The Diachronic Self, chapter 4: Technological Transformation and Assimilation

Chapter 4

TECHNOLOGICAL TRANSFORMATION AND ASSIMILATION

In the previous chapter I examined the various kinds of psychological attributes that constitute our identity and their relative importance. I argued that, even after allowing for the differing contributions to connectedness of the various attributes, we can rationally be concerned for our future phases more than proportionally to the degree of connectedness. In this chapter I will focus on various changes in or additions to the self, especially as effected through our bodies. Although normative issues will arise, most of the discussion will deal with the metaphysics of identity. In the first of three sections I will distinguish augmentative from deteriorative transformation. Then in Part II, in trying to develop principles to help decide when a new ability or a physical change becomes part of us, I will develop an account of functional integration. After considering how we assimilate changes in ourselves I will distinguish enhancement from supplementation—a distinction that can help clarify normative issues. Part III investigates whether a psychological reductionist should in any way grant intrinsic significance to a person’s body.

I. AUGMENTATIVE AND DETERIORATIVE Transformation

Augmentative vs. Deteriorative Transformation

A rough distinction can be made between deteriorative changes or transformations in a person’s characteristics, and what I will call augmentative (or developmental) changes. Deteriorative changes weaken or destroy some personal attribute without any compensating addition. Deteriorative changes include losing my ability to do arithmetic in my head; a weakening of visual discrimination; the fading away of a disposition; the loss of a memory, and so on. Augmentative changes preserve at least some of an existing attribute but add to it or alter it. Examples include having your eyes and optical center replaced with a synthetic optical system capable of seeing in a broader spectrum; the addition of foreign language skills to preexisting native language ability; and the acquisition of a new desire. I said the distinction was a rough one because there will be cases which could be classified either way; the two classes are not sharply disjoint. Consider the case above where my visual system is replaced with a synthetic optical system. If my new system maintains all the capabilities of the original while also allowing me to see into the infrared, ultraviolet, and microscopically, then this is a pure case of augmentative change. However, the synthetic system might give me these new abilities while, let us say, delivering inferior night vision or weaker color differentiation. In this latter case the change may be predominantly augmentative but also partly deteriorative. We could alter the example to change the mix along a spectrum from entirely augmentative to entirely deteriorative. More psychologically oriented examples of augmentative changes that involve some loss include a disposition for generosity becoming more selective in its objects; a system of beliefs becoming modified to more accurately reflect the truth.

            Augmentative changes are likely to result in less reduction in connectedness than are deteriorative changes in any particular aspect of self. This is because they keep some or all of the old characteristic while adding to it or altering it, whereas deteriorative changes simply take away an attribute or replace it with something unrelated. It is easy to slide from metaphysical to normative thinking in personal identity discussions, so here I will make explicit what I mean: We will generally prefer augmentative to deteriorative changes. Most of us in most circumstances would rather trade an existing ability for an altered one of greater power or range, or trade a belief-system for an altered one of higher accuracy, than have an ability or belief-system destroyed. This will be important when considering the desirability of changes in ourselves, but the question of desirability should be kept distinct from that of the actual change in connectedness. The change in connectedness between person-stages is an objective matter. The fact that I find a particular augmentative change more desirable than a particular deteriorative change in itself has nothing to do with the extent of reduction in connectedness. We should not weight the deteriorative change more heavily simply because we find it less desirable. If I lose a small amount of visual ability, the change in connectedness can be smaller than if my new visual system has powerful but only slightly overlapping abilities compared with my old. I may prefer the augmentative change even though the loss of connectedness is greater than would be the case with the deteriorative change.

            The same principle applies to changes in a person as a whole, as well as to individual characteristics: In one possible situation, I undergo a series of physical, cognitive, and emotional changes resulting in a person-stage stronger, smarter, and more emotionally well-tuned than my current stage. In a second possible situation, I suffer a gradual physical, cognitive, and emotional decline, becoming weaker, duller, and emotionally less integrated than my current stage. The degree of connectedness between current and future person-stages in the two possible situations depends only on the extent of the change; the degree of connectedness is unaffected by the direction of change. If the extent of the developmental (augmentative) changes are similar to the extent of the deteriorative changes, then the change in connectedness will be similar. In practice, developmental changes involve smaller changes in connectedness since they add to and partially modify existing characteristics. Deteriorative changes, by destroying or impairing existing characteristics, tend to involve larger changes in connectedness. These practical differences, though, have nothing intrinsically to do with the direction of change; they result from the objective nature of the changes.

            We might alter the two possible situations so that the augmentation and deterioration happen abruptly and discontinuously rather than gradually. In the case of an abrupt deterioration (perhaps resulting from a major accident or massive stroke, for instance) the later person-stage (if he were still a person) might not be a stage of the same person as my current stage. The change may have been so extensive and discontinuous that I cease to exist, being replaced by a different person (or by no person, just a living human body). The parallel between deteriorative and augmentative change is preserved even under these conditions. It can equally be true that, if I improve drastically and discontinuously rather than gradually, my current stage and that future stage will not be stages of the same person.

            Augmentative and deteriorative changes, then, should be treated the same when estimating the degree of reduction in connectedness. Despite the absence of any metaphysical priority of one type of change over another, there can be an axiological or normative difference between them. At least given the typical values people hold,[1] deteriorative changes we will see as undesirable, developmental changes as desirable. In deteriorative change we lose characteristics we value. In developmental change we willingly exchange old characteristics for new or partly new characteristics we value more highly. This is one way, as I argued in the previous chapter, that we can rationally have self-interested concern for future stages more than proportionally to the degree of connectedness between current and future stages. Most of would rather survive—maintain psychological continuity—while gradually losing some connectedness through developmental change than stay just as we are now. The question arises: Would we, or should we, be willing to change so drastically and abruptly that we have effectively been replaced by a different person, albeit one who we believe is superior to us according to our current values? I now turn to examine Raymond Martin’s argument that we would make such a choice if it were possible. Martin’s view, if correct, strongly undermines the view that the extent of our (rational) concern for our future stages depends on the degree of connectedness, at least in a range of possible situations.

Raymond Martin on Tranformation and  Replacement

This sub-section deals primarily with a normative issue. The discussion will, however, lead me to clarify some terms involved in the metaphysical issues of the next sub-section, terms such as “transformation”, and “replacement.” The critical discussion here of a paper by Raymond Martin[2] follows naturally the previous sub-section and helps prepare the way for the subsequent part. Martin’s view is especially relevant since it sounds similar to the Transformationist view I defined in the previous chapter. I will compare and contrast the two views after clarifying what Martin is proposing.

            Martin presents two examples intended to show that identity is less important than generally supposed. The first example involves a person fissioning into two, in a manner familiar from Parfit. Some will respond to this kind of example by being unimpressed. This may be on the basis that all that fission examples demonstrate is that we do not care very much about the technical question of the transitivity of identity. In other words, we do not lose our identity in fission examples except in a narrow technical sense. Others, such as Lewis, argue that fission examples fail to show that identity is not what matters. Lewis holds that there are two persons sharing one person-stage prior to fission; both of these persons retain their identity through the fission. Since the fission example may be unpersuasive in showing that identity is not as important as typically thought, Martin gives a second example. This example is intended to demonstrate that not only is it not identity that matters most to us in survival, neither is it continuity (with or without branching).

            If this second example is persuasive, it “will show that there are situations in which many of us would prefer to give up our identities and transform into the persons we most want to be rather than to retain our identities and fail to make such a transformation.” It is neither our identity nor continuity that matter most to us, Martin suggests. We value becoming who we most want to be more than either of these. His example is as follows:

            We imagine that it is possible for a person to undergo a painless, safe, and inexpensive operation in which we exchange some physical or psychological trait for a better replacement. “So, for instance, you could, through a single, almost instantaneous operative procedure, one say, that simply used sound waves and involved no cutting, become physically better—stronger, more flexible, more beautiful, and so on—[and] psychologically better—more patient, more generous, more intelligent, and so on.” [295] The only cost to the procedure is that you lose some memory of your life up to that point. “Because you have only one chance at the operation, and the alternative ways of changing yourself dramatically for the better are so onerous and unreliable, there would be a tremendous incentive to change yourself drastically in all the ways you would like to change to become the person you most want to be. However, the greater the changes, the greater the tax on your personal memory. You could change radically and become the person you most want to be (assuming you are not already close to that sort of person), but only by ceasing to be either physically or psychologically closely continuous with your current self. On most theories of personal identity, perhaps on all psychological-continuity theories, this would mean you could change radically and become the person you most want to be only by ceasing to be the person you now are.”

            Martin believes that most people, if this procedure were available, would choose to do it. He concludes that the fact that we would choose changes that would cost us our identities shows that “becoming the persons we most want to be is what matters primarily in survival.” [296] “This conclusion reveals something fundamental and perhaps also startling about our most basic values. In simplest terms, it reveals that many of us crave to be fulfilled more than we crave to be—that, paradoxically, we would choose to cease to exist if by so choosing we could realize our deepest selfish desires.”

            In order to address Martin’s view, and to prepare for my subsequent thoughts, I will pause to define some terms such as disruption, transformation, replacement, and becoming. For the present discussion, the most important contrast is that between changes to a person compatible with continuity, and changes in a person that effectively result in the loss of the original person with a new person appearing in their stead. I will use the term transformation in a way that seems to accord with Martin’s use (though he does not explicitly define it). Transformation is a spectrum of degrees of change from the slightest change in a person to total change in which not a single original attribute remains. Transformation typically implies some degree of change while leaving something of the original intact. If nothing at all remains of the original, or if the change is extremely small, we will not usually talk of transformation (because to say “x is transformed” implies tht x still exists”). Nevertheless, we can regard the cases where the term feels odd simply as extreme ends of the spectrum of transformation. We will feel differently about different points (or regions) along the transformation spectrum. Martin’s argument, and analysis of it, requires that we be able clearly to distinguish between cases of transformation in which we feel we would continue and cases of transformation where we feel that we would not continue.

