Peter asked:

Hi, here I'm sitting in the middle of Bangkok and hesitating.

What are expectations?

How do I define expectations?

This definition is going to be the key element in research done amongst tourists. I'm trying to find out
what the expectations of eco-tourists are (nature, forest, people etc). My problem is that I can't
formulate questions without a theory and the theory needs a basic. If I use your answer (which I hope
you will allow) I will refer to the source in my Masters thesis at the Agricultural University of Norway.

I have had problems with finding an answer, and for two days I have been stocked in thoughts and
boring descriptions from www.

============

Ok, nobody else did it... Let's differentiate between goals and expectations. When you do that, what
do you get? A goal is a state or situation, something like that, which you want to attain, to make
something into, etc., in other words, a goal involves, usually, it would seem, some sort of effort on the
part of an agent, in order to reach it, construct it, or attain it in some fashion. An expectation, on the
other hand, is either a state or situation to which you assign, formally or not, some valence as to
whether it will exist, usually independently of your effort, for the most part, although I imagine that the
two can overlap: one could have an expectation of reaching a goal. But that latter would imply that
you think it probable that you'll reach the goal, which again puts a kind of passivity or inevitability into
the dynamic, perhaps. Or it's just that valence.

So a goal is a kind of concrete state or thing which you have no necessary likelihood for; an
expectation is either a) the likelihood that you assign, or b) a state which you dohave a likelihood for.
A likelihood is the evaluation, the valence, which can either be a probability or an estimated
probability or merely an emotional bias which is attached to... well, whatever you want to attach it to.

How's that?

And so we can extrapolate and claim that there will be both negative and positive expectation
hierarchies, right? That we can have gradations of things more likely, and also of things less likely...
at least if we think about it. It's interesting, in a way, that we can't, really, have negative
expectations... we can say that things are less likely, down to zero, they won't happen; but the only
way we can say there is a negative likelihood is if something preventssomething else from
happening... I wonder if there's any theory for that one; I mean, that might give you negative
probabilities, mightn't it?

Steven Ravett Brown

What I want to offer you is first of all a dictionary definition of 'expectation' and then an idiosyncratic
theory of expectation as an aspect of the phenomena of 'thinking' which I will place in the context of
the concept of a knowledge schema. All of which should give you some food for thought or further
research for your Masters.

Collin's Cobuild Dictionary, a dictionary of modern English Usage, exemplifies expectation as: a
strong hope that something will happen, or a strong belief that something will happen, or strong belief
that someone will develop into having a specific identity, or a strong belief that someone should
behave in a particular way. Lawyers for example understand promises to create an expectation in the
receiver in the senses given above.

So clearly 'expectation' falls squarely into the field of the concept of belief which could take us into
well trodden territory concerning knowledge and it's pragmatic definition as 'justified true belief', a
definition used by Nonaka and Takeuchi in their seminal work, The knowledge Creating Companyin
the context of their paradigm shifting work in 'Knowledge Management'. Which I mention because you
could consider the project you are undertaking as a problem of knowledge schema creation and
management.

We could find some disagreements with this definition of knowledge and its translation into 'justified
true belief' as used in this context given that both are based on the implicit propositional content of
knowledge with its associated problems of the verification of opinion or attitude as objects in a private
language and some doubts over the logical meaning of the phrase, 'True Belief' or 'False belief'
except as metaphor masquerading as theory. Whereas there is a growing body of thought that
supports the view that non-propositional knowledge is not simply the absence of thought and the
presence of random emotional impulses but that many aspects of thinking include an inseparable
propositional and value content although the balance of one over the other may vary according to
specific uses or linguistic 'forms of life'.

Goleman, for example argues in his book Emotional Intelligencethat the view of knowledge that
governments and parents colluded in developing up to now was based on a system education and a
concept of intelligence geared towards the acquisition of propositional information and the inhibition of
'emotional intelligence' to use his umbrella term. Similarly, Stevenson argued in his work, Ethics and
Languagethat ethical thinking contains both propositional content and separately emotive content
and that both channels of thought support systematic reasoning though they always remain mutually
exclusive.

What I want to argue is that we can develop a system of knowledge evaluation called 'SIFT'. This
system contains some logic-like features and is based on the idea that there is a central
organisational unit of thought that consists of the inseparable conjunction of factual or propositional
thought together with emotive or value based thought the underlying vehicle of which is an
'expectation nucleus', though the balance of each channel of thought differs for different contexts. In
mathematical thinking for example the value channel of thought is dominated by the propositional
channel at the object level but switches round at the heuristic or problem solving level.

In this system knowledge is structured in terms of the categories we can impose on a situation. These
divide into static values and dynamic agents, both of which occur together as distinct from occurring
alternately or dependently. The static values consist of the mutually exclusive categories of
satisfaction and non-satisfaction. The dynamic agents consist of the mutually exclusive categories of
promissory value; those agents that produce satisfaction or reduce dissatisfaction and alternately
agents of threat value that produce dissatisfaction or reduce satisfaction.

The static values are not simple in that they are made up from the distribution of elements from the
primary qualitative expectation field which itself is generated from the fundamental components of the
expectation nucleus. This conjugation leads to the generation of the primary 'Qualitative Expectation
Field' shown schematically below.

