of: (a) the text it appears in and the date it was published; (b) the number it was assigned; (c) its illocutionary value; (d) its rhetorical value; and (e) the PS component in which it appears. We now turn to the results. 402 Isabel Alonso Belmonte 5. Results From the application of an updated version of Tirkkonen-Condit's method to describe the structure of argumentative texts to a sample of newspaper editorials and their corresponding op-eds, a genre-oriented characterization of the PS pattern emerges. For presentation purposes, findings are reported and discussed under the following subsections. 5.1. The rhetorical organization of newspaper editorials and op-eds The analysis undertaken shows that both newspaper editorials and op-eds can be characterized as complex networks of functionally arranged and interrelated PS sequences. More specifically, data shows that most PS sequences are chained in groups of approximately three coordinated sequences for each text.5 This tripartite structural pattern confirms previous research findings on the structure of editorials (Bolivar 1994; Riazi and Assar 2001; Ansary and Babaii 2005). Among the 201 PS sequences, the most frequent patterns are: Situation + Evaluation (44 cases), Situation+ Problem (27 cases), and Evaluation (24 cases). These are followed by Situation + Problem + Evaluation (18), Problem + Solution + Evaluation (14), Problem + Evaluation (12), Situation + Problem + Solution + Evaluation (10), and Solution + Evaluation (10). As can be seen, all sequences described revolve around the Problem and/or the Evaluation textual components, the nuclei of the PS textual pattern. The Solution component hardly appears either in the editorials (21.35%) or in the op-eds (27.55%). It seems that neither editorialists nor guest writers feel obliged to give suggestions as to how the problem should be solved; they just expose the problem, provide an evaluation of it, and send a message to the politicians, whose task it is to find solutions for readers' worries. Findings also reveal that the most common PS sequences—Situation + Evaluation; Situation + Problem, and Evaluation—exhibit a certain preference to be located in certain positions within the textual global structure of both newspaper editorials and their corresponding op-eds. For example, while the Situation + Evaluation and the Evaluation sequences tend to appear in initial, middle, and final positions, the Situation + Problem ones tend to be initially, and specially, middle placed. When an Evaluation sequence is used to conclude an article, analysis shows that it tends to be negative. Consider the following example, taken from an editorial about an anti-smoking bill that was not passed at the US Senate: The PS pattern in newspaper opinion pieces 403 (2) (USA TOD A Y 22/06/98 ED) Did tobacco money kill anti-smoking bill? (...) 14 With this kind of addiction, small wonder that party leaders killed modest campaign reform in the Senate and are trying to do so in the House. 15As long as campaigns are paid for by tobacco and other interests, not by the public, no one should be surprised that they win. (...) We may interpret that the placement of a negative evaluation at the end of an opinion text can generate a certain feeling of discomfort to the reader who sees that the problem exposed in the text is not easily solved. As for the position of the remaining sequences within the sample texts, Table 2 provides a summary. Table 2. The PS sequence positioning in newspaper editorials and op-eds PS sequences Situation + problem + solution + evaluation Situation + problem + plan Situation + problem + result Situation + problem + solution Situation + problem + evaluation Situation + solution + evaluation Situation + evaluation + problem Situation + evaluation + solution Problem Problem + situation Problem + solution Problem + evaluation Problem + situation + evaluation Problem + plan + evaluation Problem + evaluation + plan Problem + solution + evaluation Problem + solution + result + evaluation Situation + problem Situation + evaluation Solution + evaluation Solution + problem + solution Solution + problem + evaluation Evaluation Evaluation + situation + evaluation Evaluation + solution + evaluation Evaluation + problem + solution + evaluation Total Initial Middle Final Final position position position position 3 4 3 10 1 1 1 1 1 1 1 3 6 10 2 18 1 1 2 2 3 5 2 1 3 1 3 1 5 1 1 1 4 2 7 2 8 2 12 1 1 1 1 1 1 1 5 8 14 1 1 9 18 27 12 15 17 44 4 6 10 1 1 1 1 9 5 10 24 6 6 1 1 1 1 201 404 Isabel Alonso Belmonte As Table 2 shows, many sample texts are articulated around one or more coordinated PS sequences in which only the Evaluation textual component is present. This is quite common in op-eds, which take advantage of the informative framework created by the USA TOD A Y opinion page and the corresponding editorial and directly express their writers' opinion, without feeling the need of referring to the existing problem tacked in the editorial. This is the case of the following excerpt about the CIA, in which the authors firmly defend the agency: (3) {USA TODAY9/06/98 OP.ED) Agency has done much good; talent and effort will make it better. Rep. Porter J. Goss, R-Fla., who chairs the House Committee on Intelligence, and Rep. Norm Dicks, D-Wash., a ranking minority member. [1st PS sequence] (Evaluation) ^'Last one out, turn off the lights," is not a responsive policy to U.S. intelligence needs or global obligations in today's dangerous world. [2nd PS sequence] (Evaluation) 2Good sources, technical methods and special relationships take years to craft and develop. 