Analysing English Grammar: A Systemic Functional Introduction (2012)

Chapter 7: From text to clause

7.1 Introduction

Having now completed the internal view of the clause, this relatively short chapter looks at how to recognize the boundaries of the clause within text. It may seem odd that what will be the first step in analysing text is presented after the analysis of the clause but this is because, in order to identify the boundaries of the clause, it is essential to have a firm understanding of the way that the components of the clause work both structurally and functionally. This information has to be relied upon in order to recognize clauses within a text.

This chapter together with Chapter 8 provides the full set of guidelines for text analysis. This chapter, as already stated, explains how clause boundaries can be identified within a text, and Chapter 8 summarizes the guidelines that have been being developed throughout this book. These two chapters combined offer a complete view of analysing text. However, as is explained in Chapter 9, this is only the first stage in understanding language. Grammatical analysis forms the basis on which to build an interpretation of text. Chapter 9 describes how to manage the results of grammatical analysis so that the patterns of meaning in text can be understood.

7.1.1 Goals and limitations

This chapter summarizes the key details from all previous chapters which are relevant to recognizing clause boundaries. While this chapter, and indeed this book, will provide a very good basis from which to analyse the grammar of English, it has been an introduction to grammatical analysis and couldn’t possibly offer the degree of detail necessary to equip someone with everything necessary to be able to analyse any clause encountered. There is no single volume which could manage a comprehensive view of English grammar. Halliday (1994: xiii) describes his introduction to functional grammar as a ‘short introduction because, despite any illusion of length, it is no more than a mere fragment of an account of English grammar. Anything approaching a complete grammar would be hundreds of times this length.’

The goal of this book has not been explicitly to introduce the grammar but instead an approach to how to analyse the grammar within a functional framework. However, grammar isn’t something we find when we want to analyse and understand text. Language is presented to us as text, as an output of the language system. It is generally agreed that this output is organized and structured in units. Exactly what these units are may be debatable. This book is based on the premise that the main organizing unit is the clause. The challenge for the analyst interested in understanding language in use is recognizing this unit. This chapter builds on the knowledge and understanding gained so far about the clause and provides a foundation for moving from text to clause. However, it is in many ways only a starting point. Language always throws curve balls; it is dynamic not static, and furthermore it is creative and inventive. Those who find it fascinating will certainly rise to the challenges it presents. In this view, this chapter is a springboard for a continuing journey into understanding language.

7.2 Textual Themes and clause boundaries

The previous chapter considered the relationship between the clause and text by focusing on Theme within the textual metafunction. The function of textual Theme was described as having an explicit relevance function since it serves to indicate the relevance of the clause to neighbouring clauses within the text. By their very nature, when they occur they will be located at the start of the clause as the first element of the clause.

Understanding Theme and specifically textual Theme is useful in identifying clause boundaries since a clause which expresses a textual Theme will be explicitly marking its initial boundary. Of course, this doesn’t provide any information about how the clause ends. The final boundary of the clause is more challenging.

Another feature of Theme is that if there are no textual or interpersonal Themes the clause will begin with either the Subject element or an adjunct element with the exception of imperative clauses, which, as seen in Chapter 6, omit the Subject and whose first element is often the Event.

One of the particular functions of textual Theme is to link one independent clause to another (for example, from Chapter 6, I tried to send it to him but there’s still something wrong). A very common way to do this is with the use of a conjunction. In Chapter 5, the ampersand symbol, ‘&’, was used to indicate the co-ordination of any two units within the clause. However, at the clause level, a conjunction will serve to express the function of textual Theme. It may be useful to consider this use of textual Theme as a linking or co-ordinating textual Theme since it explicitly expresses the co-ordination of two clauses.

In addition to linking independent clauses, textual Themes can also show mutual relevance between clauses when one clause is functioning dependently on the other clause (e.g. I tried to send it to him since there’s still something wrong). The textual Theme in these dependent clauses links the clause so strongly that it is seen as binding it to the other clause. This use of textual Theme subordinates one clause in relation to the other, effectively binding it.

