
COMPETENCE, LEARNING AND KNOWLEDGE
MANAGEMENT
Literature
Topic 8
PLANS AND SITUATED ACTIONS:
The problem of human machine
communication
Lucy A. Suchman

andreas.diedrich@handels.gu.se
Norbert
LUNG
Table of Contents
Introduction................................................................................................
Page 2
Plans............................................................................................................Page
3
The indexicality of
language.......................................................................Page
4
The mutual intelligibility of
action............................................................. Page 4
Communicative
Resources..........................................................................Page
6
Conclusion...................................................................................................Page
7
Introduction
As a starting point of this paper we will
discuss the purposes of plans, how they are created and which relation they
have to actions. We will also discuss, as it is a crucial part of our work, how
mutual intelligibility, or shared understanding, of actions is achieved and
which methods and resources are necessary for it. In a further part we will go
into more detail and switch from the planned action as a part of the plan’s
structure to the so-called situated action, which is taken ad-hoc, responsive
to certain material and physical circumstances in the situation itself, and
which is not described in detail in the prerequisite plan. Also in this part we
will investigate how mutual intelligibility is obtained and which resources and
methods we use to achieve it. Finally, we will discuss the communicative
resources which we use in everyday conversation that are necessary to create
mutual intelligibility and shared understanding for actions, especially
situated actions.
This should be just a short overview of the
topics we will cover in this paper, from this beginning point of view there
will be for sure a need for clarification. This will be attained throughout the
paper when we go more into detail.
It is useful to give a short example to point
out the topics that we’re going to discuss in order to give an overview of
their content and an overall picture of what the paper will be about.
Let’s assume there’s a canoeist who wants to go
down a wild river with lots of rapids and obstacles. Of course, he will first
look at the river from above, study its nature and according to his observation
he will make a plan which describes how to manage to go down the river, to pass
the rapids and obstacles with the lowest risk. But he’s looking at it from
above, so he’s not able to explain in detail which actions to take at every
point of the river. He just plans more abstract which points in the river he
wants to avoid and which to head for in order to have an optimal starting
position to react on unforeseen circumstances and finally to reach the end
point. But during the expedition down the river in his canoe, he will act
intuitively according to the particular situation he’s in. He has not planned
these situated actions in advance, because he couldn’t presuppose the
circumstances from above. He’s acting according to his background knowledge,
his experience and his skills in canoeing and probably this will not fit perfectly
with the plan he made in advance, but the aim stays the same. So for other
canoeists, his behaviour and actions are understandable because they have a
shared understanding. In contrast for non-canoeists, it might be not
understandable because they don’t share the same knowledge. But it can also
happen that even a canoeist doesn’t understand his actions because he was not
in the particular situation.
As you can see, to produce or to create mutual
intelligibility, different preconditions are necessary and there are certain
methods and resources available to obtain a shared understanding.
There are
two alternative views of plans. The first one describes a plan as the
prerequisite to action which describes it at every level of detail. Mutual
intelligibility is achieved through reciprocal plan recognition. On the other
hand, the prescriptive significance of plans for situated actions is inherently
vague and abstract. This vagueness makes plans an ideal resource or
representation for action rather than a prescription of those, while the detail
of intent and action must be contingent on the actual situation itself.
Therefore,
you can say plans are prescriptions for abstract actions and at the same time a
representation of situated actions, which will be described more firmly later
on in the paper.
In order to
achieve mutual intelligibility or shared understanding of action, three major
resources or methods are required:
In the
planning model the plan is treated as a sequence of actions which aim to
accomplish a preconceived goal. The plan can be seen as the prerequisite of the
action. In this case, the production of mutual intelligibility of action is a
matter of plan recognition. This process is simply an observation of actions
and then the formation of hypotheses about the plan which could motivate and
explain those actions. But the problem is that while actions can be explained
after the fact with reference to its effect, the action’s course cannot be
predicted by knowing the actor’s plan. Therefore there’s an indeterminate
relationship of planned effect and the action’s course. As a result a statement
of intent says little about the following action and we can state our
intentions without describing the action’s actual course. To afford this, an
intentional vocabulary is necessary to be able to communicate our plans and
intentions.
Language
can be seen as a form of action. If communication is added to the planning
model, the language understanding involves an analysis of a speaker’s
utterances regarding his plan.
