Daniel Kies Department of English College of DuPage
Modern English Grammar English 2126		Contact Form

Indeterminacy in Sentence Structure

Originally published in Linguistics and Education 2, 231-258 (1990)

Abstract

Current work:   Days remaining this term: Notes: Add Note |  Log in? | Privacy | Change Name & Email

Mail this page to a friend

Earlier research in English grammar and discourse analysis demonstrates how thematic and informational structures in the English clause shape the structures of English texts, cf. Halliday (1985) and Quirk, Greenbaum, Leech, and Svartvik (1985). That is, as writers compose and revise, they must master and choose the syntactic structures that allow them to vary thematic prominence and information focus in their texts, cf. Dillon (1981). This article, based on a (150,000 word) corpus of college compositions, analyzes the discourse functions served by thematic organization and information focus in clauses of college writers. By studying thematic and informational organization in the texts of composition students, one can capture important generalizations about the relationship between linguistic form and linguistic function, discovering that for many college composition students specific linguistic forms are indeterminate between the functions of parataxis and hypotaxis. Those generalizations explain some of the sentence structure faults common in the writing style of college composition students.

1. Introduction.

Halliday (1987) describes some syntactic and semantic differences between the spoken and written modes of communication. Hoey (1986), Van Valin (1984) and Winter (1977) describe the syntax and semantics of clause connection, focusing largely on several English registers. This study contributes to both of those discussions in that the analysis of the prose of college composition students presented below begins with a discussion of differences in the functions of focus and theme between the two modes of communication and then moves toward some insights about the syntax of clause connections and the semantics of coordination and subordination.¹

The analysis here explores college composition both paradigmatically and syntagmatically in context. The paradigmatic aspect highlights the writers' choices of focus and theme in their clauses, and through the examination of choice, this study offers a stylistic analysis of the clause in college composition. The syntagmatic aspect highlights the syntax and semantics of the writers' clausal connections, and through that analysis, this study suggests that clausal connections in college composition are indeterminate, not clearly exhibiting coordination or subordination. Rather, clausal connections are gradient phenomena, existing on a cline between the function of parataxis (realized by coordinated, juxtaposed, or adjacent clauses) and the function of hypotaxis (realized by subordinated and relative clauses).

Previous research in linguistics has drawn attention to the indeterminacy of such basic concepts in the Western grammatical tradition as subordination and coordination (cf. Andersson, 1975; Haiman & Thompson, 1984; Kuno, 1973; and Van Valin, 1984). This study departs from the earlier research into indeterminate structures in several ways. First, the earlier research has attempted to demonstrate the cross-linguistic inadequacy of a Western grammatical concept (subordination). This study demonstrates that the same concept is also suspect in English. Second, this analysis presents unified description of subordination in English through the concept of gradience.

Previous research in the field of composition has also investigated language, communication, and style. Some researchers, for example, attend to the language used by college writers, learning about language development through error analysis and/or text structure (cf. Shaughnessy, 1977; Colomb & Williams, 1985; Mann & Thompson, 1986; or Connors & Lunsford, 1988). Other researchers concentrate on composition as a communicative act, uncovering the psychology and sociology underlying the composition process (cf. Brandt, 1986; Cooper, 1986; Faigley, 1986; and Flower, Hayes, Carey, Schriver, & Stratman, 1986). Still others examine writing style in college composition, studying the writer's linguistic choices so that researchers may learn about both the production and the structure of composition (cf. Strong, 1985; Williams, 1986; or Witte, Daly, & Cherry, 1986).

This study departs from much of the previous research into composition by presenting a functional linguistic analysis. The earlier research often focuses on the frequency with which certain linguistic features occur, yet another (more interesting) description focuses on the functions of those features. Functional descriptions of language — like the one offered here — are more valuable since they offer some understanding of communicative purpose and, thus, explain the use and frequency of linguistic features.

The analysis is then extended in two respects: first, to several registers of English to provide independent motivation for the notion of indeterminacy and, second, to propose a hypothesis about language development, involving para- and hypotactic structures.

More explicitly, I hypothesize

a. INDETERMINACY

The clausal relations are syntactically and semantically indeterminate, ambiguous between parataxis and hypotaxis.

 

b. GRADIENCE

Although many clausal relationships are indeterminate, clausal relations are organized along a gradient with parataxis at one pole and hypotaxis at the other.

 

c. STYLISTIC MOTIVATION

Many stylistic errors in college compositions originate through the indeterminacy of clausal relations; this relationship between indeterminacy and style is a developmental phenomenon, evidenced in several varieties of English.

I define the theoretical underpinnings of this inquiry in section 2 and then focus on two types of clauses (nonrestrictive relative clauses and adverbial clauses) in section 3 as evidence for the three hypotheses mentioned above.

 

2. Information focus and thematic structure.

2.1. Speech and writing: intonation and word order.

A number of scholars studying language from a functional point of view (cf., Danes, 1964; Firbas, 1964, 1966; Halliday, 1985; and Quirk, Greenbaum, Leech, & Svartvik, 1985) have recognized the role that the linear order of words plays in written texts to compensate for the loss of explicit clues about information focus available through intonation in the spoken language. In spoken English, emphasis is achieved by manipulating intonational contours so that informationally important elements receive nuclear stress in tone units. Since those tone units can be conveyed directly, this creates few problems. In written English, however, those tone units must be inferred from their graphic representations. Consequently, the reader has to rely more heavily on certain conventions of interpretation — especially conventions that relate the structure of the tone unit to the structure and linear ordering of the syntax. While speakers can directly convey their emphases through stress and pitch, the writer must construct a clause, indeed the whole sentence (and text), carefully — with judicious word order, punctuation, and discourse implication (about what the writer considers given or new information for the audience) — to recapture the loss of such explicit markers of emphasis in speech.

The loss of prosodic markers of information focus in written English highlights a major problem that arises for all writers who must rely on word order to carry information focus. That problem might be called the poverty of surface syntactic information, i.e., the restricted number of distinctions linear order can make as an information carrying device. Linear order can provide only two possible pieces of information: (1) two constituents can be sequential or not (i.e., linear order can describe adjacency relations) and (2) if the constituents are sequential, their order may be either X - Y or Y - X (i.e., linear order can describe precedence relations).

Through the relationships of adjacency and precedence, linear order, supplemented by morphology (and intonation in speech), provides information about grammatical relationships of subject, object, etc; about thematic structure of theme/rheme or topic/comment (the psychological subject discussed by Sandmann, 1954); about the participant roles of agent, patient, etc (the logical subject of Sandmann, 1954); and about information structure of given and new (cf., Prince, 1981). As Chafe (1976, p. 27) puts it, "A noun in its sentence plays many roles, or has the potential of doing so."

So part of the difficulty in determining the contribution of linear order to one's understanding of language results from the interplay of various language processes. Those linguistic processes conspire to determine the linear order of clausal constituents and the linear order of clauses themselves. Out of context, or in a controlled context, it is possible to isolate the functions of end-focus, thematic prominence, or euphony in determining linear order, but in vivo, as it were, it becomes more difficult to characterize precisely the contribution of individual language processes.

