Chapter Nine
Model 7: Self as the dominant context of experience and action.
"... the total self (is) partly known and partly
knower, partly object and partly subject ... we may
call one the îMeï and the other the îIï... îïI shall
therefore treat the self as known or the îmeï, and ...
the self as knower, or the îIï ... " (p. 189).
--- W. James, îPsychology: Briefer Course.ï (1893)
Unidentified Guest:
" ... There's a loss of personality
Or rather, you lost touch with the person
You thought you were. You no longer feel quite human.
You're suddenly reduced to the status of an object ---
A living object, but no longer a person.
It's always happening, because one is an object
As well as a person. But we forget about it
As quickly as we can. When you've dressed for a party
And are going downstairs, with everything about you
Arranged to support you in the role you have chosen.
Then sometimes, when you come to the bottom step
There is one step more than your feet expected
And you come down with a jolt. Just for a moment
You have the experience of being an object
At the mercy of a malevolent staircase."
--- T.S. Eliot, îThe Cocktail Partyï
(1950, pp. 29-30)
9.0 Introduction.
9.01 The self-system is opaque to direct introspection.
9.02 Conscious selfİmonitoring may be guided by the self™
concept.
9.1 Contrasting self and not-self experiences.
9.11 The wide range of self/not-self contrasts.
9.12 Self-alien experiences as evidence for self as
context.
9.2 Modeling self and selfİconcept.
9.21 Self as deep context.
9.22 The selfİconcept system controls conscious selfİmonitoring.
9.23 Model 7.
9.3 Further questions to explore.
9.31 Normal voluntary action is largely conflictİfree.
9.32 Unresolved goal conflicts may persist outside of the
conflict-free sphere.
9.33 Explaining self-alien intrusions of thoughts and
images.
9.34 What happens if the highest-level context is
disrupted?
9.4 Summary.
‹F ‹
9.0 Introduction.
It was not the original intent in this book to deal with
"self." However, there are good reasons to think that we cannot
discuss consciousness adequately without introducing some
reasonable conception of self (e.g. Tulving, 1985; Dennett, 1978;
Nagel, 1974). This makes it necessary to explore the question,
though "self" really requires a separate book. There is of course
a profound literature on the psychology of self and other, with
major scientific, clinical and philosophical contributions (e.g.,
Freud, 19xx, ego & id, etc.; Hartmann, 1958; A. Freud, 1938;
Kohut, 19xx; Horowitz & Zilberg, 1983; Markus & Sentis, 19xx;
Kihlstrom & Cantor, 1984). We cannot deal with this great
literature here with any adequacy, but in a tentative way we can
outline some ways in which the current theory may make contact
with it.
We take the viewpoint here that some notion of "self" in
psychological theory is not a luxury, not a metaphysical or
artificial issue, but a necessity for any complete psychological
framework. In this respect "self" is like consciousness, a core
psychological problem which stubbornly survives all attempts to
ignore or circumvent it. Selfİother differentiation is a central
concern in perceptualİmotor systems, in motherİchild interaction,
in the development of autonomy, and even, as recent good evidence
indicates, in the workings of the immune system (Melnechuk,
198x).
Some commentators suggest that consciousness is essentially
"the domain of access of the self." Thus Dennett (1978) writes:
"That of which I am conscious is that to which I have
îaccessï, or (to put the emphasis where it belongs), that to which
îIï have access ... "
This idea has a good deal of intuitive plausibility. It
certainly fits our ordinary language: All statements about
conscious experience use personal pronouns, as in "I saw a
pussycat," "You are only imagining that pain in your tummy," "She
smelled a rat," etc. Certainly we would be surprised if we were
unable to access consciously some vivid recent memory, some
sight, smell or taste in the immediate environment, or some well-
known fact about our own lives. The "self" involved in conscious
access is sometimes referred to as îthe self as observerï.
Similarly, all commonsense statements of voluntary control have
"self" as the initiator or actor in charge, as in "I told him to
go," "He decided to find out more," and "I am responsible for my
own actions." Again, we would be surprised and upset if we were
unable to move an arm, to stop an intended speech act, or to
control a usually controllable conscious desire. The controlling
agency for this expected domain of voluntary control is sometimes
labeled îthe self as agentï (e.g. James, 1890). ‹j ‹ċ
A number of behavioral psychologists maintain that the
notion of self is a delusion of common sense; perhaps we simply
infer a non-existent agent in charge of our actions and
experiences, creating an imaginary entity where there is none.
Certainly people sometimes make false inferences. The scientific
question is, of course: is there an underlying reality that
justfies the inference? If there is, then "self" is not
delusional; it is something we need to understand.
In this chapter we develop the idea that "self" can be
operationally defined as îthat system whose change or violation is
spontaneously interpreted as a loss of the sense of self.ï The
Jamesian "I," in other words, is only knowable directly by the
experiences that take place in moments of challenge and change.