People
maintain descriptions of the world models of other people, as part of their own
world model. Speech acts then affects the models that the people maintain of each
other. These speech acts produce utterances which aim at having some effect on
the hearer who then infers what the speaker’s plan is. Nevertheless, planning
model and speech acts are not sufficient enough to account for the mutual
intelligibility of actions. Another important resource is needed to achieve
shared understanding, namely background knowledge.
The
background of action is not the world as such, but our knowledge about the
world. This common-sense knowledge can be described as a classification of the
everyday world as types of situations. This shared common-sense knowledge are
implicit assumptions or presuppositions that stand behind every explicit action
or utterance and which helps the participants in interaction to understand each
other’s action. In other words, what is unspoken in interactions is assumed to
be part of the speaker’s and listener’s common stock of background knowledge.
As a result we can assume the intelligibility of our action and as long as our
interaction’s partner doesn’t fail to understand, we don’t need to explain
ourselves, but if necessary, the significance of our actions cannot be
explicated in so many words.
In order to go deeper in Lucy A.
Suchman ideas, one is confronted to the notion of situated action, which was
introduced for the first time in the book, so this notion dates of 1987.
Suchman defines situated action as “the view that
every course of action depends in essential ways upon its material and social
circumstances. Rather than attempting to abstract action away from its
circumstances and represent it as a rational plan, the approach is to study how
people use their circumstances to achieve intelligent action”. One could find
other definitions of situated actions but they drive us to the same point, so
to explain it in others words and in order to make it more understandable,
situated action is how people act in a situation and theses people's actions are influenced by
the context of their specific situation.
A new branch of the sociology appeared some years ago, its name is ethnomethodology. The ethnomethodology, which is mainly explained by the term situated actions, will be studied here through three points.
This part fits very well with the example of the canoe going though rapid introduced in the introduction. In this situation, the person is making a plan to go down the river with the lowest risks. During the river’s descent, the person will face many situations he did not anticipate because the situation at this place of the river was not the situation he expected and he could not have expected that. So, to cope with this problem, the person will use his experience, whatever embodied skills available and indications provided by the environment of the situation. So, during his descent, the person will meet many situations which are situated actions, because of their specificities and the unique character of the situation. It means the situation encountered is quite impossible to apply to another (we can make a parallel to the tacit knowledge which is shared with difficulty or even impossible). The plan is used just to get through the river, but after when it comes to deal with the currents and avoid rocks for instance, you drop the plan and you rest on your skills. By this fact, the goal during the river descent becomes to find the points in where you can maximise the use of your skills.
Moreover, even concerning unskilled activities, people do not really anticipate problems and do not plan alternative courses of action. It is when the action’s implementation is under way that we think about other alternatives.
This chapter is highly linked with the previous one. Indeed, one will cope now with the issue that has been underlined previously: a perturbation which occurs through a change in the situation. Indeed, previously we have seen people are mostly resting on the experience and embodied skills to cope with changes in plans or situations. But when it is impossible to use its capacities, people are aiming to solve it by using rules or procedures which are in majority explicit.
Situated actions have a central
place in the ethnomethodology because of the importance of talk and action in
the construction of the social world. Indeed, the everyday situations of
talking and living are relevant ways to learn from the world and to build it.
In others words, it helps to build a common-sense view of the social world.
The efficiency of language is crucial for our shared understanding. There are a lot of expressions in our language which can be used in many different ways and they have assigned to them conventional meanings, which hold on any occasion of their use. The significance of the expressions we are using in a particular situation lies in its relationship to the circumstances that are preconditioned in the expression itself. The meaning of the language we use, therefore, is connected or based on the embedding world. These expressions, namely actions which rely on the situation for significance, are called indexical.
For instance, an indexical expression would be:
“that’s a nice one”. This expression will have quite different significance
when a visitor refers to a picture in a photo album or a customer uses it
inside a grocery store when facing a fresh lettuce. Thus, the meaning of the
expression is very associated to the situation of its use.
In other words it can be said that expressions and also the interpretation involve an active process of focusing on and searching the situation of talk. It is to say, that language is a form of situated action. Therefore, the term indexicality is used by philosophers and linguists to distinguish classes of expressions whose meanings are conditional on the situation of its use from other expressions which are context-independent. The communicative significance, however, is always dependent on the circumstances and the embedding world of its use. When we communicate expressions we can never say everything we mean because we always include to our statements an implicit ceteris paribus clause and close with an implicit et cetera clause. We never mention all our circumstances. In this sense, time and place adverbs and pronouns are just illustrations of the fact that everything we try to communicate stands in an indexical relationship to the embedding world. Therefore, the interpretation is very important and turns on the unspoken situation of its use. Thus, instructions are often very difficult to be carried out in the way of the instructor’s intentions because they are related to a certain background. The goal of instructions is to produce action descriptions which lead to a particular interpretation and not the exact execution of the instructions. Therefore, the interpretation, or the mutual understandability is very important.