7.3 Combining clauses

All clauses have the potential to be simple or complex. The complexity of the clause is usually determined by the degree of complexity of the structural units which express it. It is generally agreed that a simple clause has one lexical verb (and perhaps one auxiliary verb) and a simple nominal group expressing the participant elements. More complex clauses will have more than one lexical verb and, consequently, this means that more than one clausal unit is involved. When analysing a text, knowing where one clause ends and another begins can be the greatest challenge. Understanding how clauses can combine is one more step towards unlocking this problem.

Perhaps it would be useful at this point to explain what is meant by the use of ‘combine’. There is an assumption being made with regard to written text at least that clause combining is something made explicit by the speaker. This is done through the sentence, which is an orthographic written unit that marks the speaker’s deliberate combination of clauses (or not in the case of a simple sentence). This is very similar to the use of indentation or white space around a section of text to indicate a paragraph. However, in one sense, it could be argued that the only true sense of clause combining is embedding, where one clause is literally combined within another. Most other ways of combining clauses, as will be shown below, are more a question of juxtaposition with explicit textual markers.

In the previous three chapters, the clause has been considered from three functional perspectives, each with its own configuration of elements. Now that a full functional–structural description of the clause has been presented, we will look at the three main types of clause in English. The clause types are distinguished in terms of how they can potentially combine with other clauses. The three types are independent, dependent and embedded clauses. In this section each one will be described in turn with a focus on how each one can combine with other clauses. The goal of this chapter is to develop guidelines for identifying clause boundaries, so the discussion of these clause types will concentrate on what these types tell us about recognizing clauses in text.

7.3.1 Co-ordinating clauses

The examples presented so far in this book have most often been what are traditionally called independent clauses (e.g. This is the house that Jack built; In a high altitude emergency, an oxygen mask will drop in front of you from the panel above; By the way did you ever pay that dollar to the post office?). In most grammar classes in school, these clauses were most often referred to as stand-alone sentences. In other words, they are fully complete in the sense that they satisfy the requirements for acceptable grammatical status.

This is a vague description but it is quite challenging to explain what makes a clause a clause and how we can tell when it is complete or not. In experiential terms, the clause would be seen as complete if all participants required by the process were expressed. From the interpersonal perspective, the clause would be seen as complete if it includes a Subject and Finite element along with an Event and any required Complements. Of course, it is possible for some of these elements to be left out and still have a complete clause – for example, passive clause structure and ellipsis permit certain elements to be left covert but recoverable, so in a sense it is not the case that they are not at all present. Furthermore, a clause may include more elements than those expected as in the case of adjuncts and certain types of Theme. The independent status of a given clause depends therefore on the specific clause in question. The determining factor is whether or not it satisfies the expectations of both speaker and addressee.

Recognizing an independent clause is simplified when the clause has only one lexical verb and expresses one process. When this is not the case, then it can be assumed that the clause combines with another clause. As this section intends to show, there are three main ways of combining clauses. When two independent clauses combine, they are said to be co-ordinated. Clause co-ordination is usually recognized by the presence of a co-ordinating conjunction such as and or but. In written language, this is sometimes indicated by a semi-colon.

In SFL, the two co-ordinated clauses are seen to be of equal status and they are referred to as being in a relationship of parataxis. Although any clause type can be co-ordinated (including finite or non-finite clauses), independent clauses combined this way are finite. However, finiteness is not a determining factor for recognizing an independent clause since subordinate clauses are also finite and some embedded clauses are finite. Identifying two co-ordinated independent clauses will be based on punctuation in written texts and the presence of potential textual Themes which explicitly indicate co-ordination. Recognizing independent clauses is based on the discussion above in terms of recognizing the core elements of the clause with respect to both experiential and interpersonal meaning.

7.3.2 Subordinating clauses

Another way in which clauses can be combined is through a relationship of subordination, or hypotaxis as it is called in SFL. When clauses are combined in this way, one clause is an independent clause and the other is a dependent clause. All dependent clauses are combined with an independent clause since they are inherently dependent on another clause, hence their label. Subordinate clauses are very similar in terms of structure and function to independent clauses but they are marked textually as being bound to another clause by a textual Theme.

In addition to being textually marked, these clauses do not meet the completeness criterion for independent clauses since, if stranded on their own, they maintain a sense of expectation. For example, if someone were to say, if I win the lottery, the addressee would be expecting more to come; they would expect a clause on which this one is dependent, such as if I win the lottery I will buy a house in Canada. The combination of subordinated clauses can be reversed: I will buy a house in Canada if I win the lottery.