3Decision-makers need the best information all the time. 4Meeting that challenge demands skills, training, strength to endure and the courage to take risks. 5Perhaps the essence of our intelligence capability is very good people. 6Should we just throw them out? 3rd PS sequence (Evaluation) 7Our existing intelligence capabilities are clearly the world's best—8just ask our allies. (...) As Table 2 also shows, the majority of the PS sequences analyzed present their components in a canonical, unmarked order—Situation —>■ Problem —>■ Solution —>■ Evaluation—since it is easier for the reader to understand, store, and retrieve information that has been presented in a linear fashion. Yet data show the existence of a minor but interesting tendency in which the writer occasionally deviates from the expected and purposely subverts the unmarked order of the PS components, producing an emphatic and hence persuasive effect. This proclivity is most manifested by two PS sequences—Situation + Problem and Situation + Evaluation— especially in the initial position, both in editorials and op-eds. Let me illustrate this with the following example: (4) (USA TOD A Y 27/07/98 ED) FAA mishandles baggage issue [Problem + Situation] The PS pattern in newspaper opinion pieces 405 (Problem) confused about how much and what you can carry on an airplane these days? 2You've got company. (Situation) 3 Consider this random and unscientific telephone survey of U.S. airlines' carry-on bag policies: 4American Airlines allows two carry-ons, if they meet specific weight and size limitations; 5laptop computers and briefcases count, but not purses. 6On Northwest Airlines, a total of three bags can be checked or carried, but purses, briefcases and computers get a free pass. 7 Southwest Airlines counts computers against its two-bag carry-on limit, but purses only if they're "real big." (...) It is claimed here that in the previous example the writer attempts to call readers' attention by consciously subverting the PS canonical order in a key position of the text. The same effect is achieved with PS sequences as Evaluation + Situation + Evaluation—instead of Situation + Evaluation —located in the initial position of the text. 5.2. The interactional characterization of the PS textual components in newspaper editorials and op-eds An interactional analysis of the PS organizational pattern has led to an identification of 863 rhetorical roles in newspaper editorials and 539 in op-eds. On the whole, it can be claimed that both USA TOD A Y editorials and op-eds are evaluative texts, in which the writers express their own position on a given usually problematic situation and elaborate on it. In fact, the most common rhetorical role in the texts analyzed is negative evaluation, which is predominant both in the newspaper editorials (173 cases) and op-eds (112 cases). It is followed in importance by elaborations (224 cases), situations (215 cases), justifications (109 cases), neutral evaluations (105 cases), exemplifications (86 cases), and positive evaluations (69 cases). The remaining rhetorical roles make up 27.27% of the total. The distribution of rhetorical roles by textual components is summarized in Table 3. The interactional analysis does not show significant differences between the two sub-samples. Editorialists make use of more subordinated rhetorical roles such as elaborations (149 versus 75 cases in op-eds), exemplifications (72 versus 14 cases in op-eds), and explanations (22 versus 9 cases in op-eds), while op-eds are built around sequences of evaluations of any kind. Here are some examples: (5) (USA TOD A Y 14/01/99 ED) Microsoft trial reaches midpoint with key charges 406 Isabel Alonso Belmonte Table 3. The distribution of rhetorical roles and sequences by PS components in newspaper editorials and op-eds Newspaper editorials Op-eds Situation Problem Solution Evaluation situation + situation + situation + (...); situation + [elaboration]; situation + [exemplification]; situation + [explanation] negative evaluation + negative evaluation + (...); negative evaluation + [elaboration]; negative evaluation + [justification] solution + solution + solution - (■■■); solution + [elaboration]; solution + [exemplification] negative evaluation -\ evaluation + (...); negative evaluation -\ [elaboration]; positive evaluation + [justification]; neutral evaluation ■+ [justification]; neutral evaluation ■+ [elaboration] negative situation + situation + situation + (...); situation + [elaboration]; situation + [reformulation] negative evaluation + negative evaluation + (...); negative evaluation + negative evaluation (...) + [elaboration] solution + solution + solution (■■■); solution + [elaboration]; solution + [justification] negative evaluation + [justification]; neutral evaluation - evaluation; negative evaluation - [elaboration]; neutral evaluation - [justification] negative (^negative evaluation + (18) [elaboration] + (^negative evaluation + (20) [ justification] + (21) [elaboration] (...) 