Introduction

The rapid advances in information and communication technologies have created tremendous changes all over the world, not least among which concern a number of new philosophical problems and ways to solve them. During the last few years social networking websites such as Facebook and Twitter seem to throw much of the traditional thinking about the self and the object into confusion. A main characteristic of social networking is to form webs of links among “persons” whose identities are there on the social networking websites. It is typical for a member of Facebook to have hundreds of “friends.” It does not matter how many of these “friends” are those whom the member actually meet and interact in real life; what does matter is that the interaction is taking place more and more online. The line between the real person (or the “offline” one) and her projection onto social networking sites (her “online” self) is becoming blurred. There are situations where a real person has multiple accounts on Facebook, each with a unique personality. For example, the person might appear as a serious professional in one account, and a completely different personality in another. These accounts, or to put it better these personae, seem to be on a par with the real person herself when it comes to the question of identity. So the questions are how one can distinguish between the offline, real-world self and her online projection onto social networking sites?; how are the two different or similar to each other?; and who is the real person behind all these personae and façades?

This paper contends that a fusion between the offline and online is taking place in the area of the self and the person. This fusion also appears to reflect the view that even reality itself is informational; that is, it is constituted by information (Wheeler 1990: 5; Floridi 2008). I will argue that both the offline and the online selves are ultimately constructions and do not have any essence of its own. This argument reflects my own Buddhist inclination, which is based on the idea of the Non-Self, namely the view that what is commonly understood to be the self is nothing more than a convenient label one puts on a myriad of mental and physical episodes.

More specifically, I would like to argue that the so-called online self, in other words the putative self existing on profile pages and updated timeline or news status on social networking sites such as Facebook, Twitter and others, is essentially no different from the real self that is already there in the “offline” world. As the offline self is a construction (as I shall show in the paper), so is its online counterpart. The online self functions as a persona, a front used by the underlying person when she faces the public world, and there is a degree of freedom within which the person can create her persona the way she likes. Moreover, the constructed nature of the online self also shows that privacy tends to be as valued in the online world as in the offline one. This seems to contradict a widely held view that users of social networking sites tend to be more open toward one another and perhaps value their privacy less in social networking situation than in others. However, it is not the aim of this paper to discuss privacy in any detail; it is only my aim to note that if the argument about the identity of the self and the person presented here is tenable, then this will have a far reaching effect on discussions of privacy.

Where there appear to be persons, there arises the traditional problem of their identity. How should the identity of a person be understood? Here I propose a more externalist version of identity where identity is constituted more by external factors rather than solely on the subject’s own mental content. Then I discuss Kant’s famous Transcendental Unity of Apperception (TUA) and try to find out whether this mechanism succeeds in securing identity of the self or the person. What I find is that Kant’s TUA succeeds only in securing unity of a person, but not her identity.

So the section arrangement is as follows: In the following section, I discuss an empirical situation of what is happening in Thailand, an Asian country that is fast embracing social networking sites into its social and cultural milieu. Then in the next two sections I discuss the philosophical problem of personal identity in the online world, arguing for an externalist version of identity construction. The last section then concludes the paper.

The Putative Self, Social Networking Websites, and Thai Culture

When we look at the popular social networking websites today, we are struck by the sheer number of the people who are connected to one another through them. Facebook has more than 500 million users at the last count (http://www.facebook.com/press/info.php?statistics), and the number far exceed the entire population of many countries. Twitter has around 75 million (http://www.numberof.net/number-of-twitter-users/) (There are certainly other social networking sites, such as Google Plus, Hi5, MySpace, and so on, but these two are the most popular). These users put up their “profile pages” on the websites, which are essentially a projection of their own identities in the online world for their peers, colleagues and friends. In most cases the profiles actually represent the persons behind them; in other words, the profiles are mostly intended to refer to the persons themselves. This can be useful when, for example, I would like to find out whether my long lost high school friend is on Facebook or not and can get reconnected with her again after I have seen her profile. In this case there is a clear link between the profile and the person. However, in some areas, especially in Thailand, the profiles on Facebook serve another function. Many Thai Facebook users opt not to show their faces or their real names on their profile pages. Instead they are very creative in inventing new names for themselves which effectively prevents anybody from knowing who the real person behind the Facebook persona is. So unless the person herself tells her friends who she really is, nobody would know who she is. Instead of putting up their own portrait on the profile as is originally intended by the very name “Facebook,” many in Thailand are putting up all kinds of pictures: Some put up pictures of their favorite pets; some put up a political banner complete with the Thai national flag; some use pictures of well known cartoon personalities such as Winnie the Pooh, and so on. Furthermore, they are not using their real names in the profiles. Some call themselves “Laughing out Loud throughout the Field,” “Red Linguist,” “Dragon from the Plateau,” and so on. A recent practice has a result of the ongoing political conflicts in Thailand is that many put up the phrase “love the King” following their names to show their support for the King. Someone else who stand on the other side of the political divide then say something like “Love Everybody” or “Love my Parents” or “Love Humanity,” to mimic those who declare their love for the King. They cannot say outright that they do not love the King because according to the draconian law against criticizing the King in Thailand this might be interpreted as insulting to the King himself.