Neutralization processes provide an analogous situation to the conspiracy described here. Neutralization rules, at any level of linguistic analysis, eliminate a potential contrast, thereby creating the potential for ambiguity. At some level of analysis, one would want to explain the ambiguity by positing different forms, which are no longer overtly distinguished at the surface level because of a neutralization process. Likewise, different sentence types seem to neutralize some distinctions between grammatical, thematic, psychological, and logical subject, for example, in order to express some distinction that otherwise may be missed. That is, in appropriate contexts, language users may need to be explicitly clear about information structure, or thematic structure, etc, for efficient language processing. For example, when speakers need to be explicit about the packaging of information within a clause, the cleft and pseudo-cleft constructions allow a distinction between given versus known information.² And as Pawley and Syder (1983) argue, certain structural strategies (or packages of information) may be better suited in certain communicative contexts because of physical or cognitive constraints, and many of those constraints vary as context varies, arguing against any notion of an ideal relationship between form and function.

To compensate for the poverty of surface syntactic information in written English, writers commonly employ two conventions or strategies of interpretation — end-focus and thematic structure.

 

2.2. The principle of end-focus.

One strategy for interpreting the focus of information from syntax is the principle of end-focus (Quirk, Greenbaum, Leech, & Svartvik, 1985, p. 1356-57). There is variation in the information value of different parts of an utterance or clause. To interpret the following question and answer pair What will the Smiths do now? They will leave tomorrow., the subject, verb, and adjunct of the answering sentence carry increasing prominence in information value. The subject, they, conveys the least information since it is pronominal and unstressed prosodically, indicating its coreference with another noun phrase earlier in the discourse, the Smiths. The verb is slightly more prominent in that it is not wholly predictable, even though its form is in part anticipated earlier in the discourse (in the mood element of the question). The information focus is on the adjunct. It is the least predictable item; it is given intonational prominence in the tone unit; it occupies the focal (clause final) position.

Although information focus highlights one item, this does not mean that the rest of the clause is without information value. Rather, the information value of the different clause elements are relative, from low information value in the pronoun they, expressing old, given information, through medium in the verb will leave, predictable in form (will + V) but still non-recoverable (in leave), to high in the adjunct tomorrow, the element receiving intonational prominence.

Thus, it is common — though not necessary — in both speech and writing for the information value of a clause's elements to increase from low to high as one proceeds linearly through the clause. The clause-final element carries the (relatively) highest informational prominence, or "end-focus" as Quirk, Greenbaum, Leech, and Svartvik (1985, p. 1357) put it: "... it is common to process information in a message so as to achieve a linear presentation from low to high information value. We shall refer to this as the principle of END-FOCUS".

Whenever a clause appears in an unmarked (i.e., normal) word order, focus is given to (the stressed syllable of) the last major constituent (open-class lexical item) in the clause. Intonational prominence is realized by a syllable, a unit that has no necessary semantic correlation or informational value. It is more accurate, therefore, to characterize the effect of prosodic prominence as the ability to draw attention to the semantic/grammatical unit in which the prominent syllable occurs. In that case, the entire unit shares the information focus.

For example, in a written sentence like John ate the apple, readers usually interpret the focus of information in that clause to be the last major constituent, the apple, because in speech the pitch slide and heavy stress associated with information prominence would occur on the grammatical object of the clause, as in ||John |ate the 'APple||. That is the usual, unmarked state of affairs. Yet in speech, prosodic features (particularly stress and pitch) can combine to signal changes in focus, although the word order remains constant. Consider the sentences in (1) below where italics indicates the pitch slide and heavy stress characteristic of focus intonation in speech.

1. a. John ate the apple.
        [focus on John, rather than Bill, Joe, etc]
   
   b. John ate the apple.
        [focus on ate, rather than saved, threw, etc]
   
   c. John ate the apple.
        [focus on the apple, rather than grapes, pears, etc]
   

In writing, however, that same sentence would most likely be interpreted as (1c), the neutral or unmarked interpretation. Sentences (1a) and (1b) are marked structures in that they exhibit contrastive stress, which highlights the elements John and ate from their respective presuppositional sets. To signal changing focus in the written language, without resorting to some graphic device like italics, a writer would resort to stylistic variants of the basic, unmarked clause in (1c) above. Clefting, for example, is one syntactic (and stylistic) device a writer could employ to signal information focus on John or ate by making those constituents the focus of their respective clauses and tone units. Consider (2a) and (2b) below.

2. a. ||It was `JOHN |who ate the `APple||
   
   b. ||What John did with the `APple |was `EAT it||

Of course, some of these alternatives in achieving emphasis are not equivalent in other ways. Some require more words and thus sacrifice economy; some rely on grammatical processes that alter the force of the original; some are aesthetically equivalent to the original; some are not; some sacrifice the rhythm of the original; and so on. Those questions of appropriate focus, economy, aesthetics, and rhythm all demonstrate the conspiracy at work again when writers choose a sentence like (2b) over (1c) or (2a).

 

2.3. Thematic structure.

Likewise, the thematic structure of the clause is equally important in interpreting the message in a clause. The initial constituent of a clause serves as "the point of departure for the clause as message" (Halliday, 1967, p. 212) or as the "starting point" from which readers interpret the message of the clause (Halliday, 1985; MacWhinney, 1977; Quirk, Greenbaum, Leech, & Svartvik, 1985).

"Theme is the name we give to the initial part of any structure when we consider it from an informational point of view. When it occurs in its expected or unmarked form, its direct relation to given information can be seen informally as announcing that the starting point of the message is established and agreed" (Quirk, Greenbaum, Leech, & Svartvik, 1985, p. 1361).

Halliday (1970, p. 161) characterizes the thematic constituent as the "peg on which the message of the clause is hung"; "The Theme is the element which serves as the point of departure of the message; it is that with which the clause is concerned" (Halliday, 1985, p. 38).

The theme organizes the clause as message, functioning as the starting point of the message. In English, the theme is realized by the element in clause initial position. Again, in English, in the indicative mood, the unmarked theme usually is the grammatical subject of the clause, with the predication of the clause serving as rheme. Thematic structure in the unmarked case correlates with information structure.

"There is a close semantic relationship between information structure and thematic structure .... Other things being equal, a speaker will choose the Theme from within what is Given and locate the focus, the climax of the New, somewhere in the Rheme" (Halliday, 1985, p. 278).

Fries (in press) provides an interesting and useful explication of the peg metaphor. Fries suggests that the thematic content of any well formed text correlates with a text's method of development, not necessarily with old, or given, information alone. Instead, themes serve an orienting function within the text, establishing a universe of discourse in which listeners/readers can more readily understand the message. His example, As my brother the doctor says, "Patients are a virtue", demonstrates a marked theme (As my brother the doctor says) that is not old, or given, earlier in context, but the theme does provide orientation, a universe of discourse, in which to interpret Patients are as intentional wording, not a mistake. This orientation function of thematic content is discussed further in Kies (1988) as the presentational function of marked themes and is part of what I call backgrounding below.

 

2.4. The relationship between focus and theme.

Some linguists (such as Quirk, Greenbaum, Leech, & Svartvik, 1985, p. 1361-64) see end- focus and theme as opposite ends of a single pole: in the unmarked case, theme provides the element of lowest communicative value while focus provides the element of highest communicative value. (For Quirk, Greenbaum, Leech, & Svartvik, 1985, the terms "communicative value" and "information value" are interchangeable.) On the other hand, Halliday (1985, p. 278-81) maintains that such a conflation oversimplifies what he calls the "textual macro-function" of language. He suggests that thematic and informational structure are interrelated, but different, systems.

It seems to me that one can capture and define the relationship between information structure and thematic structure by teasing out the differences between the informational value of an element and the communicative value of an element that Quirk, Greenbaum, Leech, and Svartvik (1985) choose to combine. I argue that some elements of an utterance serve the communicative function of background; others serve as foreground. The concepts of background and foreground describe the interrelationship between theme and focus.