17But that defense did not get off to a promising start Wednesday. 18Microsoft's first witness, MIT economist Richard Schmalen-see, argued not only that Microsoft didn't break any anti-trust laws, but also that it couldn't, since it doesn't have a monopoly in operating systems. 19That's a pretty tough sell, 20given Microsoft's 90% grip on the operating-system market and the fact that while the price for Windows has remained flat, software products that do face competition have seen prices drop sharply even as the products were vastly improved. 21A Consumer Federation of America study suggests that had Windows followed the pricing patterns of the rest of the industry, consumers would have saved a whopping $10 billion over the past three years. (...) The PS pattern in newspaper opinion pieces 407 (6) (USA TOD A Y 30/04/98 OP.ED) Industry strives for safety (3) positive evaluation + (4) positive evaluation + ^positive eval-uation (. . .) (...) Manufacturers that label products with truthful, science-based information about the effects of dietary supplements on the structure and function of the body provide a valuable tool for informed decision-making. 4The Food and Drug Administration's proposed regulations to enhance the reliability and uniformity of label statements underscore the value of dietary supplements for the maintenance of good health and nutrition. 5The FDA's proposals signal that most current label statements will continue to be appropriate and that the majority of the industry has been responsible in fulfilling the promise of the law. (...) The predominance of subordinate rhetorical roles can be interpreted here as indicative of a writer-responsible (reader-oriented) tendency in editorials: the reader's belief, comprehension, or adherence to the thesis presented is not taken for granted, hence the need, on the part of the writer, to justify, explain, exemplify, reformulate. Conversely, a minor presence of subordinated rhetorical roles in op-eds is understood as an illustration of less audience-sensitive (writer-oriented) writing. In other words, guest writers situate themselves in the informative framework created by the previous editorial and produce eminently evaluative texts without feeling the need to justify, explain, or exemplify their position. 5.3. The illocutionary characterization of the PS textual components in newspaper editorials and op-eds In total, 891 communicative acts in newspaper editorials and 549 in op-eds have been analyzed in terms of their predominant illocutionary value. A total number of 678 assertions and 541 statements have been individualized, followed by reported assertions (89) and shared knowledge assertions (38). Results indicate that in newspaper editorials, statements are slightly more predominant than assertions in the Situation components (126 cases of statements versus 36 assertions) and in the Problem ones (162 statements versus 131 assertions), while the number of assertions is slightly higher in the Solution components (31 assertions versus 28 statements) and especially outstanding in the Evaluation ones (183 assertions versus 42 statements). In the Solution component there is a minimal degree of imposition of writers to readers to take action of any kind about the issue which is under discussion. 408 Isabel Alonso Belmonte (7) (USA TOD A Y 13/08/98 ED) Tired of waiting on runways? (...) 30Congress can fix the situation, changing the law to at least give airports the ability to experiment with peak-time charges. 31 That would mean standing up to the powerful general aviation lobby. 32 But it would also bring relief to the millions of travellers cooling their heels every day on the nation's runways. This illocutionary characterization contrasts with the one used in USA TOD A Y opinion articles: guest writers express their own opinion about a given topic in a more direct and subjective way and try to convince their readers in two ways. Firstly, they criticize claims made by recognizable sources; this is achieved by concentrating reported assertions used in the Situation component (33.3% of all reported assertions in op-eds) and then providing an opinion about them. (8) (USA TOD A Y 14/08/98 OP.ED) FAA aware of abuses By Jane F. Garvey, administrator of the Federal Aviation Administration. (Situation) ^he Transportation Department's inspector general has raised concerns about the Familiarization Training program. 2The inspector general agrees the program, as conceived, "has a legitimate and valuable objective" of putting Federal Aviation Administration (FAA) personnel in the cockpit to get "firsthand experience into the operation characteristics of various types of aircraft, interface with air traffic control and gain insight into system performance." (Evaluation) 3We agree. 