I will refer to cases near the conservative end of the transformation spectrum as continuous transformations, or as sustaining transformations.

I have already defined deteriorative changes as those that “weaken or destroy some personal attribute without any compensating addition.”

I will refer to those cases where most or all of the original person’s characteristics have been destroyed, with new ones in their place, as discontinuous transformations. (Discontinuous transformations are therefore a subset of deteriorative changes; those involving a sudden, dramatic deterioration rather than a gradual deterioration.)

When the transformation is discontinuous I will say the person has been replaced.

When the transformation is continuous or sustaining I will say the original person-stage has become the later person-stage.

Replacement involves the disruption of most or all of a person’s central characteristics with new characteristics being put in their place. Replacement is a form of transformation that starts to appear about halfway along the spectrum. At earlier parts of the spectrum the person-stage is becoming another stage. The central opposition is between becoming and being replaced. We need not pretend there is any sharp line between the two. There will be many clear cases and there will be a fuzzy region where both descriptions can reasonably be applied. Parfit would place the cut off between becoming and replacement at a loss of 50% of connectedness over a day.[3]

            The question at issue in Martin’s paper can now be stated as: Would most people choose to be replaced if their replacement were the person they most want to be? The main conclusion at issue is Martin’s view that an affirmative answer shows that most important to us in survival is not identity but transforming into the person we most want to be. (This seems to be a peculiar way for Martin to state his conclusion, but he does use the phrase “important to us in survival.” His way of stating it apparently begs the question as to whether or not we survive the transformation. A better phrasing would be: “most important to us in our considerations about the future is not identity but transforming into the person we most want to be.”) I agree that most of us, or at least many of us, would choose to undergo the operation to transform into our ideal self. My view diverges from Martin’s in that I think he has described the choice in a misleading way. Also, the conclusion he draws, while accurate for some us, seems to be an overgeneralization.

            Martin presents the operation as a choice between either staying as you are or transforming into your ideal self all at once, where the latter choice involves giving up your identity. The operation involves such a massive change that psychological continuity and personal identity are breached. In fact, I will argue, undergoing this operation, in most cases, will not require us to relinquish our identity. This is because a major part of our identity is constituted by our values and we would not want to change our values. Our values form the core of our identity. This is not true for everyone, and in some cases Martin’s description will be appropriate: Some persons lack a strong core of values. These persons would give up their identity through transforming. Although it seems to me that most of us could transform into our ideal self without being replaced, the nature of the identity of persons is sufficiently vague that no strongly compelling demonstration can be made. It is at least arguable that even those of us with well-developed values would be giving up our identity in transformation. If so, this would show that it is not our identity that matters to us in survival, but that part of our identity constituted by our values. It is not true for all of us, as Martin claims, that transforming into who we most want to be is the most important thing in survival. Values differ: this will be true for some of us but not for others. Those who value self-transformation strongly can undergo more changes in other characteristics while maintaining identity.

            What is the basis for these claims? First, my claim that, for most of us, transforming into our ideal selves would not require us to relinquish identity. Martin attempts to influence our intuitions by the way he describes the operation. What you are asked to give up is your memories of your experiences. Abolishing these memories certainly would be a major loss, though the impact on our degree of connectedness would be small. In an earlier chapter, when examining the relative contribution to overall connectedness of particular types of personal characteristic, I argued that memory contributed much less to connectedness than other types of characteristic. (It only seems particularly significant if we count as memory other things like skills and abilities.) Undergoing the changes described by Martin would detract little from connectedness, since memory is only a small part of it. If the memory losses were all we had to give up, we could undergo transformation without coming near loss of identity and replacement by a new person. In addition to this, most of the other changes Martin describes involve additions to you or strengthening of existing characteristics. When you acquire an attribute there need be no loss of connectedness (unless it is incompatible with a pre-existing characteristic). We measure connectedness not by the number of the later stage’s characteristics shared by the earlier stage, but the converse. I may come through the operation stronger, healthier, more imaginative, with new abilities to play musical instruments, to comprehend abstractions previously too difficult for me, and with the addition of new desires and dispositions (compatible with the rest of my character). But these leave the characteristics of my earlier stage intact. (Gaining other psychological characteristics will mean giving up preexisting characteristics: gaining an attribute of tolerance or serenity will mean giving up anger.) Neither the loss of declarative memories nor the addition of new features removes a critical degree of connectedness. If these are what transforming into my ideal self entails, then I can become that ideal self rather than be replaced by him.

            This problem with the persuasiveness of Martin’s example can be remedied. We can imagine that to become your ideal self, you would have to cut away much more than memories. You would have to remove most of the desires, dispositions, abilities, and intentions that constituted you. This would have to mean that most of your existing characteristics are incompatible with your ideal self. Certainly this is possible. However, it seems like an uncommon situation. For the example to genuinely show the transformation to involve replacement, it would have to mean that most of our existing characteristics would not exist in the person we most want to be. This implies a high degree of self-rejection and dislike. For some of us the example then would involve replacement, but for most of us it would not.

            Another consideration corroborates the low likelihood that transforming into our ideal self would mean our replacement rather than our continuation: Our ideal selves will reflect our values, and our values form much of our connectedness. When I examined, in an earlier chapter, the relative contributions to overall connectedness of types of attribute, it turned out that values form a central and widely ramifying part of our identity. Values shape many aspects of us; they influence which (non-value) desires we act on or accept, affect which skills and abilities we acquire or use, largely determine which intentions we form, and they shape our long-term projects. If our values form the major part of our identity, then we would have to become, rather than be replaced by, our ideal self. This is because we would not choose to give up our values to transform into the person we most want to be. Our ideal self is one that accords with our ideals. It will include abilities and qualities that we do not yet have, but it cannot have values incompatible with our current values. If I now value my honesty and rationality, I will not conceive of my ideal self as a lying irrationalist. If I will not give up my values in transforming into my ideal, and my values constitute most of my identity, then I will not have to relinquish my identity to become my ideal. For something to be my value, rather than simply my desire, it must be integrated into a system of desires (as explained in an earlier chapter). Since it is the integrated person—not some errant transitory desire—who chooses the ideal self, the person’s conception of their ideal self will not involve destruction of existing values. Your “values” are not your values if you would do away with them.

            Even if we grant that values are the weightiest component of our identity, we might argue that all the other characteristics summed together outweigh the contribution of our values. If so, in transforming into our ideal we could retain our values but lose our identities. Though this could happen, and for some persons would happen, it will be quite uncommon. Values may not be able to outweigh all other types of characteristic added together in terms of contribution to connectedness, but they will contribute at least a large minority. You would only lose continuity in that case if your ideal involved the abolition of a large majority of your other characteristics. Values are more likely to be outweighed by other characteristics if the person has only weak values. An extreme case of this would be a schizophrenic. In individuals with poorly integrated desires, values exist weakly if at all. An individual with weak values but strong desires could choose an ideal self that involved a discontinuity.

            To sum up: It is unlikely that most of us would be replaced by, rather than become, our ideal self. Replacement seems more likely for certain persons such as schizophrenics and those with weakly-formed values. Though it seems unlikely to be a common result, I grant that it is arguable that even some normal persons, in order to transform into the person they most want to be, would have to give up so many characteristics other than their values that they would be replaced. Granting these possibilities, should we concur with Martin’s view that this shows that it is not our identity that matters most to us in survival (or in our thinking about our place in the future), and that what does matter most is transforming into our ideal self? Yes and no. Yes, we will have to agree that identity is not the most important thing. Some, and possibly many, of us would be willing to give up our identity to be replaced by an ideal self.

            However, it does not follow that what matters most to everyone is transforming into the person we most want to be. The example reveals that we care most about the continuation of our values—the core of our identity—rather than about our identity as a whole. This conclusion is interesting and might lead us to reevaluate some notions, such as the view that personal responsibility is tied to the persistence of identity. Perhaps responsibility (for past actions and commitments) instead is tied to the persistence of values even when identity is lost. However, considering such possible implications would take me far afield.

            When I claim that it does not follow that what matters most to everyone is transforming into the person we most want to be, I am not denying that this is the most important thing for some of us. Our values are the most important thing for each of us in our survival. Obviously Martin highly values self-improvement and self-transformation. So do I, and so do many self-reflective persons. But, in making a general claim that transforming into our ideal is the most important thing in survival, Martin overgeneralizes. This claim seems natural to Martin and may appeal to many of us, but not everyone places any substantial value on self-transformation. Such people may be self-satisfied, deluded about their own perfection, or simply minimally self-aware or imaginative. Others may feel dissatisfied with who they are but have little or no idea how they would improve if they could. What they care about is the persistence of their values, and these do not happen to include self-transformation. (Some will care about self-transformation but only when brought about by the self.) Those to whom self-transformation seems a natural value tend to reflect on themselves and have developed a relatively high degree of self-awareness. They will tend to be questioning, challenging, imaginative persons. (Or, to recognize negative motivations, some of those to whom self-transformation seems a natural value will be self-hating.)

            It is worth noting that the more strongly we value self-improvement and the more broadly that value ramifies through our behavior, the greater the transformation we can undergo before we feel that we will lose what matters in survival. We might be willing to give up not only memories, desires, and abilities in the process of transforming into our ideal but even some of our other values. When we value self-transformation, we are really holding a complex of values rather than one simple value. The drive to self-improvement (if positively rather than negatively motivated) involves optimism (you must believe that improvement is achievable), enjoyment of experimentation, appreciation of novelty, tolerance of uncertainty, a willingness to take responsibility for your destiny, enjoyment of challenge, and desires for autonomy and self-direction. Self-transformation will also link to a disposition to think critically and imaginatively.