Given its function as the starting point of a clause, and given that speakers will usually pick a theme from recoverable, known, given information, the theme becomes nonrandom in that it "is more frequent, more prevalent in its appearance to us and thus becomes the background" (Givon, 1979, p. 324). Given its function as informational prominence, the focal element expresses that which is new, non-recoverable, unpredictable, and random, hence it "is less frequent, thus perceptually more salient, standing out on the background, that is, it is the foreground" (Givon, 1979, p. 324). (Both the systems of theme and focus contribute to the background/foreground communicative function of language, and this is why I believe that many linguists combine the two systems.)

For example, writers frequently use the passive voice to ensure that the appropriate constituent is in thematic initial position and/or that focus falls on a prominent element. Consider an example like (3), where business and technical writers exploit the power of passive voice constructions to control thematic prominence in the clause (and in the discourse as a whole). In (3), the writer introduces the topic in the first clause by highlighting it through end-focus. The writer then uses the theme of the second clause to maintain the topic of discourse (i.e., the theme serves a connective function). The writer finishes the passage with a passive voice final clause to maintain the topic of discourse by thematically highlighting the pronoun that refers to the topic (Cities Services Company) and to foreground an important element in the passage (Cities Services is a "major" oil company).

3. Now for some of our measurements. Within the 20 largest oil companies, we rank 16th in gasoline sales, 17th in distillate sales, and 18th in refining capacity. We are recognized as a "major" oil company by many ... (Rivers, 1980, p. 29).

If the author of (3) chose the active voice in the last clause, the thematic element would have been many (as in Many recognize us as a "major" oil company), interfering with the topical progression of the whole text and destroying too the connective function of the first person plural subject pronoun in the second and third clauses' thematic elements.

So it seems that in the linear order of a clause, the thematic (initial) position and the focal (final) position are central for interpreting and integrating the message of the clause. The paragraphs below also illustrate the point. The paragraph in (4a) is difficult (to say the least) only because each clause has been written (using nominalizations, passive voice, and word order inversions) to interfere with the thematic sequencing and information focus of the revision in (4b).

4. a. Analytic and normative criticism are the two modes that this kind of stylistic criticism comes in. That the best of all possible texts for the content it expresses is the text before us is the assumption of analytic criticism. To explain why the text is as it is is the only task of the analytic critic. Where the writer missed matching his language to his ideas is explained by the normative critic, on the other hand. That the writer could have failed to achieve his intention is its assumption. The fame or obscurity of the author more than the intrinsic quality of a text determines which we choose (from Williams, 1985, p. 44).
   
   b. This kind of stylistic criticism has two modes: analytic and normative. The analytic critic assumes that the best possible text is the one before him and that his only task is to explain why the text is as it is. On the other hand, the normative critic assumes that the writer could have missed his intention and then explains where the writer failed to match his language to his ideas. Which form of criticism we choose is determined more by the fame or obscurity of an author than by the intrinsic quality of a text (from Williams, 1985, pp. 234-235).

3. Information focus and thematic structure in the prose of college composition students.

3.1 Foreground/background and parataxis/hypotaxis.

As I have outlined in section 2.1 above, the linguistics of information focus and thematic prominence is a complex of various processes which conspire to assign the most appropriate focus and theme for each clause in its spoken or graphic representation. Not surprisingly then, novice writers often experience difficulty learning to manipulate the numerous syntactic processes that compensate in the written language for the intonational processes that allow for information focus and thematic prominence in the spoken language. Subordination is one area where one might expect to find such difficulty.

The assignment of information focus and thematic prominence is dependent upon the basic arrangement of clauses in linear order. Clauses may be ordered sequentially without any one clause grammatically dominating the other (parataxis) as in (3) above or they may be ordered sequentially with one clause dominating the other grammatically (hypotaxis) as in (4b) above.

Paratactic sequences of clauses allow a writer multiple foci of information and multiple themes, as in juxtaposed or coordinated clauses. In (3) remember, the paratactic arrangement of the three clauses allows the writer to create a cohesive discourse by establishing a topic of discourse in the first clause through end-focus and then to use repetitive and pronominal themes in the second and third clauses to maintain textual cohesion. Schematically, the pattern of theme (Th) and focus (F) in each clause (C) in (3) would look like:

C1 [(Th1)Now for some of (F1)our measurements]

C2 [(Th2)Within the 20 largest oil companies, we rank (F2)16th in gasoline sales, 17th in distillate sales, and 18th in refining capacity]

C3 [(Th3)We are recognized as (F3)a "major" oil company by many].

Hypotactic sequences allow a writer to arrange the foci of information and themes hierarchically, as in subordinate clauses of different kinds. In (4b), the writer can arrange foci and themes to express information priorities, i.e., which elements seem relatively more important for effective communication and are hence foregrounded through end-focus while other elements seem relatively less important for effective communication at this point in the discourse and so are backgrounded as themes. Schematically again, notice the distribution of theme and focus in each sentence of (4b):

Sentence 1:

C1[(Th1)This kind of stylistic criticism has two modes: (F1) analytic and normative]

Sentence 2:

C1[(Th1)The analytic critic (F1)assumes C2[(Th2)that the best possible text is (F2)the one before him] and C3[(Th3)that his only task (F3)is C4[(Th4)to (F4)explain C5[(Th5)why the text (F5)is C6[(Th6)as it (F6)is]]]]]

Sentence 3:

C1[(Th1)On the other hand, the normative critic (F1)assumes C2[(Th2)that the writer could have missed his (F2)intention and then explain C3[(Th3)where the writer (F3)failed C4[(Th4)to match his language to his (F4)ideas]]]]

Sentence 4:

C1[(Th1)C2[(Th2)Which form of criticism we (F2)choose] is determined (F1) more by the fame or obscurity of an author than by the intrinsic quality of a text].

Again, notice that the first sentence establishes a topic for this text through end focus on analytic and normative and many of the text's themes serve a connective function. But notice also that the subordination in the clauses allows the writer to create a hierarchy of themes and foci of information, frequently backgrounding old, presupposed, known information through the themes while foregrounding the new, unpredictable information through end-focus, as in the use of the correlative conjunction more by ... than by the intrinsic quality of the text.

 

3.2. The indeterminacy of nonrestrictive relative clauses: hypotaxis grades into parataxis.

The difficulty in assigning focus in the information structure of a clause is evident in the paragraph in (5). In the following paragraph, the writer chooses to subordinate an enormous amount of information using nonrestrictive relative clauses. As one reads the passage, especially the third sentence, however, one soon discovers that the hypotactic arrangement of the clauses does not coincide with the usual functions of hypotaxis. Instead, it seems evident that the writer has some stylistic conflict between the hypotactic syntax of her sentences and the paratactic semantics of the same sentences.

5. However there were some negative experiences which I encountered. Some of the negative experiences were the inability of the foster parents to attempt to understand many of the needs of these foster children. In my opinion, many of the parents living with these children need councelling or family councelling between the children and the foster parents in order to alleviate many of the lack of communications which where revealed which councelling some of these children such as lack of empathies listening from the parent's lack of support, in many of these children's problems which often would cause the child to give other forms of substitution, many of which were relating to some criminal tendancies such as petty theft, lack of interest in school studies, lack of modivation in general, a general feeling of not being loved and understood, which often was revealed to the counselor [only italics added].

A schematic presentation of the third sentence in (5) demonstrates the difficulty more dramatically.

