To confuse imagery of the self, in the everyday sense of this expression, with the self itself is the hallmark of narcissistic disturbance (Klein 1975; Kohut 1985). The narcissist thinks of herself as an object, as viewed from an external perspective, as the protagonist of a story as read by someone else, or by oneself in the act of viewing and judging oneself as if from an external perspective (Miller, 1981). Narcissists think of themselves in third person; the events of the narcissist’s life are narrated to herself as if read from the pages of the biography of some famous person from the past, who has already been judged important enough to write books and stories about (Horney 1937, 1950).

This is not to deny that the experience of a "true self" (I shall argue that there is one) also crucially involves imagery. But this will be "imagery of the self" in a different sense. The terms "imagery"and "self," as well as the word "of" linking them, have all remained notoriously ambiguous throughout the histories of philosophy and psychology. "Imagery" is often thought of as limited to visual imagery, or imagery in other sensory modalities such as audition. But there is also sensorimotor and proprioceptive imagery. One can imagine the way a leg cramp would feel without actually having one; or one can imagine oneself performing a bodily action such as grabbing one’s left foot, without actually doing so: these are proprioceptive and sensorimotor images, or what Jeannerod, Newton, and others have called "action images." I shall argue that action images are crucial to understanding what the true self is.

Consider also the ambiguity of the term "self." Of the numerous senses that have been given to this term, it is convenient to divide them into two broad types: there is the sense of a "self-for-others," as an object of experience, and there is the sense of a "self as it exists in itself" -- as a subjective entity or process, which for various reasons may be transparent to itself to greater or lesser degrees. It was Sartre (1943/1966) who first popularized the view that the second type of self, the "self in itself," does not exist, and that there is only a narrative self, a constructed set of images that we tell ourselves are of ourselves, in which the protagonist is seen as an object, just as someone might see a person other than oneself. Hence Sartre’s (1943/1966) notorious claim that Peter knows just as well as Paul what Paul is feeling. I have rejected this argument in many places (for example, Ellis 1986, 1995, 1996, 1997, 2000), and shall continue to hold here that there is a difference between the subjective self and the constructed self. For consciousness of the subjective self, it is action imagery and imagery of others that is central, not pictures of the self-for-others as in the attempt to view images of ourselves as we might appear from an external perspective.

To offer a narrative account of myself is to present a series of objectified images that in principle are intelligible as well to others as to myself. Such a narrative thus captures primarily the self-for-others, and presents the temptation to construe myself as equivalent with this objectified series of images. As far as a subjective self is concerned, the narrative can capture it only to the extent that it reveals different subjective responses to the same objective events that others might also have undergone. Thus the events in the narrative do not touch on the nature of the subjective self per se, and less still do objective images of the self that is undergoing or creating the events.

In addition to these ambiguities in the terms "imagery" and "self," there is an equally important ambiguity of the word "of" when it connects them. There can be images "of" the self, as in looking at pictures of our bodies or stories about ourselves as seen from an objective or external point of view. These images of the self are images of the narrative or constructed self. But in another sense of "of," we could say that even these images of the narrative self exist in the consciousness "of" the true self. They are images "of" the self in a different sense – they are images had by the subjective self in the attempt to experience itself. If the images are images of the self-for-others -- the self as viewed from an external viewpoint, as an object -- then what is viewed is merely the narrative self. But if the images are action images had by the subjective self in the context of action, and at the same time this action imagery helps to capture the true self in its intended actions, then the sensory images involved are not pictures of the self-for-others, but rather pictures of the objects in relation to which one acts, would hope to act, fear to act, etc., along with sensorimotor and proprioceptive action imagery as one evolves real or imaginary action schemas from the valuational emotions that motivate the action. As these action images become very complex and are related to complex environmental affordances, the question of course becomes increasingly pressing to what degree even these action images can diverge from an accurate reference to an authentic sense of the self.

While "healthy narcissism," including a narrative self, is a vital part of the self’s functioning, too exaggerated a preoccupation with the narrative self, to the point where the narrative self is equated with the subjective self, or treated as more important than the subjective self, is a hallmark of narcissistic disturbance, and an overemphasis on it in philosophy and literature is an expression of a "narcissistic culture." This is not to say that writers and thinkers should not preoccupy themselves with such things; intellectuals must explore what is important in their culture – its dysfunctions as well as its healthy aspects. Part of the writer’s task is to immerse herself in the aberrations of her culture so as to explore them intimately and hopefully to work through and move beyond them. Cultural aberrations can be important existential barometers or guideposts, in the same sense in which the jazz pianist Ojeda Penn once remarked that it is good to make a mistake, since from mistakes we learn, and that it is therefore best, when making mistakes, to make big ones. To explore a big mistake in a constructive way can direct us to the most interesting and exciting issues that are playing themselves out in a life or a culture.

