Monday, August 12, 2013

Life, death and computation

I have been spending a bit too much time lately reading other people's blogs and (to some extent) participating in associated discussions. The main problem with this sort of activity is that – largely because the focus of discussions is always shifting – it encourages superficial debate at the expense of deep understanding.

But, interestingly, two recent blog discussions on two very different sites which I happen to follow touch on a similar theme.

Biologist and philosopher Massimo Pigliucci recently precipitated a freewheeling discussion of the relevance of computers and computing to understanding the human mind and the universe in general. In fact, Pigliucci's post on the topic prompted more than 200 comments, many of which are well worth reading.

Professor Pigliucci has a disarming tendency to rush in where more cautious academics fear to tread – that is, beyond his areas of specific expertise. (I suspect his approach owes something to the intellectual traditions of his native Italy, where academics have traditionally played an important role in the broader cultural, moral and political sphere.)

Pigliucci argues strongly against functionalist and computational views of the mind. I don't have strong views on this question, though I share Pigliucci's skepticism about some of the (as I see it) wilder claims about mind uploading and the scope of simulations etc.

I did, however, question his contention that seeing the operations of nature in computational terms is likely to lead to mathematical Platonism, commenting as follows:

My understanding is that many of the leading proponents of an information- and information processing-based approach to physics see information as physical. The bits or qubits are always 'embodied' in actual physical processes, albeit that these processes are understood at a deep level in terms of the processing of information. (There are close parallels between information theory and thermodynamics.)

So I'm not sure that such a view leads to Platonism. Seeing physical processes as algorithmic (and scientific theories as predictive algorithms) seems to me a genuinely interesting perspective: but it may well be that there is no way actual physical processes can be perfectly simulated (or predicted).

Adrian McKinty is a novelist with a strong interest in social, cultural and philosophical topics. In the comment thread of a post at McKinty's nicely named site, The Psychopathology of Everyday Life (I know – Freud got there first), a post about Philip Larkin featuring his confronting poem, 'Aubade', McKinty mentions Nick Bostrom's simulation argument: that if we accept two fairly plausible-seeming assumptions then our universe is almost certainly a 'simulated' universe created by an advanced civilization.

As I commented there:

I am ... (prompted by your comments, Adrian) having a look at Nick Bostrom's ideas. My initial attitude is skepticism, but that may just be what he would call my status quo bias jumping in.

I do think it makes sense (simply in terms of physics) to see natural processes in terms of information processing, but it is a big jump from there to thinking about beings who might have set the process going (and to calling it a simulation).

And what would Larkin make of all this? (Turning in his grave, I suspect.)

I am continuing to look into the simulation argument which I first encountered some years ago. More later, perhaps.

But regular readers will know that I am very skeptical of arguments and points of view which take their origins from a philosophical (as distinct from a scientific) base. Bostrom's main argument for the simulation hypothesis is in part statistical but basically philosophical – and far from convincing from my point of view.

I can't help feeling that people like Bostrom (and David Pearce who influenced him) are driven by a kind of religious instinct. Certainly some of the groups with which they are associated have a cultish feel.

The other thinker mentioned by Adrian in the comment thread is Samuel Scheffler. Scheffler applies 'what if' scenarios to thinking about death. What if we knew the world was going to be destroyed soon after our death? His general point seems to be that we are underlyingly less concerned about our own personal fate per se than about our fate seen in the light of a continuing social context.

This may well be, and such thinking is very much in accordance with the view that the sense of self derives from the linguistic, cultural and social context in which we grow up. But I think Scheffler overplays the extent to which future generations give meaning to our lives.

Also, I had a look at Scheffler's background. And it seems pretty clear that his being a socialist (he is apparently a disciple of the 'analytical Marxist' G.A. Cohen) would have – to some extent at least – shaped and played a part in his approach to thinking about the future in general, and about ethics.