We are now in a position to provide definitions of the three levels of informativity:
1. First order: this level is always present in a text and is typified by choices which are obligatory or almost so; 'function words' are a good example, since they contain little actual content, their role being logical and structural. So low is their informativity, that they are frequently omitted in such texts as telegrams and newspaper headlines and their function is easily inferred from the surrounding co-text and context.
2. Second order, this level represents the middle ground between first and third and arises when first-order expectations are not fulfilled i.e. where unexpected but not unlikely choices are made. For example, given a text which contains Coffee and tea are ——, a choice falling within the upper range of probability would be popular drinks (and several other possibilities which we have stored in memory). This would be true but very uninteresting; we all know that coffee and tea are popular drinks. However, if the sentence were completed dangerous drugs, we would have an example of second-order informativity. A search of our memory database for coffee and tea would, no doubt, throw up the fact that coffee and tea are drugs and are dangerous, though not in the sense or to the degree that alcohol or the 'hard drugs' are.
3. Third order. this level is attained by choices which fall outside the expected set of options and is typified by discontinuities, where information appears to have been omitted, and/or discrepancies, where what is being presented in the text fails to match with our knowledge; i.e. there is a mismatch between the text-world and the real world, as there is in the text we have just been considering. The classic poetic example is Dylan Thomas' a grief ago.
This brings us to the sixth of the standards of textuality: relevance.
Texts not only contain information, they possess a degree of relevance or situationality in so far as they exist for a particular communicative purpose and link communicative acts (discourse) to the situation in which they occur. Indeed, it is crucially important for the assessment of the appropriateness of a text to know where it occurred and what its function was in that situation. For example, what are we to make of this text?:
CHINESE TAKE AWAY FOOD
Unless we know the situation in which it occurs, we cannot work out what it is. Found in a newspaper above an item of news, the text is clearly a headline. Conversely, if the text is seen outside a shop, it is, equally clearly, a sign for a fast-food outlet.
The ability to discriminate in this way depending on the situation of occurrence is, of necessity, derived from 'real world knowledge'-knowledge of contexts of utterances, schemas, frames, etc. – and is mediated by our own personal goals, values and attitudes. Indeed, it has been argued that the 'acceptability' of a text is frequently judged not in terms of 'the "correctness" of its "reference" to the "real world" but rather... its believability and relevance to the participants' outlook regarding the situation'.
The passage we considered above is 'acceptable' in a science fiction story but not in a history text-book; part of our assessment of 'acceptability' relates to our knowledge of similar texts. There is, then, a need for a standard which recognizes this fact.
The final standard – intertextuality – refers to the relationship between a particular text and other texts which share characteristics with it; the factors which allow text-processors to recognize, in a new text, features of other texts they have encountered. What is involved here is the notion of genre or text-type and the crucial role played by knowledge of previous texts in 'making sense' of newly encountered texts (and, we might add, making texts of new sense).
We shall not attempt to build up a text-typology at this point but will limit ourselves to making the simple point that much of our frequent appeal to 'real world knowledge' has assumed, implicitly, knowledge of the forms and functions of texts. Were this not the case, we would have been unable to use the examples we have been using to flesh out the
discussion. We recognize a text such as
STOP CHILDREN CROSSING
as a direction to road-users and not as a political slogan, if we encounter it written on a circular board being carried on a black and white pole by a man or woman wearing a white coat who is stepping into the road followed by school-age children (all situationality; relevance), because we have come across such texts before. They belong to the genre 'road signs' and, for that reason, we know how to respond to them; we come to a halt before the line of children rather than rush across to them and try to prevent them from crossing the road!
责任编辑:admin