"Qualitative Expectation Field

Have and Want = Positive satisfaction

Have but not-Want = Frustration

Not-Have but Want = Dissatisfaction

Not-Have and Not Want = Negative Satisfaction"

The agents of change are not simple either in that there are two other general subtypes contained
within them, those that increase their protagonist value and/or decrease their antagonistic value,
represented symbolically as Pv+ or Tv+ and those that maintain their protagonist value or
equivalently counter the effects of reducing agents, represented as Pv0 or Tv0. One further division of
the static values allows us to complete the schema for a 'Sift' evaluation of knowledge in terms of
qualitative expectation. This comes with the introduction of a category for the objects in the
knowledge field on which we are currently not focussing our attention. This provides us with a vehicle
to which we adopt an attitude of Indifference. Static value then divide into those things on which we
are focussed or those things of which we are aware but currently have no interest. We could think of
this as a division between objects of knowledge that we currently consider essential and those that
we consider inessential.

Objects in the category of indifference are often assigned the numerical value zero while satisfactions
are assigned positive numerical values and non-satisfaction are assigned negative values, in games
and decision theory for example. There is a danger that we read the assignment of the value zero as
an indicator of non-existence, precisely because this is what it does often signify. In the context of the
analysis of knowledge in terms of qualitative expectation it is important to realise that the category of
indifference is not necessarily empty and further that we can recognise the existence of two special
kinds of indifference, labeled positive indifference, representing those objects of knowledge that we
have but to which we are currently indifferent and secondly labeled negative indifference,
representing those objects of knowledge that exist but we do not have and to which we are currently
indifferent. One final distinction needs to be made and that involves the placement into the earlier
major or top level category of non-satisfaction of the sub units of frustration and dissatisfaction.

In general then the Sift schema provides you with the following categories into which you can analyse
yours or others thoughts about situations and which represent "knowledge schemas that are built up
from past experience to make sense of the world such that our past experience leads us to form
expectations about objects or events and from which we anticipate what we are likely to encounter" (
Butler & McManus on Ulric Neisser in Psychology, a very short introduction,Oxford).

[Situation description] can be analysed into examples of [How things are], placing them into
categories of:

1.1 Positive Satisfaction

1.2 Negative Satisfaction

2.1 Frustration

2.2 Dissatisfaction

3.0 Indifference

And at the same time examples of [How these things could be changed], placing them into the
categories of change:

1. Change for the better

2. Value maintenance

3. Change for the worse.

The concept of 'expectation' has a well defined in use in games theory in which the expectation
nucleus or 'act conditioned pair' is the combination of a 'cost' and probability which is then combined
mathematically with other weighted pairs in the field of analysis to give an overall value for an event
outcome.

There is also a sense of 'likelihood' attached to the ordinary sense of 'expectation' as exemplified in
the language of promising in that an object, event or agent, including the speech event or act of
making a promise, has promissory value if it is more likely than not to bring about the desired
outcome.

Wilfred Hodges book, Introduction to Logic(Penguin) has a very interesting section in which he offers
a qualitative analysis of likelihood in terms of 'inequality' operators. So given this gradation of
definitions of expectation ranging from the mathematical, through the relational or qualitative logic and
cognitive schema to the dictionary exemplified ordinary usage what are the grounds for believing 'Sift'
analysis is an analysis of expectation? The grounds are that the concept offers a semantic as distinct
from a syntactic explication or mapping of the properties of the concept in that it creates categories
which are sensitive to the structures in the concept of expectation and represents them in a schema
through which it is possible to undertake symbolic manipulation, an essential feature of 'thought
schema's, (Luria The Working Brainand Fodor The Modularity of Mind) without over reducing them to
inappropriate logical entities. A complete field of study for example is concerned with exactly the
problem of how to represent 'expert' knowledge without over reduction in the context of knowledge
elicitation for 'expert systems', computer programs designed to reason.

The key concept of expectation has been represented in the Sift system in terms of the image of
change embodied in the mathematical, relational and usage aspects via the concept of promissory
and threat value as representatives of agents of change. A full analysis of expectation then, I would
argue leads to the development of an expectation schemaa part of which involves the assignment of
likelihood to the elements in its expectation nuclei. For example when you have classified aspects of
the knowledge field for Eco Tourism into the Sift categories mentioned above you can then also
assign likelihoods to their occurrence. You can also assign likelihoods to the promissory of threat
agents your data surveys produce and these would contribute significantly to the way in which the
primary expectation about Eco tourism are represented in marketing and advertising. (See my earlier
answer, below on analysing the Guinness advertisements). Attached to the transformational and
'marketing' aspect of expectation is an inferential aspect. For example teachers often find that setting
high expectations for students leads to higher achievement and conversely, setting low expectations
lead to lower achievement. Expectation in this sense has more to do with creating and sustaining
motivation and confidence than assigning probabilities of exam success.

Similarly, expectation has a 'look ahead' aspect that can lead us to reject our present position if we
are dissatisfied with the future position to which it leads us in a form of inference by modus tollens.

Finally and most important is the inductive effect or 'lending effect' of promissory value in which
present satisfaction is 'lent' from likely future satisfaction. 'I have satisfaction now because I believe I
will have satisfaction later'.

Again, the complete structure of an expectation schema for Eco tourism must take account of the
inferential aspects of the concept to give you some idea of the way that people could make decisions
based on the embedded inferential structures.

Neil Buckland