On Philosophy

July 19, 2007

Practical Immortality

Filed under: General Philosophy — Peter @ 12:00 am

Today, right now, there is a way to (almost) guarantee that you survive forever, with no magical technology or leaps of faith required. But before I explain how this is possible let me first explain the philosophical understanding of survival and consciousness that serves as a background for this claim (of course more detail on these topics can also be found elsewhere on this blog). While these details are admittedly boring they are necessary to understand why what I propose here might reasonably be thought to give us at least a shot at immortality.

First let’s consider survival. It is easy to think of survival as somehow dependant on the existence of an unbroken process of consciousness. But this clearly can’t be the case; it is reasonable to suppose that our consciousness is frequently stopped, only to restart again at a later time (a dreamless sleep probably has this effect, for example). And similarly it is implausible to suppose that survival depends somehow on the same physical stuff being present over time. What survival does seem to depend on is the existence of a “successor” consciousness for our current consciousness. And this successor is determined by sharing memories with the consciousness it succeeds, as well as personality, patterns of thought and so on (in other words the features that individuate our consciousness from that of other people). And this in turn means that survival comes in degrees. The consciousness that I am experiencing now (that constitutes me, as I should properly say, so as not to inadvertently imply that there is an “I” separate from my consciousness) is a very close successor of that consciousness which existed a few minutes ago, after all its experiences are almost completely intact in memory, and I have changed little between then and now. But the consciousness that exists tomorrow will be less of a successor, I will have changed slightly since then, and I probably have forgotten all about that particular experience. And obviously the consciousness that exists next year will be even less of a successor, but still some kind of successor, and thus I, the person that exists now, will still survive to some extent, an extent that I am normally happy with.

Now, briefly, let me say a bit about consciousness. Consciousness, it is reasonable to suppose, is pretty much the same from person to person. The content of consciousness, and how that content changes over time, of course varies greatly from person to person, that is what makes us different people. But consciousness itself, the nature of experience, is probably pretty similar. While you and I associate different things with the color red it probably feels about the same to the both of us. And, more importantly, as long as the overall structure of consciousness is preserved it doesn’t seem like the exact details of what it is like to have a particular experience matter much for survival. Were I to be fitted with color inverting lenses tomorrow, and all my memories also inverted in color I might be somewhat of a different person, but not different enough to make my color inverted consciousness not a successor to my pre-inversion consciousness.

Together these observations imply that we have the ability now to collect enough data to be able, at some future time when consciousness is well understood, to plug that data into a generic consciousness and thus generate a successor consciousness to our consciousness now. Not a perfect successor of course, but a successor good enough to count as (and feel like) survival. This is of course only possible because survival doesn’t depend on the perfect capture of every memory and facet of personality; our personality is continually in some state of flux anyways, and we forget much of what we experience, with our mind usually filling in the gaps by fabricating some or all of our memories. Of course we can’t gather this data by some kind of magical brain-scan, that is still in the realm of science fiction. But we can record our thoughts as text, with the occasional image and audio as supplements. In the simplest sense it is kind of like a super-diary, recording as much as possible, in little bits throughout the day. But the focus of this recording effort is different than the diary. The diary is used to make memories, and it is used selectively to record events, without focusing much on the experience of them. In contrast this effort would be to record not just the important experiences of the day, but to record the thoughts and associations that went with them, in order to capture not just what we thought, but how we thought it. And it is especially important to record when we remember things after the fact, because that reflects what made an impression on our personality. Of course better than a flat text format would be some kind of “smart” software designed especially for recording this information. Specifically this software should do its best to predict how we think, given a few starting ideas and the associations between ideas that we have recorded before. This is not to try to make the device conscious (consciousness requires more than that), but to be sure that it is capturing our personality as accurately as possible. And of course constant audio-video capture would be a huge bonus, but carrying around such a rig is just out of the realm of possibility for the moment (and would require a huge amount of storage), and would still need to be supplemented by accounts of our thought process and what we remembered. I won’t try to go into what other data should be captured at the moment, but I’ll note that some minimal biological data (such as a DNA sample and an fMRI record) would probably also be useful (but not vital).

But such a project cannot be a private effort. A community is needed, not just to make hardware suggestions, design software, and create standards for the data files. Locations to retain the data that had been saved in this way for those who have died is required (preferably with redundancy and geographical separation) and that requires people actively interested in maintaining it; checking all that data on occasion for corruption and restoring what has been damaged from the back-ups. Now that wouldn’t be too hard, a few thousand plus the original hardware left to the maintainers in the will should do it, but it does require a central effort dedicated to preserving this data to continue over possibly many generations, and that requires a large enough number of people who are interested in the project so that we can be sure that someone will always be around to do the maintaining (at least until a successor consciousness can be created using the data).

Of course this idea isn’t perfect, it would be nice to have a full copy of our current brain state made frequently, so as not to lose any data. And it would be nice not to have to manually enter all that data. But it is the best shot I can think of for immortality (and not a long shot either, it should work, of course not as well as normal survival, but hey, survival in some sense is better than no survival). We don’t have the power to copy all the relevant information about the brain, and we may never have that ability (certainly not within our lifetimes and within easy reach). And freezing people doesn’t seem like a viable option either. There is no guarantee that you will be able to get your body frozen soon enough after death, nor be able to revive it in the future. And of course there is the practical benefit of being able to use this local store as a kind of addition to your personal memory, although of course for security reasons you could never make it internet accessible without heavy encryption (because to be the foundation for a good successor consciousness it must contain all the important information, which definitely means your most closely guarded secrets). Finally, research into understanding consciousness and intelligence is well under way, and while I doubt that we will be able to “revive” people in this way within our lifetimes the science necessary to do it is well underway (because of its other uses, namely AI), and we can reasonably expect to come into existence within the next few hundred years (assuming the human race survives that long).

All I can say then is that the method I have proposed here should work, in theory, assuming I am not radically mistaken about consciousness and the nature of survival (which seems unlikely). More importantly it is about the only thing that we can do right now to try to cheat death that has a reasonable chance of working. Given that, the possible benefits seem to outweigh the costs. Feel free to steal the idea, and feel free to chip in, since on my own this project is pretty much going to amount to a giant backed up plain text file, which is probably going to generate a poor future version of me, if anyone finds it to put in a consciousness at all.

Blog at WordPress.com.