7. Analysis
7. AnalysisThe system of Theme is concerned with the organization of information within individual clauses (Martin et al. (1997, p.21)), mainly since the clause is generally considered to be the pivotal unit of grammatical meaning (Eggins (1994, p.139)). Consider Example 2:
Example 2:
If you have any questions, I would be happy to discuss them with you.
In this example the sentence is built up of two clauses, the first dependent and the second independent. If we analyse Theme at clause level there are two Themes ('If you' and 'I'), which are both italicized in the example. However, if we wish to analyse the Theme at the sentence level, the whole if-clause serves as the Theme, which has been underlined in the example. In this study, the analyses are restricted to Themes at clause level rather than analysis of Theme at sentence level.
The texts contained a total of 465 clauses (both independent and dependent). All clauses were treated equally and each clause was analysed for Theme according to Halliday's model (1994, p.37ff) to investigate what linguistic elements are typically chosen as Theme. Table 2 below presents an overview of the types of linguistic elements that are given Thematic status.
| Types of Theme | N | % |
| Mainly pronouns (Unmarked Themes which serve as subject) (Independent clauses: 247=88% , Dependent clauses: 33=12%) | 280 | 60% |
| Circumstantial, modal, textual adjuncts (Marked Themes which serve as adjunct) (Independent clauses: 98=71% , Dependent clauses: 41=29%) | 139 | 30% |
| Verbs (All independent clauses) | 22 | 5% |
| Wh-elements (Independent clauses: 11= 92% , Dependent clauses: 1= 8 %) | 12 | 3% |
| Elements used for politeness purposes (All independent clauses) | 12 | 3% |
| Total | 465 | 101% |
Table 2: Types of Theme
As is shown in Table 2 most of the Themes are either pronouns or various types of adjuncts. Furthermore, Themes represented by verbs, Wh-elements and elements used for politeness purposes [9] present a smaller proportion. To clarify the analysis further, example 1 above would have been categorised as an unmarked Theme since the topical Theme is a pronoun and serves as subject in the example.
Since mood [10] is part of the Thematic structure of a clause, the choice of Theme in a clause is related to the choice of mood. Clauses with a subject as Theme is the typical pattern for a declarative clause and are said to have unmarked Themes [11]. Declarative clauses with Themes other than a subject are said to have marked Themes. Adjuncts (adverbial groups, prepositional phrases) and Complements (nominal groups, nominalizations) are typical marked Themes. Clauses which have a verb (the predicator) as Theme are regularly labelled imperative clauses and can function as either unmarked or marked Themes (Halliday (1994, p.47)). Wh-elements which are given Thematic status are found in interrogative clauses where the Theme is constituted only by the Wh-element. Themes in interrogative clauses are mainly unmarked but occasionally marked Themes do occur (Halliday (1994, p.47)). In Table 3 the distribution of the unmarked and marked Themes in declarative clauses are shown.
| Declaratives | N | % |
| Unmarked (Independent clauses: 247=88% , Dependent clauses: 33=12%) | 280 | 60% |
| Marked (Independent clauses: 98=71% , Dependent clauses: 41=29%) | 139 | 30% |
| Total | 419 | 90%* |
Table 3: Unmarked and marked Themes in declarative clauses (* For the remaining percentage, see the discussion below)
Examples of unmarked and marked Themes from the data are shown in italics and in Examples 3 and 4:
Example 3:
I am a Ph.D. candidate in the Linguistics department … (unmarked)
Example 4:
As is illustrated in Table 3, the unmarked Themes are much more frequent with 60% of all clauses than the marked Themes with 30% of all clauses. This result indicates that the remaining part of the clauses, 10% of all clauses, have Themes which categorize the clauses as something other than declaratives.In Fall 2001 - Fall 2002, I will be teaching introductory courses in…(marked)
Table 4 presents an overview of the various types of clauses found in the data.