Dependent clauses can be as complex as any other type of clause (e.g. if this is the dog that chased the cat that ate the rat that lived in Jack’s house then it must be the same dog that ran through my garden). Therefore, the problem of recognizing them is just as challenging as for independent clauses. However, they are somewhat more easily recognized by the textual Theme, which orients and relates the subordinate clause to the clause on which it is dependent.

7.3.3 Embedding clauses

As was pointed out in Chapter 2, it is possible to find one clause within another. When clauses combine in this way, it is referred to as embedding. The example used to illustrate embedding was the ‘House that Jack Built’ example, where the humour in the story is the result of the very lengthy series of embedded clauses. In these cases, the embedded clause is not of the same status as the clause it combines with, nor is it subordinated. It is used as a resource for expressing a function within an independent or dependent clause. In fact, it can even be embedded within an embedded clause. Embedded clauses can express a variety of functions within the clause. This is what makes them a fundamentally different clause. However, this is precisely what makes embedded clauses so challenging in grammatical analysis.

A further complication in recognizing clause boundaries involves the finite status of the clause. As was explained in Chapter 5, all clauses are either finite or non-finite. All independent and dependent clauses are finite. In contrast, embedded clauses can be either finite or non-finite. In a sense, non-finite clauses, once understood, are relatively straightforward since there are only three types (e.g. to eat an apple a day can be very healthy; eating an apple a day can be very healthy; eaten once a day, apples can be very healthy). This feature makes them much easier to recognize in a clause than their finite counterpart.

If non-finite clauses can be recognized by their structure and the absence of any identifiable or recoverable Finite element, then finite embedded clauses are the problem ones. As stated above and in Chapter 5, all embedded clauses serve to express a function within a clause. Some of the most common functions that embedded clauses express are participating entities, circumstances, qualifier in the nominal group (e.g. relative clause), completive in the prepositional phrase and finisher in the adjective group. Examples of each will be discussed below. Understanding how these clauses function and what indicators they may have should help in recognizing them in a text.

Garden-path sentences are often quite useful for talking about the challenges of recognizing embedded clauses. Examples of garden-path sentences are given in (1) to (3).Garden-path sentences are misleading because they trick you in some way, often because adjacent words in certain contexts trigger a particular interpretation but then this interpretation is proved false by what follows. In this sense, these sentences lead you down the garden path. Often, when trying to understand such sentences, readers will construct a scenario where the sentence can make sense. For example, students will often interpret example (1) as referring to a thing that is called a ‘barn fell’. They assume that there must be such a thing on a farm and that a horse is able to run past it. Or, they assume that the word ‘and’ is missing, which would give: The horse raced past the barn and fell.

(1) The horse raced past the barn fell

(2) Mary gave the child the dog bit a bandaid

(3) I convinced her children are noisy

Each sentence in examples (1) to (3) is an independent clause which includes an embedded clause. However, there is no marker of the embedded clause as there is sometimes with relative clauses, for example. These sentences would suddenly not be problematic if a relative pronoun such as that had been used. See if you can resolve the interpretation by adding that to each sentence and then check your results below.

The sentences in (1) to (3) have been rewritten as examples (1′) to (3′) where the embedded clause is now clearly introduced by that and the embedded clause is underlined.There is an alternative to (1′), which is given in (1″), but it is less likely since it sounds awkward and unnatural. What these examples show is that embedded clauses can be finite and that the relative pronoun which introduces some embedded clauses can be ellipsed.

(1′) The horse that raced past the barn fell

(1″) The horse raced past the barn that fell

(2′) Mary gave the child that the dog bit a bandaid

(3′) I convinced her that children are noisy

Each embedded clause serves to express a particular function. In (1), the embedded clause, that raced past the barn, expresses the qualifier in the nominal group which is expressing the participant function (the horse that raced past the barn). The embedded clause in example (2) works in the same way as in example (1), as qualifier (the child that the dog bit). In (3), the embedded clause expresses a participant in the clause (or Complement in interpersonal meaning).

