Here’s another post that started out as some idle Twitter comments. Kind of a long one. I hope it makes sense.
It’s a pretty well-known trope in science fiction: Uploading your mind into a computer, so that you can live forever as a machine. Discarding our flesh bodies and becoming something intangible, immortal, and inhuman. It may even be possible some day, as computing power and our understanding of the brain increase.
So perhaps we can hope that one day we’ll be able to cheat death, to copy ourselves over onto the net and never have to experience death.
Well, I don’t think it’s that simple. I’d say it depends on what, exactly, you mean by uploading.
Copy and Paste
Let’s take a look at the crudest form of the idea: We build a computer with memory and processing power capable of matching a human mind, and we find a way to hook your brain up to it and copy all of the information inside. Now there’s a new you, inside the computer, with all of your memories and personality, and this new you will live on.
But it’s only a copy. The old you remains.
The new you thinks and feels that it is you, that it once was flesh-and-blood, but in reality, it is a creation that only mimics you. Your mind is still in your body, and you will still age and die.
In this scenario, your mind didn’t transfer out of your body into a machine; a new you was created within the machine, something that has all the continuity of self as if it is you, but the old you does not take part in that continuity – it has its own self, still inside your brain.
Aside: Don’t Beam Me Up
This kind of uploading, which preserves a copy of a human mind but does not save the original, parallels another classic sci-fi problem: the teleporter. When the teleporter is used, your body is broken down entirely, a signal representing the whole of your self is transmitted, and at the receiving end a new body in reconstructed in a perfect copy of the original. The new you has a continuity of self that tells it that it is you, all the memories required to be you, but again, it was created at the moment it was assembled by the teleporter. The old you was destroyed by the same device. Think of it this way: At one end, your consciousness ceased to exist, and are no longer able to perceive anything. At the other, a new consciousness was created that is unable to perceive that it did not exist before.
A successful teleportation subject appears to all observers – including the subject themselves – to have survived the journey intact, when in fact the original subject no longer exists.
In the larger scheme of things, this doesn’t seem to really matter. If to any and all observers no one died and teleportation works, then it works. But for the person stepping into that teleportation machine? Everything. Just. Stops. They do not get to wake up at the other end. This is a frightening thought.
If teleportation is ever invented, I will not use it.
So what can we do to actually move our brains onto a computer without snuffing ourselves out, or finding ourselves still left behind in our bodies after the upload is done? Let’s look at one particular example of a brain transfer.
In John Scalzi’s Old Man’s War, protagonist John Perry is a 75 year old man who has signed up to join the Colonial Defense Force. As part of the deal, Perry is provided with a brand new, modified body, to replace his old, dying one. To accomplish this, recruits like him are fitted with a sensor array that monitors their brain activity, and can “broadcast [their] consciousness”. According to Dr. Russell in the book, “consciousness can’t be recorded. It has to be live if it’s going to make the transfer.”
The specifics of how the transfer works don’t matter so much for the point I am making as the essence of it: The sensor array transmits Perry’s consciousness to a computer bank. Another connection is made with the destination brain, which was prepared using the brain mapping from the array over the previous days. And reading from Perry’s perspective, we see his consciousness synchronise with the new brain through the medium of the computer – including a few moments where he feels himself occupying both bodies at the same time, looking at himself. Once the new brain is fully active with Perry’s consciousness, the connection is severed.
So in this case, an original human, John Perry, is fully awake and conscious throughout a process that brings his active mind into another body. For this new John Perry, there is no illusion of existing beforehand: He did. The consciousness inside the new body is not a recreation but a continuation of the original self.
So, problem solved? Not so much.
The thing is, this scene of the book plays a little trick on us with perspective. We begin the scene in the mind of original John Perry, go into the synchronisation between both minds, and then end in the new Perry. But when the connection was severed from the computer bank, it didn’t leave John Perry on one side and a blank brain on the other: it left two John Perrys. Yes, this was another case of creating a copy – just a more clever one, eliminating discontinuity on one side.
For the original, when the transfer ends, the doctor “shut[s] down [its] brain”. It seems pretty clear: John Perry was still in there. They didn’t empty out his brain, just copied the signals it was making. He still had to experience death, while another him lived on.
So is there no way around the problem? I think so, but you have to be a bit more sly with it…
Enter the Cyborg
We’ve come this far: We know that by synchronising the original brain in real time with the destination – whether organic or computer – we can create a full continuity of experience that leaves no doubt that the uploaded mind is valid as more than just an imitation. But we’re still leaving the old mind to die.
The answer, I would say, is to go slow. Transferring the mind out of the brain into another vessel in one big dump is always going to be a case of making a copy. But what if we were to gradually replace the working brain’s functions with the external versions?
In the near future in Charles Stross’ Accelerando, people have wearable technology that links them directly to the net. This is a book about change over time, and you see the progression: from AI agents used to access data, to increasing use of the computer accessories to store memories, to offload mental processing, to increase the potential of the human mind by augmenting it with this technology. In one of the stories, Manfred Macx has his glasses stolen – leaving him disoriented and confused – and the thief, wearing the glasses, finds Macx’s gear altering his behaviour and personality, programs used to helping Macx think now pushing this new wearer to behave like he did.
It’s through this kind of partial outsourcing of processes that I think we could get closest to actually transferring out of the body and avoiding death. New memory banks to store things the brain forgets, for example. Brain-computer interfaces that allow the mind to increasingly rely on external processing for thought processes. A dying brain, if the body is kept going long enough, could have its mind kept in full function by replacing each damaged part with the external processes, just as over time dying cells in a living body are replaced by fresh ones.
It’ll never truly avoid killing off the original brain. The mind cannot be extracted from the flesh, because they are one and the same. But by working with the living mind, finding ways to use machines to augment and replace organic functions, it could be done so gradually and piecemeal that it is never possible to identify a distinct original and copy. The fully transferred mind would be a continuation of the original, and no original would be left behind.
Now we just need people to invent this technology. We’re already living through the years Stross depicted in the earliest stories of Accelerando, and we’re not quite there yet.
We can hope, though, right?