This practice of putting up anonymous or alternative profiles on social networking sites, then, is common in Thailand. The practice is rather different from the older one of putting profiles on web discussion forums in that social networking sites are originally intended to act as a forum where those who are already know one another interact online. When one poses a profile on Facebook or LinkedIn, a professional social networking site, for example, one in effect is telling one’s own social group of one’s presence and is inviting those in the circle to join the link. The sites are “social” in the sense that getting together in a bar or in a meeting are social—the sites are gathering places where one interacts with one’s friends. In older web discussion forums, however, these social aspects are not as much emphasized. The profile page is almost non-existent in these older forums and there is no way for a user to “update her status” in order to tell her social group of what she is up to as she can on Facebook or Twitter.

Let us look at a profile page of one Thai user of Facebook. He calls himself “Burn Out.” Then we can analyze his strategies in constructing his online identity on the social networking website:

We do not need to worry about his real identity here, because he does not use his real name on his profile page. He also uses a picture of a koala as his profile picture. The result is that, if one does not know beforehand who Burn Out really is, then there is no possibility of knowing his real identity.

So Burn Out has created an online persona that does not necessarily connect with his real life person. The name “Burn Out” is only used for his Facebook account. The koala apparently does not have anything to do with his personality (I know this because I know who “Burn Out” really is, and he is just an ordinary Thai man having no essential connection to a koala). Thus, it seems that the profile page here functions only as a place holder, a neutral platform on which Burn Out can project his thoughts and ideas to his friends most of whom already know him in real life. Furthermore, those who do not know him personally also can interact with him; they know that he is a Thai person and has certain likes and dislikes based on what he posts on his Facebook profile. Since they are all Thai, they already share quite a lot in common so there is little need to reintroduce background information as is the case when people from remotely different background need to do when they interact. The situation here represents Hall’s view of the “high context” (Hall 1976), in contrast with the “low context” situation normally found when people from different cultural background meet and interact with one another.

Contrary to the received view that Facebook and Twitter tend to reflect the true identity and personality of the users (since according to the view users tend to be more revealing of their personal information), in the Thai context at least the characteristics of the earlier anonymous online discussion tends to carry over onto social networking sites, though in a visibly different form. This is a new finding that deserves a closer look and more systematic research, which is not the aim of this paper. Nonetheless, if this preliminary finding is tenable, then it would contradict studies such as one by Zhao et al. (2008), which shows that construction of identity in social networking websites tend to reflect more of the true identity of the user. In their terms, social networking sites such as Facebook are ‘nonymous’ rather than ‘anonymous’ (Zhao et al. 2008). However, the data from the Thai use of Facebook tend to show that there are strong elements of anonymous asynchronous discussion found in earlier online discussion lists in social networking sites too (Zhao et al. 2008). What is different between my preliminary findings and those by Zhao et al. seems to be that Zhao et al. argue that the use of social networking websites is markedly different from the use of older forms of websites in that in the former the users tend to be more open toward revealing themselves to the public world. Nevertheless, what we have found in this very brief discussion of the Thai situation is that Thai users tend to protect their privacy through creative uses of profile pages in Facebook (and also in other social networking sites).