Situated language more generally in large measure constitutes the situation of its use. Here, language has a constitutive function. For a shared understanding or in other words, for a mutual understandability we need rules and norms which help us to communicate and make the purpose of actions meaningful to others. Therefore, we need something which is called known-in-common social conventions and behavioral norms. We agree on appropriate relation of actions to situations. Therewith, a particular situation can be handled by recognising it as an example of a class of typical situations. We walk into the situation, identify the features of it and adjust our actions to it. So, our tacit knowledge, rules and norms are crucial for our shared understanding. It is, however, not the case that the rules determine the actions but we use the normative rules to produce significant actions. The tacit knowledge mentioned before we do not learn explicit but we get through typification over families of similar actions and situations. To sum it up, we have many resources in communication that we need so that others understand our actions and these resources and the tacit knowledge we get from a long-term learning process within the social world.
Communicative
Resources
Finally, in this part we will discuss the
communicative resources to understand actions that are used in everyday life.
These are necessary to create mutual intelligibility and shared understanding
for actions, especially situated actions. In order to outline the breadth and subtlety
of the resources available for shared understanding, we will focus on
face-to-face interaction, the richest form of human communication, as it
incorporates the broadest range of possible resources.
Face-to-face interaction is not so much an
alternating of actions and reactions between individuals as it is a joint
action accomplished through speaking and listening. It is important to
highlight that the act of listening is essential and will determine the way
that the speaker finishes what they are saying. The listener must also act at
the correct time so that the speaker can take account of what is said and when
they continue to speak, they take account of the listener and validate their
actions. Here are the set of rules for turn taking in conversation:
1. The current speaker selects the next
speaker, e.g., by directing a question or other implicative utterance at a
particular hearer;
2. Another participant self-selects, by
being the first to start speaking;
3. The current speaker continues.
One important point to make is that
conversations are highly sensitive and adaptable to particular participants and
to unforeseen circumstances in the developing interaction. This is down to the
localness of the constraints on speakers’ constructions of turns-talk and the
contingency this has on the other speakers. Understanding the mutual
intelligibility will only come in conversation if the participants are aware of
the settings. One example that can be given is the medical examination at a
doctor’s clinic. Many problems arise in this situation because the patient does
not have the expert background knowledge to fully understand the situation.
Many people say the patient “hides” the problem, however, it may just be down
to the fact that they do not understand institutional purposes of the
interaction.
Basically, the patient in the above example
does not know the plan and is therefore only able to respond to the doctor’s
actions, the extent of this cooperation will determine how effective the
examination is. And for this, the fundamentals of conversation are required to be
used at a sufficient level. Implicit knowledge of conversation is essential and
is part of the background experience of participants in conversation.
Conclusion
In this paper we have explained the key concepts of the literature provided to us, ‘PLANS AND SITUATED ACTIONS: The problem of human machine communication’ by Lucy A. Suchman. Parts of the book are irrelevant to the subject of Competence, Learning and Knowledge Management in our context, thereforevwe have given the core lessons which can be learned.
We have discussed the purpose and importance of plans,
how they are created and the relation they have to actions. There are two
alternative views of plans, the prerequisite to action and in a more vague
sense, plans as a representation of action. We have also discussed how mutual intelligibility
or shared understanding of actions is achieved and the methods and resources
necessary for it because this is a crucial part of the literature. We discussed
in more detail, the switch from planned action as a part of the plan’s
structure to the so-called situated action, which is taken ad-hoc, responsive
to certain material and physical circumstances in the situation itself, and
which is not described in detail in the prerequisite plan. The section on
mutual intelligibility explained the importance and how this can be achieved
through a variety of methods and resources. For the situated actions section,
the branch of sociology, ethnomethodology was introduced and the five
point explaination lets us see that plans are a representation of actions and
also highlights the need for talk and action for the construction of the social
world. The topic of communicative resources brings
attention to the importance of face-to-face interaction and conversation as a
way of obtaining the mutual intelligibility gained from situated action. Using
this method, shared understanding of actions, especially situated actions is
realised.