# Introduction I will show that the JICLE students mostly did not use the Problem-Solution pattern, which is represented by Situation -Problem -Response/solution -Evaluation (Hoey, 2001), while the US students used it in a similar way to the model sequence. I will also discuss pedagogical implications of the findings. The study provides teachers with ideas about what aspects should be focused on in the teaching of the Problem-Solution pattern in EFL classes. # II. # Methodology The research question of the present study is: How do L1 Japanese students use Problem-Solution in the writing of English argumentation essays, as compared with L1 English students? # a) Text data In the present study, the text data for NNS students argumentation essays are the Japanese subcorpus of the International Corpus of Learner English (JICLE), and those for NS students' essays are the US subcorpus of the Louvain Corpus of Native English (US). NS student essays, rather than ones professionally written, are used as a reference corpus. This is because if not necessarily correct and a model for instructional application (Granger, 1998;Leech, 1998;Adel, 2004Adel, , 2006)), student essays are a specific genre of text (Lorenz, 1999), and NS essays are similar to NNS essays in several factors, such as text length, purposes of writing, and writers' age and writing proficiency levels (Gilquin, Granger & Paquot, 2007). The JICLE and the US essays comprise 366 and 176 essays, and word tokens, recounted according to AntConc (Laurence, 2012), are 202,099 and 150,530, respectively. Some topics are common in the two corpora (e.g., capital punishment, nuclear energy), but many others are corpus-specific; for example, JICLE wrote on maintaining a maiden name after marriage and employment systems, whereas US wrote on euthanasia and abortion. # b) Theoretical base The concepts of metadiscursive nouns and text patterns used in analyzing the JICLE and the US in the present study share the idea that lexical vocabulary can work as cohesive devices, and it is traced back to Vocabulary 3 (Winter, 1977), which is comprised of he present study investigates the use of Problem-Solution in student essays to identify whether or not, or to what extent, this text pattern is a source of perceived difference in NNS student essays, in comparison with NS student essays. The study is a follow-up to Tahara (2017), which compared argumentation essays written by NNS students with those by NS students, conducted from the perspective of the use of metadiscursive nouns. They are general and unspecific meaning nouns that can serve as markers of the discourse in some ways by referring to a textual segment in the texts where the nouns occur. Of 33 selected metadiscursive nouns examined in Tahara (2017), this paper reexamines the use of a noun problem in relation to the Problem-Solution pattern. The focus of the noun for the investigation of the use of Problem-Solution is because in the 2017 study (Tahara), problem very often occurred in combination with a Response/Solution-indicating vocabulary in both corpora, as in 'problem is solved'; 'consider the problem'; or 'problem should be dealt with' (underlined are vocabulary signaling Response/Solution). T Problem-Solution is a well-known English rhetorical pattern, often used in technical academic writing (Flowerdew, 2003), but it seems not to have been taught in the writing of English essays, at least in Japan. In contrast, the text pattern often used in the class is Introduction-Body-Conclusion to prepare for TOEFL/IELTS writing, along with the teaching of the paragraph structure, comprised of a topic sentence, supporting details, and concluding sentences. The use of Problem-Solution in student essays is not so much investigated either in the research literature. However, the findings from a small number of studies suggest a need to draw more attention to this rhetorical pattern: Flowerdew (2003), who analyzed the Problem and the Solution elements of Problem-Solution in the technical essays by L1 Cantonese students and professional writers, reports a difference in Problem-Solution accounted for by the preferred type of signaling vocabulary. Also, Galán and Peréz (2004) report an improvement in the quality of L2 essays, after testing some approaches to teach the Problem-Solution pattern on Spanish students. nouns, verbs, and adjectives. Vocabulary 3 is considered to form clause relations by connecting two segments, such as result being able to signal a causeconsequence relationship (Coulthard, Moon, Johnson, Caldas-Coulthard & Holland, 2000). By extension, lexical items can signal functional segments of larger discourses such as Problem-Solution, Argument-Counterargument, and General-Specific (Coulthard, et al., 2000). In case of metadiscursive nouns, as a type of lexical items, they can serve as text-organizing devices. Being abstract nouns with unspecific meanings, they recover their full meanings in reference to the text, and