This makes sense theoretically, because we can interpret self as
an enduring, dominant context, near the top-most levels of the
dominant context hierarchy. We have previously cited evidence
that contexts can be experienced only through mismatch ---
through change or violation in contextual expectations (see
x.xx). In the same way, it may be that self can only be
experienced (and hence reported) through mismatch.
The idea of "mismatch with self" leads naturally to a
contrastive analysis of self vs. not-self reports. People report
self-alien experiences in many situation that seem superficially
just like the normal, self-consistent experiences that most of
us have most of the time. Spontaneous self-alien experiences are
reported in disorders like depersonalization, psychogenic fugue,
and multiple personality (9.xx). Reliable evidence is available
about these conditions, and they are all grist for our mill. We
develop a contrastive analysis based on this evidence.
We will conclude that the "self" can be viewed theoretically
as îthe enduring higher levels of the dominant context hierarchy,
including both conceptual and goal contextsï. Thus the self-system
is more than just another knowledge representation --- it is
knowledge that provides the framework for all conscious
experience. Self, in this sense, is a perspective, a point of
view, an overarching context for the flow of conscious events. It
has perceptual-motor, evaluative, conceptual, motivational, and
social aspects. The self-system evidently mediates and creates
continuity among more local contexts.
The word "self" will be used as an abbreviation for "self-
system," and contrasted to the self-îconceptï, which is a set of
beliefs îaboutï oneself. The self-concept corresponds to James' Me.
Like any context, self has aspects that can be de-contextualized
and experienced as objects of consciousness (x.xx). These
objectified aspects of self can then be used to construct a model
of self; but, contrary to some suggestions, we suggest that this
model of ourselves is not the self itself. When T.S. Eliot's
Unidentified Guest in the epigraph remarks on the experience of
stumbling on the staircase on the way to a party, he is pointing‹j ‹
to a moment where an aspect of the self as context comes to be
experienced as conscious content or object. Stumbling is a
violation of expectations, of course, and suddenly, from being in
charge and confident of one's reception at the party, one becomes
an object "at the mercy of a malevolent staircase." We humans are
often surprised by our own reactions to a new situation,
suggesting again that the self (as dominant context) and
self-concept (as one's beliefs about oneself) are not the same.
However, in the normal course of events we are continually,
smoothly switching back and forth between self as context and
aspects of self as objects of experience.
Thus the "self" corresponds to what James called the "I,"
while the self-concept, insofar as it is consciously accessible
corresponds to James' "me." As a set of beliefs about oneself,
the self-concept is based on experiences of oneself îïîïîas if seen
From an outside perspective.ï When people are asked abo
themselves they will express some part of their self-concept, but
the self itself is not immediately available to put into words.
For this reason it is best to avoid using the beliefs people
express about themselves as evidence for the self; but it is, of
course, evidence about their self-concepts.
The self-îconceptï may function as a monitoring system within
the larger self system. One's beliefs about oneself, including
beliefs about how one îshouldï be, can serve as a tool to evaluate
and control thoughts and actions. In adults, most voluntary
activities can be quickly tested against the self-concept. That
is, most of the time adults can answer the question, "Is what I
am doing right now really what I 'should be' doing?" In the
language of psychodynamics, the self-concept includes the
ego-ideal (Freud, 19xx). Severe violations of the self-concept
are experienced as shameful, guilt-provoking, or depressing, as
we fall short of our ideal self. Matches with self-concept may be
experienced as pride, self-acceptance and satisfaction.
In sum, we explore in this chapter both a theoretical
opportunity and a necessity. The necessity comes from the fact
that any discussion of consciousness seems incomplete without
appeal to some plausible self-system; the opportunity lies in the
fact that we can develop one such concept straightforwardly from
what we already know. The resulting perspective has implications
for numerous issues, including motivational conflict, emotion,
the perceived freedom of voluntary action, the disruptive nature
of emotional upset, impulse control, and attributional ambiguity
(9.6; Mandler, 1975; Weiner, 19xx). We will now develop these
ideas in detail.
‹j ‹ċ 9.01 The self-system is opaque to direct introspection.
How could we approach the organization of "self"? First of
all, we can try to pinpoint a set of empirical operations that
presumably reflect it. There are a number of reasons to believe
that the evidentiary basis of self may be different from the
evidence for our readily available concept of ourselves. Here are
a few reasons for this distinction.
îResistance to self-knowledge.ï
Perhaps the most obvious reason to differentiate between the
self-system and self-concept are the psychodynamic barriers to
self-knowledge, the extent to which we profit from self-
deception (Holmes, 1972, 1974; Goleman, 1985). One does not need
to accept all of psychodynamic theory to believe that wishful
thinking (and sometimes catastrophic thinking) stands in the way
of an accurate self-concept. Indeed, even the scientific skeptics
do not question the existence of pervasive self-serving
distortions about ourselves (Holmes, 1978). In everyday life we
surprise ourselves so often with unexpected feelings, actions,
and images. If we knew ourselves well this could not happen.
