Analysing English Grammar: A Systemic Functional Introduction (2012)

Chapter 9: There and back again: interpreting the analysis

9.1 Introduction

Whatever the ultimate goal that is envisaged, the actual analysis of a text in grammatical terms is only the first step.

(Halliday, 1994: xvi)

Throughout this book, the focus has been on understanding the clause both functionally and structurally. Chapters 1 to 7 have progressively constructed this understanding by focusing on specific individual aspects in turn. Chapter 8 summarized all of this by presenting a set of guidelines for a multifunctional analysis of the clause. In this final chapter, we have come full circle. As pointed out in Chapter 1, one of the main objectives of analysing language in a functional perspective is to consider language in context and gain an understanding of how language is used. It is the text not the clause that is socially relevant. This chapter completes the picture by exploring how the analysis of the clause informs the analysis of text.

It may be clear by now that analysing all the clauses in a text is a considerable amount of work, but it is only the first step. According to Halliday (2010), there are two main questions we might have when analysing text. The first is ‘Why does the text mean what it does?’ and the second is ‘Why is the text valued as it is?’ The answer to the first question is that ‘it means what the linguist says it means’ (Halliday, 2010), by relating to the system and exploring how the text comes to mean what it does. In order to achieve this, as Halliday (1994: xvi) states, ‘there has to be a grammar at the base’. The answer to the second question is more difficult since ‘it requires an interpretation not only of the text itself but also of its context (context of situation, context of culture), and of the systematic relationship between context and text’ (Halliday 1994: xv). Halliday (2010) argues that there is not much point in trying to answer the second question without having already answered the first question. Therefore, irrespective of the goals one hopes to achieve in analysing text or discourse, a firm understanding of grammar is essential.

The meanings in the text are not immediately observable in most cases since they are expressed through the clause. This is why, before an understanding of the meanings in the text can be reached, it has to be segmented into its constitutive clauses so that these individual clauses can be analysed. The text is somewhat like a jigsaw puzzle in this sense; all the pieces must be taken apart and put back together before the full picture can be revealed.

9.1.1 Theory–description–use–theory cycle

Systemic functional linguistics presents both a theoretical model of language (a view of how language works) and a model of analysis (an analytical tool for describing language). There is a distinction to be made between considering the full potential of the grammar (language as system) and the perspective of ‘analysis’ when considering the actual instance of the grammar (language as text). The theory of SFL is based on the notion of choice and the modelling of the available choices in language as system networks. Language, in this view, is seen as a resource for making meaning. In theory, semantics drives the production of language as speakers are considered to be selecting from semantic options which are related in a complex network. The output of the system is grammatical form. In analysing grammar, the analyst works backwards, in a sense, starting with the grammatical form (e.g. the clause as identified in text), and attempts to deduce the functions expressed by the speaker. There is a kind of necessarily symbiotic relationship between the theory and the application of the theory. The theory drives the approach to analysis and the results of any analysis will then, in turn, inform the theoretical model. These two very different strands have been referred to as ‘theoretical-generative’ and ‘text-descriptive’ respectively (Fawcett, 2000c: 78). Ideally, they work together in balance in what Halliday and Fawcett (1987, cited in Matthiessen and Bateman, 1991: xvi) refer to as the ‘theory–description–use–theory cycle’. There should be open lines of communication between those who consider themselves architects of a theory and those who consider themselves appliers of a theory. Applied analysts need improvements to the model in order to better understand real-world concerns such as literacy, language disorders, media analysis, and so forth. Theorists need to test their model on real applications in order to adjust the theory, when necessary.

9.1.2 Goals of the chapter

The main goal of this chapter is to demonstrate how the analysis of the clause leads to an understanding of the text. This is often a difficult transition to make after focusing on individual clauses and it requires a different perspective from the micro-analysis that is done when working at the clause level. It is at this point, once the analysis of the clauses has been completed, that it is important to take a step back and look at the big picture.