            I will conclude this subsection by relating the present conclusions to the view I called Transformationism in the previous chapter. Transformationism consists of a normative claim to the effect that:

(2) Earlier stage A may reasonably care about later stage B more than proportionally to the degree of connectedness between them; i.e., continuity is significant, not just connectedness. This is because:

       (i) the person may value their life as a whole (or long stretches of their life).

       (ii) B may be closer to A’s conception of an ideal self.

       (iii) the person may hold self-transformation as a central goal.

The conclusion I have reached here, through consideration of Martin’s example, that what we value most in survival is the continuation of our values and, for some of us, our becoming the person we most want to be, adds further support to the previous chapter’s defense of (2)(ii) and (iii). Consideration of Martin’s example gives further reason to believe we can reasonably have concern for our future person-stages much more than proportionally to the degree of connectedness we have to them. By complementing the earlier chapter’s ideas, we can now see more clearly which parts of ourselves matter most in survival—which parts will most strongly support our future-concern.

II. INTEGRATION OF CHANGE

Assimilation

This section analyses the notion of assimilation in order to determine when physical additions to us become part of us. I begin by applying the idea of assimilation to modify Parfit’s criterion for continuity. I will develop an account of assimilation in terms of functional integration. It will turn out that functional integration seems to require something like exclusive, or at least interference-free, access to and interdependence with a part of the self. It does not require physical connection, conscious or direct control, nor sensory awareness of the part.

            Earlier in the chapter I expressed dissatisfaction with the criterion for continuity given by Parfit. According to this criterion, continuity persists so long as there are overlapping chains of strong connectedness, where “strong connectedness” is defined as persistence of at least 50% of the typical psychological connectedness over the course of one day. The 50% connectedness over the course of a day seems arbitrary. Why should we look at the degree of psychological connectedness over the course of a day? Why not 3 hours? Or 14 days? Or 7 months? Or 23 seconds? The 50% condition, though an obvious choice, also seems uncomfortably arbitrary. I will propose an alternative criterion for continuity, one suggested by the considerations involving assimilation which occupy this section.

            A person is a reasonably well-integrated system of beliefs, desires, values, abilities, and so on. Persons are dynamic entities, constantly changing in response to both external and internal pressures. There are limits, however, to the degree of change possible for a person to assimilate. Equivalently: there will be a maximum rate at which changes in the self can be assimilated. This will vary between individuals, with the extreme limits being set by the common genetic, neurological, and biological nature of humans. If a person goes through changes in excess of their ability to assimilate, they will disintegrate or fracture. (Disintegration may occur for other reasons, such as an extreme trauma, or neurochemical disorder.) Disintegration means that personality fragments or decomposes: the systematic interrelationships between elements of the self dissolve. This might occur because of (a) the loss of elements or aspects of a person, leaving gaps that interfere with their overall functioning. Fragmentation of this kind may be gradual, as in senility, or sudden, as in the case of a head injury or neurotoxic accident. Disintegration may also result from (b) the introduction of discordant elements that cannot be assimilated.

            Considering the actual conditions in human beings that lead to disintegration or fragmentation of self, the 50% connectedness criterion begins to look less innocuous in its arbitrariness. The 50% criterion has been chosen because it is in the middle of the range, and not because of any attention given to the psychological nature of humans. Let us imagine someone undergoing change: each day they lose 40% of their characteristics. (We can assume they receive new characteristics in their place so that they are not simply being rapidly erased.) After four days they will have only about one-eighth of their original characteristics. Since less than 40% of their characteristics are changing over the course of any one day, the same person would continue to exist, according to Parfit’s criterion. Yet, it seems unlikely that anyone could survive such a rapid transformation intact. They might survive if their core values and associated beliefs were mostly untouched, so that almost all the changes were restricted to less critical abilities, memories, beliefs, and desires. If there is no such restriction on the transformation, then it seems psychologically unrealistic to believe that a person would survive such rapid transformation. They would be unable to assimilate the changes before the next round of alterations arrived. I will not attempt to suggest a particular limit, such as 30%/day, or 15%/day change that would allow a person to continue as an integrated individual. Such an attempt at precision would be unrealistic. Nevertheless, I can now suggest an alternative criterion for continuity. It will be less precise, on its face, than Parfit’s, but more psychologically realistic:

Continuity is maintained (identity persists) so long as the changes undergone by the person are limited enough so that fragmentation (which would destroy personhood) does not occur. Rather than 50% connectedness over the course of a day, identity persists so long as, over any given period of time, the degree of change remains within the capacity of the person to assimilate.

Over any time period, a person may be able to absorb and integrate[4] changes in 10% of their characteristics, or 20%, or whatever. Whereas Parfit seems to assume that we could set the degree of change to any percentage up to 50% and still secure the person’s continued identity, my view modifies his criterion. I assume that there will be some critical point or small range after which the extent of transformation becomes excessive; there comes a discontinuity or phase change where transformation leads to disintegration and an abrupt fall in connectedness.

ASSIMILATION AND FUNCTIONAL INTEGRATION: Although I have made what will be a minor modification in Parfit’s criterion of continuity for most purposes, the idea of assimilation will lead me consider when some change in us or addition to us becomes part of us. I will examine familiar cases of psychological change, but will primarily be concerned with additions to our abilities. When does an added ability, such as provided by a device, really become part of us? Some people casually talk of their car as part of them, or their computer, or their clothes as part of them. Is there any truth to these statements? If not, could they become true as technology advances? To answer these questions I will develop an account of assimilation. This will lead to a distinction between supplementation and enhancement—a distinction with weighty implications for normative issues such as fairness, merit, and good competition. I will begin by proposing a notion of assimilation as functional integration.

            In order to assimilate a new characteristic, we need to bring it into harmony with our existing condition. We do this by functionally integrating it with us. When we first acquire a new characteristic it may seem unfamiliar, odd, or awkward. In the case of a new belief, we integrate it by coming to understand its implications, how it supports or conflicts with our other beliefs (or how it is irrelevant to our other beliefs), and (sometimes) how we came to hold the belief. Before we go through this process, the belief is only peripherally our belief. Until it is assimilated, it may have little effect on action if it is overridden by conflicting established beliefs. (A largely unassimilated, unconscious, or repressed belief may have some effects on behavior. The low level of assimilation is nevertheless indicated by the unresponsiveness of the repressed belief to the conscious beliefs.) In the case of forming—or being induced to have—a new desire, we integrate it by bringing it into our system of desires, accepting it, willingly letting it control our actions, and by working out its place in our hierarchy of desires. Assimilation of new abilities, involving physical modifications, appears to be quite different from assimilation of beliefs and desires, yet entails analogous processes of mutual accommodation. In this discussion I will be focusing primarily on assimilation of abilities and physical alterations and additions.

            Some obvious points about assimilation of physical changes can be made quickly. As Peter Unger[5] notes: “For me to survive their replacement, larger, more central parts of me will require more assimilation than smaller, less central ones. More than this, a sequence wherein all my matter changes over by several abrupt replacements of greater, more central parts will demand more assimilation than a sequence where complete material changeover is by way of the replacement of lesser, less central ones.” [152] Since Unger holds a physical not a psychological continuity theory of personal identity, I should note that a psychological reductionist would allow that continuity has been maintained despite a sweeping material changeover so long as this did not involve a sweeping change in function. Thus, if I go through the Teletransporter, all of my matter is replaced but no change is made in my functioning. Since, in the absence of teletransporters or uploaded personalities, functional changes generally are tied to physical changes, we can ignore this distinction for the purpose of developing an account of assimilation.

            Quite apart from the foregoing point, there is reason to describe assimilation as functional integration, rather than as physical or structural integration. Physical or structural integration implies a direct physical connection between parts of a single entity. Certainly, in most cases that come readily to mind, a functionally integrated entity will also be structurally or physically integrated. It would be a mistake to make the latter a condition of integration rather than a typical concomitant. We can surely imagine entities that are spatially distributed, and we can surely find them in the actual world. A corporation whose offices, resources, and personnel are scattered across a city, country, or the world, may be counted as an entity if we can find a concordance between the parts. If the vital skills, knowledge, and resources of the corporation are divided spatially so that no one location could function alone, and if communication and movement of resources between locations ties them together, then we have a spatially distributed entity. As computers and software have developed, the idea of distributed processes and entities is becoming more familiar. Increasingly, specialized, expensive supercomputers are being replaced by networked desktop computers. These may often be physically connected through permanent wire or fiber-optic lines, but may also be physically disconnected, working together by temporarily hooking up to a communication system, or even by transmitting signals through electromagnetic emissions.[6] Early in the evolutionary process nature did not have the means to secure communication and coordination between physically separate structures. Simple organisms became physically connected, as when cells formed. Technology is allowing us to separate functional integration from physical connection, so this should be reflected in our account of the assimilation of new abilities by persons.

            The kind of functional integration occurring when a person assimilates physical changes depends heavily on internal processes. The situation is quite different with artifacts. Our judgment that an artifact has survived through replacement of its parts depends significantly on external relations. Consider the standard example of the Ship of Theseus: Parts of the original ship are gradually replaced with new material until none of the original parts of the ship remain. As the old parts are removed, one by one, they are gathered and used to construct another ship, identical in form. Many who consider this case are more inclined to view the ship whose parts are being replaced as the original ship surviving if that ship remains in use throughout the process: Between each replacement, the ship is sailed. If that ship is never sailed, we may be more inclined to judge as the original the ship constructed entirely out of the original parts. We judge the ship-with-replacement parts as the continuing original partly because it continues to fulfill the function of the original, whereas the ship constructed out of the original parts is not used until completed, if then. By fulfilling a function determined by its users’ purposes, the first ship wins the title of continuer of the original.

