3. Attitude as part of Appraisal

3.1 INTRODUCTION

At the next level of delicacy, ATTITUDE is subdivided into three subsystems: AFFECT, JUDGEMENT and APPRECIATION. While AFFECT is concerned with values to do with feelings and emotions, JUDGEMENT describes Attitudes towards human behaviour and actions. Thus, both these subsystems are concerned with describing the evaluation of human targets and agents, although it is common for affectual values to be expressed towards non-human targets (or 'affectors'). Non-human agents ('emoters') of these types of Appraisal values might be considered marked, and thus may rarely be found outside metaphor and poetics. APPRECIATION, on the other hand, is concerned with values which describe objects and the attributes of 'things', or beings construed as things. Each of these subsystems is discussed in turn below.

3.2 AFFECT

Within AFFECT the main secondary entry conditions which Martin (2000b) uses are positive/negative; surge/predisposition and low/high intensity. In the analysis presented here I have not used the latter two unless it seemed particularly useful in analysing each ATTITUDE in context. The main values of [AFFECT: negative/positive] can be summarised as:

Not all of these values are realised in the two texts analysed here, but where possible, actual examples from the text have been given, together with made-up (*asterisked) counter examples for illustrative purposes.

3.2.1

Because AFFECT refers to largely individually reported subjective emotion, as distinct from JUDGEMENT which deals with social sanction and esteem, most sources of AFFECT (Appraisers) report the evaluation as 'happening to themselves', perhaps caused by some other person or action, as, for example, "I have been surprised that no-one identified..."(2:18). In this case, the Appraiser and the Appraised would seem to be one and the same, but in this type of situation, the cause of the emotional reaction is classed as the Appraised, 'target'[note1] or 'affector' of the evaluation: that no-one identified… In other cases, the source of the Appraisal would seem to be the Addresser, if, for example I say "*John looks happy". However, in this system it needs to be distinguished for purposes of analysis, that the reporter is seen as that only, and that the Appraisal values are classed as originating in the 'emoter'. If it can be recognised that all statements of any kind represent an 'assessment' of the state of the world, then this method of classing the values of AFFECT as being located in the person who feels or manifests such emotional orientation, rather than the Addresser who reports their assessment of the origin of the AFFECT, can be more usefully applied. In reporting such an assessment of the nature of someone else's evaluative positioning, the Addressor 'avers' that such an emotion "occurred". Grammatically, the proposition is non-sourced (even though understood as located in the Addresser), while the Attitude is sourced in the emoter - in the example suggested above, 'John'.

If, on the other hand, we consider an utterance such as "* I think John looks happy", John is still the source of the Attitude (the emoter), but the proposition has been framed by a grammatical metaphor of probability, and the source of the assessment can be classed as 'self' or the Addresser. "I think" here, would also function as an Engagement value, acknowledging as it does the subjective nature of the assessment which it frames. With Affect therefore, the subjectivity of the Evaluator is always either indicated or at issue (c.f. Hunston 2000).

3.3 JUDGEMENT

Within the system of JUDGEMENT, the two main distinctions are those of social SANCTION and social ESTEEM. The most succinct gloss of the differences between these two areas of description is, for me, that given by Martin (2000a: 156 inter alia):

While it seems that people make judgements of social Sanction without the implication that the so-judged will be soon involved in litigation, it does highlight the major differences between the two areas, Sanction and Esteem.

3.3.1 Social Esteeem

Social esteem is subdivided into three types of judgement: NORMALITY, TENACITY and CAPACITY. Again, these subtypes can be described as being linked to values of modality, in which NORMALITY is to [MODALIZATION: USUALITY], as TENACITY is to [MODULATION: INCLINATION], and as CAPACITY is to [MODALIZATION: ABILITY]. These values are tabulated below together with examples from the text where possible.

