A systemic functional analysis of multisemiotic biology texts 1

To frame this research, this chapter discusses the nature and use of textbooks in science education Section 1.1 and reviews some selected approaches to the description of English used in

CHAPTER ONE INTRODUCTION

The present study investigates the nature of the linguistic and visual semiotic resources deployed in selected tertiary biology textbooks written in English It also aims to explore how such multimodal analyses may contribute to teaching academic English to non-native university or college students of science and engineering, in particular the development of academic reading skills in English To frame this research, this chapter discusses the nature and use of textbooks in science education (Section 1.1) and reviews some selected approaches to the description of English used in scientific and technological settings (Section 1.2) The chapter concludes by outlining the scope and organization of the dissertation (Section 1.3)

1.1 Nature and Use of Textbooks in Science Education

As one of the means of imparting knowledge from one generation to another, textbooks play a crucially important role in the socialization and educational processes

of almost all members of modern society They are so important that many governments of today invest heavily in the commission, publication, distribution and evaluation of school textbooks, especially those for primary and secondary school pupils

In the strict sense of the term, “a textbook is a book that presents a body of knowledge in an organized and usually simplified manner for purposes of learning The textbook is frequently the most important teaching tool because it can determine

not only what will be taught but also how it will be taught” (Langenbach 1999: 563) Although television, videotapes, films and, in particular, computers are rivaling printed materials of communication, “textbooks remain major resources in schools and colleges” (1999: 563)

Although science education, especially in the industrialized countries, draws heavily on the readily available and highly diversified laboratory facilities where students are encouraged to perform hands-on inquiry-oriented activities in and outside the science class, “recent studies … estimate that as much as 75% of classroom instruction and 90% of the homework time is structured around text materials In the science classroom, the textbook dominates the curriculum” (Spiegel and Barufaldi 1994: 913) “Some researchers reported that the status of school science could be summarized with one word: textbooks” (Chiang-Soong and Yager 1993: 340)1

Specifically, in the first place, a science textbook introduces the students to the world of science It teaches them how to observe and analyze what happens around them and thus fosters an interest in and a proper attitude towards science This is particularly true if the textbooks are printed and used in series, for example, secondary school science series and junior college or senior high school chemistry, biology and physics series As Johns (1997: 46) observes, “[i]n many classrooms the textbook is the chief reading source, the single window into the values and practices of a discipline This is particularly the case in science and technology”

Secondly, a science textbook familiarizes students with “specialized vocabulary, the complicated sentence structure, and the difficult concepts that characterize higher level science writing” (DiGisi and Willet 1995: 123) It also exposes students to non-linguistic materials, for example, pictures, diagrams, and mathematical symbolism According to Molitor, Ballstaedt and Mandl (1989: 5-6),

ever since Johann Amos Comenius’s first picture textbook Orbis sensualium pictus in

1658, “graphicness has remained a central concept in pedagogy and didactics Illustrations have become an indispensable component in teaching material and other expository texts” By reading the textbooks, the students thus learn how to decipher the message contained in the non-linguistic material as well as in the linguistic text

Finally, a science textbook, if properly used, can provide students with instruction and practice in understanding science expository writing It provides examples or models of expository writing by which the students may improve their writing skills

While science textbooks are written and taught in a number of languages, in this study I shall be concerned with introductory tertiary science textbooks written in the English language Such a focus on English textbooks rather than on textbooks in other languages is primarily motivated by the growing demands from the increasing number of non-native speakers of English around the world to read English scientific texts In the next section I briefly describe the importance of the English language in international scientific and technological undertakings and survey some approaches to the description of written scientific English

1.2 Approaches to the Description of Written Scientific English: A Review

Since the end of the Second World War, the English language has been playing an increasingly important role in international scientific and technological communication The number and importance of English-medium journals, books and research papers continues to increase Baldauf (1986: 221), in a study of 338 articles published between 1978 and 1982 in four leading journals devoted to cross-cultural

psychology2, has found that “[t]hree hundred and twenty seven, or 97%, of the studies were in English, that “all but one of these articles had an English language abstract (the exception contained no abstract)”, and that out of the 8,489 citations provided for these

338 articles, 97% were in English It is not just in cross-cultural psychology that English reigns, but also in other disciplines (though in various degrees) Graddol (1997: 9) notes that “English is now the international currency of science and technology … the global language of experiment and discovery” Citing Graddol (1997), Flowerdew and Peacock (2001: 10) further observe that “[t]he international language of research and academic publication is English and anyone who wishes to have ready access to this material needs to know the language” With the advent of the computer-mediated communication in the 21st century, the English language is well on its way towards a global lingua franca