Accurate self-knowledge seems to be culturally and
developmentally sophisticated, and rare; it may always be
incomplete.
îOther source of incorrect self-knowledge.ï
But we do not have to appeal to wishes and fears,
repression, or emotional conflict to note the absence of accuracy
in many self-descriptions. A number of social psychological
studies show that people often make false claims about themselves
when there seems to be little motivational pay-off in doing so.
These studies emerge from two streams of investigation within
social psychology, one focused on errors of attribution, and the
second on the induction of cognitive dissonance. In a typical
attribution study a subject may be led to believe that the sound
of an accelerating heart-beat is the sound of his/her own heart.
This false feedback has been shown to affect subjects'
perceptions of their own emotional excitement, say, in the
context of threatening stimuli (e.g. Valins & Ray, 1967; see also
Schachter & Singer, 1962). Further, a wide variety of cognitive
dissonance studies show that subjects will evaluate an unpleasant
event more highly if they are given îinïadequate justification for
their involvement in the unpleasant event. Presumably they create‹j ‹
a higher evaluation justify their involvement îpost hocï (e.g.
Zimbardo, Cohen, Wisenberg, and Firestone, 1969). In general,
these studies show that people are consistently inaccurate in
their descriptions of their own motives.
Nisbett & Wilson (1977) claim on the basis of such studies
that human beings have no privileged access at all to their own
processes. This conclusion has been widely criticized as
overstated (e.g. White, 1982; Ericsson & Simon, 1984; viz.,
4.xx). The critics have pointed out that there are many
persuasive cases of accurate introspection, and that we really
need to know is îunder what conditionsï we can expect accurate
judgments about ourselves. Nevertheless, the evidence remains
strong that people cannot tell us about themselves much of the
time, even when common sense would expect them to be able to do
so. For example, when people choose from a display of identical
stockings, they will tend to choose the rightİmost, or the
best™illuminated stockings. Asked to explain their preference, they
will generate hypotheses with an air of conviction; but they will
not know the reasons for their action. There are numerous
examples of this kind, showing that very often people do not know
their own motives. The Langer & Imber (1979) on misinterpreting
one's own automatic performance is very much along the same line.
In general, human beings greatly overestimate the accuracy of
their self-knowledge, and seem to freely fabricate answers about
their intentions, especially when the information available to
make the judgment is inadequate.
îSelf-concept is typically oversimplified.ï
Another reason for doubting the identity of self-concept and
self is the extraordinary oversimplification that seems to
characterize our self-concept. The self-concept seem to be value-
laden, reducing the complexities of living to remarkably simple
"shoulds" and "wants." These voluntarily accessible beliefs about
ourselves often seem to be context-free and absolute. "I'm a
likable person. I'm very efficient. I have a lot of friends." In
contrast, the organization of the self-system, as we will see
below, seems to be highly complex, multilayered, and adaptive. In
the overall self-system, the "self-concept" may play a monitoring
or supervisory role (see x.x).
Research on thought monitoring by Singer and his co-workers
suggests an explanation for the remarkable oversimplification of
the self-concept (e.g. Pope and Singer, 1978). If we were to
track every bit of inner speech produced by one person, day after
day, we would quickly fill volumes. Even disregarding other
conscious events --- mental images, evanescent feelings,
percepts, and the like --- the stream of consciousness is
lengthy, constrained by numerous accidental and local factors,
often self-contradictory, and complex. When we are asked to‹j ‹
characterize ourselves in a phrase or two, we are forced to
summarize this rich lode of information. And the fact is, of
course, that people are often unable to produce accurate
summaries for great amounts of diverse information (e.g. Tversky
& Kahneman, 1973; Newell & Simon, 1974; Ericsson & Simon, 1980).
With the best will in the world, and even absent all of the
motivational distortions in self-perception, we simply cannot
describe ourselves very well.
This does înotï mean that it is hopeless to ask people about
themselves and expect accurate answers. Rather, it is vital to do
so under optimal circumstances, and not to expect people to have
access to the deeper layers of their own organization. Further,
our previous discussion throughout this book suggest that people
only learn about contextual representations by failure of those
representations; we can check reports of such violations
objectively, and see if the result accords with voluntary self-
reports. Inferences made about ourselves on the basis of these
surprising events may lead to a different understanding of
ourselves than our normal self-concept indicates.
9.02 Conscious selfİmonitoring may be guided by the self™
concept.