3.3.2 Social Sanction

Social sanction is subdivided into VERACITY and PROPRIETY, and these may be usefully linked to values of [MODALIZATION: PROBABILITY], and [MODULATION: OBLIGATION] respectively (Halliday, 1994). In other words, VERACITY is most likely applied to the arguability of the target's sincerity, while PROPRIETY would be invoked in order to suggest that the target should comply with certain socially held values, norms or rules for moral action and behaviour.

These values are tabulated below together with examples from the text where possible.

Table 3.2

3.3.2.1 Judgement: normality: Comment on some examples

Values of negative NORMALITY would be invoked if someone or group were judged to be strange or out of the ordinary, although this points out one of the problems with Judgements of Normality, especially in the context of many western subcultures, where 'out of the ordinary', even though 'negatively normal' would still constitute a positive evaluation in many cases. This highlights the fact that Appraisal theory is concerned with the operation of the discourse semantic level, and as such, evaluative meanings of positive or negative, are locally contingent, not universal. The following example: "I had anticipated curiosity,[affect] fear,[affect] jealousy,[affect] among others, but not suspicion,[affect] and particularly of my identity... this is, in my experience, unique [graduation: focus]to the Web." (2:4), shows how such values of attitude can vary with co-text. The word unique on its own might usually have a positive value of normality in an utterance such as "*She is a unique person." But, as will be discussed below, the negative normality Judgement in this example, is provoked (as distinct from inscribed: see below 2.3.3) via a series of statements regarding the types of Affect the poster expected to find in her target, the email list group. These expected types of Affect are then contrasted with the value of Affect [affect: insecurity] she ascribes to this target, which behaviour is linked in its uniqueness 'to the Web'.

3.3.2.2 Judgement: tenacity/capacity

In the case of TENACITY, values are inscribed in terms of the target's accomplishments, and any claim to have successfully done something or shown determination or willingness to sustain work towards some goal may invoke positive values of Tenacity for example.

CAPACITY is concerned with values of ability, not permission, and as such, forms values linked with Judgements of skill in carrying out an action, or knowledge in relation to some activity, rather than a willingness to comply with any request or command, which in that case, would more likely fall under values of TENACITY.

3.3.3 Provoked and evoked Judgements

As discussed previously, Judgement values are not necessarily inscribed in a text. That is to say, such evaluative judgements are not always explicitly made but may be, indeed are, usually invoked via a series of other values, such as Affect, which leads to the implication that the target may be judged according to the local norms of social interaction which in turn, sanction or value the described behaviour. Invoked Judgements may be either 'evoked', or 'provoked' (White, 1998), with the former arising implicitly via tokens realised by ideational or experiential value, which depend for their evaluative status on the assumed value system of the audience members - or that constructed as assumed. 'Provoked' Appraisal is usually construed via 'tokens' of Judgement - for example, more explicit values of Affect, Appreciation, Engagement and/or Graduation - such as counter-expectation, negation, intensification, and so on. A system of "invocations of Attitude" derived from analysis of a wider set of texts in the study is is set out in Module 3.

This thesis suggests that provoked Appraisal can be extended to include the effect of cumulative work done on the autonomous plane of discourse, where the provoked Judgement may arise implicitly via a string of explicitly inscribed values of Attitude in the text up to that point, as a cue or trigger, their status traceable to series of positioning moves in a move complex. These move complexes can be labelled strategies. The provoked Judgement is still usually signalled by values of Engagement and other indicators of textual prospection which act to link previous values in the text in setting up a frame of relevance for the (provoked) move being made. These links are regarded as acting over several clauses or clause complexes via cohesive devices such as conjunction, and repetition (lexical cohesion, matching relations).

