Content-type: text/html Downes.ca ~ Stephen's Web ~ Consciousness

Stephen Downes

Knowledge, Learning, Community

Half an Hour, Nov 14, 2017

Consciousness seems to be mysterious to most people. How does subjective experience arise? What is the relation between the perception of redness, say, or the thought that "Paris is the capital of France," and the purely physical mechanisms that philosopher Daniel Dennett believes - and I believe - constitute human processes of thought?

Dennett addressed these problems most famously in his book Consciousness Explained. Many readers suggest that the book "might better have been titled 'Consciousness Explained Away.'" The suggestion that there is no such thing as consciousness (at least as traditionally understood) is a hallmark of a materialist philosophy, but at the same time, difficult to reconcile with our own experiences of day-to-day life.

Most recently Dennett published a follow-up called From Bacteria to Bach and Back: The Evolution of Minds in which he attempts to describe how consciousness came about. The responses have been mixed, but I landed on a criticism by David Bentley Hart which takes to task not only the current work but the totality of the materialist theory of mind.

I will not be attempting a defense of Daniel Dennett against David Bentley Hart in this post, though I certainly think that Dennett is more right than wrong. Rather, my purpose here is to use Hart's article criticizing Dennett as a frame through which to offer an account of consciousness.


--- My Position

Let me give you my answer first, and then we'll work our way through Hart's critique, a critique which will allow me to position my response against most major objections (and let me be clear, by 'my position' and 'my response' I do not mean to claim originality or priority - these just happen to be what I believe, and almost certainly every point has a basis in previous writing).

My position, then, is that consciousness is sensation, no more, no less.

Now you might believe that there's more to consciousness than sensation, and that's one question we'll address.

And you might wonder how sensation arises out of physical processes, and that's another question we'll address.

But we need to begin by being clear about what consciousness is, because everything else will follow from that. And for that, we go back to the original statement that consciousness is sensation.


--- Consciousness

Consciousness is distinguished, to me, by two attributes: its wholeness, and its simplicity. Let me explain each in turn.

By 'wholeness' what I mean is that, for a person, consciousness is everything. Everything you know, everything you believe, everything you feel, everything you wish for: that's a part of consciousness. When you say 'Paris is the capital of France', that is a part of your consciousness. There may be a city called Paris out there in the world that is the capital of France, but that's beyond your consciousness.

This is important. When I'm talking about consciousness, nothing about the city of Paris, out there in the world, constitutes a part of that consciousness. I might, perhaps, refer to a city of Paris as part of an explanation of consciousness, but that reference, and that explanation, are constituted entirely of consciousness. An explanation of consciousness is a story I tell myself about where my sensations may have come from, but that story itself remains entirely within the realm of consciousness.

By 'simplicity' what I mean is that consciousness isn't some special property of things over and above everything else. To explain this, I'll draw an analogy, and then apply the analogy to consciousness.

Take fire. We all know what fire is: we see it when we light a match or a lighter, or when we create a campfire. We see flames, we feel heat. We also know that fire occurs when a combustible material is oxidized. When wood burns, for example, the complex hydrocarbon in wood is reduced to gases (steam, carbon monoxide, carbon dioxide) and char (charcol, ashes, smoke). That's it.

You might ask, "but where is the fire?" Because, after all, you see a glowing flame, a plasma, that doesn't resemble gas or ash at all. The response is, it's glowing carbon, in the form of soot. There is no 'fire' over and above the reduction of wood to gases and char via a chemical reaction. (Yes, this is very similar to Ryle's argument about category mistakes, and much of the discussion there would apply here).

It's the same thing with consciousness.

When a sensation occurs, our sensory neurons are stimulated. This stimulation proceeds through layers of neurons to produce what we commonly call 'perceptions', that is, our common everyday experience of colours, objects, people, and the like. Consciousness is this process. There's no 'fire' over and above the stimulation of layers of neurons.

Now I need to be clear about this. I am not saying that the stimulation of neurons causes sensations or perceptions. This is what might be called an eliminativist account: consciousness is the stimulation of neurons, just as sensory experience is the stimulation of neurons. Or, more accurately: consciousness is sensation, and the existence of sensation is explained via a theory about the stimulation of neurons.

But remember: this is an explanation. I'm not saying consciousness is really nothing more than the stimulation of neurons. I'm saying that consciousness is sensation, and that the best explanation (in science, in culture, in my experience) is that sensation is the stimulation of neurons.


--- Intentionality

One more thing, and then we'll get to the Hart article. And that's the question of why consciousness seems mysterious.

The problem of consciousness, really, is the problem of intentionality. That is, the problem revolves around the question of how consciousness turns out to be about something.

Put more concretely: if everything is subjective, if everything is sensation, how do we get to the point where we're talking about objects like neurons, persons, the city of Paris, or even ourselves? Because, logically, if all we have is sensation, then nothing we talk about can be anything over and above sensations.

I think there are two related responses to this concern.

The first, I think, is an error about how thought and perception work. It is a presumption, inherited from people like Descartes, and reified by the logical positivists, that our thoughts and beliefs about objects and principles and the world at large are the result of a logical inference from sensations to beliefs. The definitive statement of this argument is probably found in A.J. Ayer's Language, Truth and Logic.

We now know that there is no such inference. We can't infer from concrete experience to universal propositions (this is what Chomsky called Plato's problem). We can't generalize to laws of nature (this is the problem of induction). We can't even distinguish between logical statements and statements about experience (this is show in Quine's two dogmas). There is no logical inference from experience to objects forthcoming, and there never will be.

The second, I think, is that we experience our thoughts about objects and the like in the same way we experience sensations and the like: through the stimulation of layers of neurons. Not all of our thought processes are intentional (in the sense of being something that we did deliberately or through some sort of logical or constructive process). Much (maybe all) of our thought occurs naturally, through cascades of signals from one layer of neurons to the next.

We might say, most accurately, that our experiences and beliefs about things like objects and generalizations (and language and universals) are the result of the self-organization of layers of interconnected neurons. That, to me at least, seems to be the best explanation.

Our thoughts about objects are not representations of the external world, they are not inferred from experience, they are sensations of the external world (which J.J. Gibson would call direct perception), and are experienced directly.

This is the theory and if you want you can stop here. The rest of this post will deal with the details as raised through argumentation offered by Hart against Dennett.


--- Descartes and Dualism

Hart begins by describing "Dennett's implicit admission on page 364 that no philosopher of mind before Descartes is of any consequence to his thinking." Hart argues in response that "The whole pre-modern tradition of speculation on the matter — Aristotle, Plotinus, the Schoolmen, Ficino, and so on — scarcely qualifies as prologue."

Of course, he is right. Philosophy did not begin with Descartes. But we have to understand that it took an important turn with Descartes, and it is this turn that produced many of the problems that we are talking about today. There are different ways of describing this turn; I'll pick one that works for us, though of course we could talk about it in a lot more detail if you wanted.

Before Descartes, pretty much all of philosophy was Aristotlean, that is, based on the philosophy of Aristotle. Aristotle's philosophy (and I'm eliding a lot here) is the idea of the nature of things, or essences. The idea is that there is a single substance (the essence of things) where individual things (or types of things) are distinguished only by contingent properties (or 'accidents') of things.

This is similar to how we define things and categorize things today: what is the essence of a thing (ie., what makes it the same as other things of its type), and what is the accident of a thing (ie., what makes it different from other things of its type). Hence, for example, we say that "a robin is a bird (cold-blooded, lays eggs, flies) with a red breast."

The big debates in pre-Cartesian philosophy revolved around questions like whether essences have independent existence, or exist only as instantiated in physical objects (this is what Ockham addressed); and of the nature of spirituality as essence or accident (this is part of the problem of the trinity, "that God, Jesus, and the Holy Spirit are all the same entity, until they aren't").

Descartes introduces dualism into the mix, the idea that there are two substances, the mental and the physical. Before Descartes, the problem of consciousness doesn't exist. The spiritual part of a person might be a part of the essence of a person, say, but it would not be thought of as separate from the physical instantiation of the person.

