finer grained distinction between submeanings. Two of them correspond to the conceptual categories of epistemic modality and evidentiality, two categories which qualify commitment to the information transmitted (Nuyts 2001; Carretero and Zamorano-Mansilla 2013). The difference lies in that epistemic modality does so in terms of probability, as in the English example (3), the Spanish example (4) and its English translation (5), while evidentiality concerns the kind or source of evidence, as in (6). Epistemic modality, and evidentiality to a lesser extent, have often been treated in the literature as expressing commitment to the truth of a proposition. However, Martin and White (2005: 109) state that the main function of epistemic and evidential expressions in actual discourse is not to qualify commitment to the truth of a proposition, but to indicate dialogism, i.e. consideration of alternative points of view. I believe, however, that the meaning of epistemic and evidential expressions in terms of commitment to the truth of a proposition is not incompatible with the expression of dialogism, but rather is a subtype of dialogism; indeed, to qualify the content communicated by means of an estimation of probability or by assessing the evidence available implies the consideration of alternative states or events as other possibilities:
(3)Let us understand what our own selfish genes are up to, because we may then at least have the chance to upset their designs, something that no other species has ever aspired to. (EO_EXP_003)
(4)Pero, quizás porque se habían tomado en serio su destino de dueños del mundo, se habían embarcado en tantas empresas que estaban desbordados (SO_ESS_003)
(5)But, maybe because they had taken their destiny as rulers of the world seriously, they had embarked upon so many endeavours that they were overwhelmed. (ETrans_ESS_003)
(6)And yet there are few signs that working- and middle-class Americans are living any better than they did 35 years ago. Even stranger, productivity growth does not seem to be soaring, as one would expect; (EO_ESS_006)
It is true that these expressions display cases where the epistemic or evidential meaning is bleached into a more general dialogic function, as in (7), where the real reason for using perhaps is to save face rather than to express a qualification of probability:
(7)But briefly, and perhaps a little misleadingly, I can at least state that my point of view entails that it is our present lack of understanding of the fundamental laws of physics that prevents us from coming to grips with the concept of ‘mind’ in physical or logical terms. (EO_EXP_015)
Therefore, the specific consideration of epistemic and evidential expressions as subtypes of Entertain may well bring about differences between tendencies in the realisation of this category depending on the language and discourse type. Consequently, two subcategories have been created for these types of expressions; in keeping with the other labels in the Appraisal framework, these subcategories have been named with verbs that indicate what the speaker/ writer does with language when s/he uses them. The labels chosen are ‘Estimate’ for epistemic modal expressions, and ‘Infer’ for evidential expressions. It must be noted that Estimate and Infer only include epistemic and evidential expressions that do not express a high degree of commitment to the validity of the information: strong epistemic and evidential expressions enhance assertiveness and hence lie within the scope of Contraction, as will be seen in 2.2.2. Nor does Infer include expressions that mention the persons or things (such as reports or studies) that provide evidence; these expressions are classified under “Attribute”, as will be seen below in this section.
Examples of Estimate expressions in English are: modal auxiliaries (may, might, could, must, should) when they express epistemic modality; adverbs (perhaps, probably, maybe…); adjectives (likely); mental state verbs in the first person when they occur with verifiable utterances (I think, I believe), and expressions with nouns such as expectations, possibility or risk. Spanish correlates of all these expressions have been found in the texts. Examples of realisations of Infer are lexical verbs (seem, appear…), adverbs (allegedly, apparently, presumably, reportedly, seemingly…) and their Spanish equivalents.
Another subcategory of Entertain will include those cases in which a dialogic alternative is presented without expressing a specific degree of commitment. To this kind belong the expository questions specifically included within Entertain in Martin and White (2005: 110) although not as a subcategory, such as (8) and its Spanish translation (9), as well as conditional clauses that present the proposition without expressing commitment to its truth or falsity (10), and expressions of lack of knowledge (11). The name chosen for the category is ‘Speculate’. By contrast, pseudo-questions assuming an obvious answer belong to Contraction, subcategory Concur (Martin and White 2005: 123), as will be seen in 2.2.2.
(8)to what extent are minds functionally dependent upon the physical structures with which they are associated? (EO_EXP_015)
(9)¿en qué medida [la mente] depende de las estructuras físicas a las que está asociada? (STrans_EXP_015)
(10)If Greece does well, its creditors will receive more of their money (EO_ESS_001)
(11)It seems highly likely to us that both genes and environment have something to do with this issue. What might the mix be? We are resolutely agnostic on that issue; as far as we can determine, the evidence does not yet justify an estimate. (EO_EXP_020)
A further meaning within Entertain that could not be accommodated in the subcategories mentioned above concerns the cases where the writer indicates that what s/he is communicating is his/her own opinion, which may well coexist with other opinions. These cases, signalled with the category Opine, differ from Estimate and Infer in that the truth of the proposition is treated as non-verifiable; that is, the utterance has no objective truth and consequently the writer’s opinion may well coexist with other different opinions (Carretero and Zamorano-Mansilla 2013: 324-325). Some expressions of Opine are arguably, in my opinion or to my mind.
Martin and White also include deontic modality as part of Entertain (2005: 110-111) but, as was stated above, the deontic category expresses effective stance and is therefore excluded from the analysis. Therefore, the Entertain system devised for this research consists of four categories: Estimate, Infer, Speculate and Opine.
The other subcategory of Expansion is Attribute, which covers the representations of the proposition as grounded in the subjectivity of an external voice, thus situating it within a range of possible positions (Martin and White 2005: 98). In other words, the information is attributed to an external source and the writer does not express a high degree of commitment (the cases in which the writer subscribes to the information belong to Contraction, subcategory Endorse, as will be seen in 4.2.1.). I adopt Martin and White’s (2005) division of Attribute into the subcategories of Acknowledge and Distance.
Acknowledge comprises the cases in which the source of the information is specified, but the writer gives no clues about his/her own position. The expressions of Acknowledge include verbs of saying and of thinking with persons different from the first (he/she thinks, they say, it is said… X reports / states / declares / announces / believes / thinks…), and expressions of opinion by persons different from the writer (in X’s opinion / view…, according to X…). Distance differs from Acknowledge in that the authorial voice explicitly distances itself from the information transmitted by the specified source (Martin and White 2005: 113), by expressing or implying that this information is false or at least unreliable. Examples of Distance are (12), as well as (13) and its English translation (14):
(12)In other words, we are producing and consuming much more than our economic indicators suggest – and the creators of many of those products are not being adequately compensated. (EO_ESS_006)
(13)Sólo faltaban mil años para que los fuegos purificadores del juicio final arrasaran el mundo, según creían los hombres del siglo XV (SO_EXP_003)
(14)According to fifteenth-century man, only 1,000 years remained before the purifying flames of the Last Judgment would destroy the world (ETrans_EXP_003)
2.2.2 Contraction
Contraction differs from Expansion in that, even if it also admits