In the paper I will show that both the offline and online selves are constructs, and if there is any line purporting to demarcate one from the other, the line is also a construct. What this means is that both kinds of selves are not something that exist out there in themselves in objectivity. The self is something that emerges out of certain components and activities that constitute it. This is a strongly philosophical point which will be discussed in detail in the next sections.

Personal Identity Online, But Who or What is the Person Here?

So far we have conducted a preliminary investigation on the behaviors of some Thai users on social networking sites that appear to contradict what seems to be a received view on how social networking sites are being used to construct identities. But what exactly do these behaviors tell us about the philosophical problem of self and identity in the online phenomenon? Instead of using social networking profile to show who they really are, many Thai users are using it in a creative way, essentially to create a new persona which exists only in the online world. A reason behind this move may be due to the fact that Thailand still has very limited freedom of speech; it is a serious crime, punishable by jail terms up to 15 years, to commit lèse majesté, or insult to the king. This law has in recent years been interpreted in such a way as to suit the political agenda of the faction that is holding power in Thailand, and the result has been that more people have been charged with the crime than ever before. Many of them, moreover, have been charged simply because they have posted information and ideas on the Internet. The newly created persona, then, allows the person behind to say things in such a way that would not be possible if the person revealed who she really is to the world. Another situation is that more and more users are connected with those whom they do not know before. This is understandable given the situation where many are putting up invented semi-anonymous profiles.

In order to understand the effect of this new phenomenon on the conception of the self and the person, a philosophical discussion of these concepts is in order. In any case, what seems to be uncontroversial is that the self is not the same as the body. I, obviously, am not my body, because my body does change—I might become thinner as a result of an exercise program—but that does not mean that I become another person. However, some philosophers do argue that bodily identity is constitutive of personal identity. Bernard Williams, for example, argues that personal identity cannot be understood apart from bodily identity and continuity (Williams 1973). For Williams I can indeed become thinner as a result of exercising, but that does not mean that I become another person because it is still the same body that becomes thinner. Hence any criterion of personal identity has to include criterion of bodily identity too. However, Williams does not say specifically what exactly are the criteria by means of which a body of a person at one time is identified with a body of the same person at another time, except only that they are the same body or that the bodies at different times endure as a single entity. However, since it is well known that most of the molecules that make up the human body are lost after a period of time and are replaced by new molecules, simply identifying the body with the molecules or the matter itself does not work. Williams may reply that it is not the molecules that make up identity of a body, but its spatiotemporal location occupied by the body—the same body has to endure through time even though its constituent parts do change. But since a human body does not exactly occupy the same volume of space through time its identity then has to be inferred from other factors such as certain likenesses and other properties of the body that endure for a time, not the occupied volume alone. But as these likenesses and properties are chosen to represent the identity of the person, it seems that even in using the body as a criterion personal identity has to be constructed out of some interaction between those who perceive the likenesses and the body itself. At any rate, the perception of the likenesses of bodily properties and judging that these properties are similar enough to constitute identity would seem to be something that is external to the body itself.

Furthermore, the self is not entirely constituted solely by any of my mental events or episodes, or any collection thereof. My mental episodes change very rapidly during the course of a day, but that does not mean that I become different persons each time my thinking changes. The problem of personal identity is precisely to account for the apparent existence of the self even though analysis shows that everything that constitutes it does change over time. But is there actually something that remains the same in one person amidst all these changes?

One might object to the presentation above, arguing that some set of mental episodes do constitute the identity of the person. Memory, for example, has been cited by Locke as a means by which the person can be identified (Locke 2008). However, it is also well known that relying memory results in a vicious circle (Cf. Butler 2008; Reid 2008), for in order to me to verify that an episode of my memory, say, of a young kid playing a toy car that I seem to remember, as belonging to me, I would need an account to tie up the young kid and myself at this moment together. There must be something which is external to the young kid and to myself as of now that is responsible for the 1-year-old playing a toy car and the 48-year-old college professor to be one and the same person. But this is what precisely the memory account of personal identity denies. There cannot be anything external. But if there is not, if it is the memory itself that is responsible, then the memory account clearly presupposes that continuance of the person, which is exactly what needs to be explicated through the memory account in the first place. The proponent of the memory theory might object that memory can certainly be fallible, but my memory largely is true, which shows that what I do remember is largely true, so the memory account is tenable after all. However, the proponent still needs to account for the assertion that the memory he is having is largely true. If he relies on his memory to do that then the vicious circle spins again. Moreover, in relying on memory or any other internal mental content, one also has to compare and contrast likenesses or differences among those episodes, which means that identity does not reside in the mental content alone.