The Cartesian approach solved a lot of problems with the pre-Cartesian approach (and oh, it had a lot of problems) but it introduced with it a range of new problems. It is these new problems Dennett addresses, which is why he is not concerned about the work before Descartes.


--- Descartes and Scepticism

The other problem that Descartes introduced to the world was scepticism. Not that Descartes was the first sceptic - the ancient Greeks had Pyrrhonism -  but it was Descartes who created the problem of inferring from the mental realm (experience) to the physical realm (reality). That we could do it is evident, since we all acknowledge the existence of physical objects, but (according to Descartes) we need help. And thus was born the Cartesian argument for the existence of God, not understood as a 'prime mover' or 'first cause', but rather apprehended directly as "the mark of the Maker stamped upon his work."

The creation of scepticism creates the requirement that we infer from the mental to the physical. This requirement was addressed first through the Cartesian method (what we know today as 'analysis and synthesis'), then through inductive reasoning, and finally through to the scientific method we know today. The Cartesian approach is based on reason and rationality (and so it is known as 'rationalism'), and the idea that knowledge is founded on reason persists today (as what John Ralston Saul calls Voltaire's Bastards).

I will not for a minute dispute the historical significance - and the overall utility - of the Cartesian method. Everything we enjoy today, from central heating to feed processing to computing to space travel - is the result of the Cartesian method, the scientific method, and the application of reason to problems. Descartes, for better or worse, launched us from the position of being dependent on nature and the universe to standing astride it as its master.

But it's wrong. Language, science and reason are incredibly useful constructs, but they are artifacts, things we created, rather than statements about the essential nature of thought, perception and humanity. We do not actually perceive, learn and know through a process of scepticism, reason and construction. And there is not, in human nature, a separate mental realm that reasons abstractly about the physical realm.


--- Monism and Dualism

There is the question of the choice we have in the current debate. Hart expresses it this way: the mental "was soon demoted to sheer illusion, and the mind that perceived it to an emergent product of the real (which is to say, mindless) causal order." As a result, Dennett is "is content with the stark choice with which the modern picture confronts us: to adopt either a Cartesian dualism or a thoroughgoing mechanistic monism." Which is a shame, says Hart, because "both options are equally absurd."

Basically the choice we are presented with is as follows: either the mental (or the spiritual; pick your vocabulary) is real, and we have dualism. Or only the physical is real, and we have monism. The logical positivists (and the scientists in general, including Julie Payette), opted for monism (they even named their flagship journal after it). The non-scientists, the religious, and most of the public, retained dualism, and reconciled their philosophy with belief in a non-physical and sometimes spiritual concept of a soul or personal (or animal) spirit.

There is a third option, of course, and it is the option that only the mental is real. Today this position is usually derided as 'idealism' or 'subjectivism' or 'solipsism'. It is the philosophy, we were taught, espoused by 'the mad bishop' George Berkeley, who proposed that ordinary objects are ideas, and nothing but ideas. Berkeley (and the other empiricists Locke and Hume) were called 'sceptics'. But unlike the rationalists (and Kant, and the logical positivists) they were sceptical of the idea that the physical is separate from, and inferred by, the mental.

We need to be clear about what Descartes did, from the perspective of the empiricists. He convinced us that there are two types of things (two types of experience or perception) such that one is real and the other is not, and that we would have to infer from the one (the mental) to the other (the physical). But where does this requirement come from? 

There is no requirement to infer to the existence of objects. Objects are no more or less real than our perceptions of colour, our feelings of hot or cold, our hearing of middle C, or our smelling of maple-smoked bacon. Mmmm, maple-smoked bacon. Oh, sorry, I digress.

We might want to explain how we perceive objects. but this is no different from the task of explaining how we sense movement or touch or whatever.


--- Causal Closure

Hart depicts Dennett as a monist (or as a materialist, more accurately) and one of his critiques of the position is what might be called causal closure. Here's what Hart means, precisely:

"He still thinks it a solvent critique of Cartesianism to say that interactions between bodies and minds would violate the laws of physics. Apart from involving a particularly doctrinaire view of the causal closure of the physical (the positively Laplacian fantasy that all physical events constitute an inviolable continuum of purely physical causes), this argument clumsily assumes that such an interaction would constitute simply another mechanical exchange of energy in addition to material forces."

I'm not sure what Hart means by a "solvent" critique, but it's certainly a sound critique. But we need to separate between two ideas: first, the idea that all phenomena are material phenomena, and second, the idea that all causes (and effects) are deterministic.

The latter was called by James Gleick the "Laplacian fantasy": "Relativity eliminated the Newtonian illusion of absolute space and time; quantum theory eliminated the Newtonian dream of controllable measurement process; chaos eliminates the Laplacian fantasy of deterministic predictability" (you can read about Laplace's demon here).

But let's be clear: chaos theory is not a non-materialist theory. You can have both non-determinism and materialism together, at the same time. Chaos theory stems from the idea that very small (but still physical) causes can have very large (and unpredictable) effects. The idea that materialism means determinism stems from the idea (the analogy, really) that the mind is a machine. But there's no substance to this analogy.

But what of the first idea, what Hart calls "the causal closure of the physical?" Can there be non-physical causes, that is, causes from beyond the matter-energy continuum that is the foundation of modern physics today?

No, and for a simple reason. If science is wrong about this, then science is wrong about everything. There is no room in contemporary science for non-physical causes or non-physical explanations of behaviour. No room. Yes, there are questions that remain unanswered, but this isn't one of them. The causal structure of the universe involving the various combinations of matter and energy form a single continuous understanding from the movement of mastodons to the tiniest influences of micromatter on the standard model. What remains unexplained - if anything - is so small it cannot be detected by the most sensitive equipment in the world. It is certainly not substantial enough to create behaviour in a human without the violation of numerous natural laws of physics.

Does this mean that there's no God, there's no soul, and there's no life after death? Well, maybe. That's what I think. But I could be wrong about that. But I do know this: if there is a God, a soul, or a life after death, they're all a part of the natural world, not the supernatural world. As deGrasse Tyson argues, science doesn't automatically contradict religion. It just contradicts the version of religion as magic.


--- Explanations and Universals

It seems odd to read Hart describe Dennett's method as "essentially fabulous." He represents Dennett as though he is the one proposing some sort of supernatural explanation.

According to Hart, "he (Dennett) constructs a grand speculative narrative, comprising a disturbing number of sheer assertions, and an even more disturbing number of missing transitions between episodes."

Here's Hart's criticism: "Rather, however, than attempt to explain nature in terms of a "mind-like" order of rational relations, as Aristotelian tradition did, Dennett seeks to do very nearly the opposite: to reduce mind and nature alike to a computational system..."

This raises the question of just what constitutes an explanation of mind and nature. We'll return to this, but it's important to lay some groundwork.

Both the Platonic and the Aristotlean approach are based in the idea of universals; the difference was essentially that Plato believed they had a separate existence, and were apprehended by reason alone (this is the point of the allegory of Plato's cave) while Aristotle thought, as mentioned above, that they were embodied in substance. Either way, the nature of, and the behaviour of, things (language, logic, mathematics, physics) was explained by these universals.

This basis for explanation carried over through the Cartesian revolution and has come to underlie modern scientific method as described by the logical positivists. The idea was that we would observe regularities in nature (perhaps through experimentation or the process of trying to solve some problem) and from this infer through a process of induction (or maybe, abduction) to a universal principle. We would then test, verify, and confirm (or, at least, fail to falsify) this principle, which after standing the test of time, was elevated to the status of scientific theory, and ultimately, law of nature.

The problem is, it doesn't work. It fails in two regards.

First, there's the aforementioned problem of induction. There is no principle of logic or reason that will allow an inference from concrete experience to abstract universal. This is more than just an empirical problem, but as Nelson Goodman showed (with his 'new riddle of induction') its a problem for language and logic as well.

Second, even if we could derive a universal principle (by, say, guessing) there is no way to confirm, verify, or falsify it through experimentation of experience. This is known as the problem of confirmation, or generally, Hempel's problem. Falsification, Karl Popper's answer to the problem of confirmation, is susceptible to similar criticisms.

