Taiwan International ESP Journal, Vol.

1: 1, 5-22, 2009

Writing in the disciplines: Research evidence for specificity

Ken Hyland Centre for Applied English Studies University of Hong Kong

Abstract
Academic writing, much like any other kind of writing, is only effective when writers use conventions that other members of their community find familiar and convincing. Essentially the process of writing involves creating a text that we assume the reader will recognise and expect, and the process of reading involves drawing on assumptions about what the writer is trying to do. It is this writer-reader coordination which enables the co-construction of coherence from a text. Scholars and students alike must therefore attempt to use conventions that other members of their discipline, whether journal editors and reviewers or subject specialist teachers and examiners, will recognise and accept. Because of this discourse analysis has become a central tool for identifying the specific language features of target groups. In this paper I draw on my own work, conducted over several years into research and student genres, to show how some familiar conventions of academic writing are used in different disciplines and what these differences can tell us about the work in the disciplines themselves. Keywords: academic writing, conventions, discourse analysis, genres

1. Introduction Specificity is perhaps the most central concept in language teaching and discourse analysis today and represents a key way in which we understand and practice English for Academic and Specific Purposes. Our understanding of specificity, and how language varies in different contexts, has been greatly assisted in the last twenty years

Ken Hyland

by research in discourse analysis, which has become an invaluable tool for scholars and teachers, highlighting typical patterning and salient features of academic writing. Where once intuition and impression guided teaching, we now have evidence for language variation across disciplines, genres, modes, and languages which is increasingly informing classroom materials and practice. In this paper I want to explore something of the contribution that discourse analysis research has made to the study of academic texts, drawing largely on my own research, to examine the idea of specificity. In particular I will focus on the importance of disciplinary specific language use and give some examples of this.

2. Discourse analysis and academic specificity Discourse analytic studies show considerable variation in academic language use across a range of dimensions. Halliday (1989), for example, found greater nominalization, impersonalisation and lexical density in written compared with spoken texts. There is also a high degree of specificity in the kinds of writing that students are asked to do, so that even students in fairly similar fields, such as nursing and midwifery, are given very different writing assignments (Gimenez, 2009). Research on vocabulary shows that terminology varies enormously across disciplines and even that the same words have different frequencies, collocations and often different meanings in different fields (Hyland & Tse, 2007). Similarly, research in contrastive rhetoric (e.g. Hinkel, 2002) has pointed to cultural specificity in rhetorical preferences, so that students first language and prior learning is seen to influence ways of organising ideas and structuring arguments when writing in English at university. Perhaps most research into specificity has attended to genre, where particular purposes and audiences lead writers to employ very different choices (e.g. Hyland, 2009). Table 1, for example, compares frequencies for different features in a corpus of 240 research articles and 56 textbooks.

TIESPJ, Vol. 1: 1, 2009

Table 1 Selected features in research articles and textbooks per 1000 words (Hyland, 2005)
Genre Research Articles University Textbooks Hedges 15.1 8.1 Self-mention 3.9 1.6 Citation 6.9 1.7 Transitions 12.8 24.9

We can see considerable variation in these features across the two genres. The greater use of hedging underlines the need for caution and opening up arguments in the research papers compared with the authorized certainties of the textbook, while the removal of citation in textbooks shows how statements are presented as facts rather than claims grounded in the literature. The greater use of self-mention in articles points to the personal stake that writers invest in their arguments and their desire to gain credit for claims, while the higher frequency of transitions, which are conjunctions and other linking signals, in the textbooks is a result of the fact that writers need to make connections far more explicit for readers with less topic knowledge. Overwhelmingly, however, it is disciplinary variation which underlies most specificity and this is what I want to focus on here. Research into differences in academic practices and the texts that these produce is relatively new, partly because the notion of discipline, and its underlying reliance on the idea of community, has been difficult to pin down, and partly because of our fixation with genre in recent years. While genre has provided a significant way of understanding situated language use, its power to harness generalisations has led us to over-emphasize resemblances between texts at the expense of variation. But, as Swales made clear in 1990, we need to see community and genre together to offer a framework of how meanings are socially constructed by forces outside the individual. Research on language variation across the disciplines is rapidly becoming one of the dominant paradigms in EAP (e.g. Hyland, 2004; Flttum et al., 2006; Hyland & Bondi, 2006). Specificity here refers to what I hope is a fairly uncontroversial idea: that we communicate as members of social groups and that different groups use language to conduct their business, define their boundaries, and manage their interactions in particular