A functional theory of language should be of interest and use to non-linguists and it should give ‘insights into the way language works in social interaction’ (Bloor and Bloor, 2004: 213). This chapter intends to move towards this goal and show how the results of the clause analysis can be compiled in order to both identify the patterns of meaning in the text and interpret them. As with many of the chapters in this book, the goal is to demonstrate an approach to grammatical analysis. In this chapter the focus is not on how to do the analysis (see Chapter 7 and 8) but on what to do with the results of this analysis. This is a difficult thing to model because interpreting grammatical analysis will be affected by the texts, the cultural and situational contexts involved, and the research aims of the analyst. Therefore, the presentation here should be seen as an example of one way to bridge the gap between grammatical analysis and discourse or text analysis. The discussion in this chapter is meant to be illustrative rather than exhaustive. For this reason, additional reading is suggested in section 9.5.

9.1.3 Organization of the chapter

This chapter will rely on two texts, as explained below, to illustrate one way of compiling, managing and interpreting the results of the grammatical analysis of text. The next section introduces the two texts and then presents the discussion of the results by considering each strand of meaning in turn. Examples will be given about how the results can be interpreted to show how the text means what it does. Once the meanings for each of the three metafunctions have been presented, a discussion of referent analysis is given which shows how a multifunctional view of specific referents can inform our understanding of the text. As this is the last main chapter in the book, section 9.3 offers some concluding remarks. Following this section 9.4 provides some exercises for readers to work through on their own with sample answers provided in Chapter 10. Finally, as with all other chapters, section 9.5 provides a list of suggested readings in text and discourse analysis.

9.2 Patterns of meaning in text

The analysis of the clause can be thought of as a kind of labelling exercise. The objective is to understand how meanings are expressed. This can provide a very detailed view of the clause. However, it does not really let us see the big picture, which is the text. The results from this analysis have to be collated so that the patterns of meaning in the text can be more easily seen and interpreted. This will be illustrated below using an example.

What is most relevant in the text is semantic rather than structural and while there is no direct correlation between grammatical realizations and semantic categories, there is a relationship between the two, albeit a complex one (Halliday, 1973: 75).

In this section, each strand of meaning will be considered in terms of patterns in text, leading to how results from analysis can be interpreted for a deeper understanding of the text. Sometimes, the meanings of the text are obvious and in fact so obvious that they can be difficult to notice (Thompson, 2004: 127). However, in other cases the meanings are perhaps obscured and only brought to consciousness by detailed analysis.

In order to begin to recognize patterns in the text, it can be very useful to organize the results. There are various ways in which this can be done. Generally, the results are grouped either manually or by using a software package. There are some very useful tools available for managing the analyses. One such tool is the UAM CorpusTool, developed by Mick O’Donnell (e.g. O’Donnell, 2008) and available for free on his website.¹ This software will assist the analyst in analysing single texts or collections of texts at multiple levels (for example, by text, clause or group) by keeping track of the analysis being done and managing the results. It also presents a very useful means of describing the results for the entire text or corpus and includes a statistical analysis package.

When working manually, it can be useful to compile the results of the analysis for each clause using tables for each strand of meaning. Tables can be created easily, and examples of this are given below for each metafunction. Some students have used spreadsheet software to enter clausal analysis in tabular form and others use the table function in word processor software. By doing so, an overview of the meanings in the text or in a section of the text can be seen at a glance.

As Thompson (2004: 127) points out, it is often easier to detect patterns by comparing two texts. To illustrate one way to compile the results of clause analysis in order to identify and interpret meaningful patterns in text, two texts will be presented and analysed by considering each metafunction individually. The two texts chosen are both what might be best called explanation texts since their purpose is to explain what an earthquake is. The first, Text 9.1,² was taken from the US Federal Emergency Management Agency (FEMA) website. It is intended for school-age children who are interested in learning about earthquakes. The second, Text 9.2,³ was written for an adult audience and was taken from the British Geological Survey (BGS) website. The FEMA and BGS texts are quite similar in many respects, especially with regard to their shared immediate goal of explaining earthquakes. The BGS text is an excerpt rather than the entire text so that it would be comparable in word length and topic to the FEMA text. The BGS text is presented in an FAQ (frequently asked questions) format, and although the questions are included in the text presented here they have not been included in the analysis as only the text answering the questions is being considered.

Text 9.1 FEMA text: ‘Earthquakes’

Earthquakes are the shaking, rolling or sudden shock of the earth’s surface. Earthquakes happen along ‘fault lines’ in the earth’s crust. Earthquakes can be felt over large areas although they usually last less than one minute. Earthquakes cannot be predicted – although scientists are working on it!