Scientific explanations today are subject to principles of evaluation that have nothing to do with the inference to or confirmation of universal principles. These principles include consistence, clarity, comprehensiveness, that is, the completeness of an explanation over a variety of phenomena, and parsimony (the principle that the simplest explanation should be accepted, which is the modern equivalent to Ockham's razor).

An explanation need not appeal, therefore, to some "order of rational relations". It need not, first, because there is no need to refer to some hierarchy of Aristotlean universals, and second, because it might not be based on "rational relations" at all.


--- Teleology

The core of Dennett's ideas - an my own - lies in the idea of self-organization. Our perspectives are different. He is, if you will, looking from the outside, finding no evidence of an inside, while I'm looking from the inside, finding no evidence of an outside. But the mechanisms we describe are basically the same: the explanation of consciousness is that it is a physical process, the interaction of neurons with the world outside and with each other.

The mechanisms of evolution are in many ways the same as the mechanisms of consciousness. Evolution is the result of many small things interacting with each other and with order emerging out of this chaos. It is not the result of some design, nor is it the result of some goal of objective.

As Hart says, "Dennett seeks to do very nearly the opposite: to reduce mind and nature alike to a computational system, which emerges from 'uncomprehending competences,' as he calls them — small, particulate functions wholly unaware of the larger functions they accomplish in the aggregate."

People, including Hart, have difficulty with this idea. Even evolution is described in terms of some sort of desired outcome or goal, with adaptations serving this or that function. The very notion of "survival of the fittest" embodies the idea that we have, not only as life forms but as species, the objective of survival. Maybe this is true of humans (but given our recent behaviour I doubt it) but it is certainly not true of rabbits or bacteria, no matter how prolifically they breed.

And I think that's why Hart attributes to Dennett a kind of teleology. "For Dennett, all evolutionary developments occur because they incorporate useful adaptations." But if this is true of Dennett (and I don't think it is) it is nonetheless incorrect. Adaptation may favour the utilitarian, but there is nothing that prohibits the development of a useless trait (such as an accessory spleen, or unused genes) or an unproductive idea (such as self-mutilation, or a belief in ghosts). The utility of an adaptation might explain, after the fact, why it was retained, but not why it developed in the first place.

There's an important distinction here, and we need to draw it, between describing the purpose of something, and explaining why that thing happened. Explanations do not require purpose. They do not require intent. We could have evolved a sixth toe for no reason at all, and kept it for no reason at all, even if after the fact we observe that the evolved traits in species are also those traits that helped them survive.


--- Suddenness 

Having (illegitimately) made teleology an essential part of Dennett's thinking, Hart now depicts it as a failure. "He has no patience for talk of "spandrels" — phenotypic traits that are supposedly not adaptations but byproducts of the evolution of other traits — or of large, inexplicable, fortuitous hypertrophies (such as, say, the sudden acquisition of language) that have no specific evolutionary rationale at all."

What Dennett disagrees with is the idea of suddenness. Self-organization, emergence and evolution are slow phenomena.  We didn't suddenly come into being, consciousness didn't suddenly come into being, and nor either does an idea in our head suddenly come into being.

As described here, "You shouldn't trust your intuitions," he told the philosophers on the Rembrandt. "Conceivability or inconceivability is a life's work—it's not something where you just screw up your head for a second!" He feels that Darwin's central lesson—that everything in biology is gradual; that it arrives "not in a miraculous, instantaneous whoosh, but slowly, slowly"—is too easily swept aside by our categorical habits of mind.

But to Hart, language was a sudden and fortuitous development that creates (if you will) a 'gap' in evolutionary history. He wants us to understand here that language is something that arose out of nothing. We'll revisit the idea of language below, but for now, it will suffice to assert that self-organization can be (and is) non-teleological, not goal-directed, and not coincidentally, non-rational.


--- Self-Organization and Emergence

I don't think Hart represents the concept of 'emergent' properly.

The process of self-organization (he writes) is "an algorithmic distillation and recombination of 'uncomprehending competences.'" But "even the mental and cultural worlds were, it turns out, emergent results of such competences rather than consciously designing or designed realities."

There are two ways of talking about 'emergent results' and I think Hart picks the wrong one.

One way is to think of it as an outcome. Take some eggs and flour and chocolate and ix them and bake them and out of this emerges a chocolate cake. Or take a bunch of biological stuff, and manipulate it, and manipulate it, and out pops a rabbit at the other end. That's the sense I think that Hart means.

The other way of thinking of emergent is to see it as a pattern. Mess around with some ingredients and eventually they take the form of something you recognize as a chocolate cake. Or evolve from one form of biological life to another to another and eventually you end up with something you recognize as a rabbit. That's what I think Dennett means.

It's important to understand this distinction, I think, because it helps us understand the 'suddenness' of the emergence of language (or consciousness, or any of the other phenomena we are discussing). When you are manipulating a pattern of entities, order may 'suddenly' appear out of chaos, but what changed suddenly was not the pattern of entities but rather our perception of them.

Hart presents the process as being something that should lead logically to the result. For example, if we were to suddenly emerge on Times Square in New York, our history of that should be a history of getting closer and closer to that location. If we're in Atlanta, and then suddenly emerge in new York, then something of a miracle happened, or at least, there is a gap in our story that has to be explained.

But on the other picture, what we have is something more like Albert Einstein. What is the story of his evolution? It is not a history of his parents and grandparents getting smarter and smarter until we have one of them who emerges as Albert Einstein. No; great intelligence is just one of the possible configurations of humans, and when it appears it is recognized as such, but its emergence exists because of our recognition, not because its history was somehow leading to this.

A 'mental world' is like Albert Einstein. It's a particularly wonderful configuration of existence that we (rightly) recognize as being something special.


--- Qualia

We return now to the question of the two types of things defined  by Descartes, or as I characterized them, the two kinds of perceptions. This time we approach it from the side of subjective experience, and in particular, sense perception.

There is a notion of what it feels like to perceive something. The phenomenon of pain offers a great example. When we are in pain, it hurts, but when we observe someone else in pain, no matter how closely, it does not hurt (though the activity of mirror neurons may make us flinch a little). So clearly there is something different about the subjective experience of pain that is not observable as the physical experience of pain.

These subjective experiences, as Hart tells us, are known as qualia. They are, as he tells us, "direct subjective impressions, such as color or tone." And the problem is that "There is simply no causal narrative — and probably never can be one — capable of uniting the phenomenologically discontinuous regions of 'third-person' electrochemical brain events and 'first-person' experiences, nor any imaginable science logically capable of crossing that absolute qualitative chasm."

From where I sit, the argument here is no more amazing than the observation that I am not other people. If I were, I would feel their pain, but I'm not, so I don't. But for Hart (and many millions of others besides him) this produces an irreducible difficulty.

Where does this difficulty arise, though? I think it lies in the supposition that my observation of someone else's experience should feel (to me) the same as my experiencing of the same type of experience for myself. But why would I suppose this?

Let's make it me that is having the pain while I observe the process externally. We'll imagine that my eyes have been separated from my brain ("just think of it as stretching the optic nerves") though a process such as described in Dennett's Where Am I? So, when you prod me with the red-hot iron, causing pain, I am observing my brain from the 'third person' perspective (indeed, because my eye is now detached, I can use a microscope and get as detailed a view as I want).

Now it is going to become pretty clear that the pain that I feel and the electrochemical reaction I observe are one and the same. Every time you prod me, the neurons fire, and this firing I experience as pain. They are not two separate things. They are one and the same thing.

The problem of qualia arises only because I am making inferences about the external world. It's like saying that my eye doesn't exist because I can't see my eye directly. Strictly speaking, that's true, I can't see my eye directly - I have to use a mirror or take a picture or some such thing. But my seeing of anything is my experience of the eye; my feeling of anything is my experience of neurons firing.

The inference here is not to the qualia from the physical phenomenon. It's from the qualia to the physical phenomenon. And that inference, as I've argued above, isn't an inference at all; it's just the way we perceive the world.