Ken Hyland

ways. For EAP teachers this means focusing on communicating, and learning to communicate, as a disciplinary insider.

3. Disciplinary specificity The idea of discipline has become important in EAP as we have become more sensitive to the ways genres are written and responded to by individuals acting as members of social groups. Essentially, we can see disciplines as language using communities and the term helps us join writers, texts and readers together. Communities provide the context within which we learn to communicate and to interpret each others talk, gradually acquiring the specialized discourse competencies to participate as group members. So we can see disciplines as particular ways of doing things-particularly of using language to engage with others in certain recognised and familiar ways. Academic texts are about persuasion and this involves making choices to argue in ways which fit the communitys assumptions, methods, and knowledge. This is how Wells (1992: 290) sees matters: Each subject discipline constitutes a way of making sense of human experience that has evolved over generations and each is dependent on its own particular practices: its instrumental procedures, its criteria for judging relevance and validity, and its conventions of acceptable forms of argument. In a word each has developed its own modes of discourse. To work in a discipline, then, we need to be able to engage in these practices and, in particular, in its discourses. So disciplines structure the work we do within wider frameworks of beliefs and provide the conventions and expectations that make texts meaningful. We can see this if we picture the disciplines as spread along a cline (See Figure 1), with the hard sciences at one end and the softer humanities at the other.

TIESPJ, Vol. 1: 1, 2009

SCIENCES Empirical and objective Linear and cumulative growth of knowledge Experimental methods Quantitative methods More concentrated readership Highly structured genres

SOCIAL SCIENCES

HUMANITIES Explicitly interpretive Dispersed knowledge Discursive argument Qualitative methods More varied readership More fluid discourses

Figure 1 Continuum of academic knowledge (after Coffin et al., 2003) In the sciences new knowledge is accepted by experimental proof. Science writing reinforces this by highlighting a gap in knowledge, presenting a hypothesis related to this gap, and then reporting experimental findings to support this in a standard Introduction-Methods-Results-Discussion format. The humanities such as literature, history and philosophy, on the other hand, largely rely on case studies and narratives while claims are accepted on strength of argument. The social sciences fall between these extremes. Disciplines such as Sociology, Economics and Applied Linguistics have partly adopted methods of the sciences, but in applying these to human data they have to give far more attention to explicit interpretation than those fields. In other words, academic discourse helps to give identity to a discipline. This means that we need to understand the distinctive ways they have of asking questions, addressing a literature, criticizing ideas, and presenting arguments, so we can help students participate effectively in their learning.

4. Some example differences I want to turn now to these disciplinary differences and look at a series of studies I have conducted over the past decade or so into the features of a 1.5 million word corpus of research articles in 8 disciplines and 4 million words of student dissertations together with interviews with 30 academics. I will briefly highlight a few of the

10

Ken Hyland

disciplinary differences in these corpora, focusing on citation, reporting verbs, hedges, self-mention, directives, and lexical bundles.