Most of the time, you will notice an earthquake by the gentle shaking of the ground. You may notice hanging plants swaying or objects wobbling on shelves. Sometimes you may hear a low rumbling noise or feel a sharp jolt. A survivor of the 1906 earthquake in San Francisco said the sensation was like riding a bicycle down a long flight of stairs.

The intensity of an earthquake can be measured. One measurement is called the Richter scale. Earthquakes below 4.0 on the Richter scale usually do not cause damage, and earthquakes below 2.0 usually can’t be felt. Earthquakes over 5.0 on the scale can cause damage. A magnitude 6.0 earthquake is considered strong and a magnitude 7.0 is a major earthquake. The Northridge Earthquake, which hit Southern California in 1994, was magnitude 6.7.

Earthquakes are sometimes called temblors, quakes, shakers or seismic activity. The most important thing to remember during an earthquake is to DROP, COVER and HOLD ON. So remember to DROP to the floor and get under something for COVER and HOLD ON during the shaking.

Text 9.2 Excerpt from the BGS FAQ on earthquakes

What is an earthquake?

An earthquake is the sudden release of strain energy in the Earth’s crust resulting in waves of shaking that radiate outwards from the earthquake source. When stresses in the crust exceed the strength of the rock, it breaks along lines of weakness, either a pre-existing or new fault plane. The point where an earthquake starts is termed the focus or hypocentre and may be many kilometres deep within the earth. The point at the surface directly above the focus is called the earthquake epicentre.

Where do earthquakes occur?

Anywhere! However, they are unevenly distributed over the earth, with the majority occurring at the boundaries of the major crustal plates. These plate boundaries are of three types: destructive, where the plates collide; constructive, where the plates move apart; and conservative plate boundaries, like the San Andreas Fault, where the plates slide past each other. Earthquakes also occur, less frequently, within the plates and far from the plate boundaries, as in eastern USA, Australia and the United Kingdom.

Which countries have the largest and most frequent earthquakes?

Around 75% of the world’s seismic energy is released at the edge of the Pacific, where the thinner Pacific plate is forced beneath thicker continental crust along ‘subduction zones’. This 40,000 km band of seismicity stretches up the west coasts of South and Central America and from the Northern USA to Alaska, the Aleutians, Japan, China, the Philippines, Indonesia and Australasia.

Around 15% of the total seismic energy is released where the Eurasian and African plates are colliding, forming a band of seismicity which stretches from Burma, westwards to the Himalayas to the Caucasus and the Mediterranean.

What is the biggest earthquake that has ever happened?

One of the largest earthquakes ever was the Chile event of 22 May 1960 with moment magnitude of 9.5 Mw. Other large earthquakes include Lisbon, 1 November 1755, magnitude 8.7 Ms; Assam, 12 June 1897, magnitude 8.7 Ms; Alaska, 28 March 1964, moment magnitude 9.2 Mw. Although the magnitude scale is open ended, the strength of the crustal rocks prior to fracturing limits the upper magnitude of earthquakes.

The details of the analysis are not discussed here in order to save space and to focus on how the results of the analysis can be interpreted. Segmenting the text into individual clauses was done following the guidelines given in Chapter 7 and the analysis of each clause was done following those given in Chapter 8. Before continuing in this chapter, it might be a useful exercise to work through the analysis of these two texts. It would help the understanding of the results presented in the tables that follow.

What follows is a discussion of the patterns found in each text. This discussion is organized by individual strand of meaning. The results of the analysis have been compiled in the form of tables. However, the tables given in each case are only intended as examples for illustrative purposes. They should be adapted to the needs of the analysis in each case, adding or removing detail as appropriate. If software such as the UAM CorpusTool is being used, then it will manage this organization. There is no fixed method for organizing results and it is up to each analyst to work out the best way to work with their text(s).

9.2.1 Experiential meaning

Experiential meaning, as explained in Chapter 4, involves the representation of the speaker’s experience, including what they observe, what they think or perceive, and how they relate things. This experience is inherently subjective and consciously or unconsciously reflects how the speaker sees the world or, possibly, how they want others to see it.

Since experiential meaning is sub-categorized into six process types, it can be helpful to organize the results for each process type in separate tables so that the patterns in the text are easier to see. The results for both texts are presented below. The analysed clauses for Text 9.1 have been given in Tables 9.1 to Table 9.4 followed by Tables 9.5 and 9.6, which present the analysis of experiential meaning for Text 9.2. The tables for the interpersonal and textual analyses will be presented individually below.