4Nothing beats firsthand experience for FAA workers. Secondly, they appeal to readers' beliefs and values; this effect is achieved by concentrating shared-knowledge assertions in the Problem and Evaluation components. Out of the 22 cases of shared knowledge assertions individualized in the op-eds analyzed, 10 appear in the Problem components and 12 in the Evaluation components. (9) (USA TODAY26/2/99 OP.ED) Seizing cars will save lives (Evaluation) 7DWI is a crime that frightens everyone. 8All of us understand the potential for death and injury created by a drunken driver unable to control a 3,000-pound vehicle.(...) And finally, this effect is also strengthened by urging readers to act in the Solution component, through the use of directive speech acts, which are The PS pattern in newspaper opinion pieces 409 Table 4. The writer-reader relationship in newspaper editorials and op-eds Toward a more impersonal style Toward a more involved style Situation (statements and assertions) Situation (statements and reported assertions) Problem (statements, reported asertions, Problem (assertions, shared knowledge assertions) assertions) Solution (statements and assertions) Solution (assertions, recommendations, statements, requests, proposals) Evaluation (assertions) Evaluation (assertions, shared knowledge assertions) very scarce in the sample, but appear all concentrated in this PS compo- nent. (10) (USA TOD A Y 13/08/98 OP.ED) "Peak pricing" not the answer (Solution) (...) 20Let's solve the problem, not ask it to go away. (...) To summarize, two different illocutionary tendencies have been identified in the sample to achieve communicative effectiveness in a given context: editorialists seem to use a more impersonal style, while guest writers appeal more directly and subjectively to the reader. Table 4 summarizes the results. 6. Discussion Successful writing depends on correctly identifying an audience and the potential reactions to the message conveyed, and employing the communicative conventions to which they are most likely to respond appropriately. The results of the analysis carried out so far emphasize the powerful role of writer-reader interaction as a force that shapes information exchange in the written form. By organizing sample texts around the PS pattern, our writers already show their will to be collaborative, or reader-friendly (Thompson and Thetela 1995); at the same time, by strategically manipulating the canonical linguistic realization of the PS pattern in different ways, the writers show their desire to convey their persuasive message more effectively. For example, they can choose to build their argumentative text around several sequences of Problems, or Evaluations, according to whether the writer wants to maximize the different facets of the issue being discussed and increase the feeling that the problem is re- 410 Isabel Alonso Belmonte ally serious, or just express his/her opinion directly. Writers can visualize their readers as prone to agreement, or as potential opponents who are likely to cast doubt on the veracity of the information presented. They can either choose to sympathize with their readers, claim in-group membership, common ground, or point of view; or they can just present facts and not make assumptions about the reader's wants, thus establishing a clear distance with their audience. As shown above, the use of any of these communicative strategies has clear effects on the textual realization of the PS pattern and on the illocutionary and rhetorical composition of each one of its components. Readers, on their part, are situated at the core of the communicative situation created by the newspaper opinion page and learn from both editorials and op-eds to form their own opinion about the issue being discussed. In this regard, the importance of the Problem and the Evaluation components, the scarce presence of the Solution component, and the prominence of the negative evaluation rhetorical role both in editorials and in op-eds are interesting points to be discussed. The top rhetorical sequence that underlies most of the texts analyzed is Problem + lack of Solution + (negative) Evaluation. In other words, both editorialists and guest writers show their proclivity to expose the problem, provide a (negative) evaluation of it, and send the message to the politicians. This evaluative characterization of the argumentative process in newspaper opinion texts contrasts with Tirkkonen-Condit's original conception of written argumentation as a problem-solving, resolutive process, indifferent to the genre in which it appears. In any case, it is consistent with the theory of news values: bad news is more newsworthy than good news in the written press (Galtung and Ruge 1965). Indeed, the lack of solutions and the negative evaluations of the problems exposed may generate a feeling of frustration among the audience, which can be used as another tool to pressure the Establishment to solve the problems exposed in the press. Further empirical research is needed to find out whether the above-described sequence, which has already been defined by some scholars as the "discourse of disillusionment" (Morrison and Love 1996) in other journalistic genres, is also common in newspaper editorials and other opinion pieces appearing in different publications, such as journals or magazines. 