TABLE 1: Schematic Presentation of Sentence 3 in Example 5

MAIN CLAUSE Adverbial Subject Verb Object (Th1)In my opinion many of the parents living with these children need (F1)councelling or family councelling... INFINITIVE SUBORDINATE CLAUSE   Subordinator Verb Object   (Th2)in order to alleviate many of the lack of (F2)communications RELATIVE CLAUSE A (Th3)which were (F3)revealed RELATIVE CLAUSE B (Th4)which councelling some of these children...in many of the children's (F4)problems RELATIVE CLAUSE C (Th5)which often would cause a child to give other forms of (F5)substitution RELATIVE CLAUSE D (Th6)many of which were relating to some criminal (F6)tendancies... RELATIVE CLAUSE E (Th7)which often was revealed to the (F7)counselor.

Unlike the hypotactic constructions of (4b), the third sentence in (5) does not exhibit any control of theme or focus to background or foreground information within each relative clause (or the sentence as a whole). In (4b), the themes of each clause are mostly pronominal, anaphoric noun phrases (or clauses), or disjunctive adverbials serving a contrastive function. The focus of each clause (or sentence) falls on an item of prominent information relevant to communicate the content of the discourse effectively.

Yet the third sentence in (5) lacks those qualities of the hierarchical arrangement of themes and foci found in (4b). The nonrestrictive relative clauses in the third sentence in (5) are not hierarchically arranged to allow for foregrounding and backgrounding of information. Rather, the nonrestrictive clauses allow for multiple foci, and each focus of information is weighted the same in each clause. The theme in each nonrestrictive clause is coreferential with a lexical item in the preceding clause, creating a chain arrangement in the rhetorical development of the sentence. For example, note the multiple foci and the chaining of clausal themes:

TABLE 2: Multiple Foci and Chaining of Clausal Themes of Sentence 3 in Example 5

Relative Clause Theme Focus     [many of the lack of communications (= focus of preceding subordinate clause)] A which (= lack of communications) revealed B which (= the revealed lack of communications) many of the children's problems C which (= children's problems) other forms of substitution D many of which (= other forms) some criminal tendancies E which (= some criminal tendancies) the counselor

I want to suggest here that the nonrestrictive relative clauses in (5) are syntactically and semantically indeterminate — that despite the hypotactic syntax in (5), the semantics of the nonrestrictive clause is truly paratactic. Parataxis allows for multiple themes and foci in a sequence of clauses where no one clause dominates another grammatically. Hypotaxis allows for the hierarchical arrangement of themes and foci to background and foreground information in clauses, some of which are grammatically dominant. Examining the nonrestrictive clauses in (5) again, I suggest that they are more akin to the paratactic examples in (3) than the hypotactic arrangement in (4b). In (3), and in (5), the multiple themes were frequently coreferential, hence connective, while the focus of each clause added a new piece of information, each weighted equally with all others in importance.

The use of paratactic nonrestrictive relative clauses in (5) above is not an isolated example; consider the examples in (6) through (8).

6. Writing is one of my hobbies, which helps me to relax.
   
   
7. A mask may be front of some sort, to which true identity is hidden.
   
   
8. In high school my writing was worse because I could not write a well constructed sentence or a well constructed paragraph. Which made my essay unacceptable. In my free time, after I got out of high school, I would write a lot of sentences and paragraphs. This helped me to strengthen my writing. When I went to MATC, my english teacher helped me to strengthen my form of essay writing. Which I had thought I had improved in a lot. Now I think I can write an acceptable essay on almost any subject [italics added].

Such ambiguous syntactic and semantic correspondences between hypotaxis and parataxis should not be completely unexpected, particularly in the prose of apprentice writers. Quirk, Greenbaum, Leech, and Svartvik (1985, p. 1258) note the near semantic identity of many nonrestrictive relative clauses and syndetically or asyndetically coordinated constructions (i.e., constructions with or without an explicit marker of coordination), as in (9) or (10).³

He met the chairperson,	{	who invited him to the meeting.
		and she invited him to the meeting.

Here come the Gladstone boys	{	, whom I mentioned to you yesterday.
		; I mentioned them to you yesterday.

Thematically, of course, the corresponding clauses in (9) and (10) are not identical. As theme, the relative pronoun of (10) maintains the Gladstone boys as topic of conversation, thus functioning connectively. Also it signals that the Gladstone boys has greater informational prominence in the sentence as a whole over the focus of the relative clause, functioning to order the information foci of the sentence hierarchically — a typical phenomenon of hypotaxis. On the other hand, as theme of the conjoined clause in (10), the personal pronoun shifts the topic of conversation to the speaker, functioning presentationally. Also it signals the equality of the information foci of both clauses — a typical phenomenon in parataxis. Yet, given the rough syntactic and semantic equivalence of the corresponding clauses in (9) and (10), novice writers might use the corresponding clauses interchangeably.

3.3. Indeterminacy in adverbial clauses.

Adverbial clauses (like many other subordinate clauses) present backgrounded (i.e., presupposed, given, or old) information, cf., Quirk, Greenbaum, Leech, and Svartvik (1985); Dillon (1981); and van Dijk (1977). Given that college composition students often poorly understand or control this function of hypotaxis, the result is an awkward sentence style or the misuse of subordinating conjunctions, as discussed above. Consider examples (11) through (14), where the italicized portions of the students' sentences demonstrate hypotactic syntax between clauses, conflicting with a semantic relationship that seems paratactic (paraphrased after each example).

11. ... once I thought what I was writing then look on my paper to see if my thoughts come on paper, they haven't.
    
    [cf., I think about what I am writing and then look on my paper to see if my thoughts came on paper, and they haven't.]
    
    
12. I finally thought about this [a teacher's actions and suggestions] that night and other nights. I finally started taking into account his complex nature, that I should only expect him to be ribbing everybody just like me.
    
    [cf., ... and I should only expect him to be ribbing everybody [who was shy] just like me.]
    
    
13. There were others [writing problems] but basically I am able to write, whereas the reader is able to comprehend.
    
    [cf., ... and the reader is able to comprehend.]
    
    
14. I find myself using a word once, twice, or maybe three times within a paragraph or sentence, even though I should be able to use another related word.
    
    [cf., ... and I should be able to use another related word.]

Examples (11) through (14) illustrate the mismatch between, and the ambiguity of, students' use of a syntactic form (subordination in those cases) and its functions (hypotaxis for backgrounding information in those cases). The italicized portions of (11) through (14) do not represent old, given, or presupposed information, yet the writers (in constructing sentences that seem to have a more sophisticated, collegiate style) express the information in those italicized clauses as if the information were backgrounded.

The difficulty that students have with hypotaxis is further reflected in the corpus by describing the extremely restricted function that adverbial clauses serve in the prose of college composition students. Adverbial clauses in the corpus occur almost exclusively as marked themes in sentence-initial position serving a scene-setting function (see Bever and Townsend (1979); Bolinger (1978); Dillon (1981); Greenbaum (1969); Quirk, Greenbaum, Leech, and Svartvik (1985); and van Dijk (1977) for a discussion of the semantics and discourse functions of adverbial clauses). Note, for example, the italicized portions of the students' sentences in (15) through (20), and note also the paratactic paraphrase I provide after each example.

15. When I was in high school, I had a chemistry instructor named Mr. West.
    
    [cf., I was in high school, and I had a chemistry instructor ....]
    
    
16. As I go through my college courses, I try to use the example that I was taught and apply it to what I am learning now.
    
