Exploring evaluative, emotive and persuasive strategies in discourse. AAVV

for designing systems of options that serve as the basis for quantitative analyses.

      In order to test the hypotheses, a threefold comparison of the expressions of Engagement was carried out:

      a.All the English texts (originals and translations) versus all the Spanish texts;

      b.The original English texts versus the original Spanish texts;

      c.The argumentative versus the expository texts.

      While performing the analysis, a problem was posed by the expressions of Engagement attributed to other sources, cited by means of direct or indirect reported speech or in some other way, which do not reflect the writer’s dialogic position. In order to register these cases, two new categories were created: a distinction was made between the cases in which the writer subscribes to the cited source (‘Cited-Contraction’) or else does not consider it as completely reliable (‘Cited-expansion’). Each of these categories was in turn divided into Expansion and Contraction, depending on the dialogic position of the expression itself. Given the limited number of expressions of these categories, no further distinction was made between the subtypes of Expansion and Contraction. The resulting system is depicted in Figure 1.

Figure 1. The system of Engagement used in this paper

      In order to clarify the actual signalling of the ‘Cited’ categories, let us consider (18) and (19):

      (18)El cacique Tecum, al frente de los herederos de los mayas, descabezó con su lanza el caballo de Pedro de Alvarado, convencido de que formaba parte del conquistador: Alvarado se levantó y lo mató. (SO_EXP_004)

      English translation: ‘The chief Tecum, leading the descendants of the Mayas, beheaded the horse of Pedro de Alvarado with his lance, convinced that it was part of the conquistador: Alvarado stood up and killed him.’ (ETrans_EXP_004)

      (19)For example, research finds that women tend to be less confident and less likely to negotiate for pay rises and promotions than equally qualified men. (EO_EXX_004)

      Convencido in (18) and its correlate convinced are classified as ‘Cited expansion’, since the writer does not consider the chief Tecum as a valid source of information, and the subcategory is ‘Ex-contraction’, since Tecum is portrayed as having a firm belief. By contrast, in (19) the writer expresses alignment with the cited source, research finds, and the expression less likely expresses Expansion, thus being qualified as “Cited contraction: Co-expansion”.

5 Analysis of Unclear Cases

      The system of Engagement, described in Section 2.2., provides a clear definition of subcategories and a number of sample realisations of each; however, given that the boundaries of Appraisal categories are context-dependent and inevitably not clear-cut, cases were found where categorisation was not easy. To start with, the realisations of the different categories were sometimes more complex than the standard realisations; for instance, some examples of Estimate are expressions with nouns that point to possibilities for future events to occur, as in “growth prospects are even worse” (EO_ESS_006), or “mi hipótesis de trabajo es la siguiente” (SO_ESS_004), translated as “my working hypothesis is the following” (ETrans_ ESS_004). Even a seemingly easily detectable category such as Counter, often realised by expressions such as however, although or yet, displayed instances where the realisation was rather more complex, such as (20), where the Noun Phrase headed by paradoxes indicates a contrast between a circumstance (‘in a time…’) and the rest of the proposition expressed by the that-clause.

      (20)One of the great paradoxes of our time is that workers and middle-class households continue to struggle in a time of unparalleled plenty. (EO_ESS_006)

      Another problem was posed by the modal auxiliary will referring to future time. As is well-known (a sample reference is Palmer (1990)), there is disagreement among scholars about whether it expresses a ‘pure future’ or else its modal value is maintained in all the cases. In the present study, will has been considered as a case of Pronounce, since in the texts studied here it mostly expresses speculative predictions, as in (21), rather than ‘safe’ predictions such as ‘My sister will be fifty next month’. However, will has not been signalled when it is falls under the scope of another Engagement expression; in these cases, it borrows the value of that expression. For example, will in (22) does not express Pronounce, since it falls under the scope of the weaker Estimate expression remains uncertain, and hence it has not been considered as an Engagement span.

      (21)Addressing these problems will be difficult, but not impossible. (EO_ESS_005)

      (22)But, unless the proper policies to nurture job growth are put in place, it remains uncertain whether demand for labor will continue to grow as technology marches forward. (EO_ ESS_009)

      The remainder of Section 5 will concern two basic kinds of problems, treated in 5.1. and 5.2., respectively. The first concerns a number of linguistic devices that lie in between two subcategories of Engagement; the second pertains to the boundaries between Engagement and Attitude.

      5.1 DOUBTFUL CASES BETWEEN TWO SUBCATEGORIES OF ENGAGEMENT

      There are many cases in which a source of evidence is cited, but the way in which it is cited does not give explicit indications about the writer’s degree of commitment, so they could be classified under either of the two subcategories of Attribute (Acknowledge and Distance) or under Endorse. A typical example of these ambivalent expressions are English according to and its Spanish equivalents de acuerdo con and según. In order to determine the writer’s attitude towards the source, the reliability (prestige or authority) of the source and the text as a whole have been considered. For example, the expression starting with según in (23) has been classified as Acknowledge, since another study is cited next and the writer does not show full commitment to either, while in (24), which belongs to the same text, the prestige of the source has motivated its classification as Endorse.

      (23)Según un cálculo, a nivel mundial las mujeres ocupan alrededor del 24% de los puestos ejecutivos superiores. (STrans_ ESS_004)

      English original: ‘By one estimate, women hold about 24% of top management positions globally’ (EO_ESS_004)

      (24)Según datos del Foro Económico Mundial, hay una correlación importante entre los avances de un país en cuanto a cerrar la brecha entre los géneros – particularmente en educación y fuerza laboral – y su competitividad económica. (STrans_ ESS_004)

      English original: WEF [World Economic Forum] data suggest a strong correlation between a country’s progress in closing the gender gap – particularly in education and the labor force – and its economic competitiveness. (EO_ESS_004)

      Knowledge of the world may also play a role in expressions with the source cited. We know that Marco Polo’s appreciation of geography and raw materials was invaluable in his time but is now completely outdated. Consequently, the italicised expression in (25) has been classified as Distance:

      (25)Out of Marco Polo’s sparkling pages leaped all the good things of creation: there were nearly 13,000 islands in the Indian seas, with mountains of gold and pearls and twelve kinds of spices in enormous quantities, in addition to an abundance of white and black pepper. (ETrans_EXP_003)

      5.2 OVERLAPPING CASES BETWEEN ENGAGEMENT AND OTHER APPRAISAL CATEGORIES

      This section covers cases in which there are arguments for considering the linguistic devices as instances of both Engagement and Attitude.⁵ The most remarkable overlapping cases are those that express uncertainty tinged with emotion, called ‘apprehension’ in Lichtenberk (1995) (see also Lavid et al. 2016, 2017). An expression of this kind