Imagine the following case:
A few years in the future, neuroscience has advanced considerably to the point where it is able to artificially support conscious activity that is just like the conscious activity in a human brain. After diagnosis of an untreatable illness, a patient, C, has transferred (uploaded) his consciousness to the artificial substrate via a novel surgical procedure. His body has been buried.
There are different ethical and philosophical questions that we might ask about this technology. For example, is uploading of consciousness a good idea? What would be the moral status of such an entity? In this blog I will ask a different question. In the above case, has C survived?
The idea of uploading consciousness has been around for some time. Although the technology remains some way off, there have been a couple of different ways that it has been suggested that this might be possible. This includes
- Post-mortem scanning (copy and upload)
In the case above, after C had died, his brain would be removed and cut into extremely thin (micrometer) slices. The neurons would be imaged using electron microscopy and the resulting detailed cellular map of the brain uploaded and an electronic simulacrum of C’s brain created including a replica of all of C’s neural connections.
- Gradual replacement (copy and delete)
In this method, C’s neurones would be individually and successively replaced by electronic versions of his brain cells. These would progressively replace more and more of his brain, until he had a fully synthetic cerebrum. (Like the famous ‘Ship of Theseus’, there would be no time point at which the original consciousness ceased to exist).
There are technical as well as philosophical challenges with each of these. But I want to focus on a novel proposal.
Recently, Japanese neuroscientist Masataka Watanabe has proposed a different theoretical approach
- Split brain approach (split and merge)
C would undergo a neurosurgical procedure that is a form of “corpus callosotomy” – an existing operation that divides the two halves of the brain. (This operation is performed for some cases of severe epilepsy). In the modified procedure, an artificial electrode array would be inserted across the intersection points between the two halves of the brain (the corpus callosum, and anterior and posterior commissures). This would connect each half of the brain with a separate artificial (but empty) neural construct. C’s consciousness would merge between artificial and biological hemispheres. Subsequently, the two artificial hemispheres could be rejoined to form a united digital version of C’s mind.
The idea behind Watanabe’s approach comes from the observation of what happens when patients have a corpus callosotomy. Since the 1960s, it has been clear that patients can come through such an operation without major complications and without changes in their memory or personality. However, in more extensive forms of the surgery, patients manifest a split brain syndrome, where consciousness in the two halves of the brain appears to function independently. (For example, the left half of the brain is not able to recognise or answer questions about visual input to the right half of the brain (and vice versa)). It appears that the corpus callosum in normal circumstances serves to merge the consciousness of the two halves of the brain. Watanabe’s idea is that an artificial corpus callosum would allow consciousness to merge between biological and artificial hemispheres. If that is possible, it should then (in theory) be possible for the separated artificial hemispheres to function independently in a way that is akin to the original biological ones and perhaps to merge again via the artificial connection. His proposal seems to offer some advantages over other proposals in that it allows continuity of consciousness (post-mortem scanning doesn’t offer this), in a way that might be plausibly achieved (gradual replacement of individual neurones seems a remote prospect).
Setting the technical challenges aside, if split-brain uploading were possible, would C survive the procedure?
The first stage of the procedure is the most straightforward. Assuming that we think that patients currently survive a callosotomy procedure, it appears that C would also survive the splitting procedure. Or to be more precise, C’s consciousness would continue, but in a divided (hydrid) form – CL and CR.
But then what would happen? If the artificial hemispheres do merge with C’s biological brain, there would then be two dual-hemisphere consciousnesses (CL-cL, and CR-cR). And after the artificial hemispheres are reconnected, there would be three separate conscious entities (CL, CR, and cL=cR). (If the two original hemispheres were also reconnected, there would be two: CL=CR and cL=cR).
Whether we think that C survives this procedure depends crucially on who or what we think ‘C’ is. One common answer to this question is to say that we are human beings or human organisms.
If that is our answer, then clearly C does not survive. (Or more precisely, he survives the initial procedure, but then succumbs to his illness). The artificial consciousness cL=cR might have similar thoughts and responses to C. But it is not a human organism and it is not ‘C’.
Another answer to the question would be to say that C is a human person or human mind. We are the minds inhabiting our human bodies. That response makes sense of a number of our intuitions. For example it explains how and why we might think that someone has died whose consciousness has irreversibly ended (though their heart is still beating). It likely explains C’s motivation to upload his consciousness, when diagnosed with a terminal illness. After all, he hopes that his mind will survive via the procedure.
However, there is a problem. After the brain splitting procedure there are two and then three conscious entities or minds. Those entities might be qualitatively similar to C’s original consciousness. But they aren’t numerically identical to C. (If they were numerically identical to C, they would have to be numerically identical to each other. But if that were the case, it would not be possible for one of these entities to die and the others to continue).
There is a similar phenomenon that occurs in very early embryonic life. In about 1 in 200 early embryos, within the first couple of days after conception, the initial cluster of cells splits into two forming identical (mono-zygotic) twins. The two emergent identities are qualitatively ‘identical’. But they are not numerically identical with each other. Nor are they numerically identical with the early embryo from which they developed. Does the embryo survive twinning? Maybe. Yes, and also no. We might say that the early embryo has been replaced by two separate slightly-later embryos.
The proposed brain uploading procedure appears relevantly similar to the division that occurs in forming monozygotic twins. Indeed that is why the title of this blog refers to Watanabe’s procedure as brain-twinning. But if we recognise that similarity, then we might also acknowledge that the answer of whether or not C survives the procedure has a similarly ambiguous answer to the outcome of the early embryo. Maybe. Yes, and also no.
Perhaps we are asking the wrong question. Rather than asking whether C ‘survives’, the more useful question is whether, from C’s perspective, what matters persists. If we ask that question, then there is a more definitive answer. At least on some philosophical views, assuming that the procedure allows continuity of C’s consciousness, and that the subsequent conscious entities share C’s memories, preferences, ways of thinking etc, then what matters will be preserved. There may be several new entities that all are related to the former C, but none are exactly the same as each other, or as the previous person.
Brain uploading is some way off yet. But thinking about what it would mean to upload our consciousness might help us in the meantime think about some other life and death questions. As suggested above, that is because the question of whether or when someone is alive or dead is not actually what ultimately matters.