Embedded clauses make the analysis of English grammar challenging. Standard written punctuation may help to identify clause boundaries for co-ordinated and subordinated clauses since in these cases the clauses are in some sense separate clauses that are simply more closely textually connected than other clauses in the text. In contrast, embedded clauses are integrated as part of a clause and occur within the boundaries of an independent or dependent clause.

There have been several examples of clauses with embedded clauses throughout this book. It is actually quite difficult to find any text that only includes simple independent clauses. One example of embedding was given in Chapter 6, although the focus there was on Theme. The tree diagram for this clause (After you are wearing it securely, a tug on the hose will start the oxygen flow) is given in Figure 7.1. The embedded clause is expressing the completive element within a prepositional phrase. The entire prepositional phrase is expressing the circumstance/adjunct/Theme elements in the clause.

Figure 7.1 Embedded clause as completive in a prepositional phrase

This example is debatable because one could argue that this example is rather similar to the following: if you are wearing it securely, a tug on the hose will start the oxygen flow, where it is referring to oxygen mask. If viewed this way, one might want to suggest that after you are wearing it securely is a subordinated clause. There are reasons why I would argue against such a view, although I would like to stress that there is room for different interpretations in the application of the grammar.

It is easy when taking a retrospective view of language, as is the case in analysing grammar, to make inferences based on our understanding of patterns and how we think things work. However, the core of the functional approach to language within SFL is based on the notion of choice and, in the two possible interpretations being discussed here, the speaker either did or did not intend to include a circumstance in the situation in order to modify the process.

The question is really whether there is sufficient evidence to suggest the speaker selected a circumstance or not. The answer can be found in whether in this case after is a textual Theme introducing a subordinate clause and binding it to the independent clause, and therefore it is an element of the clause after you are wearing it securely, or whether it is a preposition element in a prepositional phrase which is expressing a circumstance function. If the completive in the prepositional phrase were a nominal group rather than a clause, there would be no need for this discussion because in that case it could not be considered as a clause at all. One test to determine the status of the clause in question is to see whether the completive could be re-expressed as a nominal group without losing the sense of the entire clause. Consider for example, after this stage, a tug on the hose will start the oxygen flow. This kind of re-expression supports the view that the unit is expressing a circumstance which functions to indicate a point in time and also simultaneously a marked experiential Theme.

Furthermore, if the clause had been non-finite, there would be no need to consider the clause as subordinated since all subordinate clauses are finite. Therefore, re-expressing the clause as a non-finite clause would also suggest that it should not be seen as a subordinated clause. Consider for example, after wearing it securely, a tug on the hose will start the oxygen flow. This suggests that the analysis given in Figure 7.1 is preferred.

Of course, we could always suggest that the clause is badly written and after is not the best expression for the meaning intended by the speaker and that once would have been more appropriate. There would be no ambiguity with once. Consider, for example, once you are wearing it securely, a tug on the hose will start the oxygen flow. In this case there is no doubt that once you are wearing it securely is a subordinated clause. However, this is not what the speaker said and we would be distracted if we started to evaluate the clause.

The next section focuses, in very practical terms, on how to identify clause boundaries in a text. It draws on the contents of previous chapters to present a set of guidelines for segmenting the text into clauses so that the analysis can be done.

7.4 Identifying clause boundaries

When first beginning to analyse text, it is a good idea to select relatively short texts. It is also a good idea to work initially with texts having a relatively straightforward or simple structural representation; in other words, the fewer the number of embedded clauses, the easier it will be to identify the clause boundaries within the text. This does not mean that such a text will be any less interesting to analyse from a functional perspective. Examples of relatively simple texts include recipes written for children, some health and safety rules, and basic instructions. However, it is often the case that texts that may appear simple to the untrained eye are in fact quite complex from the analytic perspective. Once some experience has been gained from working on these types of texts, more and more challenging texts can and should be attempted. This advice is similar to learning to do puzzles, where it is often a good idea to start off with simpler versions of the puzzle before progressing to increasingly challenging ones.

The text being analysed here is a short email from a mother to her adult daughter (see Text 7.1). It is not a particularly easy text but it is not overly challenging so it should be at the right level for our purposes here. The goal is to segment the text into clause units. In the next chapter, the full set of guidelines will be presented which will serve as a walkthrough of the approach to analysing the clause. The remainder of this section is organized in terms of the steps involved in recognizing clause boundaries.