--- Unity of Apprehension

In what was probably the greatest feat of philosophical reasoning in history, Immanuel Kant, through what he called the 'transcendental deduction', derived the existence of space and time as the "necessary conditions for the possibility of perception".

This work appears (greatly abbreviated) in Hart as the 'unity of apprehension". He writes, "there is the irreducible unity of apprehension, without which there could be no coherent perception of anything at all... It is a unity that certainly cannot be reduced to some executive material faculty of the brain, as this would itself be a composite reality in need of unification,' and so on, turtles all the way down.

The phrase "unity of apprehension" actually originates in theology and is the idea that the existence of God is inconceivable; to apprehend the self is to apprehend the world is to apprehend God, all at the same time. In the present more secular sense I'll read it as the apprehension (as suggested by Kant) of space and time and a necessary part of our own existence.

So Hart's argument here is essentially that there is nothing in the physical account of mind and consciousness that could account for our perception of the unity of space and time. "Even if we accept that the mind merely represents the world to itself under an assortment of convenient fictions, this would involve a translation of sense data into specific perceptions and meanings."

If we assume the separation of our thoughts into separate categories of (say) qualia and categories (such as space and time) then, yes, we must attempt the impossible translation from one to the other. But if they aren't separate and distinct - if there's no difference between a perception of 'redness' and a perception of 'time', then the unity of apprehension is no big deal.

Hume said we arrive at such ideas easily and naturally, through, he said, custom and habit. We no more infer to the existence of space and time than we infer to the existence of ourselves, to pain, to the colour red. It was this line of thought that prompted Kant to undertake the transcendental deduction. But it wasn't really necessary.


--- Intentionality Again 

There are two senses of 'intentionality' and Hart hits both of them in a singe paragraph (sometimes they are spelled differently, a custom I will adopt here):

First, there is intensionality, which is the 'aboutness' of our thoughts. The modern discussion of intensionality follows from Kant and is probably best represented in Husserl's phenomenology, which distinguishes between the meaning (or sense) of a word, as distinct from its reference, or its relation to states of affairs in the external world.

Second, there is intentionality, which is understood in the common sense of our 'intending' to do something. In philosophy, and in particular in the context of the problem of consciousness, it has to do with the idea of free will versus determinism. If consciousness is purely physical, how can someone intend to do something? We could include in this category other mental phenomena such as 'wanting' and 'desiring'.

As I said, Hart brings both into the discussion: "This problem, moreover, points toward the far more capacious and crucial one of mental intentionality as such — the mind's pure directedness (such that its thoughts are about things), its interpretation of sense experience under determinate aspects and meanings, its movement toward particular ends, its power to act according to rationales that would appear nowhere within any inventory of antecedent physical causes."

All of these focus on the relation between ourselves and the 'other'. We see reflections of this problem throughout phenomenological and existentialist philosophy - I and Thou (Buber), Being and Time (Heidegger), Being and Nothingness (Sartre). The relation of the self to the other is also frequently reified as the will, and in for example The World as Will and Representation (Schopenhauer) and The Will to Power (Nietzsche).

So the problem raised here by Hart has as much to do with the nature of our conception of self as it does with the nature of how our thoughts come to be about something or to express intentions and desires.

Descartes begins with the self. Cogito ergo sum. "I think, therefore, I am. What am I? A thing that thinks."

How does he know this? He doesn't. He made it up! He has created out of nothing a second substance, a thinking substancethat will think about things, that will intend things and want things, and more. But what if there is only (if you will) one substance in human nature? What if what we are, really, is a thing that feels or a thing that senses?

Now, instead of a mystery, we have a story.

Why do we talk about objects? Because we perceive them as such. Our experiences include not just sense-data or qualia, as described above, but also phenomena such as space and timemulti-dimensional objects, and more. We have the five canonical senses, but also senses of balance, of heaviness or weight, feelings, emotions, sentiments, and more.

And we have senses that are not only outwardly directed (so to speak) but also inwardly directed. We have (as Hume argued) senses of taste, of pleasure and disgust, anger and fear, happiness and pain, even abstracts such as justice and fairness. We often regard these as types of opinion - "a matter of taste" - but from the perspective of the person having the sensation (that is, you and I!) there's nothing opinionated about it. If something feels disgusting to me, it just is disgusting; the best I can accept is that some other people find it less so.

So what, then, are intentions, desires and needs, other than sensations? These are feelings, directly apprehended, no less so that the redness of an apple of the lusciousness of the moon.


--- The Senses

In this section I offer an explanation of how we have the senses we do. In addition to the caveats about explanations that I addressed above, I want to caution as well that this section is somewhat speculative, and that the results of actual human physiology research may differ somewhat in the details.

That said, here's the story: the human brain is composed of layers of connected neurons. The top layer (or outermost layer) is the sensory layer. Here neurons are stimulated by phenomena in the external world. This layer is where we find the surface layer of the retina, the oderant receptorshair cells in the inner ear, and so on.

We have sensory neurons throughout the body in the form of nerve cells. Some of these detect interactions on the surface of the skin, while others detect what happens inside the body - in the stomach, in the chest, and so on.

These neurons are densely interconnected with the next layer of neurons, and the next, and so on. In the visual cortex, for example, there are six layers of neurons connected to input from the multiple layers of the retina. And these are thence connected to even more inner layers of neurons.

Conscious experience is the firing of these inner layers of neurons.

A single neuron in this inner layer may correspond to a pattern of neurons on the retina. Or it may correspond to a collection neurons in the eye, ear and mouth. What we sense may be described as a colour, a taste, an object, a period of time. The main point here is that there is no inference taking place here. There is nothing more or less than the activations of interconnected neurons.

Now here's where it gets fun. Beyond these inner layers there are more inner layers, and these neurons are also activated. But these layers are so abstract that their activation doesn't correspond to any coherent description of the senses. It might be (and indeed is likely) that we do not sense these activations at all. But deeper and deeper the layers go, with interactions and activations constantly happening as a result of the original sensory interaction.

At a certain point, we reach the last layer, but the activations don't stop. Instead, they begin to migrate back through the layers of neurons. Now the deep abstract activations begin to interact with the inner layers. Our perceptions are informed not only by the input of the senses going in one direction, but also by the backward-propagation of activations going in the other direction.

We don't just sense what we see, feel, touch, etc.; our sensations might arise out of interactions among neurons anywhere in the brain. We have sensations that don't correspond to any coherent sensory input. We see things. We hear voices in our head. Sometimes we have hallucinations. Sometimes we dream. All of these are different types of ways the neurons in those inner layers can be activated.
 

Image from Anil Seth, The Neuroscience of Consciousness,
https://www.youtube.com/watch?v=xRel1JKOEbI

What's interesting is that, in the study of neural activation in conscious states, the more complex the interaction, the more conscious the person. Consciousness isn't simply an on-off. We can be unconscious, partially conscious (as in REM sleep), fully conscious, or even hyper-conscious (as when on certain drugs such as LSD).
 

Now this is a very abstract representation of something that is much more complex. There are many more layers. There is not just one single set of layers; the structure of the brain is modular. There are different types of senses in different parts of the brain (the famous 'lizard brain' for example).

And even more importantly, the blue dots (the neurons that are not part of our consciousness) are outside our experience. Everything we know, everything we are, everything we do, our entire mental life - all of that takes place in the red neurons, the activation of which is conscious experience. which of our neuron is a red neuron and which is a blue is a matter of empirical discovery, and even more, is probably different for every person. What we feel - what we sense - seems to vary a lot from person to person.

Why would the brain feed us sensory images that are 'not real' (that is, not caused directly by external sensations)? Hermann von Helmholtz suggested that the brain is a prediction machine. As Raese writes, "What we perceive is a combination of top down knowledge based prediction and bottom up incoming sensory evidence (signals from sensory receptors)." This is what we would expect of a neural network. This is known as the Baysean brain.