Table 9.1: Material clauses in the FEMA text

Table 9.2: Mental clauses in the FEMA text

Table 9.3: Relational clauses in the FEMA text

Table 9.4: Verbal clauses in the FEMA text

Table 9.5: Material clauses in the BGS text

Table 9.6: Relational clauses in the BGS text

These results will be discussed below in comparison with the results from the BGS text. However, some patterns may already be emerging, such as the type of experience being represented. The next set of tables (Tables 9.5and 9.6) presents the results of the experiential analysis of the BGS text. There are only two tables since only two types of process were represented in this text.

It may seem immediately obvious from these tables that the two texts are not representing exactly the same experience. The FEMA text has a relatively high frequency of mental and relational processes as compared to a much higher frequency of material and relational processes in the BGS text (see Table 9.7 below). However, identifying other patterns related to experiential meaning requires a bit more digging. Thompson (2004: 127) provides a very useful overview of how transitivity can be interpreted by presenting a set of questions that help reveal the patterns. These questions, and any others that are relevant, can be used to guide the inquiry of the text:

· What are the dominant process types? And why these?

· How do the process types match with other aspects (e.g. location in the text, appearing in commands vs. statements, etc.)?

· What (groupings of) participants are there?

· How do these compare with ‘real-world’ entities and events?

· What kinds of participants (e.g. concrete vs. abstract)?

· What transitivity role(s) do they have?

· How are any nominalizations used? Are they representing ‘hidden’ situations?

· What types of circumstances are included? Where are they in the text?

· What gets expressed as circumstance rather than in the ‘nucleus’ (process + participant)?

Although both texts rely on relational processes, the FEMA text has a broader range of experience represented. Relational processes, as discussed in Chapter 4, relate two participating entities in an abstract way. This is useful for explaining what an earthquake is (e.g. An earthquake is the sudden release of strain energy in the Earth’s crust resulting in waves of shaking that radiate outwards from the earthquake source) or for describing an earthquake (e.g. A magnitude 6.0 earthquake is considered strong). The high frequency of mental processes in the FEMA text is interesting, and in order to understand how this works in the text it is necessary to consider the participants involved in these clauses.

Table 9.7: Comparison of process types in the FEMA and BGS texts

The main participants in both texts are given in Table 9.8, where it is clear that, although both texts have earthquakes as a significant participant in terms of its representation in the texts, this is not done in the same way. The FEMA text has a much more concentrated representation of earthquakes and its main role in the text is as Phenomenon; in other words, it is represented as something that is to be experienced through the senses. The Senser in these mental processes always includes the addressee – that is, the reader of the text, who in this case is a school-age child. In contrast, the BGS text, while it does include earthquakes as a participant, does so much less frequently and it is primarily represented as the Identified participant in a relational process, although it is represented implicitly as Actor in two passive material processes (clauses [9] and [11]). The addressee is never a participant in this text and there is a much greater variety of entities participating. One main participant in this text is seismic energy, which occurs most frequently as Goal in material processes.

Table 9.8: Comparison of the main participants in the FEMA and BGS texts

The use of material processes in each text is considerably different. In the FEMA text, the role of Actor is explicitly stated (e.g. the intensity of an earthquake, earthquakes below 4.0 on the Richter scale, earthquakes over 5.0 on the scale). However, in the BGS text, the material processes are most frequently in the passive voice and the Actor is left covert (or unspecified). This suggests that a certain amount of geographical knowledge is expected of the reader. For example, it is never explicitly stated that the earthquake releases seismic energy.

A comparison of the use of circumstances is given in Table 9.9, which shows that the number of circumstantial elements is roughly equivalent in each text but that the BGS text has fewer clauses than the FEMA text. So, proportionally, the BGS text has far more circumstances per clause than the FEMA text. The use of manner circumstances differs in the texts since the FEMA text uses a manner circumstance in mental processes to describe how the phenomenon (earthquake) will be sensed (see clause [9]), whereas it occurs in material processes in the BGS text to show how something is done (see clause 6).