    [cf., I go through my college course, and I try to use the examples ....]
    
    
17. If somebody was not paying attention, he would make it so that person would pay attention.
    
    [cf., ?Somebody was not paying attention, and he would make it so that person would pay attention.]
    
    
18. My experience as a writer has not been a successful one. Whenever I am asked to write, I find it difficult to write down or express my thoughts on paper.
    
    [cf., I am asked to write, and I find it difficult ....]
    
    
19. I had trouble writing an essay in a short amount of time. When I am rushing, my mind goes blank for about 10 to 15 minutes which causes me to get grades that I am not happy with. My strength as a writer is that when my mind clears, the form of an essay becomes clear and I can write a well constructed essay.
    
    [cf., ... I am rushing, and my mind goes blank ....]
    
    
20. I have many thoughts but find it difficult to express the meaning on paper, this problem occurs even in small writing tasks. When I have trouble thinking I usually get sidetracked ....
    
    [cf., I have trouble thinking, and I usually get sidetracked ....]

By their form, with explicit markers of subordination suggesting temporal, conditional, and causal semantic relationships between clauses, the adverbial clauses in (15) through (20) seem typical examples of adverbials serving a scene-setting function. Yet one could argue that each example above possesses some semantic ambiguity, similar to the semantic indeterminacy of participial and verbless adverbial clauses, nonrestrictive relative clauses, and clauses coordinated by and: "In their indeterminacy, adverbial participle and verbless clauses resemble the versatile relationships expressed by nonrestrictive relative clauses and the connective function of the coordinator and" (Quirk, Greenbaum, Leech, & Svartvik, 1985, p. 1123). Participial and verbless adverbial clauses possess no explicit marker of subordination, thereby creating their semantic indeterminacy, which is usually resolved through context, e.g., Jason, told of his son's accident, immediately phoned the hospital [semantically equivalent to either Jason, who was told of his son's accident, ... or Jason was told of his son's accident and immediately phoned the hospital].

Although the adverbial clauses in (15) through (20) have explicit markers of subordination, suggesting specific semantic relationships between the clauses, one could argue that even here the easy possibility of a paraphrase with and as the clausal link demonstrates the writers' semantic indeterminacy in those adverbial clauses. The clausal relationship seems to be close to that of parataxis again (as in the nonrestrictive relative clauses above) rather than expressing a clear temporal, causal, concessional, or conditional relationship through hypotaxis.

Yet that is not surprising when one considers again that the adverbial clauses in the corpus serve occur almost exclusively as marked themes serving a scene-setting function. Notice that the focus of information of the italicized adverbial clauses in (15) through (20) establishes a scene, a universe of discourse, in which the main clause then elaborates and builds. The foci in those adverbial and main clauses are not hierarchically ordered as one would expect from the hypotactic syntax. They seem equally important, reflecting their paratactic semantics and explaining their easy paraphrase with and. In (19), for example, notice the parallelism and the equal informational prominence of foci that the writer is trying to establish in his contrast, (F1) rushing = (F2) blank and (F3) clears = (F4) clear:

Adverbial clause	Main clause
When I am (F1)rushing, ...when my mind (F3)clears,	my mind goes (F2)blank... the form of an essay becomes (F4)clear...

Consider also these examples in (21) below as further evidence to support the semantic indeterminacy of the apparently hypotactic adverbial clauses in (15) through (20). Some hypotactic adverbial clauses exhibit a strong temporal restriction, similar to the temporal restriction on the paratactic connector and, suggesting a gradient of taxis among clause connection.

21. a. He was lecturing to his class when suddenly a door flew open.
    b. *When suddenly a door flew open, he was lecturing to his class.
    c. He was lecturing to his class and suddenly a door flew open.
    d. Suddenly a door flew open and he was lecturing to his class.
        [not the same meaning as (21c)]
    e. The car stopped when it hit the pole.
    f. When it hit the pole, the car stopped.

As (21a) and (21b) demonstrate, some clauses with when are restricted to natural time order just as the clauses (21c) and (21d) with and are temporally restricted. But only the time sense of when in (21a) and (21b) is restricted like and. (Let when₁ represent the time sense of the word.) Consider the cause sense of when in (21e) and (21f). (Let when₂ represent the cause sense of the word.) Those examples in (21) — and the data in (5) through (8) and (11) through (20) — suggest a gradient of clause connectors between parataxis and hypotaxis: and appears to be the best paratactic connector; relative conjunctions like which seem ambiguous between parataxis and hypotaxis (medial connectors); and the two when conjunctions are closer to hypotaxis, although when₁ is closer to parataxis than when₂ is.

PARATAXIS |–—and —————————-which———-—when₁-——when₂—-|  HYPOTAXIS

See Quirk, Greenbaum, Leech, and Svartvik (1985, pp. 1258-59) for further discussion of the syntax and semantic parallels between coordination by and, nonrestrictive relative clauses, and adverbial clauses.

 

3.4. Indeterminacy in other varieties of English: Historical, geographical, and spontaneous speech.

The discussion thus far has explored the indeterminacy of nonrestrictive relative clauses and adverbial clauses only in college compositions; however, Curme (1931, p. 170-73) notes a historical dimension to the paratactic function of nonrestrictive relative clauses, citing instances of their paratactic function (and form) preceding the development of full hypotactic forms: "A good man was ther of religioun, And was a poure persoun of a town" [= who was a poor parson of a town] (Chaucer, Canterbury Tales, Prologue, 477). "I knew an Irish lady, [she or who] was married at fourteen" [where the asyndetic coordinated clauses seem syntactically and semantically identical with the developing nonrestrictive relative clause]. (Meredith, Ordeal of Richard Feverel, Ch. XXVIII, 226). "I have discovered something, [it or that/which] Concerns you nearly" (Bridges, Humour of the Court, III, 2, 2583). Similarly, Wiegand (1984) supplies evidence to argue that historical antecedents for some Modern English adverbial clauses were also indeterminate, having undergone a process of grammaticalization. In that process, syntactically distinct (paratactic) Old English clauses, adjoined by a clause-initial be, gradually became syntactically dependent (hypotactic) because the discourse functions of evaluation or qualification associated with that be structure allowed speakers to grammaticalize the semantics of those Old English clauses.

Curme (1931, p. 236) also illustrates a number of contemporary colloquial paratactic constructions that seem directly related to those historical constructions, all demonstrating the same semantic indeterminacy between para- and hypotaxis found in the prose of college composition students: Here is a little book [it/that/which] will tell you how to raise roses, or There is a man at the door [he/that/who] wants to see you.

Interestingly, Crystal (1980) and Chafe (1986) find little subordination in spontaneous speech; instead clauses are run together, frequently in ways that resemble paratactic constructions. The prevalence of similar indeterminate structures in widely different varieties of English not only independently motivates the analysis here but also offers insight into the paratactic style of clause linkage that novice writers seem to carry into their texts from spontaneous speech in the form of the fused sentence or the comma splice.

 

3.5. Indeterminacy and sentence errors: comma splices and fused sentences.

Other subordinate clauses in the corpus also exhibit paratactic relationships to their main clauses; note the comma splices in (22) and (23).

22. My teacher and I became such good friends, we call each other almost every week.
23. My tooth hurt so bad, I was afraid it would have to be pulled.