can form a cohesive flow of information by guiding readers through the text (Jian & Hyland, 2017). Metadiscursive nouns are proposed under varied names (general nouns in Halliday & Hasan, 1976; enumeration in Tadros, 1994; anaphoric nouns in Francis, 1986; carrier nouns in Ivanic, 1991; shell nouns in Schmid, 2000) that emphasize one or two aspects of the roles of the nouns. In analyzing the text data, the present study uses the shell noun (Schmid, 2000) conceptual framework, which explains metadiscursive roles of nouns in relation to several syntactic patterns (host syntactic patterns), as follows (N=metadiscursive noun): ? N-be-CL (problem is that/to-clause): Lexicalization is in the complement clause ? N:CL (problem that?): Lexicalization is in the clause adjacent to the noun ? th-be-N (This is a problem): Lexicalization is in the preceding segment ? th-N (the problem): Lexicalization is in the preceding segment. # c) Procedures The investigation of Problem-Solution is conducted by analyzing the use of problem with the AntConc (Laurence, 2012) (Hoey, 1983(Hoey, , 2001(Hoey, , 2006;;Jordan, 1984;Winter, 1986). The identification of each of the functional segments is conducted by finding the signaling vocabulary. III. # Frequencies of Metadiscursive Problem In terms of the host syntactic patterns, problem occurred the most frequently in th-N (N=22:17, LL 1.53), followed by N-be-CL (N=10:12, LL 0.34) in both corpora. For these syntactic types, there is no significant frequency difference between the two corpora. However, a third syntactic type th-be-N (N=8:3, LL 4.61) occurred significantly more in JICLE than in US. Finally, N:CL virtually did not occur in either corpus, as shown below in Table 1: (Figures are normalized; # Lexicalization of Problem and # Problem-Solution a) Problem for N-be-CL For N-be-CL, problem occurred in similar frequencies in JICLE and US (N=10:12, LL 0.34). Lexicalization was also similar in the two corpora, with the CL most often occurring as a that-clause, or a whclause to a much lesser extent. The similarity may be because the lexicalization of N in N-be-CL is a grammatical requirement, and the JICLE students used the pattern properly. However, a difference was observed in terms of to what extent the content in the CL was explained in the succeeding segment. In US, the CL tended to be followed by a segment that provides further information about the meaning of problem, as exemplified in Ex. 1 (underlined is the referent): Ex. 1. The problem content expressed in the CL (underlined) is added information in the succeeding sentence. b. The problem was that the faster and more reliable the computers became, the more speed people demanded. Then came the DSL connection. DSL is a digital based line that can? (JICLE) c. I think a big problem is how to offer readers the opportunity to find books they want to read. Publishers, wholesalers, and bookstores must make more efforts to play better intermediary roles between readers and books. (JICLE) Thus for N-be-CL, although there was no particular difference in the lexicalization in the CL, how the problem content in the CL was further explained in the succeeding segment differed between JICLE and US. b) Problem for th-be-N Ex. 2. If a student has the desire to pray at any moment during the school day he or she should not encounter any determent. Only when students (or faculty) force any students to join in the prayer does it become a problem. The act of trying to force an unwilling person to digest the religious philosophy of another may lead to an uncomfortable educational setting that would hinder learning and social growth.// Prayer in public schools may continue to gain more popularity in the United States. (US) In contrast, in JICLE, no Reason segment followed the Problem segment. Instead, the Problem segment was often immediately shifted to a Response in the sequence of [Problem -(this is a problem) -Response], as can be seen in Ex. 3: Ex. 3. In Japanese class, teachers take too much time to teach English grammar. I think that it is too enough. However, students aim an entrance examination of Japanese university. It is a big problem. In order to increase the number of children who can speak English well, the government has to change the educational system. ? (JICLE) Functionally, problem is an 'implicit' causation device; problem itself does not signal causal meanings but can serve as a causation device by referring to the preceding segment and directing it to the succeeding The analysis in this section shows how different, or similar, the Problem-Solution pattern that occurred in the JICLE and the US essays was. At the same time, the analysis reveals roles of problem as an 'implicit' causation device, which were used differently in the two corpora. In contrast, in JICLE, the content of the CL was mostly not explained in the succeeding segment, and the discourse immediately shifted to a next functional segment. In the following a), the problem