Table 9.9: Comparison of circumstances in the FEMA and BGS texts

There is much more that could be said about the experience represented in these two texts (e.g. an analysis of the embedded clauses). The brief sample discussed above should provide an idea of how the results of the clause analysis combine to show what the text means experientially. Although the topic of each text is very similar, the analysis of the text shows that the goals are different. The FEMA text includes the addressee as a participating entity and focuses on earthquakes in terms of a phenomenon that the addressee can sense. The BGS text also represents the referent as a participating entity (i.e. as an entity participating in the situation), but it draws on other related entities (e.g. seismic energy, rock and earth’s surface), primarily in relational processes, in order to define and describe earthquakes.

9.2.2 Interpersonal meaning

Interpersonal meaning was described in Chapter 5 as expressing the speaker’s personal intrusion on the language situation and the speaker’s use of language to interact with others. This involves the means by which the speaker’s personal views are expressed through modalities of modalization (probability and usuality) and modulation (obligation and inclination). Speakers also express meanings that are related more directly to interaction through the use of mood choices to ask questions, give information or make requests.

Identifying patterns in the interpersonal meanings that are found in a text can be done following an approach which is similar to the one presented above. The results of the clause analysis can be grouped in a table which presents the relevant interpersonal meanings for the purpose of the study. The results of the interpersonal analysis for the FEMA text and the BGS text are given below in Tables 9.10 and 9.11. No column for mood has been included since all clauses in both texts (with one exception) are in the declarative mood. The use of the tables should be seen to help reveal relevant patterns easily. In this sense, they are simply a tool used in this stage of the analysis. Since the Subject and Finite elements have been stressed as being especially significant to interpersonal meaning in English, they should be seen as core elements to include. However, how they are included depends on the needs and interests of the analyst. In these tables, the Finite element has been included as a single column which can make it difficult to see at a glance the different forms of expression of the Finite (e.g. as expressed by the Auxiliary or Event element of the verb group). Also, in this example, modality is grouped in a single column which does not distinguish between modal auxiliary verbs and modal adjuncts. If this were an important distinction to consider, then it would be recommended to structure the tables differently so that this information is more readily available.

Table 9.10: Interpersonal analysis of the FEMA text

Table 9.11: Interpersonal analysis of the BGS text

A set of guiding questions can also be used to help identify the patterns of interpersonal meaning in a text. A sample of such questions is listed below, which focuses on mood, modality and polarity. However it could also include, if appropriate, elements of the clause which express the roles of the speaker and addressee, indications of social status and relationships such as degrees of formality or familiarity:

· What is the dominant mood choice? How is mood being used in the interaction?

· What modality is expressed? How is it expressed?

· What polarity is expressed?

· What indicators of social status are present?

· What are the roles of the speaker and addressee?

· What indicators are there of the speaker’s opinion or attitude?

The two texts do not differ significantly in their use of mood choice as all clauses are expressed in declarative mood except clause [22] in the FEMA text. The differences among the different types of mood choice (i.e. interrogative, declarative or imperative) relate to ‘differences in the communication role adopted by the speaker in his interaction with a listener’ (Halliday, 2002: 189). The use of the imperative mood expresses an obligation on the part of the addressee and reflects an unequal status between the speaker and addressee. The rest of the clauses in both texts are giving information about earthquakes and related topics.

There is considerable modality expressed in the FEMA text, which is in contrast to the BGS text, where there is only one instance of modalization (probability). The FEMA text uses a wider range of modality and includes it much more frequently, as shown in Table 9.12. The instances of modal probability express the degree of likelihood (or probability) of the experiences represented in the text, as illustrated in examples (1) to (3):

(1) Most of the time, you will notice an earthquake by the gentle shaking of the ground (clause [7], FEMA)

(2) You may notice hanging plants swaying or objects wobbling on shelves (clause [8], FEMA)

(3) Sometimes you may hear a low rumbling noise (clause [9], FEMA)

Table 9.12: Comparison of modality in the FEMA and BGS texts

There is some evidence of the relationship between speaker and addressee in both texts. The FEMA text uses the personal pronoun you to refer to the addressee. This kind of interpersonal deixis allows the speaker to talk directly to the addressee by reference to the contextual situation even though the actual situation changes with every reader. Furthermore, as already stated, the use of the imperative pushes this relationship even further and raises the role of the speaker to one that has some power over the addressee. This means that the speaker is in a position of authority and can instruct the reader because of their knowledge and experience.