Defending the memory account of personal identity, Shoemaker (1984) has argued for the conceptions of psychological connectedness and continuity. Basically speaking, psychological connectedness is what I have when the memory states that I am having right now are caused by earlier states of an earlier version presumably of myself. Psychological continuity, furthermore, is what I have when my current states are caused by the earlier states in such a way that the causal chain can be interrupted by periods of unconsciousness such as sleep. This undercuts the vicious circle argument because I don’t have to actively remember those states I had at an earlier time. However, it is questionable how elements within the causal chain constitute an identity of the same person. It seems to be tenable that my mental states at this time can affect the content of your mental states at a slightly later time, such as when I tell you something and you are then informed by it. This seems to be a standard case of cause and effect relation of mental states, but if this is tenable, then Shoemaker’s account of either psychological continuity and connectedness would not seem to be sufficient.

Now we are prepared to discuss identity in the online world. Essentially, the problem of online identity then is: What are the criteria by means of which disparate episodes of postings, comments, video links, etc. that together constitute an online person in a social networking website are unified such that they constitute the same (online) person? Note that the online person here is a persona that has been expressly constructed on such venue as social networking sites so as to represent an offline person or the latter’s projection of a semi-anonymous entity as we have discussed earlier. An analog of the memory account of personal identity in the online world seems to be that, in order for me to verify that my past episodes of posting, for example, what I posted online a year or two ago, really did belong to me and not somebody else, is that I do remember doing so. I can look up what I did post exactly 12 months ago (it seems to me now rather difficult to do that because I post so much material on Facebook), and the reason why I know that these postings belong to me, or to my online self, is because I remember doing so. However, the vicious circle argument against the memory account of the offline personal identity should be applicable in the online world too. After all, the structure of the argument is the same, the difference only being the characteristic of the person in question, whether he is online or offline.

Here the memory account, as we have seen, does not seem to do an effective job at providing such criteria. In order for me to remember that these postings on the social networking sites do belong to my profile or my online persona, I would need to be able to relate all these episodes together in a string of memory. But that would mean that the online person requires the existence of myself as an offline person whose memory accounts for the identity of the online person in question. And since the memory account has a problem of its own as mentioned before, namely the circularity problem, relying on the memory of an offline person for accounting for the identity of an online one does not seem to work; any problem that exists with the memory account of the identity of the offline person would still persist when the problem is shifted up to the online world.

It is clear, at any rate, that in the offline world I do have my body, and bodily continuity seems to count heavily for personal identity, especially for those who espouse the somatic or animalist approaches. Even though the popular view that every cell in the human body gets replaced every 7 years seems to be wrong (nerve cells in the brain, for example, never get replaced), there are so many unreplaced cells in the body that talks about the human body itself being more like an event rather than a static thing does not seem to be too far off the mark. The human body indeed looks like an event as there are dynamic activities going on there all the time, ranging from the blood circulating within the body, the firing of the nerve cells, to all the movements of the limb. That the body is usually regarded as a thing rather than an event is due perhaps more to habit and our overlooking of these dynamic activities than otherwise. But if the body is more like an event, then the criterion for its identity and continuity needs to be different from that of a static thing. For example, an event has to have a clear beginning and ending, which in the case of the body, of course, refers to the birth and death of the body and also of the person. Then we have familiar means by which we individuate the body, such as giving it a special status of a human person, giving it a name, a place in the society and community, and so on. These are the means by which the identity of the body and of the person is fixed. That the body is a process shows that it cannot be identified with the self, simply because the self is static (I normally appear to remain the same person, the same self) even though my body is always changing (For another argument in favor of the view that persons are constructs, see Brandon-Mitchell 2011).