--- Rationality

We return to Hart, who after raising the issues of qualia, unity of apprehension, and intentionality, next turns to rationality. "There is the problem of the semantic and syntactic structure of rational thought, whose logically determined sequences seem impossible to reconcile with any supposed sufficiency of the continuous stream of physical causes occurring in the brain."

There are different ways to state this problem, but probably the most straightforward is to assert that the laws (or rules or principles) in the (objective) world are not the same as (and logically independent of) the laws (or rules or principles) of the mind. Thus, for example, if we think of a giraffe, the giraffe in our thoughts will not be subject to (say) gravity the way a real giraffe is. Meanwhile, we can think of things that do not exist in nature, such as the square root of -1.

The supposition that mental states (or representations) are governed by the same principles as physical states is what Pylyshyn calls 'the objective pull',  "the tendency to view the cognitive process in terms of properties of the represented objects (i.e., the semantics of the representation) instead of the structure of the representation itself (i.e., the syntax of the representation)." A case in point would be mental images, which may behave differently from the objects in the world they represent.

Pylyshyn thinks of the objective pull as a fallacy, and anyone who has ever thought of a pink elephant knows that there is an important sense in which Pylyshyn is right. We can have thoughts that obey nothing but the laws of logic, and ignore the laws of the real world.

But the fallacy of the objective pull counts as evidence for irreducible mental states governed by laws of language and logic only if we think that our thoughts are representational states. But what if they're not representational states? Our thoughts of objects could well be simply that: thoughts of objects. They are perceived directly; they are sensations, not representations. So they behave how they behave, and it's when we try to make sense of that that we tell stories about the objective properties of objects..

But what of language, logic and mathematics themselves? These are pure abstractions. It's not possible to sense a pure abstraction; it is by definition something with no sensory properties whatsoever.

I will grant Hart and the other proponent of a 'mental substance' this: if we can't tell a (plausible) story about mathematics, logic and language in the terms of this discussion thus far, then we will have to grant that they may have a point. But I don't think we'll ever get to that point. Non-rationalist accounts of logic, language and mathematics abound.


--- Abstractions

The difference between the principles of language, logic and mathematics on the one hand and our sensory experiences on the other hand is that the former require the use and comprehension of abstractions such as necessity, negation and universals, while the latter are concrete and contingent facts.

Hart raises this difference as follows: "there is the issue of abstraction, and its necessary priority over sense experience," and "primordial and irreducible concepts of causality and of discrete forms are required for any understanding," and "some concept of resemblancemust already be in place," and "the bare concepts of Euclidean geometry," and finally (but not finally), "orientations of the mind, such as goodness or truth or beauty in the abstract."

The best (and only) response is to deny that these principles have the characteristics Hart says they have, and/or to deny that we must prove we derive them in the sense that Hart says we do.

First, none of these concepts is necessarily prior to experience (known in philosophy as 'a priori') despite being represented as such from Plato on down. What this means, specifically, is that none of these concepts is either necessary nor universal (another way of saying the same thing is to say that 'necessity' and 'universality' are useful fictions we create for ourselves). So it's sufficient to say that we perceive instances of the concepts in question.

And, clearly, we do perceive instances of these. We see an instance of abstraction any time we use a rock to stand for a sheep. Causality, as Hume observed, can be seen in a simple game of billiards. The concept of resemblance would occur to us on seeing our first set of twins. Geometry can be as clear to us as lines in the sand. And goodness and truth can be (and, I would argue, aresentiments or emotions, in other words, directly apprehended as perceptions.

The reason for supposing that these cannot be derived from experience is that it is not possible to deduce (or even induce) such concepts of necessity and universality from concrete experience. And while on the one hand I deny that these concepts are necessary or universal, it is still required to show how we might come to think that they are. After all, the concepts of necessity and universality seem themselves to be beyond our experience.

I think that the answer lies in properly understanding how we create abstractions in the first place. The usual way of doing it is to extrapolate from concrete experience: this duck is white, that duck is white, those ducks are white, ergo (after a lot of ducks) all ducks are white.

But that's not how we get abstraction. We get abstraction through a process of subtraction. At any given point in time, our senses detect millions of things. But by feeding activation through layers of neurons, we've get this down to something manageable: a white object with orange beak over here, a white object with beak over there, and so on. Subtract a property ('over here', 'over there') and we get an abstraction ('a white object with orange beak'). Keep subtracting and you get all the abstraction you need to make rationality work.


--- Absolute Qualitative Difference

Hart accuses Dennett of what he calls the "pleonastic fallacy", which he describes as "the attempt to explain away an absolute qualitative difference — such as that between third-person physical events and first-person consciousness — by positing an indefinite number of minute quantitative steps, genetic or structural, supposedly sufficient to span the interval."

It's a bit like walking from India to China, I suppose. No matter how long the journey, no matter how many steps were taken, at some point you just have crossed over the border, and it is that step that we need to look at; all the other steps are merely window dressing. This again is a concept from theology, and in particular, the brand of theology that says there is an absolute qualitative difference between, say, the ordinary and the divine.

And this, he argues, is what the story of evolution fails to tell. "Somewhere in the depths of phylogenic history something happened, and somewhere in the depths of our neurological machinery something happens, and both those somethings have accomplished within us an inversion of brute, mindless, physical causality into, at the very least, the appearance of unified intentional consciousness."

The first thought would be to suggest that there isn't an absolute qualitative difference between being a thing that is conscious and being a thing that is not conscious. But that move has been tried and has failed, he argues:

- There's supervenience, a kind of epiphenomenalism, that asserts that a mental state is something such that no change occurs in a mental state without a corresponding change in an associated physical state.

- there's mysterianism, which is the assertion that some things about nature yet remain a mystery

- there's panpsychism, which suggests that conscious experience and thought are fundamental and ubiquitous

- there are quantum theories which suggest that things like logic (and thought and free will) are quantum phenomena

Hart is right. None of these is convincing. The most plausible theory of all, which he depicts Dennett as endorsing, is eliminative materialism. Famously endorsed by Paul Churchland, this theory (to quote the SEP) "is the radical claim that our ordinary, common-sense understanding of the mind is deeply wrong and that some or all of the mental states posited by common-sense do not actually exist."

The assertion that consciousness is sensation could count as an type of eliminative materialism, if we agree that arguing that there are no mental states, only sensory states, is eliminative. Call it 'eliminative phenomenalism', maybe.

But more to the point, what are we to make of the "pleonastic fallacy"? Well, it depends on the a priori assertion that there are two quantatively different states that cannot be crossed from one into the other without crossing a boundary. But there are many types of different states where no boundary is crossed at all.

For example, suppose our intrepid hiker was walking from south Asia to north Asia. At what point did he stop being in the south and start being in the north? There's no definitive point where this happened. The requirement here, to make a charge of  "pleonastic fallacy" stick, is to prove that there is a boundary, and not just a difference in degree. But why can't I say that rocks have no consciousness, that humans have full consciousness, and that various things (plants, animals) have varying degrees of consciousness in between?

If consciousness is sensation, it's not difficult to say that at all. Each degree, and each type, of consciousness will be the different types and degrees of sensations had by the various entities, from rocks (none) to plants (minimal) to animals (fuller) to us (fulsome). There's no difference in kind here, though we might recognize some as conscious and some not.


--- Evolution

History records the development of life from the original inanimate raw materials to the first single celled organisms to plants and then lizards and then people. Somewhere along the line, consciousness emerged.

Hart is unhappy with this explanation. "Dennett is an orthodox neo-Darwinian," he writes, "in the most gradualist of the sects. Everything in nature must for him be the result of a vast sequence of tiny steps." This, however, makes his task more difficult. "The burden of any narrative of emergence framed in those terms is that the stochastic logic of the tale must be guarded with untiring vigilance against any intrusion by 'higher causes'," which may be an impossible task where consciousness is concerned.

The difficulty, as Hart sees it, is that there are stages along the sequence where "competencies" suddenly emerge, competencies indicative of consciousness, but which do not appear to have developed out of any evolutionary process.

We looked at this argument in a preliminary fashion above, in the discussion of the two senses of emergence. In what follows, we'll see the distinction applied.