But to argue in this way also requires showing that there is something in our actual experience that the label "subjective self" points to. This in turn will bring us to a discussion of the difference between a narcissistically disturbed value system and a value system characteristic of healthy narcissism. For the healthy narcissist, intrinsic value is invested in objects external to the self – love objects, social or political objectives, and abstract ideals, all of which are ways in which the subjective self intensifies the experience of the value of being per se. In love, for example, we intensely experience the intrinsic value of being as instantiated in the other. The corresponding imagery here is imagery of the other as we engage her. The imagery of the "self" that does this engaging is not primarily objective, but consists of imagery of results of the self’s actions in relation to others as guided by these non-narcissistic values.

For the narcissistically disturbed person, by contrast, only the self and its experience can be intrinsically valuable; the experience of others, political actions, and work toward abstract ideals have only instrumental value insofar as they serve to promote images of oneself as valuable. In this case, moreover, even the subjective self is relegated to a mere instrumental value, since the main purpose is now to view oneself as valuable, through one’s own eyes and the eyes of others, as an accomplished, amusing, or in some way other-pleasing persona. The inspiration of the intense experience of intrinsic value in the other is then blocked by this confusion between intrinsic and instrumental value. The narcissist is caught up in the habit of measuring the value of himself and of others instrumentally -- as depicted in objectified imagery -- and therefore misses the intrinsic value which can be seen only through empathy with the other’s subjective being. Rather than imagery of sensorimotor action schemas as they would relate the self’s actions to the other, the imagery is caught up with the way things look, objectively.

Let me say a little about how I think we got so far enmeshed in this predicament, which is by no means unfamiliar to people of all times and places, but seems especially problematic for contemporary urban-industrial cultures. Then I shall spell out more adequately why I believe there is indeed a true subjective self underneath the facades we have constructed.

To a great extent, modernity’s pronounced tendency to confuse the subjective with the objectified (or narcissistic) self can be traced to a more treacherous time in academia than our own -- the early 1600s, when Bruno had just been burned at the stake for his scientific heresies, Galileo had been forced to recant under threat of execution, and Descartes made his resulting deal with the Devil; for we all know that in the 1600s the Pope was the Devil, and it was with the Pope that Descartes made his deal. The deal was this: I’ll keep my hands off everything that pertains to the metaphysical realm -- God, the soul, etc. -- if you’ll let me pursue my scientific and mathematical studies unencumbered by religious or philosophical dogmas, and if you will guarantee not to burn me at the stake. In return for that guarantee, I’ll ensure that none of my findings about the physical realm will be in conflict with anything you believe about the metaphysical realm. I’ll keep a strict separation between the two. The soul (the subjective) will never encroach on the turf of the body (the objective), nor vice versa. More than vestiges of the same agreement survive today, between medical doctors and ministers. Doctors take care of the physical body, and allow other professionals to take care of the soul, provided, that is, that the other professionals adhere to their tacit agreement to stay off the turf of the doctors.

This rigid turf separation between the subjective and the objective famously led to the problem of the homunculus, which was also similar to the problem encountered by Medieval "causal theories of perception." Perception was understood as a causal chain leading from the perceived physical object, through the eye, to various brain circuits, finally resulting in the creation of a "phantasm" in the brain. Descartes rigidified this mechanistic type of explanation of information processing by including the brain, but not consciousness, into the realm of the physical (Sellars 1978). The phantasm is merely a picture of the object. But of course a picture per se is not conscious; if someone were to implant your favorite painting in your head, what is there about its being located in your head that would make the painting have more consciousness than it would have if it were, say, hanging in your living room? So a "homunculus" had to be posited, a little person inside the head who looks at the little picture inside the head, and the same question arose as to how the picture of the picture, inside the head of the little person inside the head, became a perceptual consciousness of the picture, resulting in a still smaller homunculus inside the head of the original homunculus, leading to an infinite regress.