            When a person undergoes changes, it is internal processes that count in assimilation, not external functions and purposes. Even if a replacement or addition is made to a person by someone else—such as a muscle graft, or a gene inserted by a surgeon to alter neurochemical balances—it must still be internally integrated with the rest of the system, unlike a ship’s new plank that can simply be nailed into place. Functional integration is relevant both to the person and the artifact, but in different ways. The replacement part of the artifact becomes part of the artifact if it enables the artifact to continue its functional role, a role determined by external factors. The replacement part of the person becomes part of the person if the function of the part becomes interwoven with the functions of the rest of the body. What is the nature of this interweaving of function, this integration of parts?

            For something to be mine, for me to have assimilated it, need I have direct or conscious control over it? This may seem plausible as a requirement if we consider selected cases. This is my arm because I can move it through an act of will. I can also feel sensations directly with my arm. I can use someone else’s arm, but only either by asking them to move it in a particular way or by grasping it and making it move by applying external force. I can distinguish my vocal ability from that of another person because only my own can I activate at will. If we consider other cases, it will be obvious that unmediated or conscious control cannot be a condition for something to have been assimilated by me. I have numerous physical and cognitive functions over which I have little or no direct or conscious control. I cannot directly alter my body temperature, nor can I will my stomach to halt its digestive processes, nor can I switch off my ability to recognize faces. That direct control over parts of ourselves can lead us to think they are ours, reflects a deeper condition for functional integration that has not yet emerged in this analysis.

            Does assimilation of a new characteristic necessarily mean that only the person assimilating it has access to it in the standard way? This seems plausible: In a legal and moral sense, we say that someone owns something—that thing is theirs, it belongs to them—if they have exclusive rights to its use. (This might seem similar to the last question; however, I may have exclusive access to the use of parts of myself that I have no conscious control over.) Similarly, it seems plausible to require that for me to be integrated with a part (a heart, for instance), there can be no other person who is also integrated with it. Whether we consider a part of me that I can consciously control, such as my arm, or an ability or function outside my control, such as regulation of body temperature, it looks like only I can use it in the standard way.

            Again, though, exclusive access is a typical consequence of integration, but not a condition of it. In virtually all the cases familiar to us, if some organ or ability is mine—if we are functionally integrated—that organ or ability will not be integrated with anyone else. We do discover the occasional exception, such as siblings who have been physically joined at birth and who have developed in that condition. Consider two siblings joined at the hip or back to back. Neither of them have exclusive access to their legs or their heart. Yet, it seems reasonable to say of each of them both that it is their heart and that they share it with their sibling. Normally if I were to share an organ or ability with another person, my integration with it would have to be disrupted and their use of it would interfere with my use. Consequently, the more the other person had access to it, the less access I would have. In special cases like that of the joined siblings, two persons can share access without this disrupting the organ’s normal functioning. By developing jointly, the siblings both have become effectively integrated with the single heart. The situation is less clear in regard to the shared use of their legs. If both brains can send and receive nerve signals to and from the legs, conflicts will arise, and access by one sibling will interfere with access by the other.

            This suggests that exclusive access is close to being a condition for integration but needs modifying. A more accurate condition might be described as: Exclusive access, unless shared access does not interfere with the functioning of the part or the system. Stated another way: Access without interference by another person. Exclusive access without interference does not require physical attachment; the points of the above discussion of distributed entities still apply. I could have access to abilities located physically externally to the rest of me just as exclusively and just as immune to interference as more standard internally-located parts. I might communicate, control, and be influenced by an external organ or addition to my brain by means of signals, just as in the file synchronization example above. I might encrypt these signals so that no one else could access or interfere with my use of my external part. If I were just as tightly integrated with this external device as I am to an internal organ, or to either hemisphere of my brain, the (modified) exclusive access condition gives us no reason to deny it to be integrated with me.

            Functional integration, it turns out, seems to require something like exclusive, or at least interference-free, access to a part of the self. It does not require physical connection, conscious or direct control, nor sensory awareness of the part. The initial plausibility of these non-conditions stems from their typical concurrence with an underlying requirement for functional integration. This condition I will call interdependence. A system and a part can be interdependent in numerous ways, depending on their functions, so I cannot give a thorough and universal account of interdependence. A couple of examples should illuminate the condition sufficiently for our purposes. Mere dependence of a person on an ability or part will not suffice for that ability to count as functionally integrated with (and so part of) them. The whole and the part must be bound together through interdependence of function. In the case of beliefs and desires, as we have seen, the nature of this interdependence is clear in essence. A belief, to be my belief, must be interrelated with my other beliefs: they directly cause one another to be supported or undermined to the extent that they are relevant to one another). As briefly noted earlier, repressed beliefs may have effects on behavior yet be only slightly integrated. This is because the person and the repressed belief are less interdependent than in the case of other beliefs: The person may be strongly though unconsciously influenced by the repressed belief, but the belief has been pushed outside the person’s influence. The influence is therefore one way, ruling out interdependence. For a desire to be mine, it must be related to my other desires such that most of them contribute to a single course of action, and they motivate only courses of action that do not interfere with one another. (Similar comments apply here as applied to repressed beliefs.) We can expect something similar to apply to abilities embodied in physical parts of ourselves.

            What kind of interdependence of abilities is involved in functional integration? How, for instance, are my abilities to see and to lift weights interdependent? I can lift weights without being able to see, and I can see without being able to lift weights. As this suggests, the interdependence we are looking for is not a direct interdependence between one ability and another randomly chosen ability. For an ability to be functionally integrated it must be integrated with the person (their physical system) as a whole. How does my ability to lift weights using my muscles differ from my using a forklift truck, such that we say my muscles are part of me but the forklift is not? I may become dependent on the forklift truck for gathering food: perhaps I am trapped in an environment where I can only reach food by moving heavy rocks. The case of my muscles differs in that we are interdependent: I need my muscles to move myself and other things, but they also need me. To function, my muscles require a constant supply of oxygen and nutrients to generate adenosine triphosphate for energy, and they depend on my body’s waste removal system to handle accumulations of lactic acid and other metabolic byproducts. The muscles cannot perform their function without a support structure; they need to work together with my body’s skeleton, ligaments, and tendons in order to exert force. My bodily system needs the muscles for mobility, gathering food and water, and to realize my other abilities.

            An objection might be raised when I say that while I would be dependent on the forklift, it would not be dependent on me for its functioning. Surely, the objection might go, does not the forklift depend on me for gasoline? If so, we are dependent on each other to function, making the forklift, by this criterion, part of me. In response I note that anyone else could just as easily supply the truck with gasoline. It is true, in principle, that someone else could supply my muscles with oxygen and nutrients. However, they could only take over this function with enormous difficulty. If someone else could get that deeply connected into my muscular system, very probably they would disrupt my own control over and access to the muscles. So, although the line between the forklift’s dependence on me to supply it with gas, and my muscles dependence on me to supply them with energy and nutrients is not a sharp one, they are far enough apart on a spectrum to say that the former is not a case of significant interdependence whereas the second one is.

            Suppose my muscles and bones were rapidly wasting away. I replace their functionality by means of a powered exoskeleton. Could this be part of me? Even though I might depend on the exoskeleton in the same way I depended on my natural muscles and bones, the exoskeleton would remain an external device since the dependence is one way. What if I undergo a massive surgical procedure in which the exoskeleton is linked directly to my nervous system? A large part of the exoskeleton’s function now involves my nervous system. We are connected at a much deeper, more pervasive level than if it were merely strapped to me and controlled by voice commands. We might now grant the exoskeleton the status of being part of me, though not so clearly as were my muscles. If we further suppose that the exoskeleton responds to exercise and practice by becoming stronger and more accurate, then it will be about as functionally integrated with me as were my original muscles and bones.

            The kind of interdependence involved in functional integration, whatever the physiological details, entails mutual support, feedback, homeostasis, and adaptation. We saw how my muscles support and sustain the other functions and abilities of my body, and how other somatic processes sustain the muscles. Whether conscious or not, feedback is required to control and coordinate any bodily ability. Whereas the basic model exoskeleton lacked any direct feedback, the second type (exoskeleton-2), linked into my nervous system, does fulfill the feedback proviso, thereby explaining why this model more plausibly counts as part of me. Exoskeleton-2, while clearly functionally integrated to a substantial degree, lacks some of the responsiveness we are used to with our natural muscular-skeletal system. Since the Nineteenth Century it has been recognized that functionally integrated systems exhibit homeostasis—a tendency for the system to return to equilibrium after a perturbation. Homeostasis crucially requires feedback, but also the means for the parts of the system to influence one another in a coordinated manner. Homeostasis need not result only in a return to a pre-existing state: if new demands on the system are now being made routinely, restoring equilibrium between a system and the demands on it will require adaptation—growth and development (or atrophy in the event of decreased demands). We will be even more inclined to grant exoskeleton-3 the status of self than its predecessor because, in addition to mutual support and feedback, it adapts to the needs of the body.

            Assimilation is a process starting with something that is only partly integrated in these ways, and ending with something highly integrated. Should we withhold our judgment that something has become part of us until assimilation is complete? This is unnecessary for two reasons: First, there are perfectly normal and familiar parts and aspects of ourselves that are not totally integrated with us, yet we do not question their status as self. Second, something that is only partially assimilated can be part of us so long as the degree of integration is increasing in a goal-directed manner. On the first point, consider psychological characteristics such as beliefs. When we first form or are induced to hold a belief, we do not instantly realize many of its consequences, implications, and prerequisites. If we were never to think of the belief again, it were never to affect our actions, and it had no interaction with any of our beliefs or desires, it would not genuinely become part of the system or ecology of self. We can count it as part of us, even before significant integration has yet occurred, if it is subject to a goal-directed process of increasing its degree of integration.[7] Since we are not normally completely integrated, it would be inappropriate to impose on an added ability the requirement that it be fully assimilated before it can be part of us.