Given the increasing importance of the English language to the international scientific and technological undertaking and the role of science and technology in modern societies, it comes as no surprise that research into the linguistic characteristics

of English for Science and Technology (hereafter EST) has been a particularly growing area of study Linguists, language teachers, and science educators, under the influence of various schools of thought, have been attempting to discover the features

of English used in scientific and technological settings In the remainder of this section I review some of the approaches to the description of written scientific English3

1.2.1 Register Analysis

Much of the early effort at the description of EST was influenced by the notion of register As Halliday, McIntosh and Strevens (1964: 87) note, “[l]anguage varies as its function varies; it differs in different situations The name given to a variety of a language distinguished according to use is ‘register’” (see Section 2.3 for a fuller description of the register theory) At the same time, a register displays distinctive lexicogrammatical features That is to say, a register is defined both semantically by reference to the context of situation and formally by reference to the lexicogrammatical characteristics Scientific English is a functional variety, which serves to construct and transmit to and for a particular audience through a particular channel the knowledge accumulated in science and technology Linguists in the 1960s aimed to document the lexicogrammatical features of scientific English

One of the earliest large-scale systematic studies of the grammatical features of scientific English was carried out at University College London in 1964-1967 under the direction of M A K Halliday (see Huddleston et al (1968) for the final report of the research project; see also Huddleston (1971) for “a substantially revised version” (p vii) of his contribution to the 1968 report) This research was heavily influenced by the then emerging systemic framework (for a brief account of Halliday’s systemic-functional linguistics (hereafter SFL), see Chapter Two) By reference to the register variables of the field of discourse, the style of discourse (later to be known as “the tenor of discourse” in Halliday (in Halliday and Hasan 1989 [1985]: 12)) and the mode

of discourse, the researchers selected 5,000 words from each of the 27 texts belonging

to specialist journals, undergraduate textbooks and popular science works in biology, chemistry and physics The grammatical items in the systems of TRANSITIVITY,

ATTRIBUTION, MOOD and so forth were studied and how frequently they occur determined From this information the quantitative profiles were provided

Register analysis, or rather its frequency analysis component, has been useful

in indicating what lexical items or syntactic structures are more frequent in a certain variety of English The frequency analysis of the grammatical features has developed

in SFL into probability study of terms in a system (Halliday 1991; 1992; Halliday and James 1993; Matthiessen 1999; and Nesbitt and Plum 1988) Halliday (1991: 42) notes that “[f]requency in text is the instantiation of probability in the system A linguistic system is inherently probabilistic in nature” More recently, Halliday (2003: 23-24) elaborates,

…these quantitative features are not empty curiosities They are an

inherent part of the meaning potential of a language An important

aspect of the meaning of negative is that it is significantly less likely

than positive; it takes up considerably more grammatical energy, so to

speak The frequencies that we observe in a large corpus represent the

systemic probabilities of the language; and the full representation of a

system network ought to include the probability attached to each option

in each of the principal systems (…) We have not yet got the evidence

to do this; but until it can be done, grammars will not have come of age

As far as English language, and especially EST, teaching and learning are concerned, the results of frequency analysis can be used to help determine what items should be given more priority in a certain teaching syllabus As computers are increasingly used for linguistic analysis, frequency analysis will continue to have its place in the research community

In retrospect, the early studies of EST had certain limitations First, the theoretical frameworks underlying the analyses were at the embryonic stages Systemic grammar, for instance, had just begun to take shape and “[s]erious work on

registers is even more recent in origin”, wrote Halliday, McIntosh and Strevens (1964: 98) Second, linked to the first limitation, Huddleston et al (1968) did not attempt to account for the statistics resulting from frequency analysis; they did not relate it to the context of situation and context of culture in which the lexicogrammatical items occur Thirdly, practically no attention was paid to the non-linguistic materials in the corpus, the exception being the discussion of mathematical symbolisms used in the texts analyzed (Huddleston et al 1968: 682-685, Appendix C) In addition, the researchers were preoccupied with the description of the formal features of EST, rarely taking into consideration how the results they obtained could be turned into practical and useful input into language teaching or other fields of study

1.2.2 Rhetorical-Communicative Approach

In this approach to EST, the focus of attention shifted from the grammatical and lexical properties to the communicative properties of the language In the words of Widdowson (1979: 57), this approach “tells us what the [lexicogrammatical] forms count as communication, how they express elements of discourse”

As the heading “rhetorical-communicative” coined by Swales (1985: 72) suggests, there are two sub-categories in this broad approach: the communicative, upheld by Allen and Widdowson (1974, in Swales 1985: 73-87) (among others), and the rhetorical, practiced by the University of Washington (Seattle) EST Program comprising Selinker, Trimble and others (e.g Trimble 1985)

Allen and Widdowson (1974, in Swales 1985: 75) claimed that the frequency analysis of the formal features of EST is equivalent to treating “scientific discourse merely as exemplification of the language system” and that “[it] does little or nothing

to indicate what kind of communication it is” They suggest, therefore, that an English course at tertiary level should aim at developing two kinds of ability in the students:

The first is the ability to recognize how sentences are used in the

performance of acts of communication, the ability to understand the

rhetorical functioning of language in use The second is the ability to

recognize and manipulate the formal devices which are used to combine

sentences to create continuous passages of prose (in Swales 1985: 74)

The conceptualization of the “Washington” approach to EST arose from the members’ first-hand experience of teaching EST to non-native undergraduate engineering students at the University of Washington (Seattle) starting in 1967 Their teaching and research spanning 18 years is summarized in Trimble (1985) The

“Washington” school divides the total discourse (for instance, a research article) into four interrelated rhetorical levels By “rhetoric” and the derivative “rhetorical”, Trimble (1985: 10) refers to the EST writer’s process of “choosing and organizing information for a specific set of purposes and a specific set of readers” Level A is

“[t]he objectives of the total discourse”, for example, “Detailing an experiment”,

“Making a recommendation” Level B is “[t]he general rhetorical functions that develop the objectives of Level A”, for example, “Stating purpose”, “Reporting past research” Level C is “[t]he specific rhetorical functions that develop the general rhetorical functions of Level B”, for example, “Description”, “Definition”,

“Classification” And Level D is “[t]he rhetorical techniques that provide relationships within and between the rhetorical units of Level C”, for example, “Orders”, “Patterns” (Trimble 1985: 11) As far as teaching reading skills is concerned, Trimble (1985: 13) has found that Levels C and D deserve special attention Trimble (1985) also discusses the rhetorical-grammatical relationships and lexical problems in EST discourse Of

particular interest, Trimble (1985: 102-113) also draws attention to the visual-verbal relationships in EST

Register analysis and rhetorical-communicative approach may be taken as congruent with and complementary to each other Both recognize the importance of the language system: Widdowson’s “usage”, “text” and the Washington school’s attention to lexis and grammar in EST on the one hand, and Halliday’s lexicogrammar

on the other, point to a similar (if not the same) direction Both attach importance to the role social context plays in language activities: Widdowson’s communicative value, “use”, discourse, and communicative schemata, the Washington school’s rhetoric, on the one hand, and Halliday’s context of situation and context of culture on the other

At the same time, the two approaches differ in their theoretical orientations First, register analysis is based on a theory of language and context, i.e the systemic theory, which stresses “the complete interconnectedness of the linguistic and the social” (Christie and Unsworth 2000: 3), whereas the rhetorical-communicative approach does not commit itself to a particular view of the lexicogrammar The lexis and the syntactic structures that the rhetorical-communicative approach attempts to attach some communicative value to are not themselves shown to relate to each other and form a system For instance, in a study of the rhetorical functions of the passive as

opposed to we plus an active verb in two astrophysics journal papers, Tarone, Dwyer, Gillette and Icke (1998: 113) propose that “we indicates the author’s procedural

choice, while the passive indicates an established or standard procedure”, but they failed to recognize that the rhetorical “force” of a particular language form derives from a selection from the system, in this case the system of THEME (Halliday 1994) Secondly, register analysis in the 1960s placed the investigation of the social context

on the research agenda, which was later included in works such as Halliday (1978) The rhetorical-communicative approach, on the other hand, has not tried to map the

“use” and its relation with “usage” A summary of the overlap and complementariness between the two major approaches to EST is given in Table 1.1

Language as meaning

potential

Yes (system network) No

Relationship between text

and context

Yes (Realization) No

Table 1.1 A comparison of two approaches to EST

Key: “Yes” means that a particular approach adheres to the basic notion shown in the area of comparison “No” means that it does not What is included in the bracket is the area or areas in the approach that elaborates the basic notion

From the Table we can appreciate a need for the two approaches to complement each other The systemic approach has a potential of investigating EST in considerable detail and with considerable insight, as in Halliday and Martin (1993) and Martin and Veel (1998) But at a particular stage of research, it may not yet be able to provide an immediate answer to some practical problem arising, for instance, from

teaching EST to non-native speakers of English; the EST teachers and working scientists and engineers may turn out to be good problem solvers The rhetorical-communicative approach, on the other hand, may benefit from the comprehensiveness and depth of the systemic approach

1.2.3 Genre Analysis

According to Johns (2002: 5-10), there are broadly three genre schools, that is the Sydney School (introduced in Section 2.3.3), ESP (English for Specific Purposes) School and the New Rhetoric School Below I briefly describe the ESP School represented most prominently by Swales (1985; 1990)

