href="#ulink_87707492-03b2-5c56-ab80-b06d703501a7">Table 7 also shows that the distributional differences between the main categories of Expansion and Contraction are significant. In particular, Proclaim is markedly more frequent in the argumentative texts, while Entertain is more common in the expository texts, which is not surprising due to the greater need of the writers of essays to defend their dialogic position. The distribution of the most delicate subtypes, given in Table 8, indicates that the differences are significant for the subcategories of Contraction (Proclaim and Disclaim) but not for the subcategories of Expansion (Entertain and Attribute).
Table 7. Distribution of the Engagement spans by major categories in the argumentative and expository texts (originals and translations)
Table 8. Distribution of the Engagement spans by delicate categories in the argumentative and expository texts (originals and translations)
With regard to Contraction, the occurrences of Concur are almost equal in the two text types, while Pronounce and Endorse are more frequent in the argumentative texts. That is to say, the support of dialogic position is most commonly realised by emphasising one’s own viewpoint (Pronounce) or by supporting the position by authoritative or prestigious sources (Endorse). With regard to Disclaim, it is globally more common in the expository texts; the cases of Counter are virtually the same for the two registers, but the expository texts display more cases of Deny, which often refer to a contrast between people’s beliefs and the way things actually are, as in (37), a fragment that contains eight instances of this category:
(37)It is natural but wrong to visualize the singularity as a kind of pregnant dot hanging in a dark, boundless void. But there is no space, no darkness. The singularity has no “around” around it. There is no space for it to occupy, no place for it to be. We can’t even ask how long it has been there—whether it has just lately popped into being, like a good idea, or whether it has been there forever, quietly awaiting the right moment. Time doesn’t exist. There is no past for it to emerge from. (EO_ EXP_001)
With regard to Expansion, although the distribution of the subcategories in Entertain and Attribute is not significant, attentive observation leads us to note that the higher frequency of Entertain in the expository texts is largely due to the difference in number of Estimate devices, which is also related to the main difference in the overall purposes of the texts. Estimate expressions tend to weaken the writer’s assertiveness and this weakening is more at odds with the persuasive purpose of argumentative texts than with the informative purpose of expository texts. The high frequency of Speculate in both text types is plausibly due to the intellectual character that they share: writers often pose complex questions with no obvious answers, as in examples (8-9), in order to trigger reflection and then propose answers through reasoning.
7 Final Discussion and Concluding Remarks
This paper has set forth a quantitative study of the Engagement expressions in 40 argumentative and expository texts, consisting of English and Spanish originals and their corresponding translations into the other language. I acknowledge that the study has limitations concerning the accuracy of the labels assigned to the expression of Engagement, mainly because all the texts were analysed by the author of the paper: the analysis of the same texts by one other researcher (or more) and the ensuing comparison of the results would have been an extra tool for detecting possible errors, and the cases of disagreement would have led to an increase in the refinement of the criteria for assigning expressions to the different categories. Nevertheless, as its stands, the analysis may be considered to be sufficiently reliable for proving the hypotheses stated in Section 3.
The first hypothesis, namely that the expressions under study were faithfully translated to a great extent, has been partially disconfirmed by the significant distributional differences found in the major Engagement categories (Expansion, Contraction, Cited-expansion and Cited-contraction) in the English and Spanish texts. The main reason was that the Spanish translations sometimes added explicit Counter expressions that were absent in the English originals. This finding, together with the higher frequency of Counter expressions in the original English texts than in the original Spanish texts, lead us to believe that the Spanish translations of non-fictional texts had a tendency to overspecify cohesion, an issue that could be pursued in further research. On the other hand, this hypothesis has been partially confirmed by the absence of significant differences in the distribution of the main types of Expansion and Contraction and of all the subtypes of these categories.
The second hypothesis, which stated that differences would be found in the English and Spanish originals, has also been partially disconfirmed by the distribution of the main Engagement categories, but partially confirmed by the distributional differences found in all the subcategories of Expansion and Contraction, and especially in the subtypes of Attribute and Proclaim. The English originals display more cases of Entertain, especially of the subcategories Estimate and Infer.
The third hypothesis, namely that the distribution of Expansion and Contraction in argumentative and expository texts would differ due to the overall aims of both texts, has been confirmed by the results in terms of both the main categories and their subtypes: not surprisingly, Expansion was more common in the expository texts, which aim above all to transmit knowledge, and Contraction in the argumentative texts, where the need to persuade the reader is greater.
This study provides evidence of the crucial role of the expressions of Engagement in persuading the reader that the writer’s assessment is the most sensible within the array of actual or potential viewpoints. The need to be persuasive is more obvious in the essays, but is far from being non-existent in the expository texts. This persuasive role of Engagement may be noticed in many of the examples cited in this paper, such as (15), (16) and (37), to name only a few. Engagement, then, may be considered to offer a privileged perspective on the pervasive relation between persuasion and evaluation in language.
Further research on Engagement in different types of non-fictional texts, especially from parallel corpora, could be carried out in order to shed light on the extent to which the main findings presented here could be extrapolated to other types of non-fictional texts. If we begin by comparing the results presented here with those of Carretero’s (2014) study on film reviews, we find that the three categories that involve citation of a source of information, namely Acknowledge, Distance and Endorse, are markedly more common in the texts analysed here; conversely, Counter is even more frequent in the film reviews, for the reason that critics are quick to communicate the ways in which expectations created by films are unfulfilled in order to warn prospective viewers (Carretero 2014: 76). These differences are one more illustration of how different subtypes of non-fictional texts create different needs for writers to assess their position against other possible positions with regard to the information transmitted and of how writers actually cope with these needs. That is to say, the relative frequency of the Engagement categories seems to vary according to the subtype of nonfictional text, depending on the writers’ assessment of the need to place emphasis on one category or other in order to legitimise their position or, in other words, to persuade the addressee that they are legitimate sources of information.
Acknowledgements
This research has been carried out as part of the EVIDISPRAG Project (reference number FFI2015-65474-P MINECO/FEDER, UE). We gratefully acknowledge the support provided by the Spanish Ministry of Science and Innovation and the European Regional Development Fund. My thanks are extended to an anonymous referee for his/her thorough report on a first version of the paper. The remaining shortcomings and inconsistencies are my sole responsibility.
References
Bakhtin, Mikhail M. (1981). The Dialogic Imagination. Four Essays. (Translated by C. Emerson and M. Holquist.) Austin: University of Texas Press.
Carretero,