Text 7.1 Personal email text extract: email message from November 1999, a mother emailing her daughter

It’s about 6 am and I couldn’t sleep. I have to work at 8 as I went to Timmins yesterday for my mammogram so didn’t get to work. John drove me. I said if the roads were good I would drive myself but he said he would so it was nice. Went to the fishbowl for lunch and then did some shopping. I got Robert the Thomas with the case. I was going to get him clothes as I saw some cute things but the Thomas thing is cute.

7.4.1 Identifying possible clauses

When working with punctuated text (print, electronic, etc.), the main guidance for identifying clauses comes from the punctuation marks which the speaker has included. These are the explicit signals which the speaker has used to mark up his or her text. But not all texts use punctuation regularly or in a traditional way. For example, in many types of electronic texts, punctuation is often absent or irregular. Any reliance on punctuation must be carefully considered. The first step in approaching the text is to use punctuation to help identify sentences. Each sentence becomes a possible clause; a unit that is expected to include at least one clause. The list of possible clauses based on Text 7.1 is given in Table 7.1.

Table 7.1: List of possible clauses

At this point each clause given in Table 7.1 is only potentially a single clause, as explained above. The next step is to test whether or not each possible clause is in fact a single clause or whether further distinctions are required.

The way in which this is done is by considering whether or not there are any other clauses present within the boundaries of what has been identified as a possible clause. There are only three possibilities for each; it may represent:

· one single independent clause, which may (or not) have embedded clauses within it;

· more than one clause, in which case it will have to be divided further into individual single clauses and it will be necessary to decide whether the clauses are related through

o co-ordination,

o subordination, or

o a combination of both (in the case of more than two clauses); or

· not a complete clause.

If it is not a complete clause, a potential problem arises where we need to see how it relates to the surrounding text or, in other words, we need to figure out what it is doing in the text. If the text is published writing, then it is unlikely that the text being analysed will include any incomplete clauses unless you are analysing poetry, speeches, lyrics, and so on. Typically, such a clause will not have a clear Subject or Finite verb. Rather than being a clause, the expression may simply be a nominal group (for example, a minor clause such as Good morning). It may also be an exclamative clause such as What a great day!, although some might consider this to have ellipsed it is, as in What a great day it is!

It may be the case that this clause is in fact an embedded clause and, if so, it will have a role (or function) in another clause.

Another possibility is that it simply appears not to be an incomplete clause but in fact some elements have been ellipsed and it only ‘looks’ incomplete. To be certain of this, it is important to check whether anything has been ellipsed (see ellipsis in Chapter 6). Recall that ellipsed elements are retrievable. If this is a case of ellipsis, we need to retrieve the missing bits.

Once the list of possible clauses has been identified, the actual clause boundaries should be confirmed. This involves determining which of the three possibilities given above apply. The steps for achieving this are explained below. These include: restoring any ellipsed items, identifying all verbs, verifying the word class for all verbs, identifying the verb group(s), identifying the Finite element, and identifying any textual indicators of clause combining.

7.4.2 Restoring ellipsed items

The first thing we want to do is check for any ellipsed elements (refer to Chapter 6 and the section on ellipsis). If anything has been left out but can be put back, then the analysis will be much easier if we restore the ellipsed items before beginning the analysis. This way we won’t be led astray by their absence. This step involves retrieving any ellipsed elements and putting them back in the text using parenthesis or some other visual means to mark their place for your analysis later on. It is important to mark the fact that they were ellipsed by the speaker so that this distinction is kept clear.

In Text 7.1, there are a few cases of ellipsis. They are shown in Table 7.2 using parentheses to indicate what is believed to have been ellipsed. The reference to time in these clauses requires some comment. In possible clauses [1] and [2] listed in Table 7.2, there is an ellipsed element (thing) in the nominal groups expressing clock time. This way of referring to clock time is so frequent that it is almost unanalysable. However, the numeral is referring to an amount of something. In clause [1] the am means ‘in the morning’ and, even if it isn’t a prepositional phrase, it is still functioning as a qualifier. In clause [2], this specification isn’t expressed (i.e. at 8) but this isn’t considered to be ellipsed since the qualifier is not a required element whereas the thing element is. The structure of the nominal group in clause [1] is: qd th q (i.e. quantifying determiner, thing, qualifier) and even if we would probably never express ‘hours’, it is understood. The fact that 6 is a quantifying determiner rather than thing is shown by the presence of about. This is also true for at 8 since there is the potential for this kind of modification (e.g. at about 8).