4.1 Citation practices One of the most striking differences in disciplinary uses of language is in citation practices. The inclusion of references to the work of other authors is obviously central to academic persuasion. This is because it not only helps establish a persuasive framework for the acceptance of arguments by showing how a text depends on previous work in a discipline, but also as it displays the writers credibility and status as an insider. It helps align him or her with a particular community or orientation and confirms that this is someone who is aware of, and is knowledgeable about, the topics, approaches, and issues which currently interest and inform the field. But because discourse communities see the world in different ways they also write about it in different ways, with Table 2 showing that two thirds of all the citations in the article corpus in the philosophy, sociology, marketing and applied linguistics papers, twice as many as in the science disciplines (Hyland, 1999). Table 2 Rank order of citations by discipline per 1,000 words
Soft Disciplines Sociology Philosophy Applied Linguistics Marketing per 1000 words 12.5 10.8 10.8 10.1 Hard Disciplines Biology Electrical Engineering Mechanical Engineering Physics per 1000 words 15.5 8.4 7.3 7.4

Basically, the differences reflect the extent writers can assume a shared context with readers. In Kuhns (1962) normal science model, natural scientists produce public knowledge through cumulative growth. Problems tend to emerge on the back of earlier problems as results throw up further questions to be followed up with further research so writers do not need to report research with extensive referencing. The

TIESPJ, Vol. 1: 1, 2009

11

people who read those papers are often working on the same problems and are familiar with the earlier work. They have a good idea about the procedures used, whether they have been properly applied, and what results mean. In the humanities and social sciences, on the other hand, the literature is more dispersed and the readership more heterogeneous, so writers cannot presuppose a shared context but have to build one far more through citation. This rather neat explanation drawing on a hard-soft discipline dichotomy is spoilt somewhat by the fact that biology has the highest citation count per 1000 words. Interestingly, this is largely due to a very high proportion of self-citation with 13% of all citations to the current author compared with about 6% overall among other disciplines in the corpus. There does, in fact, seem to be a considerable emphasis given to recognising the ownership of ideas in biology and showing how current research builds on the work of others, which makes it unusual among the sciences (Hyland, 2004).

4.2 Reporting verbs There are also major differences in the ways writers report others work, with results suggesting that writers in different fields draw on very different sets of reporting verbs to refer to their literature (Hyland, 1999). Among the higher frequency verbs, almost all instances of say and 80% of think occurred in philosophy and 70% of use in electronics. It turns out, in fact, that engineers show, philosophers argue, biologists find and linguists suggest. The most common forms across the disciplines are shown in Table 3. Table 3 Most frequent reporting verbs.
Soft Disciplines Philosophy Sociology Applied Ling. Marketing say, suggest, argue, claim argue, suggest, describe, discuss suggest, argue, show, explain suggest, argue, demonstrate, propose Biology Elec Eng. Mech Eng. Physics Hard Disciplines describe, find, report, show show, propose, report, describe show, report, describe, discuss develop, report, study

12

Ken Hyland

These preferences seem to reflect broad disciplinary purposes. So, the soft fields largely use verbs which refer to writing activities, like discuss, hypothesize, suggest, argue. These involve the expression of arguments and allow writers to discursively explore issues while carrying a more evaluative element in reporting others work: (1) Lindesmiths (1965) classic work indicated the Davidson defends this claim on the grounds that Engineers and scientists, in contrast, prefer verbs which point to the research itself like observe, discover, show, analyse, and calculate, which represent real world actions. (2) Edson et al (1993) showed processes were induced . ... using (4) special process and design, or by adding (5), or removing (6) a mask. This emphasis on real-world activities helps scientists represent knowledge as proceeding from impersonal lab activities rather than from the interpretations of researchers. Two scientist informants commented on this kind of use: Of course, I make decisions about the findings I have, but it is more convincing to tie them closely to the results. (Physics interview)

You have to relate what you say to your colleagues and we dont encourage people to go out and nail their colours to the mast as maybe they dont get it published. (Biology interview)

The conventions of impersonality in science articles thus play an important role in reinforcing an objective ideology by portraying the legitimacy of hard science knowledge as built on socially invariant criteria. Again, it removes the author from the text to give priority to the unmediated voice of nature itself.