            To summarize: We assimilate a part or an ability by functionally integrating it with us. Functional integration requires interdependence between us and the part or ability. Interdependence involves mutual support, feedback, homeostasis, and adaptation. If we are functionally integrated with something, then we will generally have exclusive access to that thing. We can only be functionally integrated with it while sharing access if that shared access does not interfere with these four aspects of interdependence. Functional integration need not be an all-or-nothing matter, so there may be cases where we cannot say definitely whether something is or is not part of us. Interdependence implies exclusive access (or access without interference) because another person having access to the part or ability is likely to interfere with feedback and adaptation: Their control signals will often interfere with our own, and their adjustments to the part’s function will interfere with its adaptation to our needs. If the other person’s requirements for the part’s function differ from ours, then homeostasis cannot be attained.

Enhancement vs. Supplementation

The account I have developed of assimilation of abilities and additions to the person, as physically embodied, leads naturally to a way of distinguishing the concepts of enhancement and supplementation. The nature of and differences between enhancement and supplementation raise normative issues. Clarifying this distinction helps productive discussion of these normative issues (though is not sufficient to resolve them). Although I will not address any normative issues here, I will point out how they are affected by the enhancement/supplementation distinction. These normative issues primarily revolve around notions of fairness in sports, games, and tests.

            We assimilate a new ability or addition to our system by functionally integrating it with us, as described above. I will reserve the term “enhancement” for additions to our abilities that we have assimilated in a manner that produces enduring[8] effects. According to this usage, if you have been enhanced in some way then you have changed rather than it being the conditions under which you are operating that have changed. By contrast, if your abilities are being supplemented, then you have not changed in any significant sense, but the conditions under which your abilities function has changed. An example might help at this point, though several examples will be needed to distinguish these two concepts more clearly.

            Consider two people each of whom can run 100 meters in 14 seconds. Both of them want to be able to run the distance in a shorter time. The first person practices over months, gradually building up her legs muscles and her ability to push herself to the limit despite the discomfort. After six months, she is able to run 100 meters in 12.5 seconds. The second person does not practice hard, but completes the course faster than ever before in, say, 13 seconds due to a strong wind that pushes him along. I assume these are relatively uncontroversial cases, given my analysis of functional integration and enhancement and supplementation. I defined enhancement in terms of assimilating or integrating changes, whereas supplementation involved changes external to self—changes in the circumstances or conditions. The runner who puts herself through the training has enhanced her abilities. She has made enduring changes in numerous interrelated functions of her body and psychology. The runner who is assisted by the gust of wind but who did not train enough to otherwise run faster than 14 seconds has not made any such enduring internal modifications. The second runner in no way assimilates changes in such a way as to bring about an increase in his running ability. As soon as the wind ceases, or when he runs another race without the helpful gust, he can no longer run any faster than before.

            What about a variant on this example, where the second runner’s improved speed, instead of being assisted by wind is due to taking amphetamine before the race? I defined enhancements as additions to our abilities that we have assimilated in a manner that produces enduring effects. According to this, the second runner’s ability has been supplemented rather than enhanced since he has not made any enduring internal changes. He has supplemented his abilities by forcing his body to work beyond its normal limits by causing the release of neurotransmitters and hormones, depleting them in an unsustainable way. As soon as the amphetamine has passed through his system he can no longer run any faster than before. It might be objected that it is assimilation rather than the enduring nature of the effect that matters more if a change is to be a part of us. We might change, but transiently, before returning to a previous condition. The fact that the effects of the amphetamine last only a few hours therefore does not show that it is a supplement rather than a (transitory) enhancement.

            This objection has some force, though it does seem that the transitory nature of this effect makes amphetamine a poor example of an enhancer. Furthermore, I argued that assimilation or integration involves mutual support, feedback, homeostasis, and adaptation. While the amphetamine does lead to internal adaptation and homeostasis (at a temporary new state), it does not seem to be involved in mutual support or feedback in any interesting sense. The body responds to the amphetamine, but it does not respond to the body—such as by increasing in supply. The effect is one way; the amphetamine affects the runner, but the runner does not affect the drug (other than to simply use it up). In this way it is rather like the runner who is pushed by the wind. Amphetamine is a marginal case of enhancement. It sits on the borderline between supplementation and enhancement. Since a person typically thinks of their inherent abilities as those that last more than a few hours, we may be inclined to think of stimulants more as supplementing our abilities than enhancing them. However, we do assimilate the drug in some respects, so it has some claim to being an enhancer according to my criterion. If we are to count the effects of stimulants as enhancement, we should consider them a marginal rather than exemplary instance of the class.

            As with the conditions of assimilation, it is easy to conflate typical accompaniments of enhancement with necessary conditions of it. In this case, it might seem that an increased ability counts as a genuine enhancement only if gaining that ability requires considerable effort. The first runner had to practice and gradually develop a new level of ability. Genuinely having the skill to do arithmetic in one’s head requires practice, whereas supplementing one’s ability by using a calculator is much easier. We could easily find many other pairs of examples where enhancement of abilities involves great effort, whereas supplementing one’s abilities requires much less exertion. Exceptions that undermine effort as a general requirement include organ implants and muscle grafts. Once muscle has been grafted on (at least in an ideal procedure in which nerves and ligaments have been fully connected) the extra strength is part of the person just as much as the strength they had with their pre-existing amount of muscle.

            Enhancing oneself need not require hard work, struggle, or long-term persistence, although this may often be the case. So long as the new ability has been assimilated and integrated into the rest of our body and cognition, it is part of us. Cases like a brilliantly executed muscle graft aside, most enhancements we are familiar with are attained only with effort, for reasons rooted in our physiology and neurophysiology. Increases in physical abilities generally demand repeated stresses or patterns of activity to effect changes in our physical structure. Learning skills like balancing on one foot, playing the piano, or multiplying large numbers in our head, necessitates repetition to change the firing properties of our synapses.

            Apart from enhancements usually requiring more exertion than supplements to our abilities, the increase in our abilities associated with enhancements will typically endure longer than those associated with supplementation. In the case of the runners, the second runner loses the ability to run a sub-14 second 100 meters as soon as the wind ceases blowing in his direction, whereas the first runner’s improved ability will persist for longer. This persistence of enhancements as compared to supplements again reflects the former’s assimilation into the person’s whole system. Structural changes have been made to improve function, whereas a supplement produces improved performance only so long as it is supplied. However, it is the integration or assimilation of changes that essentially characterizes enhancement and not the persistence of the effect. We can imagine cases where supplementation (or marginal cases of enhancement) might have longer-lasting effects than a relevantly similar enhancement. Suppose we have the means to surgically implant a vial containing a drug which releases a chemical to supplement our natural level of certain neurotransmitters (the catecholamines) so as to produce heightened alertness and ability to concentrate. If the vial released the chemicals over a period of months or years, the effect could be as long lasting as an improvement produced by a structural change induced by practice. Despite this, we would be receiving a supplement (or at best an enhancer close to being a supplement) and not an enhancement. As the chemical is introduced to the system it pushes the body’s biochemistry in a particular direction. The body only remains in that state so long as the chemical is supplied—and supplied in increasing doses as resistance builds. Here we do not see the mutual support or positive adaptation that characterizes assimilation.

            Supplementation does not change the underlying abilities of the person. Once ended, the person returns to their previous state (perhaps with a temporary decline in ability). Of course returning to a previous state is not sufficient to indicate supplementation. Enhancements need not be permanent: If we discontinue exercise, our muscles will gradually atrophy. If we do not practice our skills, we will tend to lose them—some fairly quickly, others only partially over many years (such as knowing how to ride a bicycle). Ending supplementation can usually be expected to bring about a faster decline in ability, since (clear cases of) enhancements involve integrated changes not dependent on external supplies for their persistence. The crucial difference, then, between enhancement and supplementation comes from the former’s integration and the latter’s lack of integration rather than from one requiring more effort or being more enduring than the other.

MERIT IN CONTESTS, GAMES, AND TESTS: Consideration of some normative questions regarding sports, games, and tests may help to further illuminate the difference between supplementation and enhancement. We will also discover difficulties in applying the distinction. Although sharpening the distinction may not always resolve normative disagreements, it will provide a sounder point of reference.

            Persons who compete in sports, or who play games with some seriousness of intent, or who take tests, seek to improve their performance.[9] Spectators, sporting organizations, testing institutions, and the general public hold conflicting views as to what forms of advantage-seeking are appropriate. In the case of sporting contests, certain methods of improving abilities are seen as cheating, while others are granted legitimacy. It is thought legitimate for runners to train hard, eat special diets, engage in “carbohydrate loading”, train at high altitudes, receive expert coaching, and to wear well-crafted running shoes. It is usually judged as cheating if runners take stimulants, injure other competitors, or cover the course on bicycle or automobile. Ben Johnson, for example, was disqualified at the 1988 Olympics track events because he was found to have been taking the steroid Stanozolol. One objection to the use of certain performance improving substances and techniques claims that the user gains an unfair advantage over the other competitors. I think this argument has been decisively refuted (see Gardner, 1989). Whether justified or not, a more revealing objection to certain techniques, for our purposes, is not that an advantage is gained over other athletes but that an advantage is gained over the sport itself. The intended purpose or the obstacles of the sport are overcome. The purpose of the contest is to test the athlete. If some performance-boosting substance or aid is more responsible for the gained advantage, then we will be testing the substance rather than the athlete.