Arguing that the perception and judgments of others, or other forms of external evidence such as birth certificates and so on constitute the identity of the person shows that personal identity is fixed rather by external factors than internal ones. This does not simply mean that these external factors are useful in identifying or in keeping tab of a particular person, but these factors are constitutive of the identity of the person himself. This is so because what it is to be a person himself is constituted by external factors—there are no internal properties of a person such that they succeed in identifying the entity they are properties of as a person. This line of argument is akin to the externalism/internalism debate in epistemology. In their attempts to locate the source of justification of belief, epistemologists have traditionally tried to look at the subject’s beliefs, i.e., what lies internal to the subject’s own cognitive field, as the source of justification. Thus we find Descartes locating the ultimate source of justification of his belief in the cogito statement through the fact, evident to himself, that it is clear and distinct to him that he thinks and he exists. However, recently many epistemologists have started to look at external sources for the justification. For example, Alvin Goldman has argued for a kind of social epistemology where the source of justification of belief is located outside of an individual and among the social interaction that the individual has with her social environment (Goldman 1975, 1986, 1999). Perhaps in the same way, personal identity has traditionally been associated with internalism—factors thought to be responsible for fixing the identity have come from internal sources such as the subject’s own beliefs and memory episodes. However, one could follow the lead of the social epistemologists and other externalists in epistemology and start to argue that external factors are really the ones that fix the identity. For example, instead of trying to find the source of the identity internally, one could broaden out and try to locate the source instead outside of the subject’s cognitive domain. A candidate could well be what others think of the subject in question, what their collective behaviors are like such that these behaviors taken together succeed in fixing the identity of the subject. Suppose I am not absolutely certain if the picture of a young 1-year-old that I am holding is that of myself, I can certainly ask my mother. My mother’s testimony (usually mothers are very good at recognizing her young child even though decades have passed) will then fix the identity of the boy in the old picture and my own self today. Other clues are also possible; perhaps the picture is associated with some notes or documents that could relate back to me. These notes and documents thus serve as the external factors too. In fact these are the standard methods used by societies to identify persons in real life, such as in solving identity disputes and in courtrooms. Here trust seems to be given more by society to external factors than the merely internal ones of the self report of the subject himself.

It might be objected that these external factors can succeed only in keeping tab of a particular person through time while leaving his or her core identity intact. Suppose there are no external factors at all, the identity of a person would still be intact because the identity does not seem to have anything essential to do with the external factors, or so the objection goes. However, it is difficult to maintain how identity of a person can be understood or conceived of at all in such a situation. In order for something to be identical with anything, even to itself, there has to be a context or an environment within which the identity in question makes sense. Even if one imagines that there is only one thing in the whole universe, that thing has still to be related to the universe in order for its own identity to be conceivable. In fact this is a necessary condition for there to be a thing at all. But if that is the case, then it does not appear to make much sense to maintain that identity of a person (or any entity for that matter) could be constituted by that person alone. According to Hegel, a necessary condition for there to be a thing at all is that it has edges defined by something else existing beyond the edge. Without any edge there would be no thing. In the same vein, a person is also defined through a boundary separating the person from the outside.

In the online world, things do not need to be radically different. We can regard the moment when someone registers their profile onto sites such as Facebook and becomes known to the circle of people who are already on Facebook as the moment when that person is “born.” In the same vein, the moment when someone removes their profile from Facebook, thereby ceasing to engage in any activities that are performed by Facebook users, then this can refer to the “death” of that person online. All the activities during these two boundary marks represents those performed by the subject when she is “alive,” so to speak. And since it is very difficult in the online world to locate where the subject, cognitive domain is, external factors seem to be the only ones available for fixing identities. In Facebook there are guidelines that one needs to follow in order to be “born” or to “die” there. For example, one has to follow certain rules in order to have one’s profile picture show up; one has to register oneself, answer a number of questions, invite friends, and so on. The “birth” of a new user of Facebook can be announced publicly throughout the Facebook world, or it can be a rather quiet birth where the subject comes to be scene quietly without much fanfare. In the same vein, Facebook also has a clear policy regarding the “death” of its user. Formerly it was very difficult, if not entirely possible, to delete someone’s profile from Facebook, but after much protest Facebook then allowed someone to delete their profile rather completely. Furthermore, it also enables users to “memorialize” a deceased user. An account that has been memorialized will remain, and the user’s close friends can have access to the wall of the account to post their remembrances. Thus, in effect the wall of the deceased and memorialized user becomes a grave where close friends can drop by and pay their respect (see http://www.facebook.com/help/?faq=13016 and http://www.facebook.com/help/?faq=13941). Here, then, the identity of the person on Facebook is constituted through the information that is posed by the person herself as well as what others post about her. These are the activities that take place after the moment when the user is “born” and before she “dies” or removes herself completely from the site. Furthermore, even if she really dies in real life, her posts and comments can still be available, in the same way as the thoughts and ideas of dead persons can be available to us. The postings and comments of the dead person will remain there and there will be no new additions, in the same way as a dead person cannot write a new book.