Since that time, neuroscience has made great strides. In the Medieval brain, perceptual data were processed in the front of the brain, projecting the phantasm so that the homunculus in the back of the brain could be conscious of it. Contemporary neuroscientists are well aware, of course, that the situation is just the reverse: The perceptual data are processed in the back of the brain, in the occipital lobe, and projected onto a phantasm in the parietal lobe, resulting in a conscious registering of it by the homunculus, which we now know resides in the prefrontal lobe -- not, as the unsophisticated Medievals thought, in the back of the brain, and not, as Descartes thought, in the pineal gland.

Only those thoroughly steeped in objectivism, or in Husserl’s (1931) "natural attitude," will fail to get the joke of the previous paragraph. Obviously, regardless of whether the homunculus resides in the front of the brain or the back, the problem remains that empirical ways of knowing alone are inadequate to explain how a physical picture of the object, even if embedded in the brain, becomes a consciousness of the object. If only the physical can be seen through third person knowing, then no matter how efficiently we explain the physical, what Chalmers (1995) calls the "hard problem of consciousness" still has not been addressed: What explains why some physical mechanisms are accompanied by consciousness, whereas others are not? No set of empirical observations can answer this question, because empirical observations can explain only the physical mechanisms themselves, not why they are conscious, let alone help us understand what the conscious processes feel like and how to deal with them if we are the ones to feel them.

To fit consciousness into the picture, we must not only get beyond the empiricist criteria for meaning; we must also get beyond the contemporary version of the Medieval causal theory of the phantasm -- i.e., the stimulus-response paradigm. Rather than thinking of consciousness as the final effect in a causal chain leading from some physical object or collection of such, we must find another way to think of consciousness. If consciousness is not the final effect of a causal sequence, then another alternative would be to think of it as an active, self-organizing process. I.e., we can escape the problem of the homunculus only if we give up Descartes’ fundamental notion that nothing in nature does anything unless pushed or pulled by some other physical mechanism. Suppose consciousness is prior to rather than subsequent to the impressions of our senses, and suppose sensations must be created by conscious or preconscious processes rather than the other way around. Then we would not need the homunculus to see the phantasm. The role of the homunculus is taken care of by the fact that a self-organizing system with motivational purposes creates consciousness through its active anticipation of input rather than its passive response to it. This is not to assert, of course, that self-organization is a sufficient condition for the existence of consciousness; but it is a necessary condition. On this viewpoint, consciousness could be a physical activity of the organism, but not one whose understanding can be exhausted by explanations of the causal mechanisms that serve as its substrates, since a self-organizing biological system is capable of seeking out and appropriating the needed substrates to keep its definitive pattern going, within the limits of its finitude.

Why does this fit consciousness into the picture better than a mechanical, inorganic model? Because self-organization allows a purpose-directed element into the picture, without contradicting normal physical and chemical principles; and this in turn allows the organism’s activities to be directed toward purposes that are emotionally important for the organism’s total functioning. This introduction of the affective, emotional and motivational dimension into all conscious functions (by learning how emotion directs the selective and attentive functioning of the perceptual and information processing activities -- not the other way around) would allow us to distinguish between conscious and nonconscious information processing (Ellis and Newton 1998a). The organism must motivatedly "look for" X, or something related to X, if it is to consciously see X (as opposed to nonconsciously registering the information), and this "looking for" in itself can generate conscious imagery with or without any perceptual input (Damasio 1994; Posner and Rothbart 1992). Thus all consciousness requires a motivational and affective activity of the organism (Bernstein 2000; Ellis and Newton 2000). This phenomenological observation correlates with the empirical fact that, without emotional brain areas such as amygdala and anterior cingulate directing attention and corticothalamic loops, no amount of afferent brain activity can result in consciousness (Aurell 1984; Mack and Rock 1998; see also Ellis and Newton 1998b, Ellis 2001).

If consciousness requires motivation, which is an activity of the whole organism and must be actively executed in order to be felt, then introspective or phenomenological information would again have a role in psychology, integrated rather than competing with physical/empirical information. When we direct attention to our own experience, we have access to an aspect of it that no empirical observer can have, because we view it from a different perspective -- that of the one who is executing the self-organizing motivational processes required for any subjectivity (Newton 1996). For example, if I initially think that I am angry with my son for not taking out the garbage, I can refer directly to my bodily felt sense of the situation and realize that my son’s behavior is only a trigger for feelings about other things; I then can imagine how my total organism would feel different if those issues were resolved, and thus determine whether those issues could be interpreted in a way more resembling what my feeling is really about (Gendlin 1962/1997, 1992). To an empirical observer, this holistic perspective on the organism’s active self-organizing emotional directedness is not available; thus the empiricist’s tendency is to see the physical event, my son’s behavior, as the cause and the intentional object of my emotional "response" (as in LeDoux 1996). Whatever triggers the emotion is then confused with its object. The emotion is thus seen as a passive response to chemical events in my nervous system rather than as an active, self-organizing system that merely appropriates the son’s behavior and the available physical substrates in the causal mechanisms of the nervous system as an opportunity to actively explicate its own ongoing motivational directions (as in Ellis 1986, 1995, 1996).