| Types of clauses | N | % |
| Declaratives (Independent clauses: 345= 82% , Dependent clauses: 74= 18 %) | 419 | 90% |
| Interrogatives (Independent clauses: 11= 92% , Dependent clauses: 1= 8 %) | 12 | 3% |
| Imperatives (Independent clauses: 22= 100%) | 22 | 5% |
| Minor clauses (Independent clauses: 12= 100%) | 12 | 3% |
| Total | 465 | 101% |
Table 4: Distribution of types of clauses
The analysis shows that most clauses, 93%, are in the indicative mood. Declaratives is the largest group with 90% of all clauses and interrogatives represents 3% of all clauses. Twenty-two clauses are in the imperative mood which represents 5% of all clauses. I have added and labelled the final category, minor clauses (cf. Martin, Matthiessen and Painter (1997, p.71)), which are said to fulfil a minor speech functional meaning. According to Halliday (1994, p.43), minor clauses do not have Thematic structure and are usually left out of account when clauses are analysed for Themes. However, in my data minor clauses are typically politeness elements used to maintain harmonious relations in the discourse and can thus be seen to serve an interpersonal function in the dialogue between the writer and the reader. Minor clauses are therefore important linguistic features, which need to be taken into consideration to increase our understanding of the nature of the interaction going on between the writer and the reader of homepage discourse. Examples 5 and 6 below are some examples of minor clauses taken from the data, where the politeness element has been written in italics:
Example 5:
Thanks for stopping by…
Example 6:
Welcome to my homepage.
In the following sections the Themes in the various types of clauses will be discussed in relation to the homepage discourse. First there is a section discussing the unmarked and the marked Themes in declarative clauses followed by a section where the Themes in interrogative, imperative, and minor clauses are discussed.
As shown in Table 3 the Themes consisting of the subject make up the largest group of the declarative clauses with 60% compared to the marked Themes with 30%. The majority of the unmarked Themes consist of pronouns. Most frequent is the personal pronoun I in initial position followed by the possessive pronoun my. Other pronouns like you and we are also represented but not to the same extent [12]. Table 5 below presents the frequency of the pronouns found in the data.
| Types of Pronouns | N | % |
| I | 232 | 79% |
| my | 44 | 15% |
| you | 13 | 4% |
| we | 5 | 2% |
| Total | 294 | 101% |
Table 5: Frequency of pronouns
One possible explanation for using the same clause initial element may be that the writers make use of thematic repetition, e.g. the personal pronoun I, to create topical coherence, i.e., the topic of the homepage as a whole is I. The repetition of the same subject becomes a way for the writers to structure their text coherently.
Brown and Yule (1983, p.183) point out that it is reasonable to assume that the constituent which is thematised in a sentence is "what the sentence is about". And since the majority of the homepage writers have placed the personal pronoun I as the Theme, it underlines that their texts are about themselves, as is illustrated in examples
Example 7:
I am a 4th year doctoral student in the linguistics program…
Example 8:
I have a pretty vast research agenda that includes…
Example 9:
I have spent a good deal of time researching various…
Furthermore, Brown and Yule (1983, p.134ff) discuss the notion of Theme and 'topic entity' and suggest that Theme (following Halliday) refers to the left-most constituent of a clause while topic entity refers to Theme as meaning main character. In Examples 7 and 8 the main character is the writer him-/herself and so the writer's topic entity is the first person pronoun I. From an interactional point of view, the initiating pronoun I indicates that the writer assigns himself the role of informer in the discourse, informing the reader about him-/herself.
When the writers place the pronoun you as the Theme, they are always directing the readers of the homepage, which shows that there is an active interaction going on, a dialogue, between the writer and the reader through the homepage discourse. Examples 10 to 14 illustrate the use of the pronoun you in the data. The Theme is italicized and the pronoun is underlined.
Example 10:
You can send me e-mail at …
Example 11:
You may be interested in my resume.