            Since enhancements are changes in the person, whereas supplements are external to the person, if we want a sport to test the athlete perhaps we should allow enhancements but not supplements. This may not yield a usable policy, since it may be impractical to enforce the principle consistently, or we may be unable always to decide whether a particular means of performance improvement is more like an enhancement or a supplement. Consideration of a few cases will demonstrate that this indeed can be a difficulty.

            Consider four methods an athlete might use to improve her performance. (i) She might hire a dietician and consume a carefully-tailored diet including larger than normal quantities of amino acids, vitamins, and other nutrients. (ii) She might spent a month before a competition training at a high altitude in order to force her body to utilize oxygen more efficiently. (iii) She might engage in “bloodpacking.” In this procedure, some of the athlete’s blood is extracted, strongly oxygenated, and then reinfused shortly before a contest. (iv) She might ingest anabolic steroids over the months before a contest, gaining additional strength and power. Many people would find methods (i) and (ii)  perfectly acceptable, method (iii) dubious, and method (iv) unacceptable. It is not clear that those judgments can be justified in terms of testing the athlete rather than the performance-boosting substance or technique.

            Although nutritional modification might be thought the least controversial, it might seem similar in one way to more objectionable methods. While maintaining a highly nutritious diet will simply allow the athlete to bring out her abilities effectively, the practice of carbohydrate loading might be thought of as akin to taking amphetamine. They do differ in that one will be described as “natural” and the other as “unnatural” (though substituting the stimulant found in the ephedra herb will confound this move), but both give the athlete a very transient and unsustainable boost. Here it becomes uncertain whether we are genuinely testing the athlete or the supplementation. Carbohydrate loading is allowed probably for a combination of reasons: it is practically impossible to test for; it is counted as natural; it is not harmful; and although it may temporarily boost the athlete’s ability, it does not defeat the point of the competition or make it easier (unlike using a bicycle in a running race). Although the carbohydrate that is loaded is more plausibly seen as a supplement than an enhancement, the overall performance of the athlete and her ability to utilize the additional calories still depends to a high degree on her true capacities as determined by long-term training. (That this seems to be as true of amphetamine ingestion suggests that current rules are not actually based on any goal of primarily testing the athlete.)

            High altitude training probably provokes more suspicion of unfair advantage than does carbohydrate loading and dietary expertise, perhaps because it is less equally accessible. In terms of the goal of having a sport test the person rather than a performance-increasing agent, however, high altitude training seems more defensible. The athlete still must undergo the same training—training that gradually makes integrated changes in the athlete’s body—but the lower oxygen content of the air now provokes a stronger adaptive response. This adaptation is considerably more enduring than the effects of carbohydrate loading (or stimulant ingestion). The high altitude training does result in enhancements to the athlete’s whole system, so the contest will test the athlete in the desired manner.

            Bloodpacking or “bloodboosting” works because of effects on oxygenation, like high altitude training. Nevertheless, bloodpacking has more in common with carbohydrate loading in that the heightened performance capacity is transient. Clearly it is a form of supplementation rather than enhancement. Again, this does not necessarily mean we should rule it out in order to protect the integrity of the sport. We may judge that the sport still primarily tests the athlete, the supplementation merely adding slightly to performance without making the course significantly easier.

            Steroids present an interesting case. They are universally rejected by sporting bodies (though not by athletes) as legitimate performance aids, yet arguably they are just as much enhancers as they are supplements. Steroids alone will not greatly improve performance. They promote greater muscle anabolism and the resulting growth in strength and power only when combined with hard training and good nutrition. Their effects take months and endure for months after use is discontinued—unlike stimulants, blood-packing, or carbohydrate loading. In every way except their “naturalness” their effects appear similar in nature to high altitude training. Anabolic steroids accelerate training-induced adaptations in the athlete. These adaptations involve changes as physically-integrated as those induced by training alone. Since steroids enhance athlete’s abilities rather than supplementing them, there is no justification for their prohibition if the grounds are the integrity of the sport (rather than paternalistic concerns).

            Similar considerations arise outside sports, if we turn our attention to tests. These include physical tests, but I will focus on cognitive tests such as IQ tests, the SAT, GRE, LSAT, and GMAT, and various tests administered by potential employers. The tests usually are given with the purpose of determining a person’s abilities in order to ascertain how they are likely to perform at a job or benefit from a certain level of training. If this is the purpose of the tests, should the administrators prohibit, or at least discourage, the use of neurochemical aids such as “smart drugs” or “nootropics” and mood modifiers such as selective serotonin reuptake inhibitors (SSRI) of which the best known is Prozac?[10]

            Nootropics work differently from stimulants such as amphetamine or dexedrine, in that they do not result in depletion in neurotransmitter levels when discontinued. Nor do they generate neurotoxic byproducts. Typically they work by increasing the supply of neurotransmitter either by supplying more of them or by slowing their breakdown. Although the mechanism of action differs from the more familiar stimulants, bearing in mind the conditions of functional integration, they are clearly supplementing rather than enhancing cognitive function. If a test administrator wishes to test only the integrated abilities of test taker, she will want to exclude the use of such substances. If the intent is the related but separable goal of testing how individuals will be able to perform tasks over the long term, there will be no reason to exclude these generally non-toxic drugs whose effects do not diminish significantly over time. So, although both nootropics and stimulants are supplements, they may be treated differently if what is being tested for are sustainable abilities rather than strictly intrinsic, integrated abilities. (For the same reason, a few years after electronic calculators became commonplace, mathematical tests began to allow their use, recognizing that anyone would later have access to these devices.)

            Mood modifiers such as the SSRIs can be treated similarly to the cognition augmenting drugs. The improvement in mood and productivity resulting from supplementation will not result in functionally integrated change (generally SSRI users revert to their former condition upon discontinuance), but these substances appear to be usable over long periods of time. If, instead of using nootropics, someone undergoes gene therapy (if we can call the augmentation of normal function “therapy”) the result of which is to produce enduring and endogenous increases in neurotransmitter production or improved regulation, then they will have been enhanced rather than supplemented. Permanent changes will have been made; no infusions will be needed, and the augmented functioning will be attributable to the person rather than to a substance. In this case, whether a test is intended to measure native ability or to measure sustainable performance, there should be no objection to genetically-caused performance enhancement.

            As a final example to help draw the slightly fuzzy line between supplementation and enhancement, consider a series of devices starting with a handheld PDA (personal digital assistant) and ending with a fully integrated neuroprosthesis. Growing numbers of people are carrying around PDAs to help them remember phone numbers, appointments, perform calculations, and to communicate via e-mail and fax. Occasionally we may hear someone exclaim “I couldn’t live without it. It’s practically a part of me.” In light of the earlier discussion of the conditions of functional integration, it will be clear that such talk is hyperbole. A PDA may be more flexible and powerful than a paper-based organizer, but it is barely more integrated with the user. What if such devices shrink until they can be worn in the form of a paper-thin headband or a small earring, and become controlled by voice commands or gestures? This would merely make our control over the device more natural. The PDA would in no way adapt to us or be functionally interrelated with us in any deep way. If the evolution of such devices were to continue, the clarity of the device vs. integral part distinction might begin to blur, and eventually some descendent of these devices might count as just as much part of us as does a hemisphere of the brain and the functions it performs.

            Suppose these devices evolve beyond simple cleverness and programmability. They are built with neural networks, genetic algorithms, fuzzy systems, and artificial intelligence so that they monitor the wants and habits of the user and learn to anticipate them. The device is connected to the body in such a way that it notices changes in the skin’s electrical potential, alterations in brain wave patterns, changes in pupil size, body temperature, and so on. It learns how these changes are associated with various behaviors and demands on itself. It might even be connected to implanted physiological tuning devices that alter mood, hormone levels, or whatever. As the device becomes this interconnected with us and responsive to us, and we come to depend on it more, we may begin to feel the device to be a part of us. Taking the evolution of the device further, suppose it is implanted in your brain. Nerve growth factor is used to induce axons and dendrites to grow into the neuroprosthesis where they link up with adaptable connectors. But the assimilation of the device is not simply physical. You cannot immediately control it by issuing voice commands or gesturing at a visual display. Instead, you learn to make it work by experimentation which forges new connections and alters synaptic weightings. After a process of assimilation, your cognition becomes distributed over your neural tissue and the neuroprosthesis, so that the device cannot be removed with disrupting both its and your functioning. You sense the results of the neuroprosthesis’ processes in a manner subjectively indistinguishable from your own. At that point it no longer makes sense to distinguish yourself from the device, except in the way that you might talk about the functions of your brain’s left parietal lobe. This process of assimilation of a device proceeds in stages, so there may a range of cases in which we have no clear answer to the question: “Is it part of me?”

III. Intrinsic and Instrumental Bodily Identity

Many of the changes in self considered in this chapter are those brought about by making alterations to parts of the body. On psychological reductionist accounts of identity, what matters in survival is psychological connectedness and continuity. What exactly is the relationship between these changes made on a physical level and their effects on personal continuity? Does psychological reductionism commit us to the proposition that physical continuity, in itself, contributes nothing to survival? If so, what precisely does this claim mean?

            In considering these questions I will not examine theories of personal identity as bodily continuity, nor defend psychological reductionism against such theories. This dissertation starts from an assumption of the truth of psychological reductionism in order to explore the structure of such a view in detail. It is not my purpose to argue for reductionism itself, nor to show that the psychological criterion is more defensible than the physical criterion. Nevertheless, since changes to our personalities, especially as considered in the current chapter, work through physical causes, I find it necessary at this stage to clarify the physical/psychological distinction. The main issue here is: To what extent is psychological continuity or change independent of physical continuity or change?