7. Final remarks Data presented in this article contribute to extending existing research into various textual features of the newspaper editorials and op-eds. The PS pattern in newspaper opinion pieces 411 However, one has to be cautious about the interpretations provided here on the basis of the study of the sample texts. In-depth interviews with writers investigating their writing processes and with their readers are encouraged here to test whether the explanations provided in this paper can be confirmed or not. Besides, I acknowledge here that the texts analyzed in this paper belong to only one publication and therefore results cannot be considered representative of the whole newspaper opinion genre. Thus further empirical research on a more extended corpus of newspaper opinion texts written in English is needed to assess the extent to which the data presented here reveal only trends or generalizations. I believe that some of the results presented need to be investigated further. This is the case of subverted PS sequences; although they illustrate the variety and intricate reality of the PS pattern as it unfolds in newspaper opinion pieces, further empirical research is needed to test whether PS subversions are more common in different PS sequences other than Situation + Problem or Situation + Evaluation, in other textual positions, or in other genres. Another interesting point of research is the relative scarcity of subordinate rhetorical roles in op-eds. It would be interesting to find out whether this rhetorical tendency can be made extendable or not to other opinion articles published in English written newspapers, magazines, and journals. Finally, the importance of parameters such as individual writing style should also be investigated to determine their influence in text patterning. Some other promising lines of research open up for future analysis. An interesting one would be, for example, to broaden the number of rhetorical and illocutionary communicative strategies described here and to better their textual description. It would be enlightening to check whether they are present in other written argumentative texts or not, whether they are contextually constrained or not, what kind of relationship among writers and readers they reflect, etc. Also, an extensive analysis of the PS pattern in a multilingual corpus of different genres could be very useful not only to detect textual differences and similarities across languages but also to unravel the culturally based rhetorical influences present in argumentative texts produced by non-native writers. Notes * This article partly reproduces the results of my Ph.D. dissertation on the textual structure of a corpus of newspaper editorials and opinion articles written in English (Alonso Belmonte 2004). 1. Op-ed is the abbreviation of "opposite the editorial page." 412 Isabel Alonso Belmonte 2. These last two PS components—Result and Evaluation—sometimes conflate in a single sentence or text chunk (Hoey 1983). 3. The original version of the method also includes a macrostructural analysis of argumentative texts (Van Dijk 1980), but since its usefulness exceeds the scope of this paper, we will focus solely on the above-mentioned units of analysis. 4. As Teo (1995: 84) explains, a unit smaller than a sentence is desirable "(■ ■ ■) first, when parts of a sentence perform a different interactional role. Second, when they perform a similar interactional role but have a different illocution. Third, when they are preceded or followed by another interactional role." 5. There is only a small number of subordinated argumentative PS sequences (just 24 cases) that adopt the rhetorical roles of elaborations (15), justifications (7), and exemplifications (2). References Almeida, E. P. 1992. A category system for the analysis of factuality and non factuality in newspaper discourse. Text 12. 233-262. Alonso Belmonte, I. 2004. La expresion de la opinion en el discurso escrito: Un estudio de la estructura textual de la argumentation en elperiodismo de opinion en ingles. Madrid: Uni-versidad Complutense de Madrid. Alonso Belmonte, I. 2009. Positioning the reader: A study on the use of interactive textual patterns in English written newspaper editorials and articles of opinion. English Text Construction 2(1). 48-69. Al-Sharief, S. 1998. Analysing interaction in texts: A framework. Liverpool Working Papers in Applied Linguistics 4(1). 1-31. Ansary, H. & E. Babaii. 2005. The generic integrity of newspaper editorials: A systemic functional perspective. RELC Journal 36. 271-295. Aston, G. 1977. Comprehending value: Aspects of the structure of argumentative discourse. Studi italiani di linguistica teorica ed applicata 6(3). 465-509. Bolivar, A. 1994. Discurso e Interaction en el Texto Escrito. Caracas: Universidad Central de Venezuela. Connor, U. 1987. Argumentative patterns in student essays: Cross-cultural differences. In Ulla Connor & Robert B. Kaplan (eds.), Writing across languages: Analysis of L2 text, 57-71. Reading, MA: Addison-Wesley. Connor, U. & J. Lauer. 