As played out in terms of narrative imagery, the objective events of my life are to a great extent replaceable; the constant for each person is the way he or she organizes and directs attention and the self-organizational actions that make experience possible. When the ultra-liberal young racial de-segregationist George Wallace lost his first political election in Alabama, he determined that his racial liberalism would never again cost him an election. He then devoted the same fiery ambition to winning with segregationist rhetoric that he had previously devoted to de-segregation. We see his personality not in the specific events of his life, which were easily reversed still again years later when segregation fell out of fashion, but in the ways in which he chose opportunities to express the peculiar pattern of his being in terms of the available action affordances.

If Harriet Taylor had never met John Stuart Mill, would she have still been more or less the same person? The specific experiences of her life did not define her as who she was, since others could have had the same experiences. (After all, many other people did in fact meet John Stuart Mill.) There seems to be a sense in which the style in which the person organizes and patterns the flow of experiences is more important for defining the self than the what-content of those specific experiences. The next section will attempt to develop this idea, and to see how it can ground a concept of a subjective self that is not experienced in the mode of objectified imagery.

The subjective self arises essentially from the self-organizing system’s capacity for action as opposed to mere reaction. It anticipates emotionally significant experiences by seeking out and appropriating opportunities to express schematically patterned action potentialities. A narrative concept of self will miss its mark, erring on the side of narcissistic masks, if it seems to equate the self with the narrative, and does not capture the fact that if the same objective events were to occur in the lives of different individuals, there would be extremely different subjective responses to them and different uses of them for purposes of subjectively-motivated action. When Chopin was offered a series of concerts with the large audiences on which Liszt capitalized so successfully, Chopin sometimes took advantage of the opportunity, but he did so only reluctantly, as a sacrifice that must be made to secure the privilege of continuing to compose music and get it published. Objectively, his concert demeanor may not have appeared so different from Liszt’s. But subjective differences can express themselves across apparent objective similarities, and it is these differences more than the objective similarities -- more than the narcissistic images -- that really identify the self as a subjective unity distinguished from others.

There is a sense in which the self is different from each of its momentary conscious states in that the self is what gives unity and direction to the overall stream of states. This is why we recognize the difference between different individual personalities: The way we identify a person’s uniqueness has nothing to do with the fact that the person happens to experience this or that conscious state; most people either do experience or could experience most states, so experiencing these or those particular states is not what distinguishes their personalities from each other. Had I not experienced the pain of hitting my foot on a rock this morning, I would very likely still be the same person.

Correlatively, there is a recognizable style with which the person chooses at each moment in the stream of consciousness what to direct attention toward at the next moment. The overall pattern of these attention-directions is what we recognize as the personality. This is also what the "self" is in the sense of "transcendental ego" that Husserl (1931) developed and whose reality Sartre (1937/1969) denied -- something that is not confined to one particular conscious state, but does have a subjective status, and at the same time serves to unify all the other states into a coherent pattern.

What could serve both these purposes at once -- on the one hand, unifying all the subjective states in the stream, while on the other hand having a subjective character in its own right? A likely candidate would seem to be the person’s ongoing emotional life. An emotional motivation can unify all the particular states, by motivating the direction of attention at each moment, toward homeostasis when that is what is needed, and toward greater complexity when that is what is needed. And at the same time, emotional motivations are subjectively experienceable, so that, contra Sartre, this emotional motivation can both serve the purpose of unifying all the conscious states in the stream and at the same time be a conscious state (Ellis 1986, 1996).

The motivation referred to here can be thematized in terms of the organism’s motivation to maintain and enhance its pattern of self-organization. A self-organizing system is one that actively appropriates and replaces the substratum components needed to keep the pattern going, rather than being the passive causal outcome of the interactions of the components. Of course, this is just what living organisms do (Monod 1971; Kauffman 1993). The ontological status of the personality, conceived of in this sense, seems to be a higher-order pattern composed of the particular conscious states in the stream.

Given even just one state of consciousness, we can already discern a purpose-directed and aim-oriented dimension of the state, and this dimension could be called a person or self (Newton 1996; Wider 1997). Evidence that motivational and affective processes are necessary for conscious cognitive processes (as opposed to nonconscious information processing) has been mounting in recent years. For example, it is well known that the afferent brain areas (those which receive stimulation from external sources) -- such as the "sensory area" of the occipital lobe and the V4 visual area -- can be completely activated, in just the way they are in a perception, but with no perceptual consciousness of the object occurring (Luria 1980; Posner and Rothbart 1992). Consciousness occurs only when the efferent system, beginning with the emotional midbrain/limbic area, prompts the frontal lobe to start looking for environmental items which might be of interest to the organism, in light of its emotional-motivational purposes, and activates neurotransmitters to catalyze the corticothalamic loops needed to enhance important signals and direct either voluntary or involuntary attention to them (Ellis and Newton 1998a, 1998b).

Thus we can conceive of the person as the self-organizing system whose motivational tendencies coordinate the pattern and direction of cognitive contents by motivating, at each moment, what is important for us to direct our attention toward. Of course, this definition would extend "personhood" far beyond human beings; we could speak of cats, dogs, or even frogs as "persons" or "selves" in this sense. (B.P. Bowne even goes so far as to define personhood in terms of value-directedness, and then conclude that all of nature is personal.) What makes something a self, in this sense, rather than just an inanimate mechanical system, is that it is purpose directed, that it acts on its environment rather than just reacting, and organizes itself in order to maintain its existence rather than merely having its organization be the result of external mechanical processes. Such a pattern actively seeks out, appropriates, and replaces the material substrata needed to maintain the overall pattern into the future.

We know that Chopin is a completely different personality from Liszt because, even though they may both have used most of the same chord progressions at various times, Chopin would not have used chord progressions X, Y and Z in the same pattern and in the same contexts as Liszt would. We distinguish individual personalities by seeing, not what specific states of consciousness they have, but by seeing the pattern formed by those states.

Emotion and motivation are among the essential determinants of such a pattern. Emotion determines the direction in which we will move from one state of consciousness to the next, and therefore indirectly determines the content of each state. If the self is the pattern, then the self can be motivated to make an interesting pattern, and is not imprisoned within one particular state’s reductively driven desires. The pattern can organize its components in a coherent direction that has meaning apart from the smaller purposes of each component; this requires both homeostatic motives and "homeo-ecstatic" motives -- i.e., those whose aim is contrary to homeostasis, that strive to move away from entropy or energy-reduction of the living system and toward a higher energy level, Otto Rank’s (1929) "life force" that works in opposition to drive-reductive tendencies (the latter being merely a matter of electrostatic neutralization in the nervous system). From this standpoint, a narrative can be an expression of the self in the sense that the self is the author of the narrative, within certain plot constraints dictated by available environmental affordances.

In any given perception, there is a Gibsonian "affordance" attached to understanding the meaning of what is perceived. I identify the carpet I see by imagining what it would be like to crawl on it, run my hand over it, etc., although these imaginations are usually kept at a preconscious, habituated level. (Newton, 1996, explains this point by saying that, earlier in infancy, the infant had to sense in some way what those affordances would be like in order to appreciate the identity of the carpet, and then the affordances became habituated and sedimented.) For example, in visual perception, a landscape presents a Gibsonian affordance of possible patterns of eye movements that could be easily executed in relation to that scene (Pribram 1993), and the eyes are continually engaged in active rather than passive movements and oscillations. As every artist knows, some scenes are "easier to look at" than others, and patterns on a canvas can conjure moods by facilitating and interfering with patterns of eye movement in the viewer.

Newton (1996) supports this viewpoint by pointing out that the "understanding" that distinguishes humans from computers is based on the human’s ability to plan actions. Here again, humans act, whereas merely mechanical systems only react. To understand an action means to be able to form an imagistic/proprioceptive bodily sense of what it would be like to execute the action. To understand an object is to form a similar kind of sense of what it would be like to execute an action in relation to that object. Although we may believe that red is pasted to the surfaces of objects, in reality the red that we see arises only through our interactions with the object; as Merleau-Ponty (1941, 1942) emphasizes, if we (or our perceptual systems) did not act, we would not see.