content of the CL (underlined) is directly followed by a new Problem segment, in b) by the next topic, and in c), by a Response segment: The content of the problem (underlined), which is that too much time is spent on teaching grammar, may not be easily perceived as a problem without a cultural knowledge about Japanese English education. Besides, the referent, stating it is for the entrance examination, supports grammar teaching, and this further confuses the reader as to whether the referent expresses a problem or not. Without clearly explaining the content of problem the discourse is immediately and suddenly shifted to a Response (signaled by increase, change). ? The main problem was that it seemed to be made in haste . The judges decided the fate of this innocent fouryear-old boy in a matter of four hours?. (US) a. ? the first problem is to select which name to let them use . Second, if children's names are different from their parents', ? (JICLE) Problem for th-be-N (e.g., this is a problem) was a pattern strongly preferred by JICLE more than in US (N=8:3, LL score 4.61). The noun was similarly lexicalized in the immediately preceding short segment in both corpora. However, a clear difference was observed in the presence, or non-presence, of a Reason segment that succeeded this is a problem. In US, the Problem element was almost always followed by a Reason element as in [Problem -(this is a problem) -Reason -Response], as shown below in Ex. 2: The problem content (underlined, and signaled by force) is shifted by problem (in 'it becomes a problem'), to a Reason segment that explains why the referred content is a problem. The shift to the Reason segment in Ex. 2 occurs without a 'marker of reason' (e.g., since, because), but sometimes there was such a marker, and th-be-N (e.g., this is a problem) in US almost always moved the discourse from the Problem to a Reason segment. # c) Problem for th-N Problem for both th-be-N and th-N functions anaphorically. However, unlike for th-be-N, where the referent was almost always short and placed immediately before problem, the referent for th-N was often long and followed by a segment before the problem. Thus, problem for th-be-N often occurred in the of [Problem -(Segment) -the problem] in both corpora. # i. Features in JICLE In [Problem -(Segment) -the problem], lexicalized patterns of problem in JICLE exhibited some corpus-specific vagueness. The rhetorical sequence was also corpus-specific: Problem-Solution was often not intended, and if the pattern occurred, it mostly existed only in appearance, as explained below: JICLE feature 1 A featured lexicalization pattern in JICLE, expressed in a longer referent for th-N, was that the meaning of the problem was bi-directional, as shown in Ex. 4, below: Ex. 4. Also, there is another very big and important problem. It is "gakubatsu." I think that the groups of like this exist everywhere: in the company, government and even in the sports field. Of course, it is good that people have a friendship for those who graduated same school. But I feel it excessive. For example in a company, when two men who are same capacity and career may be able to career up. But one of them graduated famous university same as his boss. Being able to career up is only one. Then, the boss will select a man of graduating same school. I do believe that something like this can happen. Also this problem may make other new problem?. (JICLE) The problem is the practice of gakubatsu, academic cliques, in Japanese companies, and its reference seems the long preceding segment (underlined). Although a problem content is indicated by excessive, it does not provide sufficient information as to what exactly is considered a problem. Also, the referent is evaluated positively in the Evaluation segment that follows; stating can happen. This inconsistency in the writer's evaluation makes the discourse bi-directional and confuses the reader. Then, summarizes the vague problem content and shifts the discourse to a new Problem segment, but hastily and suddenly. This function of problem can be termed 'superficial generalization' (Hinkel, 2001). It refers to a role that summarizes vague content with a general meaning noun. Hinkel (2001) states that superficial generalization was often observed in NNS essays, but rarely found in NS essays. # JICLE feature 2 The JICLE students tended to explain the meaning of the problem in a narrative, where a series of events is described sequentially. In the narrative discourse, delineation of functional segments was difficult, and discourse marking roles of problem seemed weak, as shown in Ex. 5: Ex. 5. For example I visited China last month. The accident happened. The bath in my room was broken. Water could not take out. So I tried to ask how to fix the bath on the phone. I stayed in the hotel which is managed by Japanese company. Therefore I thought the staff in the hotel can speak Japanese. On the phone I asked the man by Japanese. However