Evoked appraisal, on the other hand, is dependent for its evaluative status on assumed shared values and knowledge, and is realised in texts by experiential meanings which function as tokens of appraisal via activation of community norms. The boundary between these two categories and their identifying features is obviously not clearly defined. Some analysts, for example Jordan (2001), would regard evoked appraisal as 'description' rather than as 'assessment', and hence not be classed as appraisal. While Jordan's (op cit) perspective on text organisation is more concerned to outline the conjunctive macrostructures operating within stretches of text, here I am concerned to start at the level of patterns of semantics on the surface of discourse, and so such 'factual' descriptions of objects, events and people need to be accounted for as dependent on assumed shared values, or contact/familiarity.

An example of provoked Judgement can be seen in sentence (2:4) quoted above (cf. 2.3.2.1). Here, the writer uses a series of values of Affect to set up a negative Judgement regarding the behaviour of the participants of the email list. At the same time, she signals, or prospects, a value of [judgement: negative normality] via counter-expectation ("I had anticipated curiosity, fear, jealousy, among others, but [counter-expect] not [neg.-pol] suspicion.."). As well, she signals that the Judgement may be contentious by the explicit insertion of an intra-vocalised reference to herself, in my experience (similar in function to modalization: probability). This in turn, functions to acknowledge the possibility of other assessments, and thus operates dialogistically to open the negotiating space for the declaration, "..this is, in my experience, unique to the Web", where this refers to the behaviour of listmembers which she had not anticipated.

In section 1.3.1 above, excerpts of texts 1 and 2 (Ex 1.1 & 1.2) were presented as demonstrating a possible use for provoked, evoked, and 'ambiguous' Appraisal in determining, or at least signalling, phase boundaries in expository texts. As an example, in the case of the proposed pivotal sentence 1:5, a question centres around whether it is hard for the writer to mesh all this, i.e. all of the thoughts about the nature of 'task', which is the theme of his discussion; or whether meshing all this is objectively 'hard'. In the first case, the writer negatively appraises his own Capacity, and the positioning with respect to his interlocutors might be one of deference through an expression of lack of expertise in this field (c.f. Martin 1992: 530). In the second, more likely case, he evaluates the nature of all this as 'complex' - as difficult to comprehend in its entirety. In either case, the appraisal is evoked by the term 'difficult/hard (to do s.t.)', which may depend for its negative or positive value on community-held norms regarding 'difficulty'. In this case, the positioning strategy would act to call on equal status (in terms of expertise) with interlocutors.

In terms of lexical association, something which is 'difficult/hard to do' is related to onerous tribute to be paid in sentence 1:3. This clause also acts on the autonomous plane of the discourse encapsulating all this, while at the same time, it orients to the complexity of the argument to come: on the interactive plane, all this is about to be expanded upon in other ways - as a series of personal examples of what 'task' might mean.

Turning now to an example from text2, one of the stretches of text which operate in this way is Sentence 2:5, reproduced here for convenience:

Ex 3.1:

The token of Judgement of negative propriety in this clause complex is provoked by a value of [appreciation: composition: negative] since the target is not, grammatically-speaking, human behaviour, but represented via experiential metaphor, as a nominalisation: our exchange. The negative evaluation is realised as a nominalised possession of a possible future condition, rather than an actual 'exchange'. This feature of text2 is common to the style in this text overall (c.f. Module 1) which represents the attributes of social actors as possessivated (van Leeuwen 1996) nominal groups. In this clause complex, the writer uses this way of realising attributes and processes to distance the activities of the unspecified Addressees she refers to, at the same time developing her positioning strategy through cohesive ties :

3.4 APPRECIATION

The final subsystem of ATTITUDE is that of APPRECIATION, which is highlighted when evaluations of objects, products, events, or even the products of human behaviour are made, or when anything is judged in these terms. Such a distinction is useful when people are evaluated in terms of their appearance, rather than their behaviour, for example. However, as discussed below (2.4.1), Appreciation, and its differentiation from values of Judgement, forms one of the most fuzzy boundaries within the system of Attitude, and thus local grammars of evaluation may need to be more closely applied in determining which analytic categories are most appropriate for each value. On the other hand, it may need to be accepted that due to the inter-stratal tension which inevitably obtains between the lexicogrammatical and the semantic levels, such permeability of the category boundaries may need to be accepted. The categories of [APPRECIATION: negative/positive] which I have concentrated on as the most prevalent are summarised and exemplified below:

Table 3.3

3.4.1 Appreciation: Comment on some examples

As noted above, problems with ascribing values of APPRECIATION are usually associated with, for example, the indistinct boundary between someone's activities and their skill in performing them (which would therefore normally need to be described under values of JUDGEMENT), and the product of that skill as performance or 'thing'. The example which is frequently quoted involves a skilful batsmen in a cricket match, who can be judged as a 'brilliant batsman', or that 'he batted brilliantly' (ie, [judgement: capacity: positive]), or whose performance can be described as a 'brilliant innings' (in which case, a value of [appreciation: reaction: impact: positive] may be used).

The need for such a distinction only becomes obvious in context. When analysing texts, the main usefulness of these distinctions is in the search for patterns - patterns which help to trace the rhetorical organisation and development of any text as a semantic unit, and patterns which help to characterise the preferred and typical stylistic features of any writer or register. The means by which writers use values of Appreciation as tokens of Judgement is often that of experiential metaphor, that is, an activity which a social actor may undertake is construed as the product of that activity - which can then be evaluated - rather than the social actors themselves, or their actual behaviour as process. This is one means by which a writer might set up a textual persona, or style which appears 'objective' or non-judgemental.

This can be seen in the final paragraph/closing sequence (1:28 - 1:30) of text1 where the provoked Judgement of positive tenacity is not only related to the comment regarding the writer's 'resolve' to squeeze the most out of the hours providence has provided, but in effect summarises the evaluative positioning of the writer-as-self in the whole of the text, in which the concept of 'task' and its relationship to work towards goals is consistently positively evaluated, and with which the writer actively associates himself. In section 3.3.3 above, and 4.1 below, ambiguity between values of Appreciation and Judgement in another example from the text (1:5) is associated with the identification of organisational junctures in the text.

3.4.1.1 Appreciation or Judgement?

What needs addressing again at this point, and as outlined in section 1.2 above, one of the contentious areas associated with Appraisal analysis stems from its reliance on discourse semantic values, rather than those of the strictly lexicogrammatical: classes and functions of items are viewed as activated in co-text and context. When Addressers, their audience, and the analyst are acknowledged as part of that context, then it is obvious that each reading may entail a different set of meanings, dependent on situational variables. This is because context of culture, especially taken from a dialogic point of view, includes the text itself as well as all the possible alternatives which might have been selected to make its meanings in the situation and culture of which it is part. Hence one's reading position needs to be factored in to any analysis of a text's message, which speaks to my earlier contention that an ethnographic perspective, or participant-observer status is important in dealing with texts taken from a specific community of practice. There may be 'resistant' or 'compliant' reading positions, for example, and these may even produce different 'statuses' (Hunston 2000) for the evaluation in texts, which in turn would constrain evaluation as either Judgement or Appreciation. To make a complete Appraisal analysis of any text, it would not only need to be seen logogenetically as I am concentrating on here, but also from the perspective of the analyst's reading position (ontogenetically) and from the perspective of the development of various local genres (phylogenetically) as well. In order to look at these types of patterns which might substantiate various reading positions within a cultural genre, the use of large corpora and studies of phrases in context to reveal the nature of semantic prosodies (e.g. Louw 1993) appears promising, as mentioned above (1.2).

