12/10/2019

Complexity – irreducible and otherwise

Filed under: — Bravus @ 7:47 am

This is the final post in this short series. It is the 12th, and a dozen seems like a reasonable tally. The sequence, as a set, is meant to allow someone who encounters a particular claim or meme to quickly access a clear, accurate response, written for a smart non-specialist. It’s by no means sufficient in terms of evidence, but hopefully it frames up the issues in a way that is helpful when someone goes looking for further evidence.

While I’m talking in this ‘meta’ way, can I encourage everyone, including myself, to always seek ‘disconfirming evidence’ and ‘discrepant cases’? Confirmation bias is real, and oh so tempting: to seek the evidence that confirms us in our existing views, and discount any that challenges them. The only approach that works, though, for anyone who values a life founded in truth, is to be always looking for the evidence that makes us change our minds.

This final topic is related to the earlier one on probability, but the focus is slightly different.

Complexity science is a fascinating and genuine area of study… but also one that is susceptible to being used as a ‘handwave’ in ways that are not scientific. Much like Deepak Chopra and his self-help ilk talk about ‘quantum indeterminacy’ to support their woo, “It appears complex therefore science supports the explanation I support” is not really an argument.

There are a number of different ways to think about complexity, and no definition is particularly universal. There isn’t a real convention. After having spent a fair bit of time reading, the following is just my best understanding, which reflects a number of influential perspectives in the field, but remains controversial.

In brief, the distinction – and it’s a philosophical one – is between things that are complicated and things that are complex. Lots of things are complicated: the internet, as a network of wires and a web of communications protocols, is complicated. Cells are complicated. Economics is complicated. In this definition, though, things that are complicated are able to be reductively analysed by breaking them down into simpler bits. While the complication is insane, it’s possible to understand each of the cables in a server farmer and what it does… and many server farms build a network. It’s possible to look at a data packet and know whether it’s organised with the ftp or http protocol.

Complex systems, on the other hand, exhibit emergent behaviours that are not able to be explained in terms of reduction to simpler components. We could argue that human brains are complex in this sense, for example: self-consciousness is not easy to explain in terms of neurons and neurotransmitters and potentials and neuroplasticity.

The question of whether a particular system is complicated or complex in this sense is not simple to determine in any final sense. It may just be that we haven’t yet thrown sufficient computational resources or smart enough algorithms at our reductive analysis. If resources short of infinity could analyse a system, it can be argued that it is merely complicated, not complex… and the case needs to remain open for a lot of things.

That approach is different from the concept of ‘irreducible complexity’ that tends to be used by the Intelligent Design advocates. They tend to launch from comments such as the ones Darwin himself made in ‘Origin’, about how difficult it is to imagine a process by which the eye could evolve. Darwin does not despair of it, however, and plausible sequences have been outlined. Eyes tend not to fossilise, so hard evidence is challenging to find, but there are numerous kinds of different eyes, and it appears as though eyes may have evolved multiple different times independently.

The key issue in their approach is assuming that an eye is ‘irreducibly complex’ – that unless there is an eye in pretty much its current form, with eyelids, muscles to turn it, a lens, an iris, a retina and optic nerves, rods and cones for black-and-white and colour vision, it is not an eye at all, and conveys no survival advantage. But much simpler eyes exist, right down to simple light-sensitive spots, and convey survival advantages significant enough to lead to selection. The notion of irreducible complexity is built on the misconception that complex systems must spring into existence in pretty much their current form from essentially nothing. But refuting something evolution does not predict does not refute evolution.

They have moved on from the eye, and things like the flagella that bacteria use to propel themselves, to DNA and the processes of cell division and replication – the most fundamental processes of life itself.

They are right to say that these processes are complex in a way Darwin couldn’t have known or imagined 150 years ago, and indeed, we have learned much more in just the past few decades. It’s quite amazing that our DNA has multiple independent self-repair mechanisms: when things go wrong they are often corrected, or the process aborted. Cancer would be far more common if these processes were not in place. They’re nothing we can yet reliably build in to our code, let alone to our material machines.

While fully accepting the complexity of the cell – which is wondrous – the argument that it could not have evolved, because it needs to exist in pretty much its current form to work at all, recapitulates many of the arguments about the eye and the flagellum, and is wrong for the same reasons.

There are plausible, but still quite early, proposals for simpler RNA-only replicating sequences that may have pre-dated, and evolved to form, the current very complex systems.

Life is, indeed, complex. Whether irreducibly so is a philosophical question. But arguments from that complexity for a divine Creator – by fiat ex nihilo or by tinkering at the edges – are not strong arguments.

For ease of navigation I will include links to each of the other posts in this series at the bottom of each post.

Why I think it’s important to understand evolution
Cosmogenesis, abiogenesis and evolution
Evolution and entropy
Facts, Theories and Laws
Radiocarbon dating
Radiometric dating and deep time
Four Forces of the Universe
Probability and evolution
Species and ‘baramin’, macro- and micro-evolution
Mitochondrial Eve and Y-chromosomal Adam
Transitional fossils

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.