In yet different language, we can say that the self-concept
as an object of conscious thought and experience may be
considered to be îan object-like analogue of the self-system.ï The
self-concept represents self as an object of knowledge. But self
is not in the first intance an object of knowledge; it is
contextual. We can of course monitor many aspects of self at
will: our orientation in space, the loudness of our speech, our
social acceptance by others. These events are objects of
consciousness that are usually compared to some set of criteria:
where are we compared to where we want to go? Should we be
speaking louder? Should we curry favor with others? Such criteria
are presumably part of one's selfİconcept. The self-concept is
that part that is always looking at ourselves from a real or
imagined outside point of view. At the social level, it is as if
we are always asking, consciously or automatically, what will the
neighbors say? And if not the neighbors, then parents, peers,
siblings, professional colleagues, teachers, or the judgment of
history?
Self, on the other hand, may be considered to be the cross-
situational context of experience and action. Our consistent
expectations about the world are unconscious; the more‹j ‹
predictable they are, the less they are likely to become
conscious. All of our experience is shaped and defined by these
unconscious contextual expectations: perceptual, conceptual,
social, communicative, scientific, etc. (4.xx). Even our actions
are generated and interpreted in a context of goals that are
mostly unconscious at the time we perform the actions. One way to
think of "self" is as a contextual organization that seldom
encounters contradiction, because it remains largely predictable
across the situations which we normally encounter. But once the
predictable situations of our lives change, aspects of ourselves
which previously provided adequate context for our experience are
violated, and need to be changed; these stable presuppositions
may then be perceived as object-like, even though they were
invisible components of self before.
9.1 Contrasting self and not-self experiences.
As James points out at the start of this chapter, the "I" is
difficult to know directly. Certainly just asking people about it
is problematic, because by definition we do not have direct
conscious access to it: It is, in Jamesian language, the knower
rather than the known. However, we can approach "self as knower"
empirically with a contrastive analysis, just as we have done
before with conscious experience and volition (2.00, 7.00). We
will find a great range of evidence for self/not-self contrasts
in perception and motor control, in social situations, self-
evaluative experiences, psychopathology, and the like. The
empirical evidence regarding self is actually plentiful and
well-studied, once we know where to look.
9.11 The wide range of self/not-self contrasts.
When the eyeİball is gently pressed with a finger, the world
seems to jump; but, as Helmholtz noted in the 1860's, this does
not seem to happen with normal eye movements (Helmholtz,
18xx/ 1962). Evidently the visual system can distinguish between
selfİgenerated and externallyİinduced movements. Somehow
self™generated movements are compensated for, so that the experience
of the world remains stable in spite of our movement. Selfİother
differentiation is absolutely necessary not just in the visual‹j ‹
system, but in any sensory system, natural or artificial. If a
radar dish rotates at a regular rate and detects an apparently
moving object, it must differentiate between movements due to its
own motion, and those that are due to object itself. Otherwise
the moon can be interpreted as a rapidly moving object, and a
passing flock of birds as a stationary object in space. Thus
selfİother differentiation is fundamental indeed, even in
perceptualİmotor systems.
One can easily show the same need for selfİother
differentiation in the social world, or in the realm of self™evaluation
and personality; the point is that we need some
conception of self as a multiİlayered entity with perceptual™motor,
social, personality, and other components. We will focus
here on the personality realm, contrasting selfİattributed and
selfİalien experiences, but with the clear understanding that
these layers of the selfİsystem cannot ultimately be separated.
Amputation of a limb will impact perceptualİmotor processes most
directly, but it may create major changes in personality and the
socially defined self as well.
9.12 Self-alien experiences as evidence for self as context.
Just as we have contrasted comparable conscious and
unconscious processes throughout this book, we can also compare
cases where "self" is perceived as "not-self," or as "another
self." The most radical, well-established cases involve
psychogenic fugue, multiple personality, and depersonalization
disorder. The standard psychiatric diagnostic manual DSM III is
an authoritative source on these conditions (Spitzer, 1979). We
will briefly review all three syndromes.
Note that if self can reasonably be viewed as a dominant
context, we can make some predictions about the effects of
violating it. We know well that contexts can be decomposed by
events that violate contextual expectations and intentions
(4.xx). These violative events may be either internal or external
in origin, but we should certainly expect "shocking" external
events, which are easy to observe, to trigger a disruption in
the stable dominant context. Further, since context helps to
shape, control and evoke conscious experience, some changes in
the contents of consciousness may also be expected under these
conditions. Fundamental life changes should sometimes evoke
surprising alterations in thought, images, inner speech,
feelings, and perhaps even perception. Personal values may
change, because after all values are associated with the dominant‹j ‹
goal context --- all values posit a claim that one thing is more
desirable (hence goal-like) than another. Finally, this point of
view predicts that after fundamental life-changing events, people
may lose touch with their assumptions about reality as well ---
these are after all viewed in GW theory as part of the conceptual
context. All of these predicted features are found the following
self-alien syndromes.