Example 12:
If you have any questions about myself, my research, or…
Example 13:
..., or if you have comments or suggestions about this webpage…
Example 14:
..., or if you prefer a more Metropolitan approach …
The pronoun we also appears as Theme, as in Examples 15 and 16:
Example 15:
We have a web page, which is put together by…
Example 16:
We are up to about 120 gigs of data…
Here the writers are using we when they are mentioning various projects that they are working on together with other academics. As for the use of the possessive pronoun my it is always connected to the writers' academic ideas, research, interests, papers, dissertation, address or department, as in Examples 17 to 21, where the whole Theme is in italics and the pronoun is underlined.
Example 17:
My idea is to…
Example 18:
My research focuses on…
Example 19:
My research interests include…
Example 20:
My Qualifying Paper for PhD Candidacy examined…
Example 21:
My campus mailing address: …
In seven examples does the subject in Theme position refer to the homepage itself as in Examples 22 to 24:
Example 22:
This site includes…
Example 23:
This page was last updated…
Example 24:
This is the personal web page of …
These Themes are metalinguistic and they may be said to function as spatial deictic expressions. The examples above suggest that the writer puts his/her own homepage in relation to other homepages in the vast cyberspace. One's own personal homepage is emphasized in relation to other sites, showing an awareness of the homepage as a new phenomenon in a larger context - the Internet.
The marked Themes are the Themes not conflated with the subject. It is interesting to see what kind of elements are chosen as clause initial elements besides the subject. As shown in Table 1, 30% of all Themes are adjuncts. There are three types of adjuncts, all with different functions. These are circumstantial adjuncts (prepositional phrases, adverbs of time, manner, place etc.) which add experiential content to the clause by expressing some circumstance relating to the process represented in the clause; modal adjuncts (mood, comment, vocative adjuncts) which add interpersonal meanings to the clause by creating and maintaining the dialogue between writer and reader; and textual adjuncts (conjunctive and continuity adjuncts) which add textual meaning to the clause to organize the message itself. Table 6 presents the distribution of the various types of adjuncts that were found in Theme position.
| Textual Adjunct | Modal Adjunct | Circumstantial Adjunct | Total | |
| N | 16 | 17 | 72 | 105 |
| % | 15% | 16% | 69% | 100% |
Table 6: Distribution of adjuncts in Theme position
Table 6 shows that the circumstantial adjuncts are the most frequently used adjuncts in Theme position, followed by modal adjuncts and textual adjuncts. Below follow examples of the various types of adjuncts. Some of the Themes are multiple Themes and some of them are simple marked topical Themes (fronted circumstantial adjuncts). Therefore, the whole Theme is in italics and the specific adjunct exemplified is underlined.
There were 16 instances of a textual adjunct in the data. In Example 25 the textual Theme is underlined and provides a linking relation to the previous clause. Examples 26 to 29 are first and foremost fronted circumstantial adjuncts and have ideational meaning: temporal in Examples 27 and 29, and manner in Example 26. However, they all have a text organising function in the homepage discourse and can therefore be viewed as textual. Furthermore, Example 28 also functions as a metalinguistic Theme, referring to the homepage itself.
Example 25:
In addition to Creoles, I am also interested in…
Example 26:
That way I get to combine my affinity for…
Example 27:
Eventually, we hope to make all…
Example 28:
And last but not least, I am currently…
Example 29:
After that I scooted off to Brussels…
The modal adjuncts represent 16 % of the adjuncts in Theme position and they add interpersonal meanings to the clause, i.e. they add meaning which is connected to the construction and the maintenance of the dialogue between the writer and the reader of the homepage. There are clauses which start with modal adjuncts (underlined in the examples) which convey the writers' judgement of the relevance or truth value of their messages (Halliday (1994, p.49)). Examples are:
Example 30:
Perhaps, it was the early exposure…
Example 31:
There are also other types of modal adjuncts (see Eggins (1994, p.166)). In Example 32 a comment adjunct (Halliday (1994, p.83)) is presented, which adds expression of the writer's attitude and evaluation. Example 33 is a focusing adjunct, expressing the writer's specific focus.Hopefully, my learning pursuits will make…
Example 32:
Broadly, the goal is to find out …
Example 33:
Particularly, I am interested in …
The majority of the adjuncts, 69%, are circumstantial adjuncts, which add meanings which refer to time (rather than to space, cause or matter), as in Examples 34 to 38:
Example 34:
During the summer following my graduation, I worked as a…
Example 35:
The following September I entered…
Example 36:
In April 2000 I completed…
Example 37:
Currently, I am …
Example 38:
At present I am in large part funded by…
The above examples are like Examples 26 to 29, i.e. simple marked topical Themes or fronted circumstantial adjuncts, which add ideational meaning to the clause. However, it is important to point out that Examples 34 to 38 also have textual meaning since they contribute to the organization of the discourse.
