Cycles of planning
المؤلف:
Paul Warren
المصدر:
Introducing Psycholinguistics
الجزء والصفحة:
P20
2025-10-29
35
Cycles of planning
What, then, is involved in planning unscripted speech Let us take as a basic unit of planning something that we might call an idea’. Researchers have tried to operationalise what is meant by idea’ by asking nave participants to look at a tidied-up transcript of spontaneous speech, from which the disfluencies have been deleted, and to mark where new ideas start. There is remarkable consistency in where participants mark these transition points between ideas Butterworth, 1975 .
What is interesting from the point of view of how pause patterns might reflect the processes of conceptualisation is that if we look at the patterns of speaking and pausing that correspond to ideas’, then we frequently find that the early part of each idea is marked by a lot of pausing, but later parts have more speaking and less pausing Henderson, Goldman-Eisler Skarbek, 1965, 1966. This is illustrated in Figure 2.5, from an analysis of the monologue text we saw earlier. Each horizontal segment indicates a period of speaking, and each vertical segment is a period of pausing. These alternate, since each pause comes between two stretches of speech. Two points from the text are marked on this plot. The first, at so I arrived in lines 10–11 of the transcript, is at the beginning of a period of the monologue where there is a lot of pausing compared with speaking see the superimposed line marked A. The second, at o e a o in line 11 of the transcript, is where this relationship of speaking and pausing reverses, as shown by the slope of the line marked B. In unrehearsed speech, these periods of hesitant and fluent speech alternate in this way.
There was some early criticism of the basic claim that there were cycles of hesitant and fluent periods in speech, since it was shown that nave judges would detect similar cyclical patterns in graphs based on sequences of randomly generated numbers Power, 1983; Schwartz & Jaffe,1968). However, further analysis showed that the speech cycles coincide with ideas’ Butterworth, 1980; Merlo Barbosa, 2010; Roberts & Kirsner, 2000. The greater hesitancy at the beginning of a new idea reflects the fact that the expression of the idea has not been planned in detail in advance, but has to be sorted out once it is started.
Macro- and microplanning
Getting the various ideas organised in a way that is going to best suit their communication is part of macroplanning, i.e. deciding how to achieve an intended communicative goal using relevant speech acts. A speech act is the performance of some action through saying something, such as asking for information, making a promise, and so on Clark, 1996. Part of our knowledge about how language works is that we know what speech acts work well in achieving particular goals in a given context and for given addressees.
Not all goals have single corresponding speech acts. Some goals require multiple speech acts, such as giving route directions. Some speech acts achieve multiple goals, such as saying I saw the bastard last night, which tells the listener that the speaker saw the person in question at a particular time, and also conveys the speaker’s opinion of that person. When a communicative goal requires a series of speech acts, such as route directions, then the speaker needs to both select and sequence those speech acts. This involves linearisation, i.e. choosing the order in which information should be expressed. There may be different consequences to saying walk towards the sea at the edge of the cliff and stop at the edge of the cliff and walk towards the sea.
Some ordering is largely natural – e.g. relating a sequence of events, or the route from A to B Filipi Wales, 2004; lein, 1981. Other ordering appears to follow certain conventions that keep the structure of what is said as simple and as transparent as possible, perhaps to minimise the processing and memory load for the listener. Illustrations of these conventions come from studies that have used network description tasks where speakers are asked to describe a set of interconnected colour nodes so that the listener can draw the node pattern from their description Levelt, 1982. Chapter 5 gives more detailed description of how these node structures have been used in collecting production data, and in particular looks at their use in generating data relating to how speakers monitor and repair their speech.
When speakers get to points where a node network splits into two parts, they usually describe a shorter arm first, because this means that the description can return to the choice point sooner than if they had first described the longer arm. Also, they tend to follow a principle of connectivity in describing these node patterns as in other tasks such as route descriptions, or descriptions of the layout of furniture in a room. That is, rather than making jumps, speakers tend to describe things that are near to one another or somehow connected. As well as physical connections, this connection might involve the functions of the things being talked about, which may be important in the description of the contents of a room.
Another aspect of macroplanning is the selection of which information will go into a structure of the utterance and which will go into s e structures, e.g. additional comments, asides and embellishments not central to the main communicative intention.
The speaker also needs to decide how much and what sort of information to include. This depends on the important notion of instrumentality – speakers select information that helps them to achieve their communicative goals. Equally, they leave out information if they can assume the listener can make the appropriate inferences. For instance, if we can infer that the listener will know that the Beehive is a landmark building in Wellington it is a beehive-shaped government building, then it is suitably efficient to give a route direction that does not include detail about what the Beehive is, like Turn left at the Beehive. But giving an equivalently informative set of directions to an overseas visitor to New Zealand will require the planning of additional information structures to explain what the Beehive is.
It is clear then that considerable planning is involved in speech production. Some of this planning can be carried out prior to beginning speaking as in anticipating what you might say before you go to an interview, but most will be carried out once speech performance has started. If the information you have to express is readily available e.g. visible to you at the time of utterance, then you do not have to do a lot of macroplanning to decide what to express. But a more complex task, such as planning a route, giving a narrative account, making a speech, etc., involves more extensive searches for information and more ordering of utterances. These types of discourse tend to reveal the cyclical patterns of hesitant and fluent phases of speech.
As well as sorting out the type and order of speech acts in macroplanning, speakers carry out detailed planning of each individual speech act, known as microplanning. This involves determining the perspective and information structure that is most appropriate for a given speech act, and deciding what should be highlighted as new or topical information. Macro- and microplanning should be thought of as two levels of planning, rather than temporally distinct stages. That is, once a speaker has made initial decisions about the sequence of speech acts required to achieve some communicative goal, individual acts can be planned in more detail, even before the overall plan has been finalised.
The outcome of macro- and micro-planning is still not language’. It is a pre-verbal message representing the speaker’s ideas or propositions, and which still needs to be converted to linguistic form. In terms of the sketch of a language production model in Figure 2.1, we have reached the output of the conceptualisation process.
الاكثر قراءة في Linguistics fields
اخر الاخبار
اخبار العتبة العباسية المقدسة