îDepersonalization disorder.ï DSM III describes this as "an
alteration in the perception and experience of the self so that
the usual sense of one's own reality is temporarily lost or
changed. This is manifested by a sensation of self-estrangement
or unreality, which may include the feeling that one's
extremities have changed in size, or the experience of seeming to
perceive oneself from a distance. ... the individual may feel
'mechanical' or as though in a dream. Various types of sensory
anesthesias and a feeling of not being in complete control of
one's actions, including speech, are often present. All of these
feelings are ego-dystonic (self-alien)..." (p.259) Mild
depersonalization is quite common: it is estimated to occur at
some time in 30% - 70% of young adults.
Depersonalization has many of the expected features. First,
it is often triggered by severe stress, such as military combat
or an auto accident, physical pain, anxiety, and depression. A
similar syndrome can occur after brain-washing, thought reform,
and indoctrination while the captive of terrorists and cultists
--- all cases in which routine, dominant goals and perspectives
are profoundly challenged. These facts are consistent with the
notion that disruption of the self involves deep
context-violation. Indeed, stress may be defined as a deep
violation of expectations and intentions (goal and conceptual
contexts) (ref. Horowitz, 1976). îOnsetï of depersonalization is
therefore likely to be rapid, as is indeed found, while recovery
may be slow, because it takes time to reconstruct a disrupted
fundamental context. The high incidence of depersonalization in
early adulthood is also significant, since people often establish
their fundamental goals and expectations during this period in
life, while at the same time going through major life changes
which may challenge a new, tentative integration.
Second, there are evidently changes in the way victims of
depersonalization experience themselves and the world, consistent
with the fact that contexts constrain conscious experiences
(4.xx). Along these lines, DSM III states that "derealization is
frequently present. This is manifested in a strange alteration in
the perception of one's surroundings so that a sense of the
reality of the external world is lost. A perceived change in the
size or shape of objects in the external world is common. People
may be perceived as dead or mechanical... Other associated
features include ... a disturbance in the subjective sense of
time." (p. 259) Evidently, as the self is challenged, the
perceived world may also be estranged.
‹j ‹ċ îPsychogenic fugueï provides another example of a self-alien
syndrome. It involves "sudden, unexpected travel away from home
or customary work locale with assumption of a new identity and an
inability to recall one's previous identity. Perplexity and
disorientation may occur. Following recovery there is no
recollection of events that took place during the fugue." This
diagnosis is not made if there is evidence for organic disorder.
Again, this disorder can be seen to be a result of deep
violations of the normal dominant context, followed by an effort
to create a new dominant context, free from the environment that
created insupportable problems for the original identity. Fugue
typically "follows severe psychosocial stress, such as marital
quarrels, personal rejections, military conflict, or natural
disaster." (p. 256) It seems to be related to îpsychogenic
amnesiaï, in which a loss of memory occurs after severe
psychological stress. Amnesia sometimes involves a loss of
personal identity, but no purposeful travel, and no assumption of
a new identity. Note, by the way, that we have encountered
spontaneous amnesia before, in our discussion of highly
hypnotizable people, who often have spontaneous amnesia for the
hypnotic session (7.xx). This is consistent with the notion that
high hypnotizables enter a deep, absorbed state, in which they
are guided by a context that differs radically from their post™hypnotic
context, so that there is relatively little in the way
of recall cues available to them afterwards. The issue of
spontaneous amnesia and loss of autobiographical memory is indeed
a key to the notion of self we are developing here.
The most famous example of self-altering pathology involves
îmultiple personalityï. (James, 1890; Prince, 19xx; Spiegel, 1984;
Hilgard, 1977). Here, too, an eclipsed personality reports a gap
afterwards in the flow of experience, just as do victims of
amnesia and fugue. "The essential feature," says DSM III, "is the
existence within the individual of two or more distinct
personalities, each of which is dominant at a particular time.
Each personality is a fully integrated and complex unit with
unique memories (sic!), behavior patterns, and social
relationships that determine the nature of the individual's acts
when that person is predominant. ... Studies have demonstrated
that different personalities may have different responses to
physiological and psychological measurements. One or more
subpersonalities may report being of the opposite sex, of a
different race or age, or from a different family than the
original personality ... The original personality and all of the
subpersonalities are aware of îlost periods of timeï..." (italics
added, p. 257). Sub-personalities may hear each other or speak to
each other, but often with a sense that the voice heard is self™alien ---
outside of the self of the current dominant personality
(Psychiatric Clinics of N. America, 1984).
Again, we can make an argument for a causal role for some
deep challenge to the normal dominant context of intentions and
expectations. Thus it is said that "transition from one
personality to another is îsudden and often associated with‹j ‹
psychosocial stress.ï" (italics added, p. 257). Spiegel (1984)
has made the case that multiple personality syndrome is
îinvariablyï associated with a history of severe traumatic abuse in
childhood. He suggests that children learn when abused to enter a
radically dissociated state, which develops over time into an
complete, differentiated self. It is easy to interpret these
ideas within our current framework.