Sentences (22) and (23) are ambiguous between hypotaxis (where an understood subordinator like that might introduce the second clause in each example, as in My tooth hurt so bad that I was afraid it would have to be pulled) and parataxis (where an understood coordinator like and might introduce the second clause in each example, as in My tooth hurt so bad, and I was afraid it would have to be pulled). In the corpus, many of the comma splices written by college composition students were exactly like examples (22) and (23), ambiguous between hypotaxis and parataxis.

Fused sentences, on the other hand, would seem to be obviously paratactic: they are two sentences joined together as one orthographic sentence without benefit of conjunctions or punctuation. Often they involve conjunctive adverbs functioning as coordinators between clauses. The semantics of conjunctive adverbs is indeterminate in itself, exhibiting a gradient between the functions of conjunction at one pole and modification at the other pole. Consider the last fused sentences in (24) from an impromptu essay written by a remedial English composition student.

24. First I would like to stress my weaknesses because I feel that my writing is very poor. The main problem is that I know what I want to say its just that its very hard for me to put it on paper. I know that I think much faster than I can write therefore once I thought what I was writing then look on my paper to see of my thoughts came on paper, they haven't.

The middle sentence in (24), also a fused sentence, exhibits a different structure, a structure indeterminate between para- and hypotaxis. In that sentence, the first main clause seems to function presentationally as theme of the second main clause. That is, the first main clause (The main problem is that I know what I want to say) functions to orient or present a universe of discourse in which the second main clause (its just that its very hard for me to put it on paper) can be interpreted.

These errors suggest that instruction about sentence faults needs to do more than prescribe to students about punctuation, conjunctions, or clause boundaries. Effective instruction must recognize first that students' syntax is itself ambiguous, demonstrating both para- and hypotaxis. Is the clausal relationship in (22) and (23) hypo- or paratactic? If hypotactic, the students need to learn about effective subordination, punctuation with commas, and the uses of subordinating conjunctions. If paratactic, the students need to learn about effective coordination, punctuation with semicolons, and coordinating conjunctions.

There are, moreover, a variety of constructions that are related to the fused sentence and the comma splice in that they derive the motivation for their existence from the same preference for paratactic constructions on the part of novice writers. For example, sentence (25) employs subordination to background some information while highlighting another portion of the message.

25. That Christopher Columbus discovered America is now openly debated.

The student's actual sentence, in (26), is now more understandable, if one remembers that the assignment of information focus is difficult for novice writers, who seem to prefer paratactic constructions whenever possible.

26. Christopher Columbus discovered America is now openly debated.

Granted the number of sentences beginning with subordinate clauses is small, but an inordinate number of sentences like (25) in the corpus (78% of 139 sentences) lacked an overt indication of subordination, as in (26). As further evidence for this analysis, consider sentence (27), which demonstrates the stylistic option of placing the information in the sentence in a paratactic sequence.

27. Christopher Columbus discovered America, and that is now openly debated.

It seems reasonable to suggest that examples of awkward syntax or faulty subordination like (26) are the result of a novice writer's incomplete attempt to use subordinate clauses as paratactic, rather than hypotactic, constructions.

3.6. Awkward syntax and style in college composition.

The stylistic conflict that I have discussed throughout the article can now be more clearly understood: the syntax of the nonrestrictive relative clauses and the adverbial clauses examined above is hypotaxis, but the semantics is parataxis. Further, one can now begin to explicate the pretheoretical notions of awkward syntax and sentence faults found in many college compositions and discussed in every composition handbook: awkwardness and sentence faults may be the result of a mismatch between syntax and semantics, between form and function generally, usually where the syntax suggests hypotaxis, but the semantics suggests parataxis.

I want to emphasize though that the term stylistic is important here — that the writer of (5) above, for example, is employing a stylistic strategy rather than a syntactic realization. Style here refers to the pattern, the texture, of linguistic choices made by a writer (or group of writers). The majority of students in college composition courses are linguistically sophisticated enough to employ a host of different syntactic structures available to them as speakers of English, but the author of (5) uses the nonrestrictive clauses as a choice, a style marker. In fact, the style of (5) is not uncommon in college composition. The majority (85%) of the 638 essays in the study had at least one sentence with a clause relationship ambiguous between para- and hypotaxis, as in (5) through (20) above, although the numbers of ambiguous clause relationships are (three times) greater among the sentences of remedial writers than among the sentences of the second semester advanced composition students.

The style of (5) may be part of a developmental strategy, in that novice writers often have difficulty with the linguistics of clausal relations and connections — difficulties evidenced by the author of (5). Some research on the composing process of writers at the college level suggests that the syntax and semantics of clausal connections are troublesome for many novice writers (cf., Faigley, 1979; House & House, 1980; Maimon, 1978; and Wolk, 1970). One study (Kies, forthcoming) found that remedial writers at the college level mastered the stylistic options of subordination relatively late in their acquisition of the composing process. In those cases, it may be fair to characterize the semantic ambiguity associated with clause connections in (5) above or in any of the other examples as developmental in origin.

However, realization may not be the most appropriate term here. As the term is used in systemic-functional linguistics, realization suggests that the choice of a paratactic nonrestrictive relative clause is one of several clause-level options available to speakers of English. I believe that such a suggestion would be an inaccurate description of the language generally and of the style of composition specifically because these structures are limited in their distribution, use, and frequency both in and out of college compositions.

 

4. Some conclusions about linguistic function and linguistic form and recommendations for further research.

After considering the various errors of fused sentences, comma splices, awkward syntax, and faulty subordination, one notices that in the prose of college composition students subordinate clauses function both as hypotactic structures and paratactic structures. Fluent writers match the linguistic functions of parataxis and hypotaxis with the linguistic forms of juxtaposition, tags, coordination, and different forms of subordination (nonrestictive relative clauses, adverbial clauses, and other subordinate clauses) as illustrated in (28). On the other hand, college composition students often match linguistic function and linguistic form differently. Subordinate clauses in the corpus, particularly the nonrestrictive relatives and the adverbial clauses, are semantically indeterminate, exhibit a gradient, functioning both as para- and hypotactic structures as illustrated in (29).

Fluent writers (cf., Quirk, Greenbaum, Leech, & Svartvik, 1985)

College composition students

The gradient nature of taxis suggested here receives some cross-linguistic support in Talmy's (1978) study of the universal properties of coordination and subordination. Talmy (1978, p. 507) recognized that nonrestrictive relative clauses are semantically identical with coordinate structures: "The kind of complex sentences which can appear thus in a relative formation seem generally to be the same as can construct with AND. In fact a rough surface version of the general syntactic structure of relative-clefted can be provided ... with the AND removed and the that replaced by which."

In conclusion, the analyses described in this study suggest several tentative hypotheses; each requires further study:

First, one can hypothesize that a significant number of structures in the prose of college composition students are semantically indeterminate and that taxis is a gradient phenomenon in college composition.

Second, in the language development of college composition students, there seems to be a mismatch between meaning and grammar, between function and form generally, between the semantics of parataxis and the syntax of hypotaxis in the complex sentences specifically. This mismatch between function and form here seems to be one example of a larger phenomenon in language development — overextension. The gradient nature of subordination discussed above is an overextension of paratactic semantics and function to structures with hypotactic syntax.⁴ In (5) above, remember, the student seems to recognize that longer, complex sentences are characteristic style markers of the prose of professional counseling. Yet the student does not seem to have mastered the linguistic functions of the nonrestrictive relative clauses as quickly as she was able to acquire the linguistic form. She overextends paratactic semantics to her hypotactic forms.