How the Self and the Person Get Unified

A key problem in analyzing the self and the person has been the problem of searching for a unifier that combines all the physical and mental episodes together to make up a real, substantive self. That I am a person who have a self is obvious, but it is not obvious how my physical and mental episodes are combined to make up my own self. This is known as the “Binding Problem” (see, for example, Searle 1990). Basically the problem is about how the brain combines various kinds of input so that they result in a single, unified field of vision or an episode of self-conscious thought. This act of combining different types of input by the brain is analogous to the more philosophical problem of how I gather different mental episodes to make up parts of my own self mentioned earlier. The difference may be only on which perspectives the act of binding is perceived. If the activities of the brain are discussed, then it is from the third-person perspective, but it is certainly from the first-person perspective that the latter version of the binding problem is looked at.

However, before we embark on the problem of combining various online episodes in order to make up an online self, we need to discuss first the problem of how to combine the various episodes so that they belong to one overarching self is well known. Kant posited the “Transcendental Unity of Apperception” as a means by which these episodes are combined so that they belong to one and the same subject, which would make cognition (or in his words “judgment” and “understanding”) possible (Kant 1997). However, a problem with the Transcendental Unity of Apperception (TUA) is that it is a purely formal concept, and does not contain any particular information that pertains to any particular individual. According to Kant, “it must be possible for the ‘I think’ to accompany all my representations” (Kant 1997: B131–132), meaning that it must be possible for me to be conscious of all my mental episodes; otherwise it would not be possible for me to be justified in asserting that these episodes are mine. But what is very interesting here is that Kant is not arguing here that there must be an objective self, the “I” which “thinks,” for to posit that would be to commit oneself to a metaphysical argument which lies beyond the scope delimited by the critique of pure reason. Kant is instead putting forward a transcendental argument here. A transcendental argument is one that accounts for a condition of possibility of a certain phenomenon; its point is that, for the phenomenon to be an objective one at all, or for it to be even possible, certain conditions must already obtain. The transcendental argument does not show tout court that the phenomenon exists objectively; that would run against the spirit of the critical philosophy. According to Kant, the transcendental concepts are necessary condition of their being objectivity in the world at all. That there is empirical objectivity is taken by Kant to depend on the objectivity of empirical knowledge which yields us knowledge of the empirical world. However, that knowledge depends, according to Kant’s Copernican Revolution, on the set of pure concepts of understanding which are found to operate a priori and function as the condition of possibility of the empirical knowledge in question. For example, in arguing about causation in the Second Analogy (Kant 1997: A189–211; B232–256), Kant’s point against Hume’s devastating attack on causation is that there are conditions of possibility of causation, such as if an event A were to be the cause of another event B, certain conditions need to obtain, such as that both A and B need to be able to be subsumed under the pure concept of understanding of logical relation. Kant does not rebut Hume directly; he does not argue that Hume’s conclusion is directly false. That strategy is closed to the critical philosophy. Kant cannot just argue that there is objective causation in itself because that would be to argue for some properties on the side of the thing in itself, which is not possible in his system. What he does instead is that he argues, in the Second Analogy (Kant 1997: A189–211; B232–256), that, if we are to be able to maintain objective knowledge, we need to posit the concept of cause and effect. Hence the concept does apply objectively to phenomena. In the same vein, Kant argues that it must be possible for the ‘I think’ to accompany all my representations. The move here is that objective knowledge requires that there be a consistent self who somehow combines all the different episodes of sensory input together to make a coherent whole. Without such a self then objective knowledge would not be possible in the first place. He does not argue categorically that the “I” has to exist; what he says is that the “I” needs to exist as a condition of possibility for relating disparate representations into a coherent whole, which in turn is necessary for there to be objective knowledge. What the “I” is doing here is nothing more than a place holder, a formal factor that serves to unify various representations together so as they belong to a coherent self.