he couldn't understand what I said. He said to me "I can't speak Japanese at all. Please say in English or Chinese" As I couldn't speak Chinese, I told him the problem in my bathroom by English. Because I used English, we could communicate with each other. So the bathroom would be fixed?. (JICLE) In this text, the meaning of the problem may be that water did not come out in the bathroom, expressed in a short segment (underlined). However, alternatively, the whole segment preceding the problem may be the referent. In either of the cases, the problem, used in the sentence 'I told him the problem', which describes one of the events, seems not to be working as an explicit discourse shifting device. With a blurred segment shift, there seems no intended Problem-Solution pattern or existence of the pattern in the excerpt. # JICLE feature 3 Sometimes, the Problem-Solution text pattern occurred in JICLE. Still, the pattern was irregular, and one common type of irregularity was exhibited in the shift to the Conclusion segment, as shown in Ex. 6: Ex. 6. The other day, I read about different ideas of meeting one's e-mail friend in a reader's column of a newspaper. I was shocked that the majority of contributions of the column said they could meet their mail friend. Of course most of the contributions contained additional advices such as "You # this problem The excerpt in Ex. 4, above, has a sequence of [weak Problem -positive Evaluation -new Problem (this problem)], indicating there is no intended Problem-Solution pattern. Instead, the text appears to be constructed by relying on the segment initial sentence 'there is another very big and important problem'. It is a 'frame marker' that labels the text stage and announces a discourse goal (Hyland, 2004). Then, this problem serves to terminate the discourse by superficially generalizing the vague content of the preceding segment. # ( G ) Global Journal of Human Social Science -Year 2020 © 2020 Global Journals should talk with your mail friend before you meet him directly." or "It is better to meet in company with other friends." and so on. In spite of the incident which happened only days before, many people think they could meet their email friends. They are too less sensitive to danger. This lack of a sense of impending crisis could let still more crimes happen.// Here, let's think about what we should do to prevent troublesome e-mails and disgusting incidents. In the first place, we should change our e-mail address of cellular phones from "phone number + ?¿" to what you newly think of. This is easy and effective way. I used to be annoyed by e-mail address. Then I was relieved from annoying e-mails. Thinking of the unpleasantness of annoying e-mails, it is not trouble to tell our friends of the new address.// Traders concerned should explain their customers about problems of e-mail and suggest that customers change their e-mail address. Also the government should do something with this problem, as long as we take pride in being advanced in portable electric products and its' network. One-way e-mails of invitation or advertisement somehow should be regulated. # Conclusion: important is consciousness As shown in Ex. 6, in JICLE, the Problem-Solution pattern was sometimes used, but the Conclusion was often a generalized comment. It may be a strategy to converge several elements, which were not explained or connected with each other so much, into a concluding remark. # JICLE feature 4 Sometimes, the Problem-Solution sequence in JICLE seems to have existed, but only in appearance, as illustrated in Ex. 7: Ex. 7. The meaning of this problem (underlined) (signaled only by less important) is not clear, but helped by disagree in the Evaluation segment that follows. The Evaluation is shifted to Response, which is signaled by cherish, but more explicitly by a frame marker that states 'I propose?' The discourse then moves to another Evaluation (signaled by inferior) and to Response (signaled by punishment, be strict, serve, life imprisonment). Thus, the discourse seems to consist of # Evaluation (to Problem) [this problem]: not have ? awareness The extract seems to have the sequence of [Problem -Evaluation -Response (this problem) -Conclusion]. Firstly, the meaning of this problem is to meet one's email friend, expressed in the distant referent (underlined). However, the content is barely perceivable as a problem and only helped by evaluative vocabulary shocked, less sensitive, and lack in the succeeding Evaluation segment. Then, the discourse is shifted to a Response, explicitly with the use of the frame marker, 'let's think about?.' The Response that follows is a long segment, although not comprised of a description on a focused aspect, but of several responses, with each of them not connected or explained in detail. Then, the whole discourse is summarized in the Conclusion. Characteristic about the Conclusion is that the statement, 'the most important thing is our own consciousness