Third, one can extend the hypotheses above to account for the linguistic nature of particular usage errors commonly found as style markers in college compositions: if linguistic function determines linguistic form in a given context (a central tenet in most functional analyses), then stylistic errors result from the use of linguistic forms that are not motivated by linguistic, communicative, function.⁵ This suggests that several pretheoretical concepts (like "awkward syntax") and usage errors (like "comma splices", "sentence fragments", and "fused sentences") are products of the gradient nature of subordination in college composition, the mismatch of form and function.

What remains unclear, nevertheless, is the order of acquisition. Do college composition students acquire linguistic forms (hypotactic syntax) before they master linguistic function (the foregrounding/backgrounding functions of hypotaxis)? Or do they learn of the functions/semantics (of hypotaxis) first and then struggle to find the appropriate forms to express themselves? Some of the data here suggest that the direction of acquisition is form (perhaps through imitation) first and then function, but much more research needs to be done.

Finally, one should note an important distinction between research methodologies in quantitative and qualitative corpus and textual studies. Quantitative corpus and textual studies often assume that statistically salient co-occurrence patterns between linguistic features bear witness to a shared communicative function. That is the assumption behind the quantitative textual studies of Ervin-Tripp (1972), Biber (1985) and (1988), and, in part, Besnier (1988); as Besnier (1988, p. 715) puts it "...the most fruitful quantitative approach to stylistic variation consists in measuring the degree to which linguistic features co-occur within a text...". However, that assumption may not always be warranted: in this study, for example, one might mistakenly assume that all overt subordinators between clauses axiomatically entails hypotactic function, conflicting with many examples in the data here on the semantic indeterminacy of subordination.

One might suggest therefore that all corpus and textual studies must include a careful qualitative analysis of the both the phenomena and the concepts under investigation, as in Halliday's study of metaphor (Halliday, 1987). Semantic indeterminacy (and gradience) would go unnoticed, for example, if the most basic concepts (such as subordination) went unanalyzed. Linguistic description and explanation would be that much weaker.

 

Notes

¹This article began as a paper presented at the 1987 Annual Convention of the Conference on College Composition and Communication, March 21st, in Atlanta, Georgia. I am grateful to the many convention participants who shared their comments and questions with me. I am especially indebted to Peter Fries, Barbara Couture, Richard Cureton, Charles Meyer, Arthur Palacas, Hugh Rank, William Vande Kopple and Rachel Whittaker, all of whom have made valuable contributions and suggestions on various drafts. I must also thank David Bloome and the anonymous reviewers of Linguistics and Education: the editor's insights and the reviewers' criticisms have made this article immeasurably better than its earlier drafts.

The corpus on which this article is based was collected at the University of Wisconsin - Milwaukee in September 1984. The corpus consists of 638 essays (totalling 157,900 words) written by freshmen early in the term. The freshmen were enrolled in English composition courses from the remedial level through the more advanced second semester English composition or business writing courses. The corpus is larger than most data bases in textual or stylistic studies. The size of the corpus helps to ensure that the texts are representative of the freshman English speech community.

The essays fall into two groups: impromptu essays and prepared essays. The impromptu essays were thirty minute diagnostic essays assigned during the first class session using writing prompts such as How do you assess yourself as a writer? or Why did you choose to attend the University of Wisconsin - Milwaukee? The prepared essays are the first assigned essays, all of which were prepared outside of the composition classroom. The topics for the prepared essays varied from instructor to instructor, but all fell into the rhetorical modes of narration or description. The exception was the business writing samples. Those essays were all examples of "bad news" correspondence designed to persuade the addressee.

   The composition of the corpus is summarized the table below.

   Class                 Nos. of essays         Nos. of words


                     Impromptu   Prepared   Impromptu   Prepared
                       essays     essays      essays     essays


   Remedial             116         81        25,800      20,600

   Traditional
   (including Business  313        128        78,250      33,250
   Writing classes)

   Totals               429        209       104,050      53,850

                            [638]                 [157,900]

 

²See Prince (1978) for a discussion of the discourse contexts that require such explicitness, and see Chafe (1976) for more details about the potential confusion arising through the poverty of surface syntactic information.

 

³Quirk, Greenbaum, Leech, and Svartvik (1985, pp. 1462-63) also note that nonrestrictive relative clauses are sometimes made into separate orthographic sentences in some self-consciously lively styles of writing, where the anaphoric which is equivalent to and that; e.g., "And women are far better at both these aspects of communication than men — as well as more sensitive to the nuances involved. Which is why, says Coates, a woman will feel personally slighted if a man walks past her wearing a frown ..." [Anne de Courcy, (1988, January 3) Between us: Male and female English spoken here, Chicago Tribune, sec. 6, p. 4].

 

⁴This preference for parataxis over hypotaxis is not only true of college composition students but also of developing writers at any age or level of linguistic development. Notice for example that parataxis is the dominant strategy for clause linkage among young speakers:

I got some money and I went to the store and I bought lots of candy and I ate it all up and I got sick and I went home and I threw up." [Ronnie, age 4]

Similarly, many writers of children's literature prefer parataxis over hypotaxis in their works:

The night Max wore his wolf suit and made mischief of one kind and another his mother called him "WILD THING!" and Max said "I'LL EAT YOU UP!" so he was sent to bed without eating anything. That very night in Max's room a forest grew and grew — and grew until his ceiling hung with vines and the walls became the world all around and an ocean tumbled by with a private boat for Max and he sailed off through night and day and in and out of weeks and almost over a year to where the wild things are. [Maurice Sendak, Where the wild things are, where four hypotactic relations are outnumbered by seven paratactic relations at the clause level alone.]

 

⁵This point is also illustrated by students in business and technical writing courses. As students read and write in such courses, they quickly discover various stylistic markers of business and technical prose, such compounding, nominalizations, and the frequent use of the passive voice. Suddenly, as every instructor of business or technical writing will testify, the students' prose is filled with those same stylistic patterns. However, the students are not acquiring the linguistic functions of passive voice as quickly as they are mastering the linguistic forms, and so their prose is filled with stylistic 'errors' as in the letter below, a student's attempt at a 'bad news' letter in a sophomore business writing class. The letter is filled with an inappropriate use of the passive voice creating an unnecessary and unwanted threatening tone.

Dear Mr. Bronson:

Over the past few months you have been sent letters requesting you to submit your S E Forms before or on the 10th of each month. Thus far you have failed to get them in on time. As one of Blaylock's best sales dealers, I don't want to see you get in trouble over this matter of late forms. I have been informed that if this situation is not reversed stern action will be taken such as fewer rewards or a talk with the Senior Vice President of Marketing. [italics added]

 

References

Andersson, L-G. (1975). Form and function of subordinate clauses. Gothenburg monographs in linguistics 1. Goteburg: University of Goteburg Department of Linguistics

Besnier, N. (1988). The linguistic relationships of spoken and written Nukulaelae registers. Language, 64, 707-736.

Bever, T., & Townsend, D. (1979). Perceptual mechanisms and formal properties of main and subordinate clauses. In W. Cooper & E. Walker (Eds.), Sentence processing (pp. 159-226). Hillsdale, NJ: Lawrence Erlbaum.

Biber, D. (1985). Investigating macroscopic textual variation through multifeature/multidimensional analyses. Linguistics, 23, 337-360.

Biber, D. (1988). Variation across speech and writing. Cambridge: Cambridge University Press.

Bolinger, D. (1978). Pronouns in discourse. In T. Givon (Ed.), Syntax and semantics, Vol. 12 (pp. 289-309). New York: Academic Press.

Brandt, D. (1986). Toward and understanding of context in composition. Written communication, 3, 139-157.