This is of course not a place to examine Kant’s philosophy in any detail. But if the “I” of the Transcendental Deduction here functions as a purely formal unifier, then this “I” would be devoid of all and any characteristic that would qualify it to be the “I” of any particular person whatsoever. All it can do is to perform this purely formal function, which must be the same for everybody. In short, the “I” here functions as the Transcendental Unity of Apperception (TUA). Thus, my TUA is exactly the same as your TUA, since both function in the same way and cannot contain anything unique to either me or you. In fact this is to ensure objectivity of knowledge in Kant’s system. If my TUA and your TUA are different, that would mean that the uniting functions at work in my cognition and in yours are different. A consequent of that would be that my and your TUA are different concepts, which would entail that objectivity is lost. Anything unique would be empirical and cannot be part of the TUA. If this is the case, then Kant’s TUA is too general and cannot perform the work expected of the individual self. If the TUA at work in my cognition is one and the same as yours, then there is no point in maintaining that the TUA is a foundation for a conception of a self or of personal identity since in that case you and I would be one and same person because we both are founded upon the TUA which is necessarily the same in both of us. In other words, one cannot rely on Kant’s transcendental argument about the “I think” here and use it to argue categorically that the self does exist as an objectively existing being.

It may be objected, however, that the Transcendental Unity of Apperception might succeed in identifying a person after all, since it specifies a range of possible mental episodes such that some episodes may belong to me and some may not, thus the TUA delimits what it is in fact to be me and not others. According to this point of view, the structural relations specified by the TUA are responsible for me being a person at all, and hence, according to this view, the TUA then specifies a range of states and events that do and do not constitute me. This then implies that the TUA succeeds in specify a unique person after all. However, the TUA cannot achieve this, since in order for it to be able to identify a particular person, A, and not another one, B, it has to contain empirical conditions which alone are necessary in distinguishing A from B, but the TUA, being transcendental, cannot contain any empirical input whatsoever. What the TUA can achieve is only for A to be certain that he is a unified person such that he can combine his various mental episodes under the rubric of “I think” because if he could not have done so he would not have had the objective empirical knowledge that he is certain of having. What goes for A certainly goes for B, as B can be certain of TUA at work so that he can combine his own episodes to constitute his own “I think” which combines his various representations too. The TUA, then, cannot choose which representations do belong to A and not to A, because that would presuppose the unity of A from the beginning, while in fact the only work that the TUA does is to function as the condition of possibility for A to have objective, empirical knowledge that he is certainly of having.

Hence, any attempt at finding the overall unifier of the mental episodes for a particular person falls under the empirical side of things (because once a candidate for the unifier is identified, it then falls under the category of a mental episode which is being thought of, which then requires another subject to think about it, and so on), or under the purely formal schema such as Kant’s, which is empty. An upshot, then, is that any attempt to bind up the episodes is always provisional and cannot escape from being itself yet another mental episode. When one attempts to bind up one’s own episodes, one is then conscious of yet another episode whose content is about the binding, but then that becomes another mental episode in need of further binding. Consequently, the offline self is a construct in the sense that it is not there objectively or ontologically. It is something “made up” in order to facilitate daily living of any human being. For example, it would be much easier for me to refer to you, using your proper name, if you stay relatively stable throughout some period of time, even though analysis shows that there is ultimately speaking no real “you” in the ontological sense. What I and others take to be “you” is a social construct not too dissimilar from Searle’s example of a bank note whose value is also a social construct (Searle 1997). In other words, the value of the bank note does not reside ontologically in the material itself, but sociologically through agreement among members of society that this particular type of a bank note has such and such monetary value. In the same vein, when I refer to you, calling you by name for example, I am abiding by certain social conventions that recognize that, relatively speaking, there is a certain person behind the persona that I am now perceiving.

