Book Review: "From Bacteria to Bach and Back" Dan Dennett

This is a book that takes some time to digest.  I've been reading it in short spurts over many months, allowing the deep subject matter to stew in my mind.  Dan Dennett is a Professor of Philosophy at Tufts University and he has spent his very long career seeking to explain, among other things, how the brain produces mind.  Dan's approach to this question is quite interesting.  He insists on using scientific findings to direct his thinking and he avoids metaphysical prognostication.

There is a school of thought in "philosophy of mind" studies that there exists a so called hard problem of consciousness.  The hard problem of consciousness was described by Prof David Chalmers, an Australian philosopher at ANU, as the subjective experience of awareness that transcends the physical.  Chalmers uses the label qualia to describe this subjective feeling that arises in our minds when we experience the world.  He believes the nature of qualia is fundamentally different to physical matter and that, as a consequence, consciousness must be a foundational property of the cosmos.

"Hogwash," says Dennett.  According to Dennett, the hard problem of consciousness isn't hard, but it is difficult to get your head around it.  The reason it is difficult is because our conscious experience is so compelling in its richness and detail that we are driven to believe that there is an internal observer, the self, which is the beneficiary of all the sensations and emotions we experience and which, importantly, is the author of all our thoughts, dreams and desires.  We believe so strongly in this autonomous self, a belief enforced and maintained by a culture which rewards the worthy and punishes the unworthy, that we almost instinctively avow a metaphysical theory of mind.  That is, we feel sure that there is something spiritual and ineffable about 'mind' which simply cannot be explained by the wet, meaty stuff in our 'brain'.

The apparent duality of mind and brain has led to a widespread and persistent belief in dualism.  René Descartes described dualism in a way that has resonated with Western philosophy for centuries.  According to his view, which is shared by Chalmers, reality consists of matter stuff and mind stuff.  It is quite understandable how dualism gains traction in our thinking when you consider the experience of death.  One moment the body, made of matter, is breathing, moving and speaking and a moment later it is still and lifeless.  It certainly seems obvious that some 'thing' has left the body at the point of death.  This phenomena continues to support belief in souls and spirits to this day.

Is there any evidence of a ghost in the human machine?  The short answer is no.  The evidence from controlled experiments in neuroscience indicates that the conscious mind becomes aware of decisions after the brain has already made the decision.  To use a metaphor, our consciousness is like the surface of a wave.  The surface of the wave is not the cause of the wave.  It is where it is because of everything happening beneath the surface.

Dennett's book is his latest attempt to synthesise the lessons of science into an explanation of how mind comes from brain.  Dennett draws on three "strange inversions of reason" in the history of science and philosophy.  I'll give a paragraph to each of these.

Firstly, there is Darwin's strange inversion of reason from the 'trickle down' hypothesis of Intelligent Design to the 'bubble up' theory of Evolution.  Darwin noted that there was variation of traits within a population of any given species.  He also noted that individuals within the population struggle to survive within their environment.  Darwin's key insight was that certain traits provide some individuals within that population an advantage, improving their chances of passing those traits onto future generations.  To use a construction analogy, evolution is like a crane building up from the foundation and Intelligent Design is like a skyhook.

Secondly, there is Turing's strange inversion of reason.  Turing is the father of modern computing.  What he showed, which we all now experience everyday with our smartphones and laptops, is that a perfect computing machine does not need to have any comprehension in order to do arithmetic.

Thirdly, there is Hume's strange inversion of reason.  Our conscious experience consists of a flow of moments in time.  We observe things happening all of the time and we instinctively attribute causes to the things that happen.  The cause - effect relationship is so fundamental to our experience, but Hume says that we never experience cause and effect, we only ever experience each moment.  The cause - effect relationships we perceive are actually a projection from our minds onto the world.  We do this, says Hume, because we have come to expect outcomes from certain actions.  And this is Hume's inversion of reason, that, in a metaphorical sense, we project our expectations onto the objects of the natural world.  We taste honey and say that it is sweet.  The sweetness of honey is not an intrinsic property of honey.  There is nothing about the molecules of honey that is, in essence, sweet.  We have projected the attribute of sweetness onto the honey as part of our ontology, our 'way of being'.

According to Dennett, these three strange inversions of reason provide us with the tools to understand how our conscious experience arises from brain stuff.

Evolution through natural selection (Darwin's strange inversion) builds biology from the simplest of replicating forms to the amazingly diverse range of species we have today, of which humans are the beneficiaries of large and intricately wired brains.

Our nervous system is stimulated by billions of neuron spikes in any given moment.  These spikes result in a cascade of spikes throughout the brain, known as spike trains.  Some of these spikes represent external things (i.e., sight, sound, touch, taste, smell), bodily sensations, responses from the brain's memory centres, and various layers of representations of those representations.  There is no reason to suspect that individual neurons are somehow self-aware or conscious as they go about their business of responding to stimulus time and time again.  The billions of neurons in our brain are competent without having comprehension (Turing's strange inversion).

Some neuronal connections could fire in response to say the colour blue, a baby's face, the concept of multiplication, the word 'self', or a directive to contract a specific muscle.  Every firing between two neurons strengthens the connection.  The strength of the connections between neurons is a measure of the degree of expectation regarding the likelihood of that connection firing again.  In this way, our brains develop expectations which can be triggered without any external stimulus, as happens with optical illusions.

Most of the spike trains in our brain have no apparent impact on our conscious experience (e.g., regulatory processes in the medulla oblongata for respiration and circulation).  Some of the neural spikes in our brain, however, are clearly correlated with consciousness.  Our conscious experience seems to consist of representations that have been filtered through many levels of brain processes in many parts of the brain yielding a projection (Hume's strange inversion) into "subjective space".  There is no need, according to Dennett, to ask the what, where, or how of subjective space.  Doing so is a violation of Occam's razor because the subjective space adds nothing to our understanding other than a smug satisfaction that we have control of our brains when the reality is that our brains control us.

Dennett does think the "what for" question is interesting, though.  He suggests that our conscious experience is available to us to facilitate communication with each other.  That is, having access to our own conscious experience helps us to share with other people what we are thinking, what we are planning, how we are feeling, and how we might be able to cooperate to make our lives better and enhance our chances of survival.  Consciousness, in this view, is for others; not merely for our own benefit, but for the benefit of our social species.

From Bacteria to Bach and Back is a difficult, thought-provoking read.  Dennett's style is somewhat rambling as he advances the various prongs of his argument a little at a time, before back-tracking and carrying his argument a little further.  He spends a great deal of the book attempting to point out the fallacies in our thinking that are a result of our conditioning.  While reading this book, I found that I needed to take a break from time to time, re-read sections, and cogitate.

In my view, Dennett's argument is quite convincing; however, I don't think he has captured the whole picture just yet.  For one thing, research continues on the physical structure of brains (i.e., the connectome, and neuroglia, etc.), neuroscience, cognitive psychology, and artificial intelligence.  I suspect that our understanding of consciousness will improve as the science progresses.  I'm not entirely sure that Dennett is on the money, but he does seem to be on the right track.  I will be particularly interested to see experimental results testing the validity of qualia.

I'll leave you with a fascinating 2017 experiment by Roman Yampolskiy available in pre-print on arXiv titled, "Detecting Qualia in Natural and Artificial Agents."  Using the hypothesis that the ability to be fooled by an illusion is an indicator that an 'agent' has a kind of subjective experience, which is not based on the real world, Yampolskiy shows that some deep neural networks can be fooled by optical illusions into seeing something that is not there.  I'm not convinced that this is actually a demonstration of conscious AI, but it might be a rudimentary precursor on the pathway to conscious AI.

You can buy your own copy via Amazon:
https://www.amazon.com/Bacteria-Bach-Back-Evolution-Minds/dp/0393242072.