In the analysed texts, the circumstantial elements have a very important role in that they create the background of the message of the clause. Usually, the writers want to give an account of their academic career and therefore go back in time to talk about past achievements. Then they move on to discuss their current academic situation, and finally finish by telling the reader about their academic plans for the future. The Theme is developed on the basis of the authors' achievements so far.
Figure 3 below is a screenshot of one personal homepage to illustrate the use of circumstantial adjuncts.
Figure 3: Illustrates the autobiography and education page of David Rojas.(http://www.rojasweb.net/david/autoed.html)
To present the text in the homepage more clearly it has been copied and the circumstantial adjuncts are marked in bold in Example 39:
Example 39:
Homepage of David Rojas, Linguistics, Indiana University, http://www.rojasweb.net/david/
I attended Jesuit High School, devoting myself to the Blue Jay Band and Marine Corps JROTC. I graduated from Jesuit in 1993 and then situated myself at The University of Alabama in Tuscaloosa, Alabama, where I created an undergraduate special studies program in Linguistics. During my stint in Tuscaloosa, I particpated in various music ensembles (Wind, Percussion, Jazz, and the Million Dollar Band). In May of 1997 I graduated from Alabama then moved on to Louisiana State University, from which I received my Master of Arts degree in French Linguistics in December of 1998, having worked on a National Science Foundation funded research project on Cajun French and English. After that I scooted off to Brussels, Belgium (see the Journal de Belgique pages) for a while then returned to the Big Easy for a stint at home. In August 2000 I began work on my PhD at Indiana University in the Linguistics Department and Cognitive Science Program. At present I am in large part funded by the Creole Institute, where I'm working on the Louisiana Lexicon Project (or LADICO)--link to come after I get the public site up and running.
The readers of the homepage, however, will not be able to make sense of the deictic expressions in Theme position unless the writer refers to time that can be deictically interpreted in relation to the reader's experience of time. Deictic expressions such as currently and at present can thus be ambiguous and writers must anticipate the effects of the time lag between the production of the clause and the reception (cf. Levinson's coding time and receiving time, Levinson (1983, p.73)). However, due to the asynchronous nature of the medium, homepage writers usually do not update their personal homepages by the minute. Consequently, placing adjuncts of time in Theme position of a clause restricts the actual statement or message of that clause with regards to time. This problem is sometimes solved by the writer of the homepage by stating the date [13] when the page was last updated, which helps the reader to interpret the deictic expressions in the texts. In this way the date when the page was last updated becomes the deictic centre for the reader. The deictic centre from where the deictic expressions should be interpreted originates from the writer not the reader. Hence, the deictic centre becomes of great importance for the reader's understanding and experience of the coherence of the homepage text.
In addition to declarative clauses, Table 4 above also shows that the data consist of other types of clauses. This section investigates interrogative clauses, imperative clauses and minor clauses.