Recent work with multiple personalities indicates that there
is often a "regulator personality," one that keeps track of and
mediates between other sub-personalities (Psychiatric Clinics of
N. America, 1984). Spiegel suggests on this basis that the normal
self, too, may function as a regulator, integrating experience
across different situations (personal communication, 1986). This
again is consistent with the notion of the self as a dominant
context, one that creates continuity across subordinate contexts.
Note the repeated theme of gaps in autobiographical memory
in the selfİalien syndromes. Autobiographical memory is of course
selfİattributed experience, and in a GW framework, if self is
identified with deep context, we know that it must shape and
select characteristic experiences.
The everyday example of deeply absorbed states reveals the
same cluster of phenomena: gaps in autobiographical memory, loss
of time, and a changed sense of self. It is often difficult to
remember a period of absorption later, and time seems to have
gone faster in retrospect, presumably because we can recall fewer
details of the absorbed period (Ornstein, 1969). Finally,
absorption is strongly associated with identification with
fictional characters in movies, plays and novels İİİ i.e., a
change in the sense of self.
Another common theme in the selfİalien syndromes is the
relationship between a loss of self and losing valued or
presupposed conditions of one's life. It seems as if the more we
rely upon something in dealing with the world --- upon an
assumption, a personal capacity, a skill, or a goal --- the more
its loss will lead to selfİalien experiences. This may be
particularly true when the lost element is presupposed, so that
we no longer even know that it is there as an object of
experience. If we have assumed all of our lives that we can trust
people completely, so that this assumption has simply become part
of our dealings with them a deep violation of trust will have
consequences that propagate widely throughout our selves and our
experiences. Presumably, if we rely less on this sort of thing,
violations of trust will not be so disruptive.
‹j ‹ċ9.2 Modeling self and selfİconcept.
We will now attempt to model these observations. This will
prove surprisingly easy, because we can assimilate all the
aforementioned facts to the theory we already have. Figure 9.2
shows how the deep goal contexts and conceptual contexts can be
viewed as different aspects of self. This is of course James'
self as "I" İİİ as the observer and agent İİİ rather than self as
an object of experience. In general, the deeper, more
predictable, and more fundamental levels İİİ those that lower
levels depend upon İİİ are more selfİlike, in the sense that
their violation will propogate throughout the goal hierarchy and
lead to disruptive, surprising, and selfİalien experiences and
even actions.
v * vİİİİİİİİİİİİİİİİİİİİİİİİİİİİİÉ
v * vInsert Figure 9.2 about here.É
v * vİİİİİİİİİİİİİİİİİİİİİİİİİİİİİÉ
9.21 Self as deep context.
Because the contextİhierarchy can control the Options
Contexts discussed in Chapter 8, the selfİsystem has routine
access to all sensory modalities, to immediate memory, recent
autobiographical memories, routine facts, to longİterm personal
"marker" memories, and future or fantasied images. In addition,
we have indirect voluntary access to a host of specialized skills
(like English syntax and motor control) which are not conscious
in the qualitative sense --- we do not experience our syntactic
rules directly --- but whose unexpected absence would create
great conscious surprise. This is of course the point Dennett
(1978) remarked upon in the passage quoted at the beginning of
this chapter, the notion that self is that which has access to
consciousness. A major, rapid change in the access conditions of
any of these domains may be perceived as a selfİalien change in
one's experience. Thus loss of memory should impact one's sense
of self, as should sudden blindness, or even a sudden increase in
one's ability to imagine things. Any rapid change violates
contextual predictability, but changes consistent with one's
goals should be relatively more acceptable.
The same may be true on the output side. We expect voluntary
control over our skeletal muscles; over many mental functions,
like the ability to recall this morning's breakfast, or the
ability to express our thoughts; over many objects in our
environment; over people, to some extent; and, within limits,
over some social institutions. We can even control autonomic
bodily functions through mental images. A loss in any area of
expected control may be perceived as a profound change in self.
These functions are controlled by the goal hierarchy, and are‹j ‹
normally selfİattributed. Recall that English sentences
describing an act of voluntary control always take a personal
pronoun as subject (9.0). Hence people should experience a loss
of control over one's body, over the social domain, and even over
deeply presupposed possessions İİİ such as a car or a house İİİ
as a selfİalien experience. Notice that such a selfİalien
experience is not just a sense of loss, sadness, or mourning İİİ
these feelings may all be selfİattributed. It is rather a sense
of things being out of control, of surprising thoughts and
feelings and images, that characterize selfİalien experiences.
9.22 The self-concept system controls conscious selfİmonitoring.
Next, we attempt to model the self-concept. The selfİconcept
presumably emerges from many conscious experiences of self™monitoring,
and comparing the results to real or imagined
evaluations from other people. It has been said (rather
cynically) that conscience is the still small voice reminding us
that someone may be looking. It may be the fleeting ideomotor
image that says, "What would the neighbors think? What will Daddy
say when he comes home? My friends would laugh at me if they saw
me now. Any person of the opposite sex simply must admire my
looks, strength, wit, and intelligence." Such sentiments are
utterly commonplace and must surely have an effect on one's
self-concept. Indeed, an effective therapy for depression is
based on the assumption that such rapid, nearly automatic
thoughts are the key to depressive self-denigration (Ellis, 1962;
Beck, 1976).