TIESPJ, Vol. 1: 1, 2009

13

4.3 Hedges Devices like possible, might, likely, and so on, collectively known as hedges, also diverge across fields. These function to withhold complete commitment to a proposition, implying that a claim is based on plausible reasoning rather than certain knowledge. They indicate the degree of confidence the writer thinks it might be wise to give a claim while opening a discursive space for readers to dispute interpretations (Hyland, 1996). Because they represent the writers direct involvement in a text, something that scientists generally try to avoid, they are twice as common in humanities and social science papers than in hard sciences. So, we tend to find more statements like this: (3) The existence of such networks did not go unnoticed by contemporaries (see, Rocke, 1989), and it seems sensible to assume the men concerned were probably not unreflective about this patterned conduct either. questionnaire bilingually. (Sociology) With hindsight, we believe it might have been better to have presented the (Applied Linguistics)

One reason for this is there is less control of variables, more diversity of research outcomes, and fewer clear bases for accepting claims than in the sciences. Writers cannot report research with the same confidence of shared assumptions so papers rely far more on recognizing alternative voices. Arguments have to be expressed more cautiously by using more hedges. In the hard sciences positivist epistemologies mean that the authority of the individual is subordinated to the authority of the text and facts are meant to speak for themselves. This means that writers often disguise their interpretative activities behind linguistic objectivity. They downplay their personal role to suggest that results would be the same whoever conducted the research. The less frequent use of hedges is one way of minimising the researchers role, and so is the preference for modals over cognitive verbs. This is because modal verbs can more easily combine with inanimate subjects to downplay the person making the evaluation. So we are more likely to find examples like (4) in the sciences and those with cognitive verbs in the soft discipline

14

Ken Hyland

fields (5): (4) For V. trifidum, ANOVA showed a significant increase from L to L and FI, which could be interpreted as reflecting the dynamics of fungal colonization. noise measurements. (5) I think this would be a mistake. the result. ( Biology) (Electrical Engineering) (Sociology) (Marketing) The deviations at high frequencies may have been caused by the

We suspect that the product used in this study may have contributed to

Scientists tend to be concerned with generalisations rather than individuals, so greater weight is put on the methods, procedures and equipment used rather than the argument. Modals, then, are one way of helping to reinforce a view of science as an impersonal, inductive enterprise while allowing scientists to see themselves as discovering truth rather than constructing it.

4.4 Self- mention Self-mention is another important feature which varies across disciplines. This concerns how far writers want to intrude into their texts through use of I or we, or to use impersonal forms. Presenting a discoursal self is central to the writing process, and we cannot avoid projecting an impression of ourselves and how we stand in relation to our arguments, discipline, and readers. To some extent we have to see this as a personal preference determined by seniority, experience, confidence, personality, and so on, but the presence or absence of explicit author reference is a conscious choice by writers to adopt a particular community-situated authorial identity. However, my 240 research articles, once again, show broad disciplinary preferences with 2/3 of cases in the social sciences & humanities papers (Hyland, 2001b). Table 4 presents the distribution of the use of self-mention across the eight disciplines in my research article corpus.

TIESPJ, Vol. 1: 1, 2009

15

Table 4 Self-mention in research articles per 1,000 words

Soft disciplines Philosophy Sociology Applied Linguistics Marketing Average 5.5 4.3 4.5 5.5 5.0 Physics Biology Mechanical Engineering Electrical Engineering Hard disciplines 4.1 3.4 1.0 3.3 2.9

Now it is clear that writers in different disciplines represent themselves, their work and their readers in different ways, with those in the humanities and social sciences taking far more personal positions than those in the sciences and engineering. The reason for this is again that the strategic use of self-mention allows writers to claim authority by expressing their convictions, emphasizing their contribution to the field, and seeking recognition for their work (Hyland, 2001b; Kuo, 1999). It sends a clear indication to the reader of the perspective from which statements should be interpreted and distinguishes the writers own work from that of others. Successful communication in the soft fields depends far more on the authors ability to invoke the sense of a real writer in the text, emphasizing their own contribution to the field while seeking agreement for it. (6) I argue that their treatment is superficial because, despite appearances, it relies solely on a sociological, as opposed to an ethical, orientation to develop a response. (Sociology) I bring to bear on the problem my own experience. This experience contains ideas derived from reading I have done which might be relevant to my puzzlement as well as my personal contacts with teaching contexts. (Applied Linguistics) So self-mention can help construct an intelligent, credible, and engaging colleague by presenting a confident and authoritative authorial self. In the hard sciences, as I noted earlier, researchers are generally seeking to