of the problem', is a superficial, uncontested comment, not drawn from the preceding argument. Shown below is a schematic chart of Ex. 6: ? Today, I assume that almost all the people look upon animals as less important than human beings. I strongly disagree with this idea that most people have. I propose that we have to cherish animals as well as our family or friends. I think that people who kill or animals are inferior because animals cannot speak a word and they are nonresistance. They are weaker than us. I also think that the punishment of cruelty to animals should be more and more strict. For example, a person who abused or killed a great number of animals have to serve a sentence of a life imprisonment or something like that. People, as a whole, do not have a keen awareness of this problem. // (JICLE) At first glance, the sequence is similar to the model English Problem-Solution pattern. However, these elements do not form a linear sequence, but most of the latter elements refer back to the initial Problem, resembling the hierarchical 'topical network' of Japanese texts. In the topical network, 'the main discourse topic operates as a pivotal point of reference, providing the starting point for related topics' (Maynard, 1998: 39). Also, the Conclusion, 'People, as a whole, do not have a keen awareness of this problem', is a superficial generalized comment, as also found in Ex. statements without support and? an unsubstantiated general statement? in the conclusion' (p.149) in L2 expository writing by L1 Japanese students. # Summary In JICLE the meaning of problem for th-N was sometimes bi-directional, and sometimes expressed in the narrative discourse, and these features contributed to the vagueness of the meaning of problem. Concerning the use of the Problem-Solution sequence, the pattern often seems not to have been intended in JICLE, or if intended, the sequence was irregular, as exhibited in the shift to the Conclusion. Also the Problem-Solution sequence on appearance had a structure similar to the hierarchical 'topical network' often found in Japanese texts (Maynard, 1998). ii. Features in US In the US corpus, the meaning of problem in the referent was clearer, and the Problem segment occurred in a sequence that was similar to the Problem-Solution model pattern (e.g., Problem -Response/solution -Evaluation -Conclusion). This can be seen with Ex. 8, shown below: Ex. 8. A basic right of a human being living in a democratic society is that they are entitled to life. Food and shelter are fundamental needs a person must have in order to survive. A percentage of the population of the citizens of America lack these necessities. Homelessness has become an intense problem in the United States that must be solved. # The problem refers to a homeless situation that represents people's lack of basic necessities in the US ( There is an agreement between all opposing viewpoints in government that something must be done that (sic) to combat homelessness. The necessary method to be used to understand homelessness is under careful scrutiny. Advocates for people without homes tend to get so wrapped up in the size of the homeless population that they ignore the fact that it is not right that this discrepancy exists. What is true is that in the United States it is continually growing, and action must be taken immediately to alleviate the problem?. (US) underlined Situation: food and shelter are the basic rights of human beings ). The reference is followed by a Response (e.g., signaled by agreement, combat, method, scrutiny). Ex.9: // Students come to school with the hope of furthering their education so that they may one day practice a career in the field of their major. Students can read and study textbook theories and examples, but it helps them to have access to guides. These guides are their professors. They pay not only for books every semester, but for tuition. Tuition money is supposed to pay for the professors to teach the students. Students who pay for their tuition expect professors to teach them in return. When professors can't speak English, they can't communicate with English speaking students; therefore they cannot teach them. Students need someone to guide them through wordy material in a book. Students need someone to lead labs and give them hands on experience. When the student is paying for his needs, for what reason should he not have them fulfilled? When a person buys an item from a store, he pays and receives the item he paid for. When students pay tuition, they should also receive what they're paying for-a teacher who can teach them. Sometimes a teacher's assistant can substitute for a professor, and this is what should happen if a professor can't speak a language well enough to teach. This is one solution, but we are confronted with another problem. Even though teaching aides can help, students don't benefit from them as much as they do a professor. A professor has a Ph.D. and is a trained expert. Students