The speaker in both texts is in the role of expert who has specialist knowledge and therefore the status between speaker (writer) and addressee (reader) is necessarily asymmetrical. The BGS text maintains distance between the speaker and addressee due to the use of the declarative mood, the relative lack of modality and the absence of reference to speaker or addressee. These combine to contribute to a sense of detachment that is not found in the FEMA text. In the FEMA text, however, the detachment typically found in written texts is reduced by the use of addressee reference, modality and imperative mood.

9.2.3 Textual meaning

As was seen in Chapter 6, the textual metafunction has the function ‘of creating text, of relating itself to the context – to the situation and the preceding text’ (Halliday 1978: 48). The Theme element provides the link between the clause and text and, consequently, looking at the contents of Theme throughout the text will provide insight into the development of the text. Thompson (2004: 165) explains that one of the main functions of Theme is to signal ‘the maintenance or progression of what the text is about’. In addition to this, other textual patterns can be identified by considering cohesion.

In terms of organizing the results of the textual analysis of the clause, the main organizing element will be Theme and the various types of Theme possible. The tables presenting these results may also include notes about cohesion, thematic progression or anything else related to the textual metafunction. The summary of the results for the FEMA and BGS texts are given in Tables 9.13 and 9.14. There were no interpersonal Themes in either text so this information has not been included.

Table 9.13: Summary of the textual analysis of the FEMA text

Table 9.14: Summary of the textual analysis of the BGS text

Given that the texts are both about earthquakes, it would be reasonable to expect that most Themes would be expressed by nominal groups referring to earthquakes. Considering the content of Themes in a text is one way to identify the patterns in the text. There are other considerations. Here is a list of suggested questions for identifying patterns in textual meaning:

· What is the most frequent Theme?

· What is the use of any marked Themes (e.g. signalling a shift or transition in the development of the text)?

· How are any textual and/or interpersonal Themes being used?

· What is the text function of the use of any special thematic constructions?

· What are the patterns of thematic progression?

· What cohesive strategies are identifiable in the text?

· What are the main strategies for reference (e.g. endophoric/exophoric reference)?

In the FEMA text, Theme is expressed by an expression referring to earthquakes generally or to a specific earthquake in 11 of the 21 clauses (52.4 per cent). This is in contrast to only 4 such Themes in the BGS text (26.7 per cent), with Themes referring to seismic energy forming 20.0 per cent of the Themes in the text. This suggests that the FEMA text maintains the topic of earthquakes much more than the BGS text does. It also points out that the BGS text uses Theme to progress the topic throughout the text. Both texts are quite similar in how they use multiple Themes. The use of textual Themes is very similar, and neither text includes any interpersonal Themes. Marked Themes were not common in either text.

It is interesting to note that the addressee (you) is only explicitly thematized in one clause (see clause [8] in Table 9.13), even though it expresses Subject in four clauses and Senser in six clauses in total. Therefore while the addressee is represented in the text in a significant way and he or she is addressed directly by the use of the imperative in clause [22], this reference is not an important thematic element. The text is not attempting to include the addressee as a topic. The use of this kind of reference does contribute to the creation of text through exophoric reference to the context of situation by referring deictically to the addressee (reader).

9.2.4 Referent analysis

There are many ways to organize and interpret the results of a full multifunctional analysis of the clause. As shown above, one way to do so is to focus on individual strands of meaning. However, it can be revealing to concentrate on specific referents in the text by isolating them and considering the meanings they express throughout all strands of meaning. The approach presented here is adapted from Martin’s (1992) reference chains and Halliday and Hasan’s (1985) identity chains. Referent analysis combines all three metafunctions and considers the view the text has taken for a specific referent.

In this example, the referent is explored and this means that all expressions referring to an earthquake or earthquakes will be included. The results are presented in Tables 9.15 and 9.16, where any embedded reference to earthquakes is listed in italics. These tables include the actual expression used, the clause number in the text, the referring strategy (e.g. definiteness of the expression and phoricity, such as anaphora), the experiential function of the expression, any interpersonal meaning, including modality and connotation, and textual meaning, which includes thematic role, cohesion and deictic functions.

Table 9.15: Referent analysis for in the FEMA text

Table 9.16: Referent analysis for in the BGS text

A similar approach could be taken to comparing and contrasting different referents in a text or across texts. For example, it may be of interest to compare the references to political candidates in the media or in their own manifestos. In this example, these tables make it very clear that the main referent, , is not used in the same way in each text. As the last two columns in each table show, is primarily expressed as functioning as Subject/Theme. However, the main differences are found in the frequency of these referring expressions and in the experiential representation of the referent, as was already mentioned above.