The characteristic function of an interrogative clause is to ask a question (Halliday (1994, p.45)). When a question is asked, an answer or response is usually required. This is the basic pattern of an interaction between speaker/hearer and in this study between writer/reader. In the data, 3% of the clauses are interrogative clauses which have Themes composed of question words, or Wh-items such as Where, Why, When, How (Martin et al. (1997, p.30)) which solely constitutes the Theme in the clause. In English the writer or speaker does not actually choose to place the Wh-element in Thematic position, it is part of the system of the language. It is in the very nature of a question in English to have the Wh-item in initial position of the clause (Halliday (1994, p.46)).
The interrogative Wh-elements can add both interpersonal and topical meanings to the clause (Halliday (1994, p.54), Table 3(7)). However, to illustrate the degree of interaction going on in the homepage discourse I will merely consider the interpersonal meanings that the Wh- elements add to the clause. Examples of Wh-elements (Themes) from the data are in italics in the following examples:
Example 40:
Why bother?
Example 41:
What is an American anyways?
Example 42:
Who am I?
Example 43:
What are they?
Example 44:
Where would you like to go now?
In the above examples, the writer is asking questions which are directed to the reader of the homepage. At the same time, however, in stating these questions the writer is not really expecting the reader(s) to contact him/her and provide an answer to the questions, hence the interrogatives in the above examples can be viewed as rhetorical. Yet, this interpretation does not minimize the value of the interactive contribution of the interrogatives to the homepage discourse. The questions demonstrate that there is a degree of involvement of the writer in the reader of the discourse.
As shown in Table 4, the imperative clauses constitute 5% of the clauses in the data and can be divided into two main groups based on the function of their Themes in the discourse. There are imperatives with Themes adding meaning of politeness to the clause and there are Themes adding meaning of physical action to the clause.
The basic function of imperative clauses is either 'I want you to do something' or 'I want us (you and me) to do something'(Halliday (1994, p.47)), and this function is indicated by the Predicator (or the verb) which is given Thematic status in the clause. The majority of the imperatives in the data encourage the reader(s) to perform a physical action of some sort connected to the design and the structure of the homepage. Examples of Themes where the writer requests the reader(s) to carry out an action are illustrated in Examples 45 to 47 below (the Themes are in italics).
Example 45:
Click here for our web page.
Example 46:
Click here for my CV.
Example 47:
Browse the menu to your left for more information.
The examples above also have a metalinguistic character since the writer is giving the reader(s) instructions of how to surf the homepage itself, to get the information wanted.
The other group of imperatives are clauses where the Theme mediates a sense of politeness directed towards the reader. According to Thomas (1995, p.157f) politeness strategies are used to achieve certain goals. Consider the following examples:
Example 48:
Please feel free to check out some of my work.
Example 49:
Please enjoy some of the images I've put up of some of my work.
The goal in this context is for the writer to create a polite and harmonious relation with the reader through the discourse by using polite imperatives ('feel free', 'enjoy') intensified by the initial use of the politeness marker 'please'.
The final clause category shown in Table 4 is that of minor clauses, which represents 3 % of the clauses in the data. Following Martin et al. (1997, p.71) and Halliday (1994, p.95f), minor clauses are said to have a semantic role in dialogue and hence serve an interpersonal function. In the homepage data these types of clauses are to a large extent greetings and exclamations. Consider the following examples:
Example 50:
Hello!
Example 51:
Hello, everyone.
Example 52:
Hello, my name is…
Example 53:
Welcome to the homepage of…
Example 54:
Welcome to my webpage.
Example 55:
Thanks for visiting my website.
In the above examples there is clear evidence of the writer's concern for creating and maintaining a polite and friendly atmosphere in the discourse. These minor clauses can be compared to greeting clauses used in traditional writing, e.g. letter writing. In traditional letter writing we would not find clauses like those in Examples 53 to 54 which are commonly used in this new medium. This is much due to the fact that on the Internet a reader can actually visit different places, one of which may be a personal homepage. The result is that the writer must adapt the text to the context and therefore finds it necessary to invite the reader(s) to the personal homepage, almost the same way as when inviting someone to a home (hence term 'homepage' [14]).