need to be taught by professors when they are taking advanced courses that apply to their major field of study. A professor is an asset to upper class students finishing courses for their degrees. To summarize, this section has shown that Problem-Solution in US occurred in a similar sequence to the model English rhetorical pattern, even when causal relations were embedded. It has also shown that the lexicalized content of problem tended to be easily perceivable as a problem due to sufficient information with the use of signaling vocabulary. # d) Summary The comparison between the JICLE and the US essays revealed differences in the use of problem as a marker of the discourse and also in its relations to the Problems-Solution pattern, as follows: c. Additionally, it has emerged that, in JICLE, problem was little used as a causation device that can form a causeresult clause relation. As a result, the JICLE students did not include a Reason segment in their writing. Causeresult is an important rhetorical pattern and will need to be addressed in the teaching of argumentation essays. V. # Pedagogical Implications One argument about L2 writing is that NNS students do not need to be taught 'correct' or 'acceptable' style of essays, as there are many Englishes, and English by non-native speakers is as legitimate as English by native speakers (e.g., Kachru, 1985;Jenkins, 2007;Mauranen, 2010;Seidlhofer, 2011). However, some of the JICLE features that were not exhibited in the US essays (e.g., an abrupt shift of discourse, superficial conclusion, very little use of causal relations) were confusing to the readers, and could hinder understanding by the reader. Such features would need to be addressed in the teaching of the EFL writing, so that NNS students can write argumentation essays that are acceptable internationally. Concerning these JICLE features, the previous studies suggest an influence of the writer's L1 culture and writing conventions. For example, bi-directional argumentation may reflect an L1 social value where outright contradiction to other people's opinion is not appreciated (Oi & Kamimura, 1997;Oi, 1986;Natsukari, 2012), and the hierarchical 'topical network' in JICLE is similar to a typical L1 Japanese writing style (Maynard, 1998). Also, preference for frame markers a is similar to L1 Japanese writing style; Saijo (1999in Maynard, 2005) reports that the readers of Japanese texts written without using frame markers had a hard time to understand the messages in the texts, indicating importance of frame markers in constructing Japanese texts. About a generalized conclusion in L2 essays by Japanese students, this feature was found common in Japanese editorials written by professional writers in the study of Ushie et al. (1997). A possible transfer of L1 language use to the JICLE essays suggests that without some explicit instruction, pervasive L1 features may remain in L2 writing. A question is how rhetorical patterns can be taught. Is it effective to provide students with readymade template of Problem-Solution in argumentation essays as Hoey (2001) suggested, or do such patterns not need to be emphasized in teaching? To this question, a study by Galán and Peréz (2004) with L1 Spanish students indicates the benefit of teaching signaling vocabulary, rather than a ready-made template, stating that providing the students with vocabulary triggered the Problem-Solution pattern. Such vocabulary-centered teaching may be an approach to be tested. It might work well with Japanese students, as the JICLE students used much fewer signaling nouns, in comparison with the US students. # VI. # Future Research The present study examined the Problem-Solution pattern in NNS argumentation essays, using NS essays as a reference corpus, focusing on problem as a metadiscursive noun. The methodology that uses problem has proved effective to explain how the Problem-Solution pattern occurred in English essays, and thus, can be valuable tool for a further investigation of this textual pattern in students' argumentation essays. The Problem-Solution pattern is an essential consideration in the teaching of argumentation essays, and this line of inquiry should be pursued further. Also, the methodology which utilizes the conceptual framework of metadiscursive nouns seems to have a potential to examine other types of text patterns, as well as clause relational patterns, and should be explored more for the study of the discourse. 