--- Language

Humans speak - and think - in language. This might be one of the more remarkable stages of evolution. It is certainly one that distinguishes humans from rocks, plants and lizards. And according to Hart (and no few rationalists generally) it is a place where Dennett's account fails.

Hart argues (as noted above) that Dennett is a gradualist, which means he does not recognize such evolutionary short cuts such as an innate capacity for language (as proposed by Chomsky). From my perspective, Chomsky needs a short cut, because he supposes that human language embodies (at least in part) a universal grammar, which as we've seen above is not going to be derived from experience.

"For Dennett, language must have arisen out of social practices of communication, rooted in basic animal gestures and sounds in an initially accidental association with features of the environment. Only afterward could these elements have become words, spreading and combining and developing into complex structures of reference."

Nor does Dennett recognize "the vital evolutionary saltation between pre-linguistic and linguistic abilities to a single mutation" such as "the elementary computational function called 'Merge,' which supposedly all at once allowed for the syntactic combination of two distinct elements, such as a noun and a verb."

Either of these is the India-China version of language acquisition, a story where at one point we didn't have a capacity to use language, and then the next, we did. There is, as I suggested above, no reason to believe language-acquisition (or consciousness generally) developed that way.


--- Grammatical Constraints and Powers

As noted above, Dennett's depiction of the development of language is gradualist. As Hart describes Dennett's position, "Language must have arisen out of social practices of communication, rooted in basic animal gestures and sounds...  Only afterward could these elements have become words, spreading and combining and developing into complex structures of reference... 'proto-languages' that have since died away."

Not so, according to Hart. "There is no trace in nature even of primitive languages, let alone proto-languages; all languages possess a full hierarchy of grammatical constraints and powers." Nor is it possible for language to have developed out of proto-language. "It is logically impossible even to reverse-engineer anything that would qualify as a proto-language. Every attempt to do so will turn out secretly to rely on the syntactic and semiotic functions of fully developed human language."

But wait a minute. Just what is "a full hierarchy of grammatical constraints and powers?" How is it that language has them? From where I sit, the error is not in gradualism, it is in the description of language (this is why Wittgenstein is so important).

Language, mathematics and logic have long been assumed to be special cases of cognition. I've alluded to some of of these special attributes above: necessity and universality. Language also provides the basis for reference and representation. It couldn't be any more unlike sensory experience. These are the "grammatical constraints and powers" Hart refers to.

There is almost no end to the list. For example, Hart talks of "what linguists call 'structural proximity' and 'linear proximity'." Take, for example, the sentence "Whose cake have you been eating?" The words 'cake' and 'eating' are not structurally proximate - they are not beside each other in space or time. But they are semantically proximate: what we are talking about in this sentence is 'cake eating'. According to Hart, "Without such a disjunction, nothing resembling linguistic practice is possible; yet that disjunction can itself exist nowhere except in language." Both types of proximity, however, are observational proximities. The association between 'cake' and 'eating' is a natural outcome of a neural network that has analyzed a large body of expressions (this is something you can test for yourself).

What would be a surprise would be to find some element of language that cannot be reproduced in neural networks. After all, language is learned by neural networks. The argument that there is some mystical non-natural aspect to languages is rapidly being proven false in the domain of natural language processing today.


--- Semantics, Again

Semantics is a hoary beast in language that won't go away. We have considered it above several times under the heading of 'intentionality'. Ultimately, semantics has to do with what we call the 'aboutness' of a word, sentence or sequence of sentences. The problem of semantics is that the 'aboutness' of something from the inner or mental state seems to require, or at the very least, presuppose, an outer or external state.

That's why the reduction of language to its constituent sounds and shapes seems so inadequate. The semantical elements seem seem to be in some sense 'over and above' the sound and the shapes. That's why Hart writes, "The repeated sound of a given word somehow embeds itself in the brain and creates an 'anchor' that functions as a 'collection point' for syntactic and semantic meanings to 'develop around the sound.'" And this creates something that is ineliminable. "The only possible organizing principle for such meanings would be that very innate grammar that Dennett denies exists — and this would seem to require distinctly mental concepts."

The diagram above will suggest my response to this assertion. Semantics is association. It is the association between the sounds and shapes with each other and with the other sensory phenomena that occur in our day to day experience. There is no 'standing for' or 'representation' over and above simple association. This gets back to the wholeness and simplicity I referred to above. If you want an explanation for the association that takes place, it can be found in the interactions in the neural network that constitutes our brain. But the description of semantics consists of nothing over and above association.

But what about the relation where one thing can 'stand for' the other? There is nothing mysterious about that relation, and we need no special powers to depict the relation. It appears in our own experiences with as much regularity as the similar 'stand on' relation. We can use a chair to stand for something; we can use a chair to stand on.

And, in general, we experience words the same way we experience chairs. They are physical symbols and audio sounds that often go together. These images and sounds are associated with other experiences that happen at the same time, though as we experience them over and over the association with other things becomes more and more abstract, Again, it doesn't make sense to say that a these images and sounds 'stand for' anything. And each person associates them with different experiences over time.


--- Conventions and Physicality

Consider, again, precisely how Hart describes language and semantics. "All semantic information consists in the interpretation of signs, and of conventions of meaning in which signs and references are formally separable from one another, and semiotic relations are susceptible of combination with other contexts of meaning."

convention is not an inner mental phenomenon. It is a thing in the world, just like companies, high income brackets, and jet-skis. A convention is something a lot of people do such that, if you asked someone, they would say that this is something you should do. Meanings, signs and references are things people do. The 'language' aspect of language is an external phenomenon, a thing in the world, that we have to learn about just as we have to learn about physics and geography.

It is important here to keep in mind that I am not proposing some sort of operationalism - I am not saying that mental entities stand for measurements or operations such as counting sheep or assessing profits. What I am saying is that a language, properly so-called, is something that exists in the natural world along with other things, and that the properties of a language do not automatically (or at all) become properties or our consciousness.

This assertion is the core of Wittgenstein's private language argument. There is no property of a word (or sound, or shape) that is inherently the 'meaning' of that word (or sound, or shape). When we have an experience of the word 'porcupine' it does not automatically carry with it any sort of reference or intent. These (insofar as they exist at all) are the result of the way the word is used by a population of speakers. Meaning is use.

The same thing with signs. Hart writes, "Signs are intentional realities, dependent upon concepts, all the way down. And between mere accidental associations and intentional signs there is a discontinuity that no gradualist — no pleonastic — narrative can span." A sign is a part of the natural world. Everything leaves signs (or traces, or tracks, or indications, or whatever). It takes no feat of the imagination to picture an ancient hunter spotting a broken twig and associating it with his prey.

Was there a point in ancient history where the first recorded instance of using something to stand for something else happened? The first cave painting, say, or the first inukshuk? Sure, just as there was a first war, a first city and a first cooperative bank. But it doesn't follow that these practices had to originate in the mind. A lot of things just happen when a group of people interacts and organizes itself. It is only recognized after the fact that this is what we have done.

A sign is an association, not a representation. It is something that leads us to think of something else. This is an association we have learned over time and which is now a part of our own experience. When we look at a rock and think to ourselves "Kareem Abdul-Jabbar" is is something we feel, not something that we reason about.

The importance of this cannot be overstated. While we typically talk of semantics in terms of reference and representation, the domain of semantics extends well beyond these. It includes truthvaluesobjectivesmeaningpurposebeliefknowledge, and more. The statement that I am making here is that these are all things that we feel.

That's why we have differences of opinion about them. We say that something is true when we feel that something is truth, and this feeling may originate from any of a million antecedent experiences, none of which will be shared by anyone else. Then begins the effort to develop a social convention around whether or not this or that should be called 'true' - but it takes a very significant experience for this effort to change what you feel is true.


--- Memes

Hart throws in an aside about memes at this point. A 'meme' is "a virally-transmitted cultural symbol or social idea." A meme can be thought of as something non-physical, but in no case has it transmitted non-physically. Today the internet is the primary means of propagation. "A link to a YouTube video of Rick Astley, a file attachment with a Stars Wars Kid movie, an email signature with a Chuck Norris quote... these are a few examples of modern meme symbols and culture spreading through online media."