Table 7.2: Possible clauses with ellipsed items retrieved

7.4.3 Identifying all verbs

In this step, all the verbs are identified for each possible clause. This is because every clause has one and only one main verb (see Chapter 2), and if a possible clause has more than one verb it may mean that there is more than one clause. In Chapter 2, some guidelines were given for identifying verbs. The list of possible clauses is repeated in Table 7.3 with the verbs underlined and any items that might be verbs indicated in italics.

Table 7.3: Possible clauses with verbs underlined

7.4.3.1 Verify word class of verbs

This is a sub-step to identifying the verbs in the clause. Here, we simply want to make sure each word identified as a verb is really functioning as a verb and not something else like a noun or an adjective or anything else that might initially look like a verb. There are two instances where this step applies in the text we are analysing. The first is the use of work in clause [2] and the second is the use of shopping in clause [5].

Clause [2], I didn’t get to work, is ambiguous at first glance because it could be possible that the speaker is saying that she couldn’t work, as in I didn’t manage to work (cf. I didn’t get to see him). If this were the case, workwould clearly be a verb, expressing the process of working. The context of the clause indicates that the speaker was not able to go to her place of employment (i.e. I didn’t get to my office). Therefore, work in this context refers to an entity. Consequently, work is in this instance a noun and not a verb.

The second item we need to test is shopping. It occurs in the following context: I did some shopping. The ‘-ing’ suffix would normally indicate a verb. In this instance, though, shopping is being used to refer to a thing and consequently it is seen as a noun. We can be confident about this since it is preceded by a determiner, some. We could test it further by seeing whether it can be modified since this is another determining principle of nouns (seeChapter 3). Since it is possible to say I did some great shopping the other day, we can feel very confident that this instance of shopping is functioning within a nominal group as a noun rather than as part of a verb group.

7.4.4 Identify verb groups

An essential part of identifying a clause is understanding how verbs work together as a group. As explained in Chapters 2 and 5, each clause has only one main verb. If any of the possible clauses have only one verb then it is usually easy to sort out the clause because it means that there can only be one clause represented. However, if we find more than one, it can be quite tricky because we need to figure out how the verbs relate to each other, and specifically whether they are working together as a group (i.e. in a single clause) or whether they are in fact part of separate verb groups (i.e. involved in different clauses). In order to resolve this, we need to identify any verb groupings.

In Chapter 5, we saw how the English verb system works and generally there are some rather regular rules for how verbs combine with auxiliary verbs. For example, any verb following the auxiliary have will occur in the past participle form, as in John has eaten the pie. We use this knowledge to recognize verb groupings.

There are verb groupings that are less straightforward of course (see Chapter 4 for complex expressions of the process). These involve the ways in which we can extend the verb group. There are generally two types. The first are instances of phrasal verbs, where a preposition or particle attaches to the main verb and alters its main lexical meaning. An example of this type is John ran up a huge telephone bill. In this case, ran up constitute a single lexical item; in other words, there are two words working together to express the sense of ‘incurred’. In an example such as John ran up a huge tree, the meaning is quite different and up in this case does not work with the verb ran but rather with a huge tree to indicate the location where John ran. The second example shows instances of a complex of verbs and these involve more than one lexical verb. These cases can be debatable since we are trying to infer which lexical verb represents the process for the speaker. An example of this is John tried to phone her. In this case, we need to determine whether the process is try or phone. If it is try then to phone her would be seen as working as an embedded clause which expresses the Complement of try. However, if it is phone then try to phone is seen as a complex verb group which expresses the process of the clause. These problem cases will generally appear in the form verb + to + verb. In some cases, it will not be obvious which approach to take in the analysis. It is a question of the complexity of the lexical item expressing the process (experientially) and the Event (interpersonally). The relevant chapters for this are Chapters 2, 4 and 5. The key to a good analysis is to use the tools available and to be as consistent and systematic as possible.