But if this is the case, then it is also similar for the online self. We have seen that the online self as a kind of persona that an individual makes up as a front to present himself or herself to the world, and sometimes the individual may intend itthat the persona assume an identity of its own, without being able to refer back to the real person behind, as we have seen earlier from the Thai examples of online persona in social networking sites. This online self, then, is nothing more than a representation that the offline self creates for a variety of purposes. One is reminded here of those who construct various versions of their selves online, such as one for the family, another for the workplace, yet another for his highschool buddies, and so on. The online self, then, is not a collection of memories and thoughts of a particular person which, according to Singularitarians such as Ray Kurzweil, can be uploaded onto a giant server and represent the essence of an individual person (Hodgkinson 2009). Nonetheless, the online self is also made up of physical and mental episodes. The physical episodes are easy enough to understand–bits of electron working together to present images, sounds, and texts on screen. But the mental episodes are also there, as we can gauge what the persona (or in other words the online self) is thinking or feeling through her use of language and other symbols (such as emoticons) through the Internet. These episodes also need to be connected together in order for us to form a more or less coherent picture of a self working behind. Here one also finds an analog of the Transcendental Unity of Apperception in the online world too. Just as in the offline world, the analog of the TUA in the online world functions to bind the different episodes of postings and comments together so that they belong to one person. It thus functions more as a regulative agent working as a condition of possibility of there being a coherent self behind the various representations constituted through images and texts that are posted online. So the analog is something like this: “It must be possible for the ‘I think’ to accompany all my postings of links, images, videos, comments, etc. on the social networking site; otherwise no coherent self does not emerge which is necessary for there to be social networking at all. In short, the Transcendental Unity of Apperception is a condition of unity of a person, but not of identity.

In other words, the TUA does not seem to succeed when it comes to securing uniqueness of a particular self or of a particular person. What the TUA does is that it gives me, for example, a means by which I can be certain that I am a coherent self and that all the representations that are flitting across my brain or my cognitive field are indeed mine. However, what the TUA does not do is to identify me as someone who is distinctly different from another person. Since the TUA is a purely formal apparatus, it cannot do this job, because what makes me a unique person, such as someone who is teaching philosophy and who is interested in many subjects and so on, cannot be contained in the TUA function that I have. Identity is not the same as uniqueness. There can be several things each of which are of course identical to itself, but without uniqueness all these things are just a bunch of entities sharing all properties in common but having no unique identity of its own (i.e., the characteristic of being itself alone and none other). This is not something that can be accomplished by the TUA. Hence, in order to account for my uniqueness, external factors need to be considered too. For example, in order for me to be certain that I am unique, I usually refer to the set of characteristics that only I have and are shared by no one else. Since the TUA functions in exactly the same way in all the selves and all the persons, it cannot specify uniqueness.

Conclusion

We have seen from the discussion above that the most intimate thing that we can have, our own persons and our own selves, are being affected significantly by the technologies. Many people are constructing their own alternate personas online; even in social networking media, which are assumed to be a place where one reveals oneself to others, are also being used in such a way as to present entirely new personae to the public. These personae do share deep seated metaphysical affinities with the real-life, offline individuals, and the strategies used by those in the offline world to construct their identities are also used in the online world. I have argued that personal identity is constituted more by external factors such as social perception and various sorts of documentation and physical traits than by the internal ones such as memory and the subjective feeling of being oneself through time. However, this does not seem to carry over for the identity of the self, since this is more a matter of being referent of the first-person pronoun, which points deeply to the sense of being the subject of the various thoughts and feelings. Kant’s view on the Transcendental Unity of Apperception might at first sight be able to explain how the identity of the self is fixed, but as we have seen Kant’s view succeeds only in fixing identity, but not uniqueness. It seems that external factors are still required for the latter.

In the online world, things are again similar. We can find an analog Kant’s TUA in the online world. There must be something functioning as the ‘I think’ that binds up all of the various texts and images posted online as belonging to one and the same self. This binding, again, does not have much to do at all with the content of what is posted. For that we need the external factors to construe their meanings and how they are received and perceived by the community of other online users, who all together form the social network. But if all this is tenable, then the two worlds—online and offline—seem to collapse into one, and we cannot really tell this collapsed world to be either strictly one or the other.