國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
group in the clause” (p.69). Snow (2010) points out that this process plays a role in determining the conciseness in academic writing and states that “[s]cience teachers are not generally well prepared to help their students penetrate the linguistic puzzles that science texts present” (p.452). Nominalization is sometimes accompanied by a postmodifying device that specifies additional information to the head noun. In other words, we would expect a heightened frequency of of -constructions in academic writing.
The third reason for investigating of -constructions in academic writing is that a growing number of studies in academic writing have identified the lexico-grammatical patterns of various constructions in detail (e.g., Biber, Conrad & Cortes, 2004; Hyland, 2008; Oakey, 2002) but have not yet paid full attention to the of -construction. For example, Groom (2005) examines the introductory it as found in the grammar patterns [it v-link ADJ that] (e.g., It is clear that…) and [it v-link ADJ to-inf] (e.g., it is important to compare…). Another study considers the lexico-grammatical pattern of if -conditionals in medical discourse (Carter-Thomas & Rowley-Jolivet, 2008). These studies demonstrate that some textual functions are genre-specific, implying a need to take into account of the text type, genre, and even disciplinary variation when analyzing textual functions. In the current study, we consider the of -construction as equivalent to a lexico-grammatical pattern in the form of [N1 of N2] where N1 and N2 are variables to be applied to the construction. How different nominal groups are employed by academic writers in the construction is what we are seeking to find.
These three reasons account for the need to investigate of -constructions in academic writing. The first research question of this thesis is therefore to ask which nouns do academic writers commonly use in of -constructions and what semantic relations do these nouns create. In the next section, we will consider the functions that of -constructions have in academic writing, from which we will introduce the second and third research questions.
1.4 Functional Considerations
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
In addition to the need of of -construction in academic writing, previous studies have investigated of -containing expressions from a functional perspective of linguistic devices which happen to contain of -constructions. One of these linguistic devices is a special group of nouns referred to as shell nouns (Hunston & Francis, 1999; Schmid, 2000) that “can act like “empty shells” in certain contexts because they can enclose or anticipate surrounding discourse” (emphasis original, Aktas & Cortes, 2008: 4). To put it in another way, the meanings of shell nouns vary depending on their context and, thus, could function as cohesive devices. In a study of shell nouns as cohesive devices, Aktas and Cortes (2008) analyze the distribution of 35 highly prevalent shell nouns in seven lexico-grammatical patterns among which two are of -constructions ([definite/indefinite article + shell noun + of prepositional phrase], e.g., the process of administering the service, a process of empowerment) (p.10). The results show that some shell nouns are more favorable to certain patterns. For instance, of -constructions that contain shell nouns such as effect, result, and process were found to be used in a higher rate in non-native graduate writers’ research papers than in professional writing of journal articles. Yet, an opposite trend was found for the shell noun problem. Aktas and Cortes also point out that shell nouns in the of -constructions function to
“semantically characterize a piece of experience in a general way” (p.11) as exemplified in (1.4).
(1.4) One technique, which is often used, is the Round Trip Time (RTT). The problem of this technique, [is that] it does not work if the mirrors are protected by firewalls. (student –Computer Science) (Aktas & Cortes, 2008: 11, emphasis added)
According to the authors, the problem is the shell noun that functions to characterize additional information to be attributed to the postmodifying prepositional phrase. In addition to characterization, there are also two other textual functions identified in this study, namely, concept-formation and linking. Concept-formation is a term adopted from Schmid (2000) who describes it as “[t]he recurrent association between the linguistic form and the idea results in the formation of a more or less stable concept”
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
exemplified in (1.5).
(1.5) The effect of a change will vary with the values of the independent variables.
(published - Economics - APFEC) (Aktas & Cortes, 2008: 12, emphasis added)
Specifically, the of -construction in (1.5), the effect of a change, represents a temporary concept serving at the subject position of a sentence. Linking is another textual function discussed by Aktas and Cortes. The authors point out that the pattern [th- + N] was found to express this function. Example (1.6) illustrates this function with This fact (underlined).
(1.6) For all the clay samples, the only fraction of chemisorbed ammonia was converted into the DeNOx products, while majority of NH3 descorbed from the clay surface. This fact suggests that only part of chemisorbed ammonia is activated properly to be converted into DeNOx products. (published -Environmental Engineering -ENVIR) (Aktas & Cortes, 2008: 13, emphasis added)
While this work draws our attention to pedagogical considerations on shell nouns as cohesive devices, it also inspires us further to look closely at shell nouns in the second nominal position in the of -construction.
Another study that mentions of -constructions among other discussions is Hunston’s (2010) work on evaluative language. For example, one of her corpus analyses demonstrates that the concordance lines containing the construction a shred of tend to connote with words like truth, dignity, decency, and hope, all of which appear in a context associated with negatively evaluated words (e.g., not, without, scarcely, and lacking) (p.60). Furthermore, Hunston also identifies modal-like expressions from several grammatical patterns among which one particular of -construction, [N + of + -ing] (e.g., the problem of finding free dates), can be found (pp.78-9). What is significant about these expressions is that they are independent of the modal auxiliaries (e.g., can, could, may, might, will, and would) or modal adverbials (e.g., I think, in fact, and possibly) but serve similar modal functions in expressing possibility, obligation, and inclination. Hunston points out that this finding is relevant to writing instructions as these modal-like expressions may serve as
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
alternatives to modal auxiliaries that are frequently used by native speakers of English.
L2 Learners of English, on the other hand, may not be aware of this type of linguistic resources and be limited strictly to using modal auxiliaries. An example to illustrate her point is shown in (1.7).
(1.7) a. She did not cry out for help for fear of being further assaulted.
b. If she had cried out she might have been assaulted again. (taken from Hunston, 2010: 87)
Example (1.7a) is a corpus example and (1.7b) is its paraphrase. Specifically, the former containing the modal-like expression for fear of can be paraphrased as the latter by using the modal auxiliary might. One of the important findings from Hunston’s research is that non-native English speakers’ misuse of of -constructions could have been compensated by their underuse. This implies that we might need to examine L2 writer’s writing carefully.
According to the studies discussed above, the of -construction may have several textual functions including a cohesive function and an evaluative function. In addition to what Hunston has found with the use of evaluative words such as problem at N1, evaluative adjectives (e.g., central, important, main, and peripheral) as discussed in Swales and Burke (2003) could also be used as premodifiers for N1 and N2. Our second research question is therefore to ask what textual functions the of -constructions carry in academic writing.