The verb groups for each possible clause are listed in Table 7.4. Since each verb group represents a clause, we can quickly see how many clauses each possible clause has. From a glance at our revised possible clause list in Table 7.4, it is easy to see that only possible clauses [3] and [6] are single clauses. The clause boundaries for the remaining five clauses still have to be determined. To do this, we need to consider the relationships amongst the various clauses. This involves determining the finite status of each clause, which is the subject of the next step.

Table 7.4: Possible clauses with verb groups underlined

7.4.5 Finding the Finite element

The key to understanding the finiteness of the clause, as was seen in Chapter 5, is the Finite element of the clause. In order for a clause to be finite, it must have an identifiable (or recoverable) Finite element. Any clause without such an element is either a non-finite clause, or a minor clause or some other unit of structure. If the clause is non-finite (see Chapters 5 and 7), it will be embedded in another clause; in other words a non-finite clause cannot function as a main or independent clause.

As explained in Chapter 5, the Finite element is always expressed by the first element of the verb group. If a modal verb (e.g. may, can, shall, will) occurs in a clause, it expresses the Finite element. If there is no modal verb, then the Finite element will be expressed by a primary auxiliary verb (e.g. be, have, do) or the Event (e.g. eat, drink, sleep).

It is important to remember that it can be challenging to recognize the Finite in the imperative mood since it is often ellipsed. As explained in Chapter 5, it is expressed by an empty verbal element but it is easily recovered by negating the clause (e.g. eat your food don’t eat your food).

The revised list of possible clauses given in Table 7.5 now includes a subscript F to indicate which item is expressing the Finite element in each verb group.

Table 7.5: Possible clauses with Finite identified by subscript, F

7.4.6 Verifying clause boundaries

We now have all the information needed in order to verify the boundaries of the possible clauses. As stated above, there are three outcomes for each possible clause. Recall that we initially identified these clauses using the speaker’s original punctuation. What we see from the clause list above is that most cases (all but clauses [3] and [6]) have more than one verb and more than one verb group, which represents more than one clause. Therefore, we have to determine how the multiple clauses relate to each other. There are three ways in which two clauses can be related: co-ordination (one clause relates to the other through parataxis); subordination (second clause relates to the other through hypotaxis); or embedding (one clause relates to the other through embedding). The following aids can be useful for recognizing these types of clauses.

· Co-ordinated clauses can often be identified by linking textual themes such as and, but, or, or punctuation such as ‘;’ or ‘:’.

· Subordinated clauses will usually be introduced by binding textual themes such as if, as or because.

· Embedded clauses will fill a function of an element of a higher unit (e.g. participant, circumstance, qualifier, completive, finisher) and may be introduced by a relative pronoun (e.g. that).

Using the information we have gained from the previous steps and our knowledge of how clauses combine (through co-ordination, subordination or embedding), we can now verify the actual clause boundary for each possible clause, as shown in Table 7.6.

Table 7.6: A comparison of possible clauses and verified clauses, with textual indicators in bold

What can be highlighted from this step is that there are no verb groups with non-finite verbs in the text which is being analysed. This means that there are no instances of non-finite embedded clauses. In order to show that this step is needed when the text does include a possible clause with a non-finite verb group, we will look briefly at another clause from a separate email message (also from November 1999). We will return to the problem of verifying all clause boundaries in a moment.In (4) above, in order to verify the clause boundaries, we need to consider whether we find more than one verb group which has a Finite verbal element. In this example, there is only one Finite verbal group (’s) and therefore we can conclude that the possible clause here does in fact indicate a single main clause. This verifies that the possible clause boundaries are in fact the actual (or confirmed) clause boundaries. However, we are left with the problem of what to do with the other verbal groups indicated by to sit, reading and sleeping. We know that these constitute embedded clauses because all non-finite clauses are embedded and have a role in the main clause. Determining the role of the embedded clauses is handled at a later step (as shown in Chapters 4, 5 and 6) when we use the process test (see Chapter 2) and work out the internal boundaries of the clause (see Chapter 3) in order to complete the analysis.

(4) It’s hardly normal for an 18 year old to sit at home reading and sleeping all the time.