Chafe, W. (1976). Givenness, constrastiveness, definiteness, subjects, topics, and points of view. In C. Li (Ed.), Subject and topic (pp. 25-55). New York: Academic Press.

Chafe, W. (1986). Writing in the perspective of speaking. In C. Cooper & S. Greenbaum (Eds.), Studying writing: Linguistic approaches (pp. 12-39). Beverly Hills, CA: Sage Publications.

Colomb, G. & Williams, J. (1985). Perceiving structure in professional prose. In L. Odell & D. Goswami (Eds.), Writing in nonacademic settings (pp. 87-128). New York: Guilford Press.

Connors, R. & Lunsford, A. (1988). Frequency of formal errors in current college writing. College composition and communication, 39, 395-409.

Cooper, M. (1986). The ecology of writing. College English, 48, 364-375.

Crystal, D. (1980). Neglected grammatical factors in conversational English. In S. Greenbaum, G. Leech, & J. Svartvik (Eds.), Studies in English linguistics (pp. 153-166). London: Longman.

Curme, G. (1931). A grammar of the English language, volume II: Syntax. Essex, CT: Verbatim.

Danes, F. (1964). A three-level approach to syntax. Travaux linguistiques de Prague, 1, 225-240.

Dillon, G. (1981). Constructing texts. Bloomington, IN: Indiana University Press.

Ervin-Tripp, S. (1972). On sociolinguistic rules: Alternation and co-occurrence. In J. Gumperz & D. Hymes (Eds.), Directions in sociolinguistics: The ethnography of communication (pp. 213-250). New York: Holt, Reinhart and Winston.

Faigley, L. (1979). The influence of generative rhetoric on the syntactic maturity and writing effectiveness of college freshmen. Research in the teaching of English, 13, 197-206.

Faigley, L. (1986). Competing theories of process: A critique and a proposal. College English, 48, 527-542.

Firbas, J. (1964). On defining theme in functional sentence analysis. Travaux linguistiques de Prague, 1, 267-380.

Firbas, J. (1966). Non-thematic subjects in contemporary English. Travaux linguistiques de Prague, 2, 239-254.

Flower, L., Hayes, J., Carey, L., Schriver, K., & Stratman, J. (1986). Detection, diagnosis, and strategies of revision. College composition and communication, 37, 16-55.

Fries, P. (In press). Patterns of information in initial position in English. In P. Fries and M. Gregory (Eds.), Discourse in society: Functional perspectives. Norwood, NJ: Ablex Publishers.

Givon, T. (1979). On understanding grammar. San Diego: Academic Press.

Greenbaum, S. (1969). Studies in English adverbial usage. Coral Gables, FL: University of Miami Press.

Haiman, J. & Thompson, S. (1984). "Subordination" in universal grammar. In C. Brugman and M. Macaulay (Ed.), Proceedings of the tenth annual meeting of the Berkeley linguistics society (pp. 510-523). Berkeley, CA: Berkeley Linguistics Society.

Halliday, M. A. K. (1967). Notes on transitivity and theme in English. Part 2. Journal of linguistics, 3, 177-274.

Halliday, M. A. K. (1970). Language structure and language function. In J. Lyons (Ed.), New horizons in linguistics (pp. 140-164). Harmondsworth, England: Penguin.

Halliday, M. A. K. (1985). An introduction to functional grammar. London: Edward Arnold.

Halliday, M. A. K. (1987). Spoken and written modes of meaning. In R. Horowitz & S. Samuels (Eds.), Comprehending oral and written language (pp. 55-82). San Diego: Academic Press.

Hoey, M. (1986). Overlapping patterns of discourse organization and their implications for clause relational analysis in problem-solution texts. In C. Cooper and S. Greenbaum (Eds.), Studying writing: Linguistic approaches (pp. 187-214). Beverly Hills, CA: Sage Publications.

House, E. & House, W. (1980). Some similarities and differences between compositions written by remedial and non-remedial college freshmen. Paper, Annual meeting of the Conference on college composition and communication. ERIC Document 186 931.

Kies, D. (1988). Marked themes with and without pronominal reinforcement: Their distribution and meaning in discourse. In E. Steiner and R. Veltman (Eds.) Pragmatics, discourse and text (pp. 47-75). London: Pinter Publishers.

Kies, D. (forthcoming). What remedial writing at the college level reveals about writing assessment and the writing process. English quarterly.

Kuno, S. (1973). The structure of the Japanese language. Cambridge, MA: Harvard University Press.

Maimon, E. (1978). Measuring syntactic growth: Errors and expectations in sentence-combining practice with college freshmen. Research in the teaching of English, 12, 233-244.

Mann, W. & Thompson, S. (1986). Relational propositions in discourse. Discourse processes, 9, 57-90.

MacWhinney, B. (1977). Starting points. Language, 53, 152-168.

Pawley, A. & Syder, F. (1983). Natural selection in syntax: Notes on adaptive variation and change in vernacular and literary grammar. Journal of pragmatics, 7, 551-579.

Prince, E. (1978). A comparison of wh-clefts and it-clefts in discourse. Language, 54, 883-906.

Prince, E. (1981). Toward a taxonomy of given-new information. In Cole (Ed.), Radical pragmatics (pp. 223-255). New York: Academic Press.

Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1985). A comprehensive grammar of the English language. London: Longman.

Rivers, W. (1980). Business reports: Samples from the "real world". Englewood Cliffs, NJ: Prentice-Hall.

Sandmann, M. (1954). Subject and predicate. Edinburgh: Edinburgh University Press.

Shaughnessy, M. (1977). Errors and expectations. New York: Oxford University Press.

Strong, W. (1985). Linguistics and writing. In B. McClelland and T. Donovan (Eds.), Perspectives on research and scholarship in composition (pp. 68-86). New York: Modern Language Association.

Talmy, L. (1978). Relations between subordination and coordination. In J. Greenberg (Ed.), Universals of human language, volume 4: Syntax (pp. 487-513). Stanford, CA: Stanford University Press.

van Dijk, T. (1977). Text and context. London: Longman.

Van Valin, R. (1984). A typology of syntactic relations in clause linkage. In C. Brugman and M. Macaulay (Ed.), Proceedings of the tenth annual meeting of the Berkeley linguistics society (pp. 542-558). Berkeley, CA: Berkeley Linguistics Society.

Wiegand, N. (1984). Creating complex sentence structure. In C. Brugman and M. Macaulay (Ed.), Proceedings of the tenth annual meeting of the Berkeley linguistics society (pp. 674-687). Berkeley, CA: Berkeley Linguistics Society.

Williams, J. (1985). Style: Ten lessons in clarity and grace. Glenview, IL: Scott, Foresman.

Williams, J. (1986). Non-linguistic linguistics and the teaching of style. In D. McQuade (Ed.), The territory of language (pp. 174-191). Carbondale: Southern Illinois University Press.

Winter, E. (1977). A clause relational approach to English texts: A study of some predictive lexical items in written discourse. Instructional science, 6, 1-92.

Witte, S., Daly, J., & Cherry, R. (1986). Syntactic complexity and writing quality. In D. McQuade (Ed.), The territory of language (pp. 150-164). Carbondale: Southern Illinois University Press.

Wolk, A. (1970). The relative importance of the final free modifier: A quantitative analysis. Research in the teaching of English, 4, 59-68.

^Take Note! | Table of Contents | Syllabus | eForum | Search

The HyperTextBooks | The English Main Page | College of DuPage