There are also differences in the referring strategies used by the speaker in each text. The BSG text uses almost exclusively indefinite referring expressions. This is noted by the use of the indefinite article an and the use of the plural. This is directly related to the fact that no specific earthquake is being referred to. Other than through lexical repetition there is no overt cohesive tie amongst these clauses, with the exception of the contrastive use of other in clause [13], which ties it to the preceding clause. The expression in clause [12] is interesting because, although it is an indefinite expression, it is related in this clause to a specific earthquake. The FEMA text includes two definite referring expression, and these are used to refer to specific earthquakes rather than to earthquakes in a general sense. Both texts also include the use of cohesive reference (see the use of they in clause [4] in FEMA and clause [6] in BGS).

9.2.5 Summary

The objective of this section was to outline a method for identifying and interpreting the patterns of meaning in text. The discussion of these two texts has been necessarily brief but a detailed analysis would be able to reveal much more. A more detailed example of the interpretation of analysed text is provided in Chapter 10 in the sample answer to Exercise 9.1. As Halliday states, ‘whatever the ultimate goal that is envisaged, the actual analysis of a text in grammatical terms is only the first step’ (Halliday, 1994: xvi).

The interpretation of the results of text analysis as presented here has been done in abstract terms. In reality, this kind of text analysis, through grammar, would be done with a specific aim in mind. This might be, for example, to understand children’s writing, the speech of a stroke victim or the meanings of political speeches. It may also be done with the aim of better understanding a particular theoretical area of the grammar such as the role of Theme in imperative clauses or the use of multi-word lexical items, such as phrasal verbs. The specific research aims will guide the explanation and interpretation of the grammatical analysis (see Chapter 10 and Exercise 9.2; see also Bloor and Bloor, 2004 and Coffin, Hewings and O’Halloran, 2004).

9.3 Final remarks

In writing this book, I set myself two challenges. The first was to write an introduction to analysing English grammar in a functional framework by focusing on the practical, ‘how-to’ aspects of analysis. There is an inherent paradox in grammatical analysis; to understand the clause (i.e. know and recognize it), you have to first understand the clause (i.e. know its components and configurations). This is why the presentation in this book compartmentalized the various aspects to analysis into manageable stages in order to construct a systematic approach to analysing grammar. The approach promoted here was built up incrementally so that the multiple functions of the clause could be clearly related to each other and directly associated to the structural units that serve to express them. For newcomers to functional analysis, this approach will be very useful as it will provide a practical set of guidelines for analysing English grammar. Those more experienced in systemic functional linguistics will have noted a hybrid approach which does not adhere strictly to one particular version of the theory but rather draws from various existing descriptions of the theory of systemic functional grammar, most notably from Halliday and Matthiessen (2004) and Fawcett (2000c).

The second challenge was to try to write it in such a way that grammatical analysis would seem less mysterious and more enjoyable. The funny thing is that speakers generally do not have any trouble understanding language but, as soon as analysing it involves terminology and theory, things quickly become more intimidating. It is important to stress that fluent speakers of a language already know how language works since they use it successfully every day. However, developing skill at grammatical analysis offers a key to unlocking some of the mysteries. It is like becoming a professional language analyser with all the advantages that professional skill has over everyday general knowledge. It means you have a deeper and broader understanding of not only how language works but also how it can work. It means that you are in a specialist position for working on real problems related to language such as, for example, language learning, language disorders, applications such as translation and also the ways in which language is used to suppress, control or benefit people. There is very little we do in this world that does not involve language. As I said in the first chapter, if you end up enjoying grammar even just a bit more from having read this book then it will have been a great success.

9.4 Exercises

The exercises for this chapter focus on interpreting results. However, in obtaining an interpretation, all three stages of analysis are required: segmenting the text, analysing the clauses and interpreting the results. The text selected for these exercises comes from an assignment completed for my undergraduate course at Cardiff University. Only the text is given here; each exercise listed below asks you to complete one stage of the analysis for this text. A sample answer for each is given in Chapter 10. The analysis and discussion was completed by David Schönthal (2009), a former student, and it is reproduced here with permission.

