Swampman

For once in a while I have been moved to try a bit of philosophy, prompted by either Bing or Google turning up reference 1 for some now-forgotten reason. I might say that I am prejudiced against philosophy, certainly modern philosophy, which mostly comes across to me as more or less inaccessible and certainly unhelpful. How do these chaps get past the academic funding committees, which these days mostly have their eyes firmly on the main chance?

However, never let it be said that my mind is closed, as I am finding reference 1 rather interesting, even if I do not agree with a lot of it.

One of the points of interest is a chap called Swampman, not the chap snapped above, rather an invention of one Donald Davidson, a perfectly respectable US philosopher from the second half of the last century. An invention located in the paper at reference 2. Third page, or page 531 for the curious. My copy of the paper being chapter 33 of something rather larger.

More properly a thought experiment, a what-if story. Now there are those who think that pondering about something which is not and which cannot be is a waste of time. Nothing can come from nothing. A point of view with which I disagree and today I cite in evidence the use of the probably false hypotheses used in ‘reductio ad absurdum’ proofs in mathematics, whereby starting from a bit of nonsense you are able to prove a useful theorem. And one could go on. Indeed, I have posted about this sort of thing before in connection with questions like ‘how would things have turned out had Napoleon won the battle of Waterloo’. See, for example, reference 4.

But back with this story, Davidson is standing beside a dead tree in a swamp when lightning strikes. He is rendered down into his constituent molecules, while by some freak of probability the molecules of the dead tree are rearranged into a replica of him. The replica looks and behaves exactly like the real thing. It exits the swamp and gets on with Davidson’s life. But it is a zombie, a robot. You might replicate the body, but you can’t replicate the soul. At least this seems to be Davidson’s point – which makes him a closet dualist who believes in the separation of body and soul, this term being almost a term of abuse to the author of reference 1. 

As a monist, I don’t go for this at all. If you replicate the body, the soul gets thrown in as well, whether or not the soul is an epiphenomenon – a phenomenon with no effect on anything else, a phenomenon without point or function – or not. 

On the other hand, I am also a cynic and sometimes suspect that consciousness – seemingly one of the hallmarks of being a human – might indeed be an epiphenomenon. That if one were to build a robot to mimic a human, one might well end up with something that was indeed a zombie. Replicating the appearance and behaviour of a human does not entail consciousness.

Davidson dodges this particular issue by going for a replica rather than a robot. But I do see some other problems, difficulties and points of interest. These follow, in no particular order.

There is a lot to be done

One might feed a three-dimensional printer suitable information about the external geometry of a human and get it to print out a life like model. A model which would, for present purposes, amount to much the same thing as a marble version knocked out by a sculptor. And such printing would take a while.

But if one is aiming to replicate the inside as well as the outside, at an atomic level, rather than at a visual level, there is clearly a great deal more to be done. A lot of orders of magnitude more to be done. With the elapsed time correspondingly longer.

So maybe, for the present at least, we might just as well rely on some freak of probability. Or on the deity.

Boundaries

The boundary between a human individual and the rest of the world is a rather grey area. It is not clear exactly what it is that we are going to replicate or copy.

What about the contents of all those body cavities which are, strictly speaking, outside the body? Thinking here particularly of the lungs and the alimentary tract. Then what about all that stuff which is either on the way out – for example dead cells on the outer surface of the skin – or on the way in – for example toxins or micro-organisms working their way through an outer surface of the alimentary tract to the interior, perhaps to the nervous system.

It probably would not do to exclude the whole lot from the copy, so some compromise needs to be effected.

Then what about all the sensory input, for example all the activation of the cells of the retinas. Do we copy that too? And being pedantic, I note that a copy of Davidson which is near but not coincident with him will get similar but not identical input to eyes and ears. Will get similar but not identical input through the feet, as no two patches of swamp are going to be identical underfoot. A problem which will be worse if both are barefoot. Will get similar but not identical sensations in the skin arising from ambient air and breezes.

But maybe near enough for any conscious hiatus to be put down to the lightning.

In any event, any clothing had better be included in the replication.

Time

Probably best to make the copy when Davidson is in a dreamless sleep. No conscious activity at all. His brain will still be active, but it will be less active than it might otherwise be. Any errors at the margin are going to be less salient.

If we allow a timeout of a few seconds to remove the original Davidson and replace him with the replica, the replica will then wake up in due course and is unlikely to notice that anything has happened. Any slight discontinuity in brain or body activity will be lost in sleep.

A waking Davidson is slightly more tricky, but when the replica comes online he will likely put down anything odd, any slight hiatus, down to the lightning bolt. Perhaps something like coming round after being knocked on the head. One does not expect to be in exactly the same position, just in the same place.

Physicists

I dare say physicists could argue that replication of atoms and their dynamic properties is just not possible, is just too far-fetched. Maybe they would invoke the impossibility of replicating just the right sort of quantum uncertainty. Maybe they would point to the time that it would take to make the copy. Or even to some principle which said that in order to know enough about that which was to be copied, you would have to disturb or damage it; that the business of measurement would have to be intrusive.

But I believe that the replication could be near enough – and that we might allow the thought experiment a bit of latitude as to the time needed to make the copy.

None of which touches the reproduction of the genome – which might for these purposes be regarded as fixed – and the gross structure of both body and brain. The reproduction of the structure of all the neurons in the brain, the axons, the dendrites, the synapses and all the associated machinery. But maybe it would touch the detail of all electrical and chemical activity in and around all those neurons.

So, given that I am among those that believe that the business of the brain will be explained at the level of the electrical impulses generated by the billions of neurons, at frequencies ranging up to a few hundred Hertz (Hz), maybe that business would be undisturbed by copy. While for those who believe that it will be necessary to move down to a lower, quantum flavoured level to get that explanation, maybe copy does not look so good.

Copy quality

In the real world, in copies on this scale there are going to be copying errors. Sometimes they will be serious.

Fast copy processes are usually unintelligent, have no regard for what it is that they are copying, with the result that all kinds of bizarre errors can occur when using the copy that results. 

Some of these errors might be trapped by devices like bit counts, checksums and hashcodes. Or, at some expense in time, one might check that the copy was identical to the original.

Other copy processes might take account of structure, might, for example, insist that the copy respects cells and require most cells – most human cells have exactly one nucleus – to have a nucleus containing valid DNA.

Markov property

A human being is a physical system which we suppose to have something like the Markov property. Which is to say that its next state is a function of its state now, it does not depend on its state in the past. So if we replicate a human being at some point in time, say T, the replica will just roll forward from T as if it were the original. With all the same memories as the original.

So we are not saying that the present state might not code up the past in some way. Our system has memory. So some past trauma, physical or mental, is likely to figure in the state now. Knowledge of such past trauma might well help us in making predictions, in helping a patient, but that is a different matter.

Animals with brains do quite a lot of this, store quite a lot of history in one form or another. So past history is very much part of their present state. While inanimate objects do rather less, do not have memory banks to which they can write more or less anything at all. But we do have the sedimentary strata, once at the bottom of the sea or of a lake, which sometimes preserve a good deal of history in the form of trace elements, fossils and such like. The history of the lake, or at least some of it, is written into the strata. Or there is the aeolian dust of references 5 and 6.

[Water flows from a vessel held by a god in this seal of an Akkadian king from around 2200 BC, about the time that a drought affected the empire. Credit: The Art Archive/Shutterstock]

A curiosity from reference 6.

Activation

With a car, we assemble it and then turn the ignition on. Hopefully, the car springs into life.

With a computer, something similar happens.

But with a biological system, things are rather more complicated. Things start to work as they are assembled; as soon as parts are close enough together they start to interact, as soon as your chemicals are mixed up they start to react with each other. There is no ignition, no waiting until you turn it on. Although if we were able to assume more or less instantaneous assembly, maybe it is near enough and the replica will be able to finish the breath that the original has started.

Maybe for present purposes, we can ignore this difficulty.

Two Davidsons

I have already suggested that doing the replication while Davidson was asleep might be a good idea. 

Varying the experiment slightly more, we might replicate Davidson without destroying the original. I would think that after a while, as intelligent chaps, they will probably come to realise that they are a variation on identical twins and will start to learn to live with it. And it is likely that both their bodies and their brains will drift apart over time. Perhaps they will opt to live apart, as real identical twins often do, to give their sometimes fragile egos a bit of space.

Conclusions

This paper, reference 2, appears to be about whether we can know what we think without actually saying it out loud; a question which is interesting, but not very accessible to this layman. In which this thought experiment is an interesting diversion, but I have yet to connect the diversion to what I take to be the question. Maybe if I get deeper into the paper, I will start to see Davidson’s point in offering it to us.

PS 1: Swampman has a relative, invented by Hilary Putnam, called Twin Earth. I find her even less accessible. Maybe I will get through to her later today.

PS 2: it seems that real swamp men, made famous by a television series, work the swamps of Florida. Know all about alligators and such.

PS 3: I notice that the new-to-me Windows 11 rendering of blog posts is not the same as that of Windows 10. The typeface is not quite the same and the bold is not so bold. Maybe I will tweak my house style to adapt to the new world.

References

Reference 1: The nature of qualia: a neurophilosophical analysis – Carlos E B de Sousa – 2009.

Reference 2: Knowing one's own mind – Donald Davidson – 1987.

Reference 3: https://en.wikipedia.org/wiki/Donald_Davidson_(philosopher). 

Reference 4: https://psmv2.blogspot.com/2015/12/napoleons-new-years-party.html. 

Reference 5: Precise timing of abrupt increase in dust activity in the Middle East coincident with 4.2 ka social change – Stacy A. Carolin, Richard T. Walker, Christopher C. Day, Vasile Ersek, R. Alastair Sloan, Michael W. Dee, Morteza Talebian, and Gideon M. Henderson – 2018.

Reference 6: Did a mega drought topple empires 4,200 years ago: People abandoned thriving cities in Mesopotamia, the Indus Valley and farther afield at about the same time as a decades-long drought gripped parts of the planet – Michael Marshall – 2022. Not been able to get at this one, but maybe it builds on reference 5.

Reference 7: As Cases Soar, ‘Dementia Villages’ Look Like the Future of Home Care: A new generation of treatment facilities is aiming to integrate dementia patients with the communities around them, blurring lines between home and hospital - Joann Plockova, New York Times - 2023. An advertisement for which turned up in my email yesterday. Villages which look all well and good, but rather expensive. Can’t see the Tories going for something which is both a public service and expensive. And only slightly more likely that the other lot would.