16

Ken Hyland

downplay their personal role in the research to highlight the phenomena under study, the replicability of research activities, and the generality of the findings. Scientists, then, try to distance themselves from interpretations in ways that are familiar to most EAP teachers. They accomplish this by either using the passive voice (7), dummy it subjects (8), or by attributing agency to inanimate things (9): (7) This suggestion was confirmed by the observation that only plants carrying the pAG-I::GUS transgene showed a gain of GUS staining in leaves of clf-2 plants. shear strain when subjected to thermal fatigue... (Mechanical Engineering) (9) The images demonstrate that the null point is once again well resolved and that diffusion is symmetric. (Physics) (Biology) (8) It was found that a larger stand-off height would give a smaller maximum

By subordinating their voice to that of nature, scientists rely on the persuasive force of lab procedures rather than the force of their writing. As this biologist told me: I feel a paper is stronger if we are allowed to see what was done without we did this and we think that. Of course we know there are researchers there, making interpretations and so on, but this is just assumed. Its part of the background. Im looking for something interesting in the study and it shouldnt really matter who did what in any case. (...) In theory anyone should be able to follow the same procedures and get the same results. (Biology interview) In contrast, in the humanities and social sciences, the first person allows writers to strongly identify with a particular argument and to gain credit for an individual perspective: Using I emphasizes what you have done. What is yours in any piece of

TIESPJ, Vol. 1: 1, 2009

17

research. I notice it in papers and use it a lot myself. (Sociology interview) The personal pronoun I is very important in philosophy. It not only tells people that it is your own unique point of view, but that you believe what you are saying. It shows your colleagues where you stand in relation to the issues and in relation to where they stand on them. It marks out the differences. (Philosophy interview) By marking your views with the first person, you leave readers in no doubt of your stance while claiming credit for what you are saying. It is a powerful way of demonstrating an individual contribution and establishing a claim for priority.

4.5 Directives Another feature which supports the idea of disciplinary specificity is directives. These are devices which instruct the reader to perform an action or to see things in a way determined by the writer (Hyland, 2002). They are largely expressed through imperatives (e.g. consider, note, imagine) and obligation modals (such as must, should, and ought). Overall, they direct readers to three main kinds of activity: textual, physical and cognitive acts. Textual acts direct readers to another part of the text or to another text (e.g. see Smith 1999, refer to table 2) Physical acts direct readers how to carry out some action in the real-world (e.g. open the valve, heat the mixture). Cognitive acts instruct readers how to interpret an argument, explicitly positioning readers by encouraging them to note, concede or consider some argument or claim in the text. Generally, explicit engagement, where writers address readers directly in a text (Hyland, 2001a) is a feature of the soft disciplines, where writers are less able to rely on the explanatory value of accepted procedures. Directives, however, are a potentially

18

Ken Hyland

risky way of seeking to build a connection with readers as they instruct them to act or see things in a way determined by the writer (Hyland, 2002). They may therefore be seen as assuming unwarranted authority and so encourage a hostile response and a rejection of the claim being made. As a result, most directives in the soft fields are textual, directing readers to a reference or table rather than telling them how they should interpret an argument. So examples like these are common in the social sciences: (10) see Steuer 1983 for a discussion of other contingencies effects. (Marketing) Look at Table 2 again for examples of behavioristic variables. For transcription conventions please refer to the Appendix. (Applied Linguistics) Two of my social science respondents noted this about their writing in their interviews: I am very conscious of using words like must and consider and so on and use them for a purpose. I want to say Right, stop here. This is important and I want you to take notice of it. So I suppose I am trying to take control of the reader and getting them to see things my way. (Sociologist interview) (Marketing)