The idea of the meme was popularized by Richard Dawkins. It's a nice way of reframing our way of thinking about identity. We usually think of interaction (and intercourse) as a means of reproducing ourselves, to produce little versions of ourselves, known as children. But intercourse could be equally well thought of as the way a gene reproduces it self - this is the idea Dawkins explores in The Selfish Gene. But if we think of a gene as an expression of an idea, well then, we have the idea of an idea reproducing itself, and thus the concept of meme was born.

The internet meme is a faithful representation of that concept, says Dawkins. "The meaning is not that far away from the original. It's anything that goes viral. In the original introduction to the word meme in the last chapter of The Selfish Gene, I did actually use the metaphor of a virus. So when anybody talks about something going viral on the internet, that is exactly what a meme is."

Hart dislikes the use of the virus analogy to depict the spreading of ideas. "When Dennett claims that words are 'memes' that reproduce like a 'virus,' he is speaking pure gibberish. Words reproduce, within minds and between persons, by being intentionally adopted and employed... depressingly substantial part of Dennett's argument requires not only that memes be accorded the status of real objects, but that they also be regarded as concrete causal forces in the neurology of the brain, whose power of ceaseless combination creates most of the mind's higher functions. And this is almost poignantly absurd."

There's a large body of thought that views not just ideas but other content such as 'concepts' and 'information' as irreducibly and necessarily non-physical. In truth, there is nothing more mysterious about ideas, concepts and information than there is about categories (and, arguably, ideas, concepts and information are no more than the modern instantiation of categories).

The meme doesn't exist at all unless it is sensed as a meme (that's why there are always some people with blank faces when you repeat the phrase "I can has cheezburger?"). It we can say that it is 'sensed as a meme' only if it elicits related experiences. The idea of a meme is that it relates sometimes surprising combinations of experiences through a non-traditionally linguistic expression (if you want more, I discuss this phenomenon in Hacking Memes and Speaking in LOLcats).

Hart attributes to memes some sort of immaculate conception. "What could memes be other than mental conventions, meanings subsisting in semiotic practices? As such, their intricate interweaving would not be the source, but rather the product, of the mental faculties they inhabit; they could possess only such complexity as the already present intentional powers of the mind could impose upon them."

Does a meme have to have been produced intentionally? Well - no. Quite the opposite - it's actually very difficult to 'think up' a meme; they are usually observed in the world as their (typically unintended) associative effect is recognized by an observer. You have to see the meme before it can even become a meme. Like this:
 


One more thing, which is a bit of an aside here, so I won't linker on it. When we look at memes we can see in a very direct way how people (and objects, and media) are connected to each other, and we can observe what we'll call a 'signal' propagate from one person to the next to the next. It allows us to think of society as though it were a giant mind.

It isn't, of course, but much of what can be said of said of human consciousness can also be said of social consciousness. When commentators say things like "The stock wants..." they are observing phenomena in society that are very similar to phenomena in people and, as we do in such cases, applying intentional language to them. Of course, the stock market does not 'want' any such thing.


--- A Kind of Computer

The discussion of things that we have no experience of is essentially impossible, which is why we're reduced to the employment of metaphors, appropriate or not, which discussing them. That's why people use the metaphor of a computer to describe the human mind. They appear to be doing similar things - managing information, drawing conclusions, remembering - and so one naturally leads to the thought of the other.

The 'mind as a computer' model is an easy target for Hart. He writes, "it would be no less apt to describe the mind as a kind of abacus. In the physical functions of a computer, there is neither a semantics nor a syntax of meaning. There is nothing resembling thought at all. There is no intentionality, or anything remotely analogous to intentionality or even to the illusion of intentionality."

Let's be clear that the mind is not a computer, if only because they are very differently constructed. Compared to a human mind, even the most powerful computers are hopelessly simple. A human brain has around 86 billion neurons and 85 billion non-neuronal cells, and the number of connections between them is orders of magnitude larger. This is far more complex than any computer. Nor is there any particular evidence that human brains employ the same solutions as those discovered by computer programmers to manage data, such as algorithms, programs, data buffers, or central processing units. At minimum, the evidence suggests that the human brain is what we would describe as a massively parallel processor.

But Hart's argument here is not, to my mind, successful. It resembles in structure John Searle's Chinese Room argument. In a manner resembling a computer, a person as a library of symbols (data) and instructions on what symbols to produce when given other symbols as input (program). Unknown to the human, these symbols are actually Chinese characters. Assuming the instructions were correct, the person-in-a-room would appear to speak Chinese. But clearly, the person isn't speaking Chinese; they don't understand it at all, they're just following instructions.

There are numerous responses to Searle's argument, but my response is simply to embrace the conclusion. "There is no intentionality, or anything remotely analogous to intentionality or even to the illusion of intentionality." Right. Just as in the human brain, it would be a fiction to apply intentionality to the Chinese room. There is no 'representation' of an external reality; everything in the person-in-a-room's awareness relates to the input and output that constitute the totality of his experience.

When we say that there can be no computer intelligence because they could never attain the same degree of mental consciousness as us, the clear (and urgent) response is: what if they can? It is far more likely that there is nothing special about humans with respect to other animals and, eventually, computers than that there is. Hart says of computer intelligence that "It is a bewitching illusion, but an illusion all the same." But it is more likely that it is the properties he ascribes to humans that are illusions.


--- The Manifest Image

"From at least the time of Galileo," writes Hart, "a division was introduced between what Wilfrid Sellars called the 'manifest image' and the 'scientific image'." The scientific image is the external world filled with objects and causes and laws of nature, while the manifest image is "the phenomenal world we experience."

For monists, materialists and most scientists, the scientific image is the only image of reality. "The manifest image, by contrast, is a collection of useful illusions, shaped by evolution to provide the interface between our brains and the world, and thus allow us to interact with our environments. The phenomenal qualities that compose our experience, the meanings and intentions that fill our thoughts, the whole world of perception and interpretation — these are merely how the machinery of our nervous systems and brains represent reality to us, for purely practical reasons."

The careful reader will be able to see where I will disagree with this statement. There is no sense to be made of the idea that "the machinery of our nervous systems and brains represent reality to us." This ascribes, first of all, a sort of intentionality to our nervous systems and brains that they simply can't have. There is no sense that it is the purpose of  nervous systems and brains to represent reality. But more to the point, there's no us over and above our nervous systems and brains to which the representation would be presented. There's no internal 'computer screen'. I don't see Dennett arguing this way, and I certainly don't.

The manifest image is the image. Consciousness is sensation, no more, no less. The scientific image, as described by Sellars (and later, famously, by Bas van Fraassen) is a physical construction, if it exists at all, out there in the world. It's an artifact, no more or less real than the Hubble Telescope, and we should be no more likely to expect the Hubble Telescope to be a mysterious inner part of our cognition. But the only reality we experience, either of science, of Hubble, of a chair, or of anything else, is the reality of sensations.

I wish it were different; I really do.

Dennett's version of things (in Hart's retelling) is the pessimist one. "Dennett's is simply the standard modern account of how the mind relates to the physical order... that consciousness itself, understood as a real dimension of wholly first-person phenomenal experience and intentional meaning, is itself only another 'user-illusion.'" It's pessimistic because it suggests (much like the existentialists) that all of our hopes and dreams and ambitions and desires don't amount to a hill of beans. And, if we take the perspective of the scientific image, that's probably true.

On the other hand, the way I describe the same story, "consciousness itself is the real dimension of wholly first-person phenomenal." All that we are - all that we see and feel and experience and hope and dream - is the most important thing is the world, because that's all that exists. That's why we fight for survival, why we work to make this a better world, why we take the time out for art and sports and leisure.

It's not correct to say that consciousness is an illusion; that very statement presupposes some third-part observer that is fooled by the illusion. But it is correct to say that consciousness, as some intrinsically purely mental phenomenon independent of our experiences and possessed of the attributes of necessity and universality, and the capacities of intention and representation - that is an illusion.