1of problem in Schmid (2000) syntactic patternsJPUSLL scoresN-be-CL/CL-be-N10 (20)12 (18)0.34N:CL1 (2)02.23th-N22 (47)17 (26)1.53th-be-N8 (17)3 (5)4.61Sum43 (86)33 (49)2.29For each of the host syntactic patterns, otherand occurrence of the Problem-Solution sequence arethan for N:CL, the form of lexicalization of problemexamined in the following sections. Problem: incidentEvaluation: shocked, lack, less sensitive? Frame marker to Response (Let's think about whatto do?)Response:(we): prevent, change ? positive Evaluation: effective,easy(traders): explain, suggest(government): do something, regulate; [this problem] b. Concerning the use of the Problem-Solution pattern, inUS, it occurred in a sequence that is similar to the modelsequence (e.g., Problem -Response -Evaluation -Conclusion). However, in JICLE, Problem-Solution oftenseems not to have been intended. Also, if Problem-Solution occurred, the pattern was irregular and often onlyin appearance. For example, the Conclusion segmentwas formed with little meaning connection to thepreceding segment. Also, the functional elements ofProblem-Solution, seemingly arranged in a linearsequence, actually did not form a successive meaningconnection. Instead, the meaning expressed in each ofthe succeeding functional segments referred back to thetopic expressed in the initial segment (see Ex. 6 & Ex.7). © 2020 Global Journals * Selecting quantitative data for qualitative analysis: A case study connecting a lexicogrammatical pattern to rhetorical moves AAdel Journal of English for Academic Purposes 16 2004. 2020 * Metadiscourse in L1 and L2 English AAdel 2006 John Benjamins Amsterdam/Philadelphia * Written Discourse MCoulthard RMoon AJohnson CCaldas-Coulthard BHolland 2000 Birmingham The Centre for English Language Studies, The University of Birmingham * A combined corpus and systemic-functional analysis of the problem-solution pattern in a student and professional corpus of technical writing LFlowerdew TESOL Quarterly 37 3 2003 * Anaphoric Nouns: Discourse Analysis Monographs,11 GFrancis 1986 Birmingham English Language Research, University of Birmingham * The Problem-Solution Pattern: A Tool for the Teaching of Writing?' BELLS: Barcelona English Language and Literature Studies AGalán MPeréz 2004 12 * Learner corpora: The link in EAP pedagogy GGilquin SGranger MPaquot Journal of English for Academic Purposes 6 2007 * SGranger Learner English on Computer. London 1998 Longman * MA KHalliday RHasan Cohesion in English 1976 Longman * Matters of cohesion in L2 academic texts EHinkel Applied Language Learning 12 2 2001 * On the surface of discourse MHoey 1983 George Allen and Unwin London * Textual Interaction: An Introduction to Written Discourse Analysis MHoey 2001 Routledge London * Problem-Solution Patterns MHoey Encyclopedia of Language & Linguistics 2006 Second Edition * Disciplinary interactions: Metadiscourse in L2 postgraduate writing' KHyland Journal of Second Language Writing 13 2004 * Nouns in search of a context: A study of nouns with both open-and closedsystem characteristics RIvanic International Review of Applied Linguistics in Language Teaching 29 1991 * English as a Lingua Franca: Attitude and Identity JJenkins 2007 Oxford University Press Oxford * Metadiscursive nouns: interaction and cohesion in abstract moves' English for Specific Purposes KFJian KHyland 2017 46 * Rhetoric of everyday English texts MPJordan 1984 Allen & Unwin London * Standards, Codification and Sociolinguistic Realism: The English Language in the Outer Circle BBKachru English in the World: Teaching and Learning the Language and Literatures RQuirk HGWiddowson Cambridge, UK Cambridge University Press 1985 * AntConc, version 3.2.4w ALaurence 2012 * GLeech Learner English on Computer. viv-xxLondon Granger, S. 1998 Longman * Adjective Intensification -Learners versus Native Speakers: A Corpus Study of Argumentative Writing GRLorenz 1999 Amsterdam/Atlanta; Rodopi * Features of English as a lingua franca in academia AMauranen Helsinki English Studies 6 2010 * Principles of Japanese discourse -handbook SMaynard 1998 Cambridge University Press Cambridge * Danwa Hyogen Handbook (Discoursal Expressions Handbook) SMaynard 2005 Kuroshio Shuppan Tokyo * Use of I in essays by Japanese EFL learners SNatsukari JALT Journal 34 1 2012 * Cross-cultural differences in rhetorical patterning: A study of Japanese and English KOi JACET Bulletin 17 1986 * A pedagogical application of research in contrastive rhetoric KOi TKamimura JACET Bulletin 28 1997 * Danwa ni Okeru Meta Gengo no Yakuwari (Roles of Metalanguage in Discourse) MSaijo 1999 Kazama Shobo Tokyo * English Abstract Nouns as Conceptual Shells: From Corpus to Cognition HSchmid 2000 Mouton de Gruyter Berlin * Understanding English as a Lingua Franca BSeidlhofer 2011 Oxford University Press Oxford * Predictive categories in expository text ATadros Advances in Written Text Analysis MCoulthard New York Routledge 1994 * The Use of Nouns in Japanese and American Student Writing NTahara 2017 University of Birmingham E-thesis * Roles of Metadiscursive Nouns in L2 English Writing -Comparison with L1 Writing' NTahara International Journal of Languages, Literature and Linguistics 6 2 2020 * The usage of generalisation in EFL writing of Japanese university students YUshie DHNagatomo ESchaefer MNishio 1997 * Ochanomizu University Studies in Arts and Culture 50 * A clause-relational approach to English texts: A study of some predictive lexical items in written discourse EWinter 1977 6