Swales (1990) is a continuation and modification of the communicative approach, but with an emphasis on the text-types, or genres, in research and academic settings Swales (1985: 212) proposes the following definition for “genre”:

rhetorical-By genre is meant a more or less standardized communicative event

with a goal or set of goals mutually understood by the participants in

that event and occurring within a functional rather than a social or

personal setting Well-established genres are reports of laboratory

experiments, scientific papers, testimonials and job references, sermons,

cross-examinations, medical case reports and so on

The aim of genre analysis is to investigate the conventions (social and linguistic) associated with a particular genre and help the prospective participants, especially the non-native speakers of English, to perform well

Swales (1990) explicates this notion of genre and examines “research-process genres”, such as research articles in English (in terms of the linguistic features of their

component parts of Introduction, Methods, Results, Discussions and Conclusions), abstracts, research presentations, grant proposals, theses and dissertations, and reprint requests

Swales’s work (1985; 1990) is both theoretical and practical It seeks to identify the typical linguistic features of a functional component of a genre on the basis

of analysis of authentic data and relate the features to the conventions prevalent in the research community This may provide practical guidance to those new to the research genre

However, like the rhetorical-communicative approach briefly discussed above, genre analysis (in the sense of Swales (1985; 1990)) does not seem to be founded upon

a particular coherent linguistic theory For instance, the linguistic items that have been studied (see Swales (1990: 131-132) for an overview of the textual studies of the English research articles) are not related to each other in a systematic framework It is obvious that by explaining to a scientist or engineer in training what functional components he or she should include in a technical report or what tense he or she should use for a particular section of the report, Swales has given him or her invaluable help However, as shown in my analyses below (Chapter 4), research into the discourse of science would be more revealing if it could be founded upon a systematic theory of language This may be one of the strengths of systemic-functional approach

to EST

1.2.4 Systemic-Functional Approach to EST

Compared with the other approaches to the study of EST reviewed briefly above, register analysis is apparently characterized by frequency analysis of particular

lexicogrammatical items But frequency analysis is only one part of register theory; another part of it is the “metafunctional hookup” between language and its context (Halliday 1996a: 323) The technique of frequency counts and other statistical procedures can also be applied to other disciplines, for example, economics, education, and so on The grammatical theory underlying the register analysis is the systemic-functional linguistics, or SFL, known in the 1960s as category and scale grammar, proposed by Halliday

We need note that ever since its inception in the 1960s SFL has always been developing, taking in new materials and enriching itself A summary account of the current SFL interpretation of language is presented in Chapter Two As well, in Chapter Two I review SFL research on grammatical metaphor in scientific English and

on important genres in science texts (e.g Halliday and Martin 1993; Martin and Veel 1998)

It is also important to note that, inspired by O’Toole (1994) and Kress and van Leeuwen’s (1996) extension of SFL to visual semiotics, Lemke (1998a), O’Halloran (1996, 1999a, 1999b, 2000 and 2003), Thibault (2001), Kress et al (2001), Jewitt and Kress (2003) and Baldry (2000a) have begun to research multimodality in science writing and teaching Their work in visual semiotics is incorporated in Chapter Three

1.2.5 Science Educators’ Research on Science Textbook Articles

In addition to linguists’ and applied linguists’ efforts at EST, science educators, particularly secondary science educators, have been paying close attention to the various features of textbooks For instance, they explore what types of words in science texts cause the pupils difficulties Summarizing major research in this area,

Wellington and Osborne (2001: 23) point out “non-technical words (often taken for granted) can be at least as problematic as the technical, specialist terms of science Equally, the logical connectives used to link sentences and ideas can present a barrier

to the reading and understanding of science” Besides vocabulary difficulties, Bulman (1985: 22- 23) draws our attention to unfamiliar sentence structures, types and forms

of verbs and lack of motivating factors in science texts that may baffle the pupils To gauge the difficulty level of a particular science text, researchers have devised various readability formulae including the Flesch formula and the FRY readability graph (Wellington and Osborne 2001: 142-143)

In addition, Strube (1989) discusses the notion of style in physics textbooks

He claims that there are four rules that govern the language of physics textbooks The first rule is that “the author as an individual must be distinct from the prose written; his

or her personal beliefs, attitudes, attitudes, idiosyncrasies, and personal speaking voice

must be absent” “A second rule demands that the writer be as precise – that is, unambiguous – as possible” The third rule is that the discourse takes place “within a

narrow, specified context” The fourth rule is “that of limited syntax” and suggests that

“only a limited range of sentence types will predominate in a given type of textbook” (1989: 294; original emphasis) According to Strube, in analyzing science textbooks,

we should consider “the writer’s view of science as a discipline, and the effect of that view on the rhetorical style of the textbook” “In particular, the type of reasoning or argument allowed is determined by the established methods that give validity within science as a discipline” (1989: 294)

Unlike linguists who have begun to realize the importance of non-linguistic resources in science, scientists and science educators have long attached great importance to the non-linguistic elements in their work This is obvious in the