Exercise 9.1

The text below is an extract from a text entitled ‘Chief Seattle’s Speech’ and is generally accepted as being attributed to screenwriter Ted Perry in 1971. This version of the now famous text was taken from a web page that is no longer available, but a similar version can be found at www.ilhawaii.net/~stony/seattle2.html.

Segment the text below into a list of individual clauses (see Chapter 7) and analyse it fully using the guidelines presented in Chapter 8. In Chapter 10, the full clause list is given for this text and a sample three-strand analysis is provided in the form of box diagrams. Only a selection of clauses will be analysed for illustrative purposes due to the length of the text.

Chief Seattle’s Speech

How can you buy or sell the sky, the warmth of the land? The idea is strange to us. If we do not own the freshness of the air and the sparkle of the water, how can you buy them?

Every part of this earth is sacred to my people. Every shining pine needle, every sandy shore, every mist in the dark woods, every clearing and humming insect is holy in the memory and experience of my people. The sap which courses through the trees carries the memories of the red man.

The white man’s dead forget the country of their birth when they go to walk among the stars. Our dead never forget this beautiful earth, for it is the mother of the red man. We are part of the earth, and it is part of us. The perfumed flowers are our sisters; the deer, the horse, the great eagle, these are our brothers. The rocky crests, the juices in the meadows, the body heat of the pony, and man – all belong to the same family.

. . .

The rivers are our brothers, they quench our thirst. The rivers carry our canoes and feed our children. If we sell you our land, you must remember, and teach your children, that the rivers are our brothers, and yours, and you must henceforth give the rivers the kindness you would give any brother. We know that the white man does not understand our ways. One portion of the land is the same to him as the next, for he is a stranger who comes in the night and takes from the land whatever he needs. The earth is not his brother but his enemy, and when he has conquered it he moves on. He leaves his father’s graves, and his children’s birthright is forgotten. He treats his mother, the earth, and his brother, the sky, as things to be bought, plundered, sold like sheep or bright beads. His appetite will devour the earth and leave behind only a desert.

. . .

There is no quiet place in the white man’s cities. No place to hear the unfurling of leaves in the spring, or the rustle of an insect’s wings. But perhaps it is because I am a savage and do not understand. The clatter only seems to insult my ears. And what is there to life if a man cannot hear the lonely cry of the whippoorwill or the arguments of the frogs around a pond at night? I am a red man and do not understand. The Indian prefers the soft sound of the wind darting over the face of a pond, and the smell of the wind itself, cleansed by rain or scented with the pine cone.

The air is precious to the red man, for all things share the same breath: the beast, the tree, the man, they all share the same breath. The white man, they all share the same breath. The white man does not seem to notice the air he breathes. Like a man dying for many days, he is numb to the stench. But if we sell you our land, you must remember that the air is precious to us, that the air shares its spirit with all the life it supports. The wind that gave our grandfather his first breath received also his last sigh. And if we sell you our land, you must keep it apart and sacred, as a place where even the white man can go to taste the wind that is sweetened by the meadow’s flowers.

link to answer

Exercise 9.2

Based on the analysis done in the previous exercise, use the approach developed in this chapter to organize and interpret the results of the functional analysis. A sample presentation of the tables of results is given in Chapter 10 along with Schönthal’s (2009) discussion as a sample of a possible interpretation of the grammatical analysis.

link to answer

9.5 Further reading

Bartlett, T. forthcoming. Analysing Powers in Text. London and New York: Routledge.

Butt, D., R. Fahey, S. Feez, S. Spinks and C. Yallop. 2001. Using Functional Grammar: An Explorer’s Guide. 2nd edn. Sydney: NCELTR.

Coffin, C., A. Hewings and K. O’Halloran, eds. 2004. Applying English Grammar: Corpus and Functional Approaches. London: Hodder Arnold.

Halliday, M. A. K. and R. Hasan. 1985. Language, Context, and Text: Aspects of Language in a Social-Semiotic Perspective. Oxford University Press.

Hasan, R. 1985. Linguistics, Language and Verbal Art. Deakin University Press.

Martin, J. 1992. English Text: System and Structure. Amsterdam: John Benjamins.

Thompson, G. 2004. Introducing Functional Grammar. 2nd edn. London: Arnold.

Threadgold, T., E. Grosz, G. Kress and M. A. K. Halliday, eds. 1986. Semiotics, Ideology, Language. Sydney Association for Studies in Society and Culture.