Figure 9.22 shows a case of such conscious selfİmonitoring
as an Options Context with a variety of subİcontexts, including
tracking of one's own performance, calling upon the selfİconcept
for comparison, and adjusting the selfİconcept up or down in
value, depending upon the results.
v
vÉ İİİİİİİİİİİİİİİİİİİİİİİİİİİİİİ
v * v Insert Figure 9.22 about here.É
v **vÉv * vİİİİİİİİİİİİİİİİİİİİİİİİİİİİİİÉ
The selfİconcept system can then be treated as a highİlevel
context, operating within the selfİsystem, and making use of the
kind of conscious selfİmonitoring shown in Figure 9.22 to control
and evaluate one's own performance. Over time, conscious
self-monitoring experiences, like any other predictable‹j ‹
experiences, must become contextualized. Thus selfİconcept
apparently begins to function as a part of the larger self-system
(x.xx). The contextualized aspects of the self-concept are of
course less available to voluntary retrieval, just as any other
context is hard to recall voluntarily. Further, aspects of
oneself that are not acknowledged in the self-concept are
notorious for influencing human actions and experiences. The
entire psychodynamic tradition of the last hundred years is
devoted to the study of these phenomena. Perhaps all human
beings have potential conflicts between those aspects of the self
that match our self-concept, and those that are disavowed.
Given reasonable conceptions of self, conscious self™monitoring,
and selfİconcept, we can now show an integration of these ideas in
Model 7 (Figure 9.23).
9.23 Model 7.
As suggested above, the selfİconcept system can be treated
as a goal context that makes use of selfİmonitoring and
evaluation in order to move one's performance closer to the
ideal, as shown in Figure 9.23. Notice that this goal encounters
competition from other goals; perhaps there is resistance to the
effort demanded by the selfİconcept system to reach its goals;
perhaps there are goals aiming to obtain more immediate rewards
than are allowed by the selfİconcept; perhaps there are goals
expressing impulses that are not consistent with the self™concept.
All these goals systems may compete with the self™concept system.
The goal hierarchy that is consistent with the
selfİconcept comes very close to the idea of the "conflictİfree
sphere of the ego" articulated by the egoİpsychologist Heinz
Hartmann (1958). That is, it is a domain in which the system
that always tries to control our actions, the selfİconcept
system, coincides in its aims with other major goals. Thus there
is no destructive competition between different goal systems.
Presumably most of our normal, voluntary actions are guided by
this conflictİfree sphere of the goal hierarchy.
v * vİİİİİİİİİİİİİİİİİİİİİİİİİİİİİİÉ
v * vInsert Figure 9.23 about here.É
v * vİİİİİİİİİİİİİİİİİİİİİİİİİİİİİİÉ
9.3 Further questions to explore.
9.31 Normal, voluntary action is largely conflictİfree.
These considerations create a new perspective on the issues
of conscious experience and voluntary control. In particular, the
conflictİfree sphere provides an "executive" of sorts that is in
control of most voluntary action. That is, there is a set of
goals and expectations within the larger self-system that are
acceptable to the self-concept, and within this shared domain,
actions can be planned and carried out without inner conflict.
For example, while we are physically perfectly able to slap our
close friends and colleagues in the face, most of us rarely do
so. If we were to do so, most likely we would shock not only
others but ourselves, and the action would have immediate
repercussion for our self-concept. Thus even intense anger is
likely to be expressed in a form that is a compromise between
fantasy revenge and the self- concept. Voluntary control is
profoundly shaped by these considerations. It seems as if we are
always attempting to earn our own self-respect first of all.
Now we can reconsider Dennett's remark that consciousness
is the domain of access of the self. After all, by means of Options
Contexts, the dominant context hierarchy can indeed gain access
to all domains of consciousness: to the senses, to immediate
memory, to voluntary effectors, to imagination, and the like. The
conflictİfree sphere as an executive can presumably access any of
these domains without internal resistance.
9.32 Unresolved goal conflicts may persist outside of the
conflict-free sphere.
However, outside of the conflictİfree domain, competing goal
contexts may persist, conceivably for many years. This allows GW
theory to represent typical impulseİcontrol problems, where
people may successfully resist the temptation to express anger
for a period of years, and then, perhaps when the dominant goal
hierarchy becomes less dominant, the suppressed anger may emerge
overtly. The research evidence for this type of phenomenon may be
controversial, but naturalistic evidence seems persuasive. In any
case, the existence of such persistent competition is implied by
the fact that goal contexts can compete for access to
consciousness. From a theoretical point of view the possibility
of persistent unexpressed goals costs nothing in added
theoretical conceptions, and it may allow for future expansion of
the theory into an important domain of human motivation.
Multiple personality represents the most spectacular case of
such persistent conflict between different "selves," or in our
earlier terminology, between different context hierarchies.
Whenever one goal hierarchy dominates consciousness, it is
presumably able to access the senses, immediate memory, voluntary
musculature, as well as options contexts that allow one to access
shortİterm memory, to monitor and evaluate oneself, etc. This
model also allows us to interpret the evidence cited in Chapter 7
for conflictful states in which people disavow sentiments which
they demonstrably hold (7.xx). Disavowal is a voluntary action
based on a lack of voluntary access to the sources of information
which still continue to "prime" involuntary phenomena like slips
of the tongue or externally attributed projections. If this is a
reasonable analysis, then multiple personality syndrome is just a
more extreme case of rather normal conflict.
9.33 Explaining self-alien intrusions of thoughts and
images.
Chapter 7 describes the common phenomenon of selfİalien,
unwanted thoughts and images, as one typical psychopathological
event. Multiple personality patients often complain of such self™alien
intrusive thoughts, coming from a more dominant self. There
are nonİpathological examples of this as well, as in persistent
mental repetition of commercial jingles, which may continue even
against our best efforts to stop it. In Model 7 we can explain
such selfİalien intrusions of internal origin as due to
competition between different goal systems, including perhaps the
disavowed goals discussed above (9.32). Momentarily a competing
goal image may gain access to the global workspace, perhaps
especially if the normally dominant hierarchy is preoccupied with
other matters.
Since selfİalien intrusions are obviously not under
voluntary control, they presumably engage the same kinds of
automatic attention mechanisms that control any involuntary
intrusion (8.xx). As we noted in 8.xx, automatic attention can be
controlled by the goal hierarchy (a part of the self-system).
Thus the hypotheses about self developed in this chapter seem
consistent with the proposals about attention advanced in Chapter 8.
9.34 What happens if the highestİlevel context is disrupted?
‹j Early in this book (4.xx) we suggested that surprise may
disrupt the context hierarchy, and that the disruption will
propagate downward to other parts of the hierarchy that depend
upon the violated context. Now we can ask another question: If
severe surprise can break apart a context, leading to
re-organization under a higher-level context, what happens if the
highest-level context is disrupted? In general, it would seem to
be much more difficult to reintegrate the disrupted context
hierarchy, because there is no unviolated "umbrella" context
under which one can reintegrate action and experience. If one's
car breaks down, one can take the bus. But if one's fundamental
assumptions about life break down, all actions, ideas, and
experiences that depend upon these assumptions must be
reintegrated under a new set of topİlevel goals. Thus
reintegration would take more time, if the violation occurs at a
deeper level. This is indeed a model for severe stress and
decompensation, for "breakdowns" in the personality.
İİİİİİİİİİİİİİİİİİİİİİİİİİİİİİ
Insert Figure 9.34 about here.
İİİİİİİİİİİİİİİİİİİİİİİİİİİİİİ
9.4 Summary.
The review of self-alien syndromes lends support to our
decision to deal with self at all: it seems clear that
alterations in self create changes in conscious experience. That
includes the quality of experience, but even more clearly, it is
the îcontinuityï of autobiographical (self-attributed) experience
and memory that is affected by the self system. In multiple
personality, in fugue, amnesia accompanied by depersonalization,
and multiple personality, there is often a gap in
autobiographical memory corresponding to the period in which the
recalling self was eclipsed. For this reason we can think of self
as a contextual organization that supports memory retrieval
(among many other functions). By contrast, the îselfİconceptï,
defined as the beliefs people hold about themselves, seems less
fundamental. Many people constantly monitor their actions and
experiences by comparison with a set of belief and values that
show how things should be; this selfİconcept is evidently a kind
of supervisory system operating within the larger selfİsystem.
The selfİconcept seems to guide îconscious selfİmonitoringï, which
in turn shapes the selfİconcept. If we feel proud or ashamed of
some achievement, the selfİconcept would be adjusted
appropriately.
If self is a context, then there is a fundamental difference
between self and the objects of consciousness. William James'
"me" is the self as viewed from an outside perspective; but the
"I" is presumably necessary for the "me" to be experienced at
all. The contextualized aspects of the self are then only to be
known through violation, change, and conflict. Of course, our‹j ‹
normal experience smoothly moves from contextualized aspects of
self, to objectified aspects, just as T.S. Eliot's Unidentified
Guest, in the epigraph, moved smoothly from a contextualized self
to a jarring encounter with the bottom of the staircase, which,
just for a moment, revealed him to be "an object at the mercy of
a malevolent staircase."
FootnotesÉ
1. This chapter has been influenced by ideas proposed by Michael
A. Wapner, David Spiegel, Werner Erhard, and David Galin,
who are hereby gratefully acknowledged.