--- The First-Person Vantage

In my life, at least - I can't speak for you - what might be called the 'first person vantage' is pretty much an irreducible. It is not something I derive or infer from other things; it is something of which I have an immediate awareness. That's why Descartes begins with the cogito and why Hart calls Dennett a "fanatic" when "willing to deny not only the analytic authority, but also the actual existence, of the first-person vantage." It seems to fly in the face of our everyday experience.

But what Dennett actually denies, I think, is the idea that the first-person vantage is a special source of truth about the external world. As Hart says, "He rejects the very notion that we 'have 'privileged access' to the causes and sources of our introspective convictions,' as though knowledge of the causes of consciousness were somehow germane to the issue of knowledge of the experience of consciousness." And - in fact - we don't. That is not to deny that we have sensations of redness or feelings of ennui, but rather, that these do not constitute evidence that other people can examine and weight and measure.

Hart raises the zombie question in this context. Put the question to Dennett: could we know that we're not a zombie? "Dennett's reply is a curt 'No, you don't'— because, you see, 'The only support for that conviction is the vehemence of the conviction itself.' This is only to say, though, that we cannot deduce that we are not a zombie.

In fact, we do not feel that we are zombies. But this feeling does not come from nowhere; it is the result of the totality of our experiences (which include our knowledge of science and physiology and the rest). In a world where zombies are real possibilities, and where convention allows that they walk and talk and look just like the rest of us, then there is no mechanism available to us, over and above our own feelings, that we are not zombies.

Or to put the same point another way: you could raise a child and convince him that he is a zombie. You could raise a child and convince him that he is a brain in a vat. Or that he is a spiritual being. Or that he is any number of other metaphysical states. The child, from an internal a priori sense, has no defense against that; he would have at best an internal feeling that he is wrong. That's when indoctrination works, even when we feel in our hearts that it shouldn't.


--- Illusions

Let's take this a step further. Hart imagines that the zombie argument is indefensible because "a zombie could not unwittingly imagine anything, since he would possess no consciousness at all, let alone reflective consciousness; that is the whole point of the imaginative exercise." Let us suppose that this is true. It is simply the contrapositive. If you have no sensory experiences at all, you have no way to know you are not a zombie. The only way you could know that you're not a zombie is that you have the feeling that you're not one.

Hart is asserts that "you cannot suffer the illusion that you are conscious because illusions are possible only for conscious minds." This sounds like the same point I have just made, but it is not, because his idea of what constitutes a 'conscious mind' is very different from mine. His conscious mind includes necessity and intentionality and all the rest. But do you need all that to be fooled by an illusion? No.

Artificial intelligence (and in particular, neural networks) can also have illusions. There was one covered just recently where Google's artificial intelligence thought that a turtle is a purse. "It's likely possible that one could construct a yard sale sign which to human drivers appears entirely ordinary, but might appear to a self-driving car as a pedestrian which suddenly appears next to the street," write labsix, the team of students from MIT who published the research."

There's nothing special about being fooled by an illusion at all. Cats are fooled all the time. So are thermostats (which 'thinks' a house is cold, thus firing up the furnace, when I open a window). So are photo-sensors (famously, by mirrors, at least in the movies).

And just so, it is possible for us to be fooled about our own consciousness, to associate properties of the physical world to the nature of our own experience. We see one billiard ball strike another and cause it to move, and we suppose that the same thing happens in our thoughts. We see a mathematical proof showing the necessity of an axiom, and we suppose the same thing of our thoughts. There's no reason to suppose these things, of course. We just naturally associate things that appear similar.

Hart says, "the limpid immediacy and incommunicable privacy of consciousness is utterly unlike the composite, objective, material sequences of physical causality in the brain, and seems impossible to explain in terms of that causality — and yet exists nonetheless, and exists more surely than any presumed world 'out there.'" But that's just another illusion. It's not impossible to explain in terms of science and causality - that's Dennett's point. And when sensation ceases, so does consciousness: that's mine.


--- Emergent Properties

We have discussed the concept of emergence previously. Above, we described the idea of emergence in the sense of coming into being or production, in contrast to the idea of emergence as a pattern that arises out of complex phenomena. It is important that we distinguish between the two lest we make the mistake of thinking of emergence as a type of cause or force.

Dennett's account appears to contain elements of both, and this leads to confusion. "The point of From Bacteria to Bach and Back is to show that minds are only emergent properties of our brains, and brains only aggregates of mindless elements and forces," writes Hart. "But it shows nothing of the sort." Well, it does, but it tells two stories in one, one for each sense of emergence.

Here's the official version of the story: "mind and consciousness are no more and no less mysterious than other natural phenomena, such as gravity. Granted the right chemical and physical conditions, life forms will emerge from the primeval slime, and granted the right conditions, life will evolve large-brained organisms such as humans (who) communicate, cooperate and compete with their fellows."

This is a classic statement employing the first sense of the word 'emergence'. We're telling a causal story here as a way of explaining how a conscious and intelligent being came to come into existence. It's a claim that doesn't need special mental powers; on this story, we did not 'will ourselves' into existence, nor could we have, without violating the totality of what we know about nature and physics.

Dennett's argument also contains this bit: "Consciousness is a system property, and is not reducible: he takes issue with those hard-line molecular biologists, notably DNA pioneer Francis Crick, who seek to locate consciousness in particular ensembles of neurons in specific brain regions."

When we say "Consciousness is a system property" what we mean is that consciousness is an emergent property. Specifically, here, the assertion is that the consciousness we experience is an emergent property of the interactions between and activations of neural systems." Whether or not we say it arises from all neurons (as suggested here) or some neurons (as I suggest above, consistent with Crick's view) is a matter of empirical investigation (specifically: is it possible to have the same conscious sensation given different neural configurations).

What's important is that, as stated above, emergent properties do not have independent or inherent existence. Something constitutes an emergent property only insofar as it is recognized as such. That is why we say that things like language, logic, meaning and representation are physical properties, not mental properties. They are not 'in' the mind as somehow innate, but are recognized as such perceptually, as the social constructs containing theme were developed, generation after generation, through history.

That's why their appearance seems so sudden. They emerged, rapidly, as cultural artifacts. "Slow biological evolution has been superseded by fast cultural evolution. This too advances through natural selection, and the agents on which such selection operates are memes, the cultural analogues of genes." The social network operates on essentially the same principles as the neural network, and through the process of stigmergy, one concept piles up on another, until out of complex interaction, we are suddenly able to see a pattern, a mosaic, a fantastic construct.

The properties or attributes of consciousness we talk about on an everyday basis aren't actually there. They are, like memes, patterns of expression or thought that have characterized social discourse over time. All that there is actually to consciousness is sensation. Everything else we think we 'know' about it is a social construct (up to and even including the names of the senses).


--- Free Will

I want to attend briefly to a subject not explicitly addressed in Hart's review, that of free will.

We've addressed the dilemma of free will and determinism above by suggesting that 'determinism' is a representation of the way the world and our bodies work. We can have both materialism and non-determinism at the same time.

But what about the idea that there an individual, an ego, an I, that has fears and hopes, that wants things, that makes decisions, that evaluates things and makes judgments, and expresses all the other elements of what we might cluster under the heading of 'free will'.

There are two elements to this story.

The first is that the concept of identity, the self, and the individual, is very much a social construction. Not all people in the world today, let alone all people in the world through history, conceive of the self in the same way. This is no more evident than in the differing ideas of the relation between the self and society in western and oriental societies. There is a significant literature and domain of study about the cultural construction of self.

The other part concerns the actual feelings we have of want (or satiation), of worth (or worthlessness), of hope (or fear), and more. These are feelings, they are sensations, our having of them does not justify an inference that there is a 'self' that is having them, only that they are a part of our consciousness.

It is a familiar cognitive exercise (to me, at least) to identify the feeling that accompanies what I describe as a hope or a fear or whatever (mainly fear). the exercise is, essentially, to feel the feeling, recognize it as a sensation, and to realize to myself "the mind is a lying bastard". Cognitive therapy may not extