Short works

Books : reviews

Douglas R. Hofstadter, Daniel C. Dennett.
The Mind's I: fantasies and reflections on self and soul.
Penguin. 1981

rating : 1.5 : unmissable

A collection of essays, including Raymond M. Smullyan's marvellous "Is God a Taoist?" and Thomas Nagel's classic "What is it Like to be a Bat?", each accompanied by Reflections of the editors.

... a sentient being without free will is no more conceivable than a physical object which exerts no gravitational attraction.

In particular, John Searle's infamous Chinese Room argument, here explained in his "Minds, Brains and Programs" essay, gets treated to a long, and opposed, Reflection.

Daniel C. Dennett.
Brainstorms: philosophical essays on mind and psychology.
Harvester Press. 1981

rating : 2.5 : great stuff

Daniel C. Dennett.
Elbow Room: the varieties of free will worth wanting.
Clarendon / Oxford. 1984

rating : 3.5 : worth reading

Daniel C. Dennett.
Content and Consciousness: 2nd edn.
Routledge. 1986

2nd edition of the author's D.Phil. thesis. (1st edition published in 1969.)

Daniel C. Dennett.
The Intentional Stance.
Bradford / MIT. 1987

rating : 3.5 : worth reading

Describing systems "as if" they have intentionality can be helpful.

Daniel C. Dennett.
Consciousness Explained.
Allen Lane. 1991

rating : 1.5 : unmissable

'Multiple drafts' theory of consciousness

Daniel C. Dennett.
Darwin's Dangerous Idea: evolution and the meanings of life.
Allen Lane. 1995

rating : 1.5 : unmissable
review : 1 June 1996

Dennett, a philosopher well known for his writings about consciousness and cognition, looks at evolution -- of genetic life, of memetic intelligence (including AI), of morality.

Imagination is cheap if we don't have to bother with the details.

His thesis is that evolution is powerful enough to do all this, that complexity can evolve from simplicity, by slow adaptation and accumulation of existing designs. Such building might need to be 'lifted', using ground-based 'cranes' (and cranes upon cranes), but no magical or mystical 'skyhooks' (elan vital, strange essences, or as-yet-unknown quantum gravity effects) are needed at any stage.

Greedy reductionists think that everything can be explained without cranes; good reductionists think that everything can be explained without skyhooks.

He uses the metaphor of exploring a rugged landscape of 'design space' to show how evolution can indeed work, but why some design solutions might never be found: because they are not accessible from the current solution,

If a single step in the genotype can produce a giant step in the phenotype, intermediate steps for the phenotype may be simply unavailable, given the mapping rules.

or simply because design space is so Vast, and explored solutions can be only a Vanishingly small subset of all possible solutions.

We don't need laws of biology to "prevent" most of the physical possibilities from becoming actualities; sheer absence of opportunity will account for most of them.

This use of the capital 'V's is a clever technique to keep reminding the reader just how vast 'Vast' really is and just how vanishingly small 'Vanishing' is, without belabouring the point every time. For example, in describing heuristic AI search algorithms:

The search space is Vast, so the method of search must be "heuristic" --- the branching tree of all possible moves has to be ruthlessly pruned by semi-intelligent, myopic demons, leading to risky, chance-ridden exploration of a tiny subportion of the whole space.

Dennett finishes off with the point that Dawkins also made, at the end of his 'The Selfish Gene', we humans are the first species on this planet not to be at the mercy of our genes.

It must be true that there is an evolutionary explanation of how our memes and genes interacted to create the policies of human cooperation that we enjoy in civilization ... but this would not show that the result was for the benefit of the genes (as principal beneficiaries). Once memes are on the scene, they, and the persons they help create, are also potential beneficiaries.

I enjoy all of Dennett's writing --- although I'm not sure I necessarily completely understand all his subtleties (I'm not a philosopher by training, and he himself admits this is a 'difficult book'). His aim is to convince people that evolution is sufficient: I'm not sure how successful he is, since with me he is 'preaching to the converted'. But it's a good read, with many new arguments (especially against the non-Darwinian arguments of Stephen Jay Gould, Roger Penrose, and John Searle) and may arguments expanded and improved from his previous books.

Has it ever occurred to you how lucky you are to be alive? More than 99 percent of all the creatures that have ever lived have died without progeny, but not a single one of your ancestors falls into that group! ... Not a single one of your ancestors, all the way back to the bacteria, succumbed to predation before reproducing, or lost out in the competition for a mate.

Other quotations:

Daniel C. Dennett.
Kinds of Minds.
Weidenfeld & Nicolson. 1996

rating : 3 : worth reading
review : 4 October 1998

In this slim volume Dennett explores the different kinds of minds that animals might have, from simple programmed unconscious behaviour, through to full self-conscious thinking, building on ideas from his earlier works Consciousness Explained and Darwin's Dangerous Idea. One of the reasons he is interested in whether animals can be considered to have minds is a moral one: can they suffer?

He takes the intentional stance, which describes certain entities as if they were rational agents making choices based on beliefs, desires and goals about the world. He emphasises that this needs to be done with care, but that it can provides a convenient shorthand for describing the behaviour of such entities. And if talking of beliefs and desires is just a convenient shorthand for describing some agents, can we be that sure that it is not also just a convenient shorthand for describing ourselves? He is also careful to point out that the intentional stance is different from anthropomorphism, unless we are careless: we expose the underlying anthropomorphism of the intentional stance: we treat all intentional systems as if they were just like us --- which of course they are not.

He goes on to describe various levels of minds, including hard-wired, reactive, and anticipatory.

A mind is fundamentally an anticipator, an expectation-generator. It mines the present for clues, which it refines with the help of materials it has saved from the past, turning them into anticipations of the future. And then it acts, rationally, on the basis of those hard-won anticipations.

One fascinating point Dennett makes is that our current models of the body's role in mind are probably too simple. A machine has transducers (converters of sensory or other environmental data into a form that can be transferred within the system) at one end, connected, via some communication and processing system, to effectors at the other. The nature of the transducers and effectors are pretty well fixed, because of their interaction with the external world, but the nature of the communication medium is much more irrelevant: anything that moves the signals appropriately would do. However, Dennett points out that the way nerves work, using electrical signals along them but neurotransmitters between them, seems to include transducers everywhere; so maybe the communication medium is much more intimately linked with the transducers and effectors, and hence less substitutable, than we had previously been assuming.

The idea that the network itself --- by virtue of its intricate structure, and hence powers of transformation, and hence capacity for controlling the body --- could assume the role of the inner Boss and thus harbor consciousness, seems preposterous. Initially.

As he works his way up his hierarchy of minds, Dennett notes that higher forms arrange their environment to enhance their minds and relieve the burden of detailed memory (dogs marking territorial borders, people writing shopping lists).

... old folks removed from their homes to hospital settings are put at a tremendous disadvantage.... They often appear to be quite demented.... Often, however, if they are returned to their homes, they can manage quite well for themselves. How do they do this? Over the years, they have loaded their home environments with ultrafamiliar landmarks, triggers for habits, reminders of what to do, where to find the food, how to get dressed, where the telephone is, and so forth. ... Taking them out of their homes is literally separating them from large parts of their minds --- potentially just as devastating a development as undergoing brain surgery.

There are some interesting ideas here, but, since the book is quite slim (170 pages of quite large type), his topic is not explored in the same level of detail as in his earlier works. Which could make it a good starting point for new readers to get a feel for his style of argument before plunging into his deeper works. But, don't expect answers. Dennett, as a philosopher, is interested in finding good questions. It's great fun exploring these questions with him.

Daniel C. Dennett, ed.
Brainchildren: essays on designing minds.
Penguin. 1998

rating : 3.5 : worth reading
review : 30 December 2006

This is a collection of Dennett's papers on Philosophy of Mind, and Artificial Intelligence, previously not readily available, here annotated with brief introductions to provide context, and the occasional more substantial postscript. It shows both the evolution of his ideas over the decade or so covered, and also how some of those ideas were present right from the start. Also, his experimental mindset is made evident: it's not enough just to philosophise: the arguments need to be consistent with (or at least constrained by) reality.

One underlying theme that comes through clearly is the importance of taking the reality of evolution into consideration. The intelligent beings on this planet that we are seeking to artificially emulate were not arbitrarily plonked down here: we evolved here, and many of the features of our intelligence are constrained by the requirements of survival. We are not tabula rasa: capabilities we have work because they have to (or at least, had to) work, and so they have evolved to work.

Although all these papers were originally written for academic journals or conferences, each individual paper is, for the most part, eminently readable -- there is a minimum of jargon, and absolutely no pretension.

Probably not the best place to start your introduction to Dennett's ideas, but an excellent way to fill in some of the gaps.


Artificial Life as Philosophy. 1994
In this manifesto in the first issue of the Artificial Life journal, Dennett explains why the study of Artificial Life is important as a new philosophical method.
Nicholas Humphrey, Daniel C. Dennett. Speaking for Our Selves: an assessment of multiple personality disorder. 1989
Examining multiple personality disorder in terms of Dennett's ideas of consciousness. It doesn't sound implausible, but there seems to be far too little data on the phenomenon for much rigorous theorising -- and, as the authors point out, the phenomenon is not really amenable to ethical experimenting.
Can machines think?. 1985
How hard the Turing test really is, and what it would actually mean for a machine to pass it. I particularly like the term he has coined for purely verbal, disembodied computers taking this test: "bedridden". The 1997 Postscript describes Dennett's experiences chairing the Loebner Prize Committee.
Do-it-yourself understanding. 1992
A fairly technical paper, discussing, and rebutting some of, Dretske's ides on "meaning", about what it "means to understand something". Does a sniffer dog who wags its tail when it discovers drug understand that it has discovered drugs? How is the correspondence between observed and mental patterns, and what they mean, established? This is Dennett, so evolution (as well as learning) plays a part in the argument.
Two contrasts: folk craft versus folk science, and belief versus opinion. 1991
About the new Connectionism.
Real patterns. The Journal of Philosophy 87. 1991
As Dennett says in the introduction, the ideas here are utterly central to his thinking. It covers the ontological status of abstract objects, including patterns, particularly those obscured by more or less noise. It moves on to patterns in the Game of Life, and a nice description of his three stances: physical (for describing the underlying cellular automaton with cells changing state), design (for describing gliders, glider guns, etc: moving patterns), and intentional (for describing what a GoL CA emulating a chess playing program is doing at the level of "wanting to play chess"). This leads to a discussion of the reality of the intentional patterns. Dennett's position is somewhere between the (implausible) extremes of Realism and Instrumentalism.
Julian Jaynes's software archeology. 1986
This is a discussion of Jaynes' 1976 book The Origins of Consciousness in the Breakdown of the Bicameral Mind. Although many of the claims in there are known not to be true, Dennett argues that the work needs to be taken seriously because of the important questions it raises for the first time. (Consciousness Explained is Dennett's own attempt to answer some of these questions.)
Real consciousness. 1994
Responses to some critiques of Consciousness Explained. Again, the explanation revolves around the fact that the critiques assume the model must be one of a certain set of alternatives, but that they have missed out one of the alternatives.
Instead of qualia. 1994
Another argument about qualia. [I don't understand it, because I don't understand "qualia". As far as I understand it, "qualia" describes what it really means to have an experience, as opposed to the simple objective facts of the experience -- but I don't understand what "really means" really means. It sounds to me a bit like discussing the "vital spark" left over when you take all the physical matter away from life: a long, convoluted argument about the properties of the elements in the empty set. All unicorns are pink (and invisible).]
The practical requirements for making a conscious robot. 1994
A discussion of the beginnings of the MIT Cog project, to build a socially interactive, embodied robot.
The unimagined preposterousness of zombies: commentary on Moody, Flanagan, and Polger. 1995
A rather technical piece on zombies (just like us, but not "conscious") and "zimboes" (an advanced zombie capable of internal reflection, but still not "conscious"). A bit like the "Can machines think?" paper on the Turing test, on how difficult the problem really is (or here, how preposterous the assumed creature is).
Cognitive wheels: the frame problem of AI. 1984
An early philosophical paper on the "frame problem": when (and how) to stop inferring consequences, and just do the right thing. Just telling the machine to ignore irrelevant inferences is no good: to do this it must first infer them, then infer their irrelevance, in order to ignore them. How to program in sensible default assumptions without resorting to cognitive wheels? (A "cognitive wheel" is Dennett's perjorative term for an elegant designed cognitive ability that doesn't, maybe couldn't, occur biologically.) Again, evolution might provide the answer (for us): our default assumptions are the ones that have worked, have allowed our (or our ancestors') survival.
Producing future by telling stories. 1996
More on the frame problem, and competing story-schemas.
The logical geography of computational approaches: a view from the east pole. 1986
An amusing categorisation of AI philosophy and research, with MIT and "High Church Computationalism" at the East Pole, and hence everywhere else West of there.
Hofstadter's quest: a tale of cognitive pursuit. 1996
A long and discursive "review" of Fluid Concepts and Creative Analogies.

"Anything you can do I can do meta-" was one of Doug's mottoes, and of course he applied it, recursively, to everything he did. [of Hofstadter's GEB]
Foreword to Robert French, The Subtlety of Sameness. 1995
Dennett's forward to French's discussion of the "Tabletop" system for investigating fluid analogies. [Based on this, and the previous "Hofstadter's quest" paper, I went and ordered The Subtlety of Sameness from Amazon 2nd-hand. This is precisely why my "to read" pile never shrinks.]
Cognitive science as reverse engineering: several meanings of "top-down" and "bottom-up". 1994
Both top-down and bottom-up approaches to understanding cognition have advantages and problems. In particular, top-down approaches tend to assume the machinery is optimally designed, rather than evolved, and tends to design against emergent properties.
When philosophers encounter artificial intelligence. 1988
A riposte to Putnam: why AI is important to, and for, philosophy.
Review of Allen Newell, Unified Theories of Cognition. 1993
A long and discursive review of the ideas in Unified Theories of Cognition.
Out of the armchair and into the field. 1988
It's easy for philosophers to conduct thought experiments. How readily could these be realised? Dennett travels to Kenya to visit Robert Seyfarth and Dorothy Cheney as they investigate vervet monkeys. He discovers that such realisation is almost impossible.
Cognitive ethology: hunting for bargains or a wild goose chase. 1989
A discussion of how studies of animal behaviour might influence AI research, including a recap of the vervet experience recounted in "Out of the armchair and into the field".
Do animals have beliefs?. 1995
It all depends on what you mean by "belief": the word has such a wide meaning in cognitive science that the answer must be "yes". So Dennett dissects different meanings of "belief". (Also includes an interesting discursion on how it is difficult to ask the same question in French, since "croyance" has a much narrower meaning that almost obviously makes the answer "no".)
Why creative intelligence is hard to find: Commentary on Whiten and Byrne. 1988
Based on his realisation of the difficulty of doing real animal cognitive experiments, outlined in "Out of the armchair and into the field", Dennett explains why it will be essentially impossible to demonstrate "creative intelligence" in animals.
Animal consciousness: what matters and why. 1995
Current discussions about animal consciousness are too distorted by political agendas.
Self-portrait. 1994
A survey of Dennett's philosophical ideas and contributions.
Information, technology and the virtues of ignorance. 1986
This paper is not about the philosophy of consciousness and cognition: it is about the negative impact of technology on our ability to live a fulfilling and good life. The first example, which is meant to be illustrative of a wider problem, is about increasing technology in medicine, particularly the use of expert systems. They may lead to a situation where the life of the "country doctor", if ethical -- making best use of the technology for the benefit of the patients -- is no longer fulfilling -- no "heroics", or even mastery, just plugging readings into a machine. (However, in the two decades since the paper was written, expert systems haven't exactly taken over in any convincing way.) The second, more chilling, example is how technology allows us to be aware of so much more suffering than we can have any conceivable chance to do anything about: what should we do to do right? This effect presumably explains the amazing response to events life like Live Aid, and the Boxing Day tsunami: finally, a focus, almost a feeling of relief, an obvious good thing to do. (One suggestion I've heard to solve this problem is to delegate: pay one's tithe to church or state, and let them decide which charity. But does this really fulfil the moral obligation? And do individual donations do overall good? Buying a goat as a present (feel-good for both giver and receiver) via Oxfam has resulted in copious follow-up "junk" mail from that organisation.)

Daniel C. Dennett.
Freedom Evolves.
Allen Lane. 2003

rating : 3 : worth reading
review : 25 May 2003

I love Daniel Dennett's writings. I think Consciousness Explained is fascinating and illuminating (even if it doesn't do exactly what it says on the tin), and I think Darwin's Dangerous Idea is full of wonderful imagery and explanation. Yet several years ago when I read Elbow Room, about free will, I just didn't get it. Well, here is his new book about free will -- and I still don't get it.

Dennett starts off in the usual way, with the argument that determinism is incompatible with free will. He uses a big Game of Life to support his arguments and definition, such as those of necessary and sufficient cause, including the idea that some effects need not have a cause: things are complicated enough that no single cause is either necessary or sufficient to ensure the effect.

Dennett explains that deterministic does not mean inevitable -- because inevitable just means unavoidable, and even in the Game of Life (since it is computationally complete), patterns, or agents, can "evolve" to be good "avoiders". He also explains that, in order to avoid, the agent needs both sufficient information on which to base its action -- where being in a deterministic universe helps it determine (geddit?) the consequences of its planned actions -- and sufficient time in which to act.

But I feel this is merely a bit of verbal sleight of hand: are we talking about avoiding things in general, or avoiding this specific event? Evolution makes things good avoiders in general, but in the deterministic Game of Life rules, if this agent pattern is going to get eaten by that one this time, there is nothing it can "do" to avoid it.

Dennett shows that adding a dash of indeterminism, maybe in the form of some random quantum chance, whether external or internal to the brain, doesn't solve the problem. But this really is a "no-brainer": if I'm at a fork in the path, then tossing a coin, or consulting a quantum event, to "choose" the branch to take can't possibly be what anyone means by free will. Not only must it be that if the "tape was run again", a different outcome was possible; but I have to choose that outcome. Now, personally I have no idea how this can possibly be the case, how I, following either deterministic or random rules, can make that choice. I can't even formulate a coherent question. But that's what philosophers are for, and so I would have liked more discussion of this point. There is some discussion of what something being "up to me" might mean, but it is all mixed in with a quantum indeterminism red herring. However, there is a nice discussion of what what the process of "making a choice" might be like, in terms of two fairly evenly balanced dynamic networks "fighting it out": whichever you eventually choose, you wanted to do it, because you wanted to do both.

I do have a problem with the Game of Life automaton as a pump primer for intuition about determinism and so on. The real world is not a cellular automaton (although some physicists might disagree), but rather is open in some not very well understood sense. Also it is quantum: this has effects above and beyond mere indeterminism: do entanglement and decoherence change the intuitions? It might be that this model being used to prime our intuition is just too simple to capture reality. I stress that I don't mean this in any spooky "ghost in the machine" way -- heaven forfend! -- but maybe the properties of the richer model are qualitatively different?

Dennett talks about Game of Life patterns that appear at higher levels: things like gliders and glider guns, not individual cells: he is taking a stance above the level of the underlying cellular rules. It feels like he is about to claim that free will is an emergent property, a property that only holds at higher levels, and is not even a meaningful concept at lower ones. As I read it, I felt a faint inkling this might work, this might make sense, so I turned to the next chapter, eager to see how it would pan out. But I was caught in a sort of mental whiplash, because he immediately goes on to talk at this higher level, about kinds of free will worth wanting, as if the problem is settled, without that bridging chapter I so dearly needed. Maybe the link is there, clear for others to see, but I couldn't see it.

The final part of the book, on how free will evolves, and how people have more and richer kinds of free will worth having, by virtue of our consciousness, is fascinating. A lot of it relies on ideas from Consciousness Explained and Darwin's Dangerous Idea, with the necessary concepts being ruthlessly summarised where required. It has a good discussion of why, even as we find out more about the brain and its mental conditions, we will never "empty the set" of people held responsible for their actions: people will accept those responsibilities in order to gain the benefit of the accompanying rights.

All in all, this is a fascinating book. The first half is a careful discussion of the meaning of determinism. The second is a naturalist's argument about free will in reasoning evolving agents. If only it had had that extra chapter in the middle, explicitly linking the two halves for me, so that I could finally "get it".

Daniel C. Dennett.
Sweet Dreams: philosophical obstacles to a science of consciousness.
MIT Press. 2005

rating : 3.5 : worth reading
review : 5 June 2008

This is a collection of lectures, essays, and articles where Dennett responds to comments on, and criticisms of, his "Multiple Drafts" theory of consciousness as laid out in Consciousness Explained, and expands on and refines some of the ideas there. As he says in the preface [p.ix] I didn't get it all right the first time, but I didn't get it all wrong either. Since it is a collection of individual essays, there is some repetition. In fact, Dennett morphs this bug into a feature:

[p.x] The Multiple Drafts Model of consciousness is also a model of my academic life during the last dozen years. Giving several dozen public lectures a year on consciousness to widely different audiences encourages a large amount of adaptation and mutation of previously used material. In this volume I have attempted to freeze time somewhat arbitrarily and compose a "best" version of all this, trying to minimize repetition while preserving context. That is just what we do, according to my theory, when we tell others---or even our later selves---about our conscious experience.

Most of this is written in his usual highly accessible style. For example, I laughed out loud (something that always gets me strange looks on the train) when I read:

[p9, footnote 10] Incurable optimist that I am, I find this recent invasion by physicists into the domain of cognitive neuroscience to be a cloud with a silver lining: for the first time in my professional life, an interloping discipline beats out philosophy for the prize of combining arrogance with ignorance about the field being invaded.

Despite this accessible style, it is quite technical in places, particularly as some of the essays are ripostes (to ripostes to ripostes) to other articles I haven't read, or to other philosophical positions I'm not aware of. So I was particularly grateful for this description of (philosophical) zombies:

[p15] More pointedly, if you still think that Chalmers and I are just wrong about this, you are simply operating with a mistaken concept of zombies, one that is irrelevant to the philosophical discussion. (I mention this because I have found that many onlookers, scientists in particular, have a hard time believing that philosophers can be taking such a preposterous idea as zombies seriously, so they generously replace it with some idea that one can take seriously---but one that does not do the requisite philosophical work. Just remember: by definition, a zombie behaves indistinguishably from a conscious being---in all possible tests, including not only answers to questions [as in the Turing test] but psychophysical tests, neurophysiological tests---all tests that any "third-person" science can devise.)

So, I will stop being generous. From this, it appears that the philosophical zombie is identical to a conscious being in every way except that it isn't conscious. So, this strikes me (taking my newly ungenerous stance) as like saying a that a "living dead" movie zombie is identical to a living being in every way except that it isn't alive. It doesn't have the vital spark, the elan vital, the quintessence, of a living being. How quaintly 19th Century! So, presumably, philosophical unconscious zombies should be used to prove, in exactly the same way, that there is nothing to consciousness beyond physical properties?

But, from reading this book, it appears not. Instead, it appears that many philosophers (Dennett excluded, naturally) engage in games as follows. First, imagine some hypothetical being with stated powers. [I'm imagining, I'm imagining very hard -- this hypothetical being would have to be more than superhuman to actually have the stated powers, though. Okay, so I'm now imagining the requested superhero. (Except in the case of zombies, where my imagination does actually fail.)] Next, say the philosophers, put that being in some scenario. [Okay, done. That was easy.] Then, hey presto, as you will have realised intuitively, they'll have such-and-such a reaction, which proves something spooky about consciousness. [Um, no, sorry. Because that intuition is based on what your unenhanced reaction would be, and you are not that superhero, you are an ordinary philosopher. The being with the superpowers you require is so enhanced, so unintuitive, that you cannot use your own intuition to, well, intuit anything about their capabilities. You have failed to follow through with your own imagined scenario.]

Dennett is very good at showing where this imagining breaks down. In other writings he has given intuition about how Vast the library of Babel is, and how enormous the actual Chinese Room would be (so enormous that the little man pottering around inside is essentially invisible). Here he pays the same game with "Mary the colour scientist" (a person who knows everything there is to know about colour, despite never having seen it: does she experience anything new when she does finally see it?) to dismiss intuitions about qualia. He does this by imagining RoboMary, a robot who knows everything there is to know about colour, despite never having seen it (a situation somewhat easier to set up). This particular argument doesn't need to discuss ideas of consciousness (the "RoboMary isn't conscious, therefore doesn't experience qualia, so it's a different case" counterargument). What it does is show just how much RoboMary (and hence, by analogy, person Mary) would have to know to know everything about colour without having seen it -- and it's a lot, a superhuman lot, including knowledge of detailed brain states. So why on Earth should we feel it reasonable to use our intuitions to reason about (Robo)Mary's experiences?

In fact, Dennett is quite scathing throughout about certain philosophers insistence on using their intuition to "reason" about properties of consciousness.

[p108] I had no idea philosophers still put so much faith in the authority of their homegrown intuitions. It is almost as if one thought one could prove that the Copernican theory was false by noting that it "seems just obvious" that the Earth doesn't move and the Sun does.

There's lots else that's fun and interesting here, too. In particular, the details of how The Tuned Deck card trick work are fascinating in the context of explaining an evolved, "kludged" system (as opposed to a designed system). Want to know why? Then read the book!

Daniel C. Dennett.
Breaking the Spell: religion as a natural phenomenon.
Allen Lane. 2006

rating : 3.5 : worth reading
review : 2 October 2006

Dennett does not start out questioning the truth of religion, or of religious beliefs. And in fact, by the time he gets to that point, he has put enough in place to show that such a question is almost unstatable in any kind of coherent manner, and certainly unanswerable, in the current situation, and so gives it short shrift. Instead, his main thrust is questioning the validity of some of the other claims made of religion, such that it "gives meaning to life", makes people "good", makes the world "a better place", and so on. He points out that these are very important claims, and so they should be carefully investigated: if they are true, secularism is in danger of destroying something important; it they are not true, and religion does more harm than good, then this is also important to know. With the emphasis always on "know", on having evidence, not on being based on (blind) "faith".

Dennett takes, not surprisingly, a philosophical stance to these questions. In particular, he insists that there should be no questions ruled "out of bounds" to enquiry, and that any attempt to do so should be gently but firmly rebuffed. However, he notes that many religions have built a strong shell of protection from this very sort of enquiry:

[p207] If I were designing a phony religion, I'd surely include a version of this little gem---but I'd have a hard time saying it with a straight face:
If anybody ever raises questions or objections about our religion that you cannot answer, that person is almost certainly Satan. In fact, the more reasonable the person is, the more eager to engage you in open-minded and congenial discussion, the more sure you can be that you're talking to Satan in disguise! Turn away! Do not listen! It's a trap!
What is particularly cute about this trick is that it is a perfect "wild card," so lacking in content that any sect or creed or conspiracy can use it effectively.

His olive branch to the religious for doing all this questioning is that, if we are to demonstrate that religion is "true" (whatever that means, and it appears to mean something different to everyone), we first have to discount possible natural explanations for religious behaviour and experience. To prove something is a miracle, first demonstrate that it couldn't have been caused by natural means. Unfortunately, even if one can demonstrate a possible natural explanation, this wouldn't, of course, constitute a proof that the supernatural claim was false, merely that is was unnecessary. No matter how many magicians demonstrate that they can bend spoons into pretzels perfectly readily, there's always the riposte "yes, but I'm doing it by mind power" (or, worse still, "they are using mind power too, but don't realise it!"). It takes a particular starting point to rule out this kind of thing as a valid argument. The fact that this is a rational starting point doesn't help when arguing with the irrational. Dennett admits that we cannot have a conversation with such people.

The discussion is in two main parts. The first is an historical account of how religion may have arisen, based on evolutionary and memetic ideas. Dennett is not claiming that this is how religion did arise, merely that it is a plausible account of how it could have, based on what we know of evolution, anthropology, and history, deserving of further investigation, either to refute it, or to firm up the details. The second is about belief in belief in God (analogous to belief in democracy, or belief in justice -- and the claim that it is a "good thing" in its own right to believe in such things, because such belief helps to maintain or establish a desirable situation).

One of Dennett's conclusions (if one can say a philosopher ever reaches such a state) is that the case for moderate religious belief is still open (it might do more good than harm, but that has yet to be demonstrated either way), but that the immoral actions of fundamentalists should be not be tolerated in the slightest, and that they should be roundly condemned for such actions (of harming other people, and in particular, of harming their own children by denying them the possibility of informed consent). He also insists that the onus of condemnation rests squarely on the shoulders of the moderates whose religion they are hijacking, but who appear not to be doing so because of the insidious memetic protection that all religion has gained, even these overtly toxic ones. Religion may be deserving of respect, but not merely because it is religion.

Dennett claims to be writing this book to help with some of the problems that (fundamentalist) religion, in a variety of flavours, is causing around the world today. Despite the underlying urgency, this is a calm, well-argued, thoughtful, and, in places, humorous account, and I found it an interesting and thought-provoking experience. However, I am most assuredly not a member of Dennett's target audience. I do rather doubt whether any members of his target audience, even if they managed to get past the title, would get past even the first page. He is simply starting from an impossibly different axiom base: that reasoned enquiry is the only reasonable way to advance and achieve understanding. (I would like to think that I fully subscribe to this axiom base myself, that nothing is "sacred", nothing is beyond question. However, I am awaiting the day when I have it brought home to me that I do in fact have some inviolable core of "truths" that I believe should not be questioned -- will I recant this axiom base, or joyously (or uncomfortably?) realise I must start a new level of questioning? Or has this event already happened, and I have some powerful mimetic coat of protection that has shrugged it off, without my even noticing? A sobering thought.)

Matthew M. Hurley, Daniel C. Dennett, Reginald B. Adams, Jr..
Inside Jokes: using humor to reverse engineer the mind.
MIT Press. 2011

rating : 3 : worth reading
review : 4 February 2014

Some things are funny—jokes, puns, sitcoms, Charlie Chaplin, The Far Side, Malvolio with his yellow garters crossed—but why? Why does humor exist in the first place? Why do we spend so much of our time passing on amusing anecdotes, making wisecracks, watching The Simpsons? In Inside Jokes, Matthew Hurley, Daniel Dennett, and Reginald Adams offer an evolutionary and cognitive perspective. Humor, they propose, evolved out of a computational problem that arose when our long-ago ancestors were furnished with open-ended thinking. Mother Nature—aka natural selection—cannot just order the brain to find and fix all our time-pressured misleaps and near-misses. She has to bribe the brain with pleasure. So we find them funny. This wired-in source of pleasure has been tickled relentlessly by humorists over the centuries, and we have become addicted to the endogenous mind candy that is humor.

This started life as Hurley’s dissertation, but fear not: it is not some dull stodgy academic treatise. Despite being peppered with jokes, however, it is also not a side-splitting read. It is instead a clearly written in-depth account of the authors’ evolutionary cognitive theory of humour. The overarching argument is excellently summarised in the back-cover blurb (above); in a little more detail, it runs:

This thesis is introduced in the context of an emotional and cognitive model of how we think. Emotion is clearly important in things like “fight or flight”, but also for many other aspects of our life. Certain forms of brain damage that inhibit emotions make it difficult or impossible for those people to make decisions: not because they can no longer reason, but because they have no urge to make a decision. Thinking itself is such hard work, we need emotions to encourage us to do so.

p79. Boredom has its place in driving us out from cognitive malaise. Though curiosity inspires our cognitive apparatus into detailed exertion surrounding particular as-of-yet-unexplained regularities, we would scarcely commence toil at all without the dull pain of boredom to keep us from the simple irresponsibility of just doing nothing. If there is no pressing topic to think about, we still think, and incessantly so, because it hurts not to.

Although we think incessantly, at times we need to think quickly, and to make a decision before we have all the information. There is an evolutionary advantage to literally jumping to conclusions: those that pondered more deeply were eaten. But if we don’t have all the information, inevitably we will make mistakes, no matter how well evolution has honed our conclusion-jumping heuristics into rational emotional behaviours.

p82. Choosing how to behave under uncertainty requires a heuristic choice process. Good heuristics give excellent approximations much of the time. But, in the (restricted-by-design) areas where they fail, they give predictably—even pathologically—poor results. The emotions are rational, but the system is a heuristic driver of behavior that operates on incomplete information; so we must accept that the emotions will fail us in some ways, such as overreactions and addictions, that are irresolvable.

Mistakes can be dangerous, as they will include false information about the world, which later will pollute the very reasoning we need in order to survive. So we have evolved mechanisms to help us correct these mistakes.

p120. The need, then, is for a timely and reliable system to protect us from the risks entailed by our own cleverness. Discerning and locating these mistakes would have the immediate payoff of allowing current reasoning to progress without an error (before we act on such errors), but would also provide a legacy for the future, keeping a fallacious conclusion from becoming registered as verity in long-term memory. A mechanism for consistency checking is indispensable for a system that depends crucially on data-intensive knowledge structures that are built by processes that have been designed to take chances under time pressure. Undersupervised and of variable reliability, their contributions need to be subjected to frequent “reality checks” if the organism that relies on them is to maintain its sanity.

However, this error correction is expensive, and is competing for the same resources that the original thought uses. Something expensive needs some reason to happen. This is the key to the authors’ model: the reason is the evolved reward mechanism of mirth. Sweet foods are not intrinsically “sweet”; we experience them as sweet (a pleasant emotional response) because we are evolved to find them so: sugar was a rare and valuable energy source. Similarly, jokes are not intrinsically “funny”; we experience them as funny (a pleasant emotional response) because we are evolved to find them so: error correction is a valuable cognitive function.

Having introduced their thesis, the authors then subject it to various challenges. It needs to account for the diverse range of things we find funny, and yet should also explain why closely related things are not funny. They do this in considerable detail, picking apart and analysing a wide range of humorous and related event. Of course, picking apart a joke destroys its humour; interestingly, the theory even explains why it destroys the humour.

This is a fascinating and well-argued account of a particular aspect of our evolutionary heritage. Recommended. (Some of the example jokes included are even funny.)

Daniel C. Dennett.
Intuition Pumps and Other Tools for Thinking.
Allen Lane. 2013

Daniel C. Dennett.
From Bacteria to Bach and Back: the evolution of minds.
Allen Lane. 2017

What is human consciousness and how is it possible? These questions fascinate thinking people from poets and painters to physicists, psychologists and philosophers.

This is Daniel C. Dennett’s brilliant answer, extending perspectives from his earlier work in surprising directions, exploring the deep interactions of evolution, brains and human culture. Part philosophical whodunnit, part bold scientific conjecture, this landmark work enlarges themes that have sustained Dennett’s career at the forefront of philosophical thought.

In his inimitable style, laced with wit and thought experiments, Dennett shows how culture enables reflection by installing a profusion of thinking tools, or memes, in our brains, and how language turbocharges this process. The result: a mind that can comprehend the questions it poses, which has emerged from a process of cultural evolution.

An agenda-setting book for a new generation of philosophers and thinkers, From Bacteria to Bach and Back is essential for anyone who hopes to understand human creativity in all its applications.

Papers/Articles : reviews

Daniel C. Dennett. Murmurs in the Cathedral. TLS. 29 September. 1989

Dennett's substantial 4000-word review of Roger Penrose's The Emperor's New Mind is complimentary about the scientific themes covered in the book:

... a pedagogical tour de force, with some dazzling new ways of illuminating the central themes of science ... His discussion of phase spaces, for instance, and his development of the rationale for the second law of thermodynamics, are particularly refreshing.

However, Penrose's main thesis, for which all this scientific exposition is mere supporting argument, is that algorithmic computers cannot ever be intelligent, because our mathematical insights are fundamentally non-algorithmic. Dennett is having none of it, and succinctly points out the underlying fallacy, that, even if there could not be an algorithm for a particular behaviour, there could still be an algorithm that was very very good (if not perfect) at that behaviour:

The following argument, then, in simply fallacious:
  1. X is superbly capable of achieving checkmate.
  2. There is no (practical) algorithm guaranteed to achieve checkmate,
  3. X does not owe its power to achieve checkmate to an algorithm.
So even if mathematicians are superb recognizers of mathematical truth, and even if there is no algorithm, practical or otherwise, for recognizing mathematical truth, it does not follow that the power of mathematicians to recognize mathematical truth is not entirely explicable in terms of their brains executing an algorithm. Not an algorithm for intuiting mathematical truth - we can suppose that Penrose has proved that there could be no such thing. What would the algorithm be for, then? Most plausibly it would be an algorithm - one of very many - for trying to stay alive, an algorithm that, by an extraordinarily convoluted and indirect generation of byproducts, "happened" to be a superb (but not foolproof) recognizer of friends, enemies, food, shelter, harbingers of spring, good arguments - and mathematical truths.
it is disconcerting that he does not even address the issue, and often writes as if an algorithm could have only the powers it could be proven mathematically to have in the worst case.