1988. Cross cultural variation in persuasive student writing. In Alan C. Purves (ed.), Writing across languages and cultures: Issues in contrastive rhetoric, 138- 159. Newbury Park, CA: Sage. Crombie, W. 1985. Process and relation in discourse and language learning. Oxford: Oxford University Press. Edmonson, W. 1981. Spoken discourse. A model for analysis. London: Longman. Flowerdew, L. 2000. Using a genre-based framework to teach organizational structure in academic writing. ELT Journal 54. 369-378. Flowerdew, L. 2003. A combined corpus and systemic-functional analysis of the problem-solution pattern in a student and professional corpus of technical writing. TESOL Quarterly 37(3). 489-511. Flowerdew, L. 2008. Corpus-based analyses of the problem-solution pattern. Amsterdam & Philadelphia: John Benjamins. The PS pattern in newspaper opinion pieces 413 Galtung, J. & M. Ruge. 1965. The structure of foreign news: The presentation of the Congo, Cuba and Cyprus crises in four Norwegian newspapers. Journal of Peace Research 2. 64-91. Grimes, J. E. 1975. The thread of discourse. The Hague: Mouton. Hoey, M. 1979. Signalling in discourse (Discourse Analysis Monographs). Birmingham: ELR, University of Birmingham. Hoey, M. 1983. On the surface of discourse. London: Allen & Unwin. Hoey, M. 1986. Overlapping patterns of discourse organization and their implications for clause relational analysis of problem-solution texts. In Charles R. Cooper & Sydney Greenbaum (eds.), Studying writing: Linguistic approaches, 187-214. Newbury Park, CA: Sage. Hoey, M. 2001. Textual interaction. London: Routledge. Jordan, M. P. 1980. Short texts to explain problem-solution structures and viceversa. Instructional Science 9. 221-252. Jordan, M. P. 1984. Rhetoric of everyday English texts. London & New York: George Allen & Unwin. Jordan, M. P. 2007. Interactive solution-problems: A set of structures in general and scientific writing. In Christopher S. Butler, Raquel Hidalgo Downing & Julia Lavid (eds.), Functional perspectives on grammar and discourse: In honour of Angela Downing, 279-300. Amsterdam & Philadelphia: John Benjamins. Koester, A. 2004. The language of work. London: Routledge. Morrison, A. & A. Love. 1996. A discourse of disillusionment: Letters to the editor in two Zimbabwean magazines 10 years after independence. Discourse and Society 7(1). 39-75. Navas, C. A. 2005. Analyzing a problem-solution pattern in the transcription of a conversation: Suggestions for the ELF classroom. Actualidades Investigativas en Education 5(2). http://revista.inie.ucr.ac.cr/articulos/2-2005/conversation.php (accessed 1 November 2007). Paltridge, B. 1996. Genre, text type and the language learning classroom. ELT Journal 50(3). 237-243. Paltridge, B. 2001. Genre and the language learning classroom. Ann Arbor: University of Michigan Press. Riazi, A. & F. Assar. 2001. A text analysis of Persian newspaper editorials at macro and micro levels. Journal of Social Sciences and Humanities of Shiraz University 31/32. 175-192. Searle, J. R. 1979. Expression and meaning. Studies in the theory of speech acts. Cambridge: Cambridge University Press. Scott, M. 2002. Picturing the key words of a very large corpus and their lexical upshots—or getting at the Guardian's view of the world. In B. Kettemann & G. Marko (eds.), Teaching and learning by doing corpus analysis, 43-50. Amsterdam: Rodopi. Swales, J. 1990. Genre analysis: English in academic and research settings. Cambridge: Cambridge University Press. Teo, A. 1995. Analysis of newspapers editorials: A study of argumentative text structure. Ann Arbor, Ml: UMI Dissertation Services. Thompson, G. & P. Thetela. 1995. The sound of one hand clapping: The management of interaction in written discourse. Text 15(1). 103-127. Tirkkonen-Condit, S. 1985. Argumentative text structure and translation. Studia Philologica Jyraskylaensia, Univ. Tyvaskyla. Van Dijk, T. A. 1980. Macrostructures. Hillsdale, NJ: Erlbaum. Winter, E. O. 1982. Towards a contextual grammar of English: The clause and its place in the definition of sentence. London: Allen & Unwin. 414 Isabel Alonso Belmonte Winter, E. O. 1994. Clause relations as information structure: Two basic text structures in English. In Malcom Coulthard (ed.), Advances in written text analysis, 46-68. London: Routledge. Isabel Alonso Belmonte is Associate Professor at the University Autonoma de Madrid (Spain), where she teaches English as a Foreign Language. Her Ph.D. dissertation analyzed the expression of opinion in English written newspaper editorials and opinion articles. Her research concerns media discourse analysis, pragmatics, and the development of the discourse (and pragmatic) competence in foreign language teaching, areas in which she has extensively published. Address for correspondence: Departamento de Filologias y su Didactica, Facultad de Formation de Profesorado y Education, Universidad Autonoma de Madrid, Campus Universitario de Cantoblanco, C/Tomas y Valiente, n. 3, 28049-Madrid, Spain . Copyright of Text &Talk is the property of Walter de 6ruyter GmbH &Co. KG. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.