This is not to say that any reading of a text is possible, or that texts are subject to a radical polysemy, as Simpkins (1996) suggests. Certain attitudinal lexis, such as the item corrupt, no matter what the co-text, entails either negative or positive evaluation - unless the positioning strategies in the same co-text have worked very hard to introduce a socially marked value for this term (what is sometimes referred to as the relative saturation of a term). Other expressions of evaluative attitude may be less strictly negative or positive, and depend on co-textual signals to set up particular readings. While there may be multiple possible readings of stretches of text, these will therefore be constrained in systematic ways - which may need to make reference to factors outside the text itself, in what Thibault (1999: 561) refers to as global intertextual resources. Nevertheless, accounting for the potential interpretations of texts via their meaning-making resources in the co-text is regarded as the goal for discourse analysis of this kind:

In terms of how Appraisal views the text as instantiation of sets of possible meanings, Macken-Horarik observes:

As has already been stated, this thesis adopts the view presented above through a focus on texts produced in a written community of practice, as a means of tracing the 'mechanisms' through which possible meanings are made.

3.4.1.2 Appreciation or Judgement: comment on some examples

In the examples of Table 3.3 above, the fine line between the categories Appreciation and Judgement is evident in some of the examples. As an example, the difference between disrupted and dysfunctional in relation to the target family may seem very slim, but in this case is linked to the generalised, unmarked types of targets for the evaluative adjectives involved, in which dysfunctional evaluates the behaviour of the family members as an inter-relating group in process, and disrupted evaluates the overall composition of the group in toto, as a product of past behaviour. In general, a local grammar, or probe, will help to distinguish values of Appreciation from values of Judgement when the lexical item in question is adjectival: 'It was adjective of you + to (non finite clause with material, or mental process)'. So that, 'it was dysfunctional of you to do that', sounds possible in the grammar; whereas, 'it was disrupted of you to do that,' does not.

A similar problem of differentiation is highlighted in the uses of the lexical item 'plan' which generally functions as either a process or as a nominal head, thus lending its noun function a connotation of having been produced by human activity. The evaluative adjective is again more significant than the thing evaluated in these cases, such that, to use as example the utterances cited above, fluid is normally used to make assessments and evaluations of objects, events, or the finished product of human behaviour; whereas caring refers to an affectual response in which human agents are implicated: 'plans', having no subjectivity, may not be attributed with Affect. The very metaphoricity of this relationship signals a type of markedness useful in assigning category membership, as well as using the local grammar probe as test cited in the previous section. The choice of lexical item, especially when it is an evaluative adjective, thus has rhetorical significance in that values of Appreciation may often be used to provoke 'implicit' Judgements, and so help to 'distance' the evaluator, usually the Addresser, from the Judgements of behaviour they may wish to imply.

Sentence 1.11 (cited above: Text1) is also problematic in this way. Here, 'day' is given a value of Appreciation, while the writer claims only that his goal is to have the day he describes. The grammatical identifying relationship is between my goal and to have + [appreciation: positive] day. In the counter example given, the actual positive Judgement of Tenacity implied is brought to the surface: the identifying relationship in this case is between my goal and to be, thereby locating more closely his goal as part of his own inclination. These distinctions are useful in analysis of textual identity and ideological stance, since preferences for evaluating phenomena in terms of either behaviour or as object are one means of accounting for stylistic differences by different writers.

In order to trace patterns of both the sources and targets of Attitudinal value, and the Addresser's preferred orientation to such targets, this distinction between Appreciation and Judgement values needs to be made. Within Appraisal, the rhetorical functionality of the resources and options of the lexicogrammar in their relation to discourse semantic positionings are the focus of analysis. As in the case of values of Judgement, with Appreciation, an Appraiser need not be specifically mentioned, as the values of the Appraisal are located in the thing or person evaluated. This is in contrast with values of Affect which are located in the subjective experience of either an Appraiser, or 'emoter' (c.f. 3.2.1).

[note1] In fact, the 'target' of the appraisal in which Affect values are a token in Judgement prosodies, may be the 'emoter' rather than the cause (affector) of the Affect. see Module 3.

Next: Section 3: 3.5 Using Attitude in Textual Analysis

Back: Section 1: An Outline of Appraisal Theory

References

Index