I am aware of the effect that an imperative can have so I tend to use the more gentle ones. I dont want to bang them over the head with an argument I want them to reflect on what Im saying. I use consider and lets look at this rather than something stronger. (Applied Linguist interview)

Argument in the hard knowledge fields, in contrast, is formulated in a highly standardised code. Succinctness is valued by both editors and scientists, and directives allow writers to cut directly to the heart of key issues in the text. Because of this we find a high proportion of cognitive directives here which explicitly position readers by leading them through an argument or emphasising what they should attend to:

TIESPJ, Vol. 1: 1, 2009

19

(11) What has to be recognized is that these issues.. (Mechanical Eng) Consider the case where a very versatile milling machine of type M5... (Electrical Eng) A distinction must be made between cytogenetic and molecular resolution. (Biology) My informants noted this in their interviews: I rarely give a lot of attention to the dressing, I look for the meat - the findings - and if the argument is sound. If someone wants to save me time in getting there then that is fine. No, Im not worried about imperatives leading me through it. (Electrical Engineering interview)

Im very conscious of how I write and I am happy to use an imperative if it puts my idea over clearly. Often we are trying to work to word limits anyway, squeezing fairly complex arguments into a tight space. (Mechanical Engineering interview)

4.6 Bundles The final example of disciplinary specificity I want to mention is lexical bundles, or frequently occurring word sequences. These are a key way of shaping text meanings and contributing to our sense of distinctiveness and naturalness in a register. So collocations like as a result of and it should be noted that, help identify a text as belonging to an academic register while in pursuance of, and in accordance with mark out a legal text. Using a corpus of 120 research articles and 120 post-graduate dissertations in four disciplines I found that the most common bundles in this academic corpus of 3.5 million words were on the other hand, at the same time and in the case of, all of which occurred over 100 times per million words (Hyland, 2008). There are, however, some interesting disciplinary differences. The electrical

20

Ken Hyland

engineering texts contained the greatest range of high frequency bundles and also the highest proportion of words in 4-word bundles. Biology, on the other hand, had the smallest range of bundles, the fewest examples, and the lowest proportion of texts comprised of words in bundles. So the electrical engineering texts were most dependent on prefabricated bundles and used many sequences not found in the other disciplines, perhaps because of the fact that technical communication is relatively abstract and graphical. This means that language constructs an argument by linking data or findings in routinely patterned, formulaic ways with the same forms used repeatedly. There are also considerable differences across disciplines in the 4-word bundles. The top 20 most common ones are shown in their rank order within disciplines in Table 5 with items that occur in all four disciplines marked in bold and those in three disciplines are shaded (Hyland, 2008). Table 5 Most frequent 4-word bundles (bold = in 4 disciplines; shaded = in 3 disciplines)
Biology in the presence of in the present study on the other hand the end of the is one of the at the end of it was found that at the beginning of as well as the as a result of it is possible that are shown in figure was found to be be due to the in the case of is shown in figure the beginning of the the nature of the the fact that the may be due to Electrical Eng on the other hand as shown in figure in the case of is shown in figure it can be seen as shown in fig is shown in fig can be seen that can be used to the performance of the as a function of is based on the with respect to the is given by equation the effect of the the magnitude of the at the same time in this case the it is found that the size of the Applied Ling on the other hand at the same time in terms of the on the basis of in relation to the in the case of in the present study the end of the the nature of the in the form of as well as the at the end of the fact that the in the context of is one of the in the process of the results of the in terms of their to the fact that in the sense that Business Studies on the other hand in the case of at the same time at the end of on the basis of as well as the the extent to which the end of the significantly different from zero are more likely to the relationship between the the results of the the hang seng index the other hand the in the context of as a result of the performance of the hong kong stock market is positively related to are significantly different from