            The question of the relation between the physical and psychological aspects of self seems especially pressing if we accept physicalism—as I do. On any version of physicalism, psychological characteristics are ontologically dependent on some physical embodiment. A person consists of an embodied psychology. All that exists is physical stuff and its relations and organization.[11] How then do we decide which characteristics are psychological and so constitutive of identity, and which are merely physical and so only instrumentally important to identity? If our behavior and personality depend on the structures in our brains and bodies, which were shaped by evolutionary pressures, we face the challenge of usefully separating the intrinsic from the instrumental aspects of persons. If we modify a brain, or alter a hormonal balance, or change the shape of a body, it might seem that we would necessarily alter the associated personality.

            Certainly we must grant that the continuity of many aspects of our bodies will contribute in a major way to the continuity of our personalities. In the most general terms, without a body of some kind, it seems incoherent to imagine existing as persons at all. We would have to exist without existing in any place or time. We could have no point of view or sensations or perceptions, since these would have no location or field of input. Possessing a body must be at least of great instrumental significance.

            Apart from the need to have some kind of body, it also seems clear that the particular body we have will deeply affect our identity. Our personalities express themselves through our bodies. To express our values, to act on our desires, to carry out our intentions, we need to use our limbs, voices, facial expressions, and physical capabilities. For most psychological characteristics a wide range of embodiments would suffice. The expression of generosity, anger, or joy are compatible with almost any human body, and with many possible non-human bodies. Some bodies will serve our expressive needs better than others. If I were unfortunate enough to have an accident that deprived me of all control over my facial muscles, I would find it more difficult to express feelings of many kinds. I might find it more difficult even to experience some of those feelings. Some psychologists have argued that part of feeling an emotion (at least consciously) is its expression.[12] (Can you be angry while your muscles are relaxed, your blood pressure and cortisone levels low, and your physical motions non-threatening?) If these psychologists are correct, emotional continuity will limit the range of bodily configurations compatible with maintaining a particular psychology. Even without being able to somatically experience an emotion, we might experience it in a weaker form so long as the emotion can have hormonal and neurological effects. Bodily form may therefore limit the intensity or clarity of emotional ability more than limiting the range of emotions we can experience.

            Among the personal characteristics I have listed as contributing to personal identity, I included abilities. Aside from the most cerebral, abilities confront us with another reason to grant the importance of the body’s contribution to personal continuity. To be a particular person with a particular life requires specific physical abilities. If all the passion in my life is bound up with testing experimental aircraft, I will need a body with keen vision and fast reflexes. If my life revolves around running marathons then, should I suddenly become paraplegic, I may feel that I am no longer quite the same person (connectedness has dropped considerably). Even our bodily appearance, because of others’ reactions to it and our responses to those reactions, powerfully affects our sense of self, encouraging us to develop one kind of personality rather than another. Two persons with similar bodies, say obese bodies, may be influenced in quite different directions depending on the way they choose to think, yet both may develop differently than they would have had their bodies been different.

            Given these considerations, what sense is left of the psychological in psychological reductionism? We need not deny any of the foregoing grounds for assigning great importance to the body’s contribution to our identity or continuity. Here is my thesis and its explanation:

THESIS: The contribution of bodily features to personal continuity is entirely of instrumental importance. Parts of a body gain their instrumental importance from their functional roles. The particular matter constituting a body, and even the specific form of a body, have no intrinsic significance for personal identity.

            On a psychological reductionist view, it will be clear that our survival does not depend on the persistence of the same matter constituting our bodies. We can see this even without invoking imaginary possibilities. None of us worry that we will cease to exist within months or years as a result of the turnover of cells and atoms through the normal processes of metabolism and catabolism. Even those who believe continuity of the body to be essential to personal survival do not attribute intrinsic importance to the persistence of the very same matter in a body. I will not argue further for the insignificance of material continuity; to do so would be to repeat the arguments of Chapter 2, “Causal Conditions for Continuity” and Chapter 3, “The Terminus of the Self.”

            Although the substance of the body lacks significance for our continuity surely our bodily form has intrinsic importance? If my body suddenly transformed into something radically different, into something with different kinds and numbers of limbs, scaly skin, five times the mass, and a frightening appearance, wouldn’t I have changed? Obviously this would bring about a change in me that I would reasonably count as a reduction in connectedness. However, we need not grant any intrinsic significance to bodily form. To see this, notice first that the changes in form gain their significance from associated changes in function and, second, that we can have changes in form that do not have associated changes in function. On the first point, suppose that the bodily transformation happened instantaneously—objectively (a miracle) or subjectively (I awake from a brain transplant). At first I would be strongly connected to my pre-transformation stage. My memories, intentions, dispositions, and values would remain intact, as would some of my abilities. One immediate change would be the loss of abilities incompatible with my new body and the gaining of new abilities. My personality might begin to change after the transformation in response to the form of my monstrous new body. If others feared me or attacked me on a regular basis, I might become more taciturn, less sociable, and experience a change in my feelings about my body from positive to negative. Numerous everyday observations support the idea that the form and appearance of individuals’ bodies affect their personality. Spotty teenagers tend to be more shy than those with clear skin; unusually short people may develop defensive attitudes or seek compensatory achievements; the beautiful may be more confident or vain.

            None of this shows that bodily form, in itself, constitutes our identity. My monstrous, reptilian body functions differently than my former human body. It is this change in function, and the changes it leads to in my personality that matter. Without hands with opposable thumbs, I will lose the ability to accomplish certain tasks: the change in form brings a change in function. If these tasks were important to me I will be unable to express part of who I was. If I could compensate for this loss with tools, that part of my new form need have no effect on my personality. The old and the new forms do not, in themselves, make me who I am. Being healthy may, over time, affect my attitudes, desires, intentions, and projects. The particular form of my immune system has only instrumental importance to my health. If I can maintain my health with differently formed organs, then the characteristics that constitute me remain intact.

            It may be objected that the form of internal organs differs in significance from the external form and appearance of my body. In the case of internal structures, the objection might say, it is true that the form is irrelevant. However, if my limbs change their form their function will necessarily also change. I may no longer be able to run swiftly or jump high. Also, as in the case of the spotty teenager, or when I become reptilian, others will react to the change in appearance. If I were a model or an actor, the change in form would necessarily bring with it a change in function. My massive, scaly body simply cannot effectively present an Armani suit. This objection seeks to collapse the distinction between form and function, at least in some cases (however that range is delimited), in order to make us grant that bodily form can matter intrinsically.

            We can reply by holding that even if there were cases where form were inseparable, in principle, from function, it would still be the function that we really care about. The distinction would seem merely academic if form and function always were inseparable. If, as in reality, the two are at least sometimes separable, it seems reasonable to pick out function as what matters even in those cases where we cannot see how form and function could be parted. We can reject the objection more firmly by denying the existence of any relevant instances where form and function are inseparable in principle. Suppose I am proud of my legs because they support me well and enable me to move swiftly. A transformation into thick stubby legs would continue to perform the same function if I relocated to a place with a stronger gravity field. If the slim shape of my body affected my personality by being attractive to others (and pleasing to myself), a change in form need not change the associated function; the standards of attractiveness might change so that my new, more adipose, shape had the same effects as the old. Our bodily form, we can conclude, gains its significance from its effects on our personality and from its enabling or restricting what we can do. The specific effects of a bodily form and the abilities it gives us depend on our environment and circumstances. Bodily form has instrumental but not intrinsic significance.

            I have argued that neither the matter composing my body nor the form of my body have any intrinsic significance for my identity. That leaves only bodily functions and abilities as candidates for a way in which our bodies may have intrinsic importance to our identity. Even here we should be careful in granting more than instrumental importance. Only some bodily functions have intrinsic significance for identity. To distinguish these from the others we should note that in saying some bodily functions are intrinsically significant, I’m saying that some functions that are embodied have intrinsic significance. Putting it this way stresses the importance of the function rather than its particular embodiment. This follows from my rejection of the intrinsic significance of bodily matter and form. Some of the intrinsically important functions or abilities afforded by our bodies include the ability to communicate (through gestures, facial expressions, sounds and words, posture, etc.), to move through space, to have an effect on the world by affecting objects, to perceive with our senses, and the sustenance of our life and consciousness. These functions clearly are high-level and their functional role is characterized quite abstractly. Each of these functional roles is tokened or embodied in specific physical structures. These complex physical structures can be broken down into collections of functional components. For example, our ability to affect objects can be broken down into functional subsystems such as the muscular-skeletal system, the nervous system, proprioceptive senses, and energy production. Each of these can be further broken down. Our body’s energy-producing function can be broken down into functional parts such as digestion, hemoglobin, ATP, mitochondria, and so on. The particular embodiment, the occupant of the functional role, has instrumental significance because it is what actually embodies the high-level function. However, the particular embodiment, the lower-level functions, lack intrinsic significance for our identity. What matters to our personal continuity is, for instance, our ability to perceive the world. In the world as it is, we do this through things like muscular contractions of the lens of the eye, oxygenation of tissues, the vibration of tiny bones in the ear, and so on. If the same high level functions and abilities came to be embodied in other ways yielding the same level of sensory ability and acuity, the functions crucial to the expression of our selves would continue as before.