We have now completed the goal of identifying the clause boundaries. The original seven possible clauses have resulted in fifteen main clauses, as shown in Table 7.7.

Table 7.7: List of verified clause boundaries

What has been presented here may well represent the most difficult part of grammatical analysis. The clause itself is almost impossible to define, making it difficult to recognize in text. Its composition and configuration alters with each metafunction, which is why a full understanding of the clause is needed before it can be identified and described in text.

7.5 Summary

Now that the clause boundaries have been verified, the analysis of each clause can begin. To illustrate this, Chapter 8 presents the full set of guidelines for grammatical analysis. These guidelines were introduced and developed in Chapters 4, 5 and 6. In a sense, everything that has been covered to this point has been in preparation for the grammatical analysis. Before a text can be functionally analysed, it has to be segmented into clauses, but before it can be segmented into clauses, the components of the clause must be understood so that they can be relied upon in the identification of the clause as a unit.

The relationship between clause and text is similar to the relationship between functional elements and structural units that we have been discussing throughout this book. Just as the meaning of Actor is expressed or realized by a nominal group, the meaning of the text must also have a form in order for its expression to be realized. As explained by Halliday and Hasan (1976: 2), ‘a text does not consist of sentences; it is realized by, or encoded in, sentences’. This is why a functional analysis of the clause leads to an understanding of text. The next chapter will offer a summary of the approach to grammatical analysis that has been developed in this book. Then, Chapter 9 will demonstrate an example of how grammatical analysis can be interpreted in terms of the meaning of the text. In this sense, these three chapters – Chapters 7, 8 and 9 – combine to form a full cycle of analysis and interpretation, moving from text to clause to text again.

7.6 Exercises

These exercises give you practice in working with text and identifying the clause boundaries. For each text, use the guidelines developed in this chapter to list all the finite clauses.

Exercise 7.1

I always get to this computer later at night. John is out golfing and Jane is at a sleepover birthday party where they are sleeping outside in tents and Sue has three friends over for a sleep over. They are watching a movie now.

(Excerpt from a personal email written by an adult female to a female friend, June 2005)

link to answer

Exercise 7.2

When it first happened, there was a big thunderstorm that shook the house and the rain fell really fast. My brother was startled because he was outside. Now the water is knee-high but we’re alright. We went canoeing to a nice park which is really fun! We saw some iguanas today, and we even had a black snake at our house and I saw a snake on a canoe too! Every time I go out we go out in a canoe or our dad carries us because me and my brother don’t like going out in the water because of the snakes. We should be going back to school in three weeks. It’s a long time off.

(BBC, CBBC Newsround. 2011. Press Pack Reports: I’m stuck in the Australian floods. http://news.bbc.co.uk/cbbcnews/hi/newsid_9340000/newsid_9341900/9341995.stm)

link to answer

Exercise 7.3

The future must see the broadening of human rights throughout the world. People who have glimpsed freedom will never be content until they have secured it for themselves. In a truest sense, human rights are a fundamental object of law and government in a just society. Human rights exist to the degree that they are respected by people in relations with each other and by governments in relations with their citizens. The world at large is aware of the tragic consequences for human beings ruled by totalitarian systems. If we examine Hitler’s rise to power, we see how the chains are forged which keep the individual a slave and we can see many similarities in the way things are accomplished in other countries. Politically men must be free to discuss and to arrive at as many facts as possible and there must be at least a two-party system in a country because when there is only one political party, too many things can be subordinated to the interests of that one party and it becomes a tyrant and not an instrument of democratic government.

(Roosevelt, Eleanor. 1948. The Struggle for Human Rights. [speech] Paris, France, 28 September 1948. http://edchange.org/multicultural/speeches/eleanor_roosevelt_rights.html)

link to answer

7.7 Further reading

Although no book specifically addresses guidelines for recognizing clause boundaries, the following books include sections which discuss how clauses combine.

Bloor, T. and M. Bloor. 2004. The Functional Analysis of English: A Hallidayan Approach. 2nd edn. London: Arnold.

Halliday, M. A. K. and C. Matthiessen. 2004. An Introduction to Functional Grammar. 3rd edn. London: Hodder Arnold.

Martin, J., C. Matthiessen and C. Painter. 1997. Working with Functional Grammar. London: Edward Arnold.

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