TIESPJ, Vol. 1: 1, 2009

21

The table clearly shows the extent of disciplinary specificity with just two forms in all four disciplines (on the other hand and in the case of) and a handful in three fields. In fact, over half of all items in the top 50 bundles in each discipline do not occur in the top 50 of any other discipline. Perhaps unsurprisingly, the greatest similarities are between cognate fields, linking the soft disciplines together and the hard disciplines together. Business studies and applied linguistics share 18 items in the top 50 with on the basis of, in the context of, the relationship between the, and it is important to exclusive to these two fields. Similarly, biology and electrical engineering have 16 bundles in common, with it was found that, is shown in figure, as shown in figure, is due to the, and the presence of the not found in the social science list at all.

5. Conclusion Discourse studies reveal that the features I have presented here all occur and behave in dissimilar ways in different disciplines. The fact that writers in different fields draw on different resources to develop their arguments, establish their credibility and persuade their readers means that EAP teachers need to take the disciplines of their students, and the ways these disciplines create texts, into account in their classroom practices. Such considerations, moreover, are not confined to student writing but underlie the conventions which guide scholarly writing for academic publication. The value of discourse analysis is not that it merely produces a list of the features of disciplinary discourses, but that it can uncover more sophisticated understanding of disciplinary communities. It provides a richer picture for academics and for teachers of EAP and so helps us to improve the ways we prepare our students for their academic studies.

22

Ken Hyland

References
Biber, D. (2006). University language: A corpus-based study of spoken and written registers. Amsterdam: John Benjamins. Coffin, C., Curry, M., Goodman, S., Hewings, A., Lillis, T., & Swann, J. (2003). Teaching academic writing: A toolkit for higher education. London: Routledge. Flttum, K., Dahl, T. and Kinn, T. (eds). (2006) Academic voices - Across languages and disciplines. Amsterdam: John Benjamins. Gimenez, J. (2009). Beyond the academic essay: Discipline-specific writing in nursing and midwifery. Journal of English for Academic Purposes, 7 (3), 151-164. Halliday, M. A. K. (1989). Spoken and written language. Oxford: OUP. Hinkel, E. ( 2002). Second language writers texts. Mahwah, NJ: Erlbaum. Hyland, K. (1996). Writing without conviction? Hedging in scientific research articles. Applied Linguistics, 17 (4), 433-454. Hyland, K. (1999). Academic attribution: Citation and the construction of disciplinary knowledge. Applied Linguistics, 20 (3), 341-267. Hyland, K. (2001a). Bringing in the reader: Addressee features in academic articles. Written Communication, 18 (4), 549-574. Hyland, K. (2001b). Humble servants of the discipline? Self-mention in research articles. English for Specific Purposes, 20(3), 207-226. Hyland, K. (2002). Directives: Power and engagement in academic writing. Applied Linguistics. 23, 215-239. Hyland, K. (2004). Disciplinary discourses. Ann Arbor, MI: University of Michigan Press. Hyland, K. (2005). Metadiscourse. London: Continuum. Hyland, K. (2008). As can be seen: Lexical bundles and disciplinary variation. English for Specific Purposes, 27 (1), 4-21. Hyland, K. (2009). Academic discourse. London: Continuum. Hyland, K. & Bondi, M. (Eds.) (2006). Academic discourse across disciplines. Frankfort: Peter Lang. Hyland, K. & Tse, P. (2007). Is there an academic vocabulary? TESOL Quarterly, 41 (2), 235-254. Kuhn, T. (1962). The Structure of Scientific Revolutions. Chicago: Chicago University Press. Kuo, C-H. (1999). The use of personal pronouns: Role relationships in scientific journal articles. English for Specific Purposes, 18 (2), 121-138. Swales, J. (2004). Research genres. Cambridge: Cambridge University Press. Wells, G. (1992). The centrality of talk in education. In K. Norman (ed.), Thinking voices: The work of the national oracy project. London: Hodder and Stoughton.