            Changes in high-level functions and abilities, such as our ability to communicate, to perceive, and physically to affect the world, directly change us. A change in my ability to perceive or to communicate will affect the degree of connectedness between my phases. By contrast, a change in some lower-level function such as the working of my retinal cells, will only affect connectedness if it results in a change to the higher-level function. If my failing retinal cells are replaced or supplemented with a synthetic implant so that I continue to see as before, the change in function has no significance for my connectedness. The intrinsic significance of changes in the high-level functions is due to their directly affecting what I can do and who I am; since they are the top-level functions, when they change I must also change. When lower-level functions change, I might change but may not. This is not to say that the effects on my identity of changes in high-level physical functions does not depend on anything else, while changes in lower-level functions does. Certainly, the overall effect on my identity will typically depend not only on the change in physical function but also on how I respond to it. Suppose I suffer a loss in hearing in the range normal for conversation. Depending on the personality I already have, in response to this change I might (a) become more assertive in having people repeat things and in seeking compensatory strategies; or (b) become more shy and unwilling to socialize; or (c) become a more angry person due to focusing on my frustration. This partial dependence of the results of a change in physical function on existing personality, however, does not detract from the intrinsic significance of the function that has changed. It is merely a result of interdependence of personal characteristics. Interdependence is just as true of the most purely psychological of attributes. Interdependence of significance is not the same as instrumentality of significance.

            At this point, it seems appropriate to note that the distinction between bodily functions and psychological attributes may not be sustainable in every instance. It may be that certain psychological attributes or modes of experience not only happen to be embodied in certain brain structures and chemical systems but have to be so embodied. Thinkers like Roger Penrose and John Searle argue that features of persons such as consciousness may necessarily require the very physical mechanisms that we find in human beings. According to this view, brains made of silicon or optical processors would not have these characteristics of persons. I strongly doubt this view but need not argue it here. I will simply note that if they were right then some bodily functions (such as neuronal function, or quantum mechanical effects within synapses) would have intrinsic significance since they would be type-identical with psychological attributes. In that situation, we could say that only psychological attributes have intrinsic significance only if we also granted that some physical functions were inseparable, even in principle, from some psychological features.

            Even excluding the last possibility, I have granted intrinsic significance to certain high-level physically-embodied functions which do not seem plausibly or usefully described as psychological: the ability to communicate through physical means, to move through space, to physically affect objects, to perceive with our senses, and so on. All of these functions have numerous and complex interrelations with psychological attributes and abilities, but are not themselves psychological. This suggests that describing my form of reductionism as “psychological reductionism”, or saying that I hold a psychological criterion for reductionism, may be a little misleading. On the other hand, these physically-embodied functions are not essentially physical. Although they must have some bodily instantiation, what matters is the function and not its particular embodiment. So it would also be misleading to say my criterion for reductionism was partly psychological and partly physical. Since I do not have a better term, and the existing one captures most of what I mean, I will continue to describe my version of reductionism as psychological.

I’ll conclude this section by considering possible objections to my denial of the intrinsic significance of bodily matter or form for identity: What if someone were to insist that without this very nose they now have, they would lose part of their identity? Or they might insist that being exactly six feet tall was itself intrinsically part of their identity. Viewed instrumentally, these features could plausibly be seen to be significant to a person’s identity. Perhaps that height is privileged in their culture, enabling them to engage in activities otherwise unavailable. Instrumentally, someone’s nose might matter, perhaps because its shape reminds that person of someone they admire and so serves as a constant reminder to live up this person’s standards. If we leave aside all instrumental and functional considerations, what sense can still be made of insisting on the intrinsic significance of these features for identity?

            Those committed to the intrinsic significance of their somatic features might try to turn the question around and ask: “Why hold memories or desires or dispositions to matter to identity? Can any reasons be given, or is it just a brute fact of what actually matters? If the latter, why can’t I believe my height or nose matters without further justification?” Rather than trying to make a complicated argument in response, it seems adequate to state that, yes, it is a brute fact. When we ask what matters in this context, we are asking what is it that must continue for a person rather than an object to survive. What I have shown is that physical features, in themselves, do not constitute us. They do contribute to our identity, but do so in virtue of their functions and the way in which we make use of them. If we make no use of a bodily feature, and it has no effect on us, then it makes no sense to claim that it still matters to our continuation.

            Another approach might be taken by an advocate of the intrinsic significance of the body. This approach claims that we can rationally attach intrinsic importance to a particular body because of its history. Such a claim may be made persuasively by a close analogy. Suppose you own an original sketch by Leonardo da Vinci. Probably you would not be happy if I were to take it away and replace it with a reproduction. If you are like most people, you would want the original da Vinci. We might even suppose that, using spectroscopic analysis and molecular nanotechnology, I have duplicated your original exactly at the atomic level. My reproduction contains the same number of the same elements (and each particle is indistinguishable) arranged in precisely the same ways. If the original and the duplicate were mixed up, there would be no way that anyone could tell which was which. Nevertheless, the objection runs, most people would, perfectly reasonably, want to have the original. The reason for this lies in the history of the original. Only the original was actually touched and handled by the great engineer and artist. By possessing and connecting with the original, because of its direct causal connection with da Vinci, we are able to connect with him (even if in a tenuous sense). This is something we can reasonably value in itself. The argument, continues the objector, can easily be extended to our bodies. We can rationally attribute intrinsic significance to the very bodies we have because of their history. This consideration persists despite the fact that a perfect copy or functional replacement could fulfill all functions of the original.

            I am not convinced of the reasonableness of preferring an original artifact over an atomically exact duplicate. (In the case of informational artifacts, such as software and algorithms, such a preference becomes completely baffling.) However, I do feel some affinity for such a preference, and cannot demonstrate it to be irrational. If the preference is maintained even after one understands that there is no practical difference, perhaps we cannot criticize such an unconditional preference as irrational. However, the case of granting special status to original artifacts (or natural objects) does not support granting intrinsic significance to our bodies. The feeling that it does arises from a confusion of the normative and metaphysical senses of  “what matters” in survival. Just as I might prefer to have da Vinci’s original sketch, I might prefer to keep my original body, rather than having a brain transplant or go through a Star Trek transporter. If I knew I was to have the brain transplant, I might care less about my future phases than if I were to continue to inhabit the same body. If I held to this preference after understanding all the facts of the situation, perhaps my preference would not be irrational. Nevertheless, the fact that I have the same body, or that I prefer to keep the same body, has no effect on the actual degree of connectedness. My body’s history is a fact about it, not a constitutive characteristic of it. If I had a tremendously strong preference for keeping the same body, I might not care at all about the post-transplant phase. I seriously doubt the rationality of such a pattern of concern, since it seems too far detached from the facts. Even if we granted this concern not to be irrational, it would not change the fact that I would survive the procedure. I might not care that I would survive, but I would survive nonetheless. All my psychological characteristics would persist, and all my physically-based abilities and functions would persist. The way I feel about the procedure has no effect on this.

Conclusion

In this final chapter I have filled out an account of psychological reductionism so as to take account of the fact that our psychologies are physically embodied, and that changes in our physical nature can have effects on our identity. I have tried to grant the significance of our physical existence while maintaining the primacy of the psychological or functional level of our identity. Without this consideration of our physical nature, and the relationship between our physicality and our psychology, a psychological reductionist view of identity would be in danger of falling into a faulty dualism.

            While much of this dissertation has followed the traditional path of ascertaining the conditions under which identity is preserved, in this chapter and the previous one I have focused on ways in which we can change and the significance of this to us. While some of the examples used in formulating principles for continuous and discontinuous transformation and assimilation have been imaginary or speculative, many have drawn on self-transformative practices and technologies commonly employed. Over the last few centuries, at least in the Western world, and particularly in recent decades, increasingly humans have sought ways to transform themselves. No longer content with their given identity, and faced with more alternatives than ever, we have devised technologies, lifestyles, fashions, and beliefs that increasingly allow us to transform ourselves—to create new identities in the image of what we value. Given this continuing trend, and the accelerating advances in genetics, biotechnology, neuroscience, computing, and other fields, the normative issues touched on in this chapter—and related questions about the propriety of altering the natural order—will receive increasing attention and urgency. This dissertation was motivated by a desire to contribute towards an improved understanding of these issues.

---

[1]I include the qualification because there will be some persons who will prefer deteriorative or destructive changes. An extreme case would be one who finds life unbearable and who wants to die or to slip into a persistent vegetative state where no decisions or actions will be required or possible.

[2]Martin, Raymond.  (1991).  Identity, transformation, and what matters in survival.  In Kolak and Martin, eds. (1991).

[3]This criterion seems uncomfortably arbitrary. In the next subsection I will attempt to develop a more useful criterion based on a notion of assimilation of changes to central characteristics.

[4]Integration comes in degrees, so questions will arise such as: At what point has a person integrated a change? Suppose someone undergoes a horrific experience, an experience too painful for them to cope with. If they repress the memory of the experience but it still has effects on their behavior, should we say they have integrated the experience? Conditions for assimilation will be developed below. Here, I will say that there has been partial integration, since the experience can have reliable effects on the person, but clearly the integration has not been complete. Fuzzy cases like this warn us not to try to force a yes-or-no answer to every question about integration.

[5]Unger, Peter. (1990).

[6]Some personal digital assistants (PDAs) and portable computers are able to synchronize their files with those on a desktop computer without any physical connection: The PDA is held near the computer, they recognize each others transmissions, and automatically exchange information.

[7]McInerney suggests this condition in McInerney (1985), p.200.

[8]Obviously terms like “enduring” are not precise. As my examples will show, there is no clean line separating enhancements from supplements. Therefore I do not seek to make my definition less vague than what it represents in reality.

[9]In some discussions this is referred to as “performance enhancement.” See, for example, Gardner (1989) and Shapiro (1990). However, that usage covers both what I call enhancement and what I call supplementation.

[10]Apart from Prozac (fluoxetine), other SSRIs include sertraline and paroxetine.

[11]We could accept mathematical Platonism without affecting this issue, since persons are not numbers or sets.

[12] Wilhelm Reich, for example, used this now widely accepted hypothesis in his therapy. Reich’s “body armoring” idea suggested that repressed emotions manifest themselves as areas of tension. By releasing the bodily tensions the emotions were released and experienced.