We can thus explicate this "transcendental-subjective" concept of the self-as-author by answering this question: When I reflect on my "self," what am I reflecting on? The answer would be that I am reflecting on my embodied sense of what the affordances would feel like. I.e., when I "inhabit" my left hand as it is feeling what the right hand as an object feels like, my "being" or "self" is "in" the left hand rather than the right; then I can shift to inhabiting the right hand, and sense what the left as an object feels like (Merleau-Ponty 1948/1968). So the difference between reflective and nonreflective experience can be understood in the following way: In nonreflective experience, we focus attention on the object of the experience rather than on the subjective affordances that make the experience possible; whereas reflective experience focuses attention on the bodily sensing of the affordances themselves. When I touch a tree trunk and feel the texture of the bark, I am focusing attention on the bark. When I allow my attention to inhabit the hand itself, feeling the bodily sensing of the texture, I am inhabiting my "self," and thus reflecting on the self, the subjective side of experiencing. In phenomenological terms, we could say that to reflect on the self is to inhabit the noetic rather than the noematic side of any given experience (Husserl 1931). This concept has the advantage of making the self into an embodied being rather than an abstraction; the self is already an aspect of the body at any given moment, and can be reflected upon as such.

The concept of the embodied self suggests that, when we reflect on consciousness, what we are doing at the concrete level is to pay attention to our own intentional action in relation to the object of consciousness. Since understanding an object always involves an implicit imagining of how it would be possible to act in relation to the object (the object’s "affordances" for us), we can reflect on the body’s orienting itself to the object in this way. On the other hand, when we have "unreflective" consciousness of an object, what we are doing is to intentionally act in relation to the object in this same sense, but in such a way that we do not pay attention to the acting that our bodies are doing. When we speak of the self as the embodied self, we mean that our self can inhabit any part of the body that we pay attention to. Whenever this embodied self is in a certain place in our bodies, that means that we are paying attention to how we act (as opposed to reacting) in that place.

But to be an actor rather than a mere reactor means to be a system which readjusts its own parts in order to maintain and enhance the continuity of the functioning of the whole, i.e., to be a higher order pattern. Parts of systems can re-act, but if the system as a whole is a self-maintaining Gestalt (what Merleau-Ponty, 1942, calls a "psychophysical form"), then it can "act" rather than just react. Of course, to "act" just means to behave in a way that is determined by the tendency of the whole to readjust its parts rather than to be pushed in partes extra parts fashion. Thus consider even a fairly complex mechanical system, i.e., one in which everything that happens to any given part can be described exhaustively in terms of some specific other part’s effect on it, without reference to a tendency of the whole to maintain its overall pattern: Such a mechanical system still cannot "act." The tendency toward self-organization is the form of inertia that counteracts the inertia that describes objects’ tendency to seek a lower energy level or conserve energy; we might call it an inertia of "action" -- a tendency for patterns to maintain themselves by appropriating and replacing the needed substratum elements to facilitate the continuation of the pattern.

To reflect on the embodied self, then, is to reflect on this acting on the part of the body (as opposed, again, to merely reacting), to focus on oneself as acting in relation to the object of experience rather than focusing on the object itself, even though it becomes an object for us only through our acting on it, whether we focus on this acting or not. We constitute the appearance of the object by acting in relation to it; but only when we "inhabit" the acting of our bodies do we "reflect" on our embodied self.

Yet the ability to act is itself a higher order process, an organization in terms of a complex, self-regulating system. Thus the self can be both an aspect of the body -- an embodied self -- and a transcendental self, i.e., a higher-order pattern than most specific states of consciousness are, which serves to organize and direct the specific states, and which could still have the same identity even without any particular one of those specific states. Moreover, the specific state could not exist except within the context of a higher-order pattern motivated toward maintaining and enhancing the pattern, regardless of whether certain momentary states have to be included or excluded in order to achieve that.

There are a number of different senses in which the term "self" can refer to a real phenomenon, and the narcissistic self is one of these senses. The narcissistic self is the self as seen by others and evaluated in terms of its instrumental values such as admirability, pleasingness and accomplishment. Our concepts of our own narcissistic selves tend to center around objectified imagery and narratives. But if we attempt to construct a real self based on such imagery, we fall into the trap of narcissistic disturbance. As we have seen, the real self must also include a subjective self, whose imagery centers not around objectified sensory imagery, but around complex sensorimotor and emotional action schemas as related to the action affordances of the world, and as we act in relation to others whose intrinsic value we can symbolize and experience in terms of the actions we can take or even imagine in relation to them. If there is a streetcar to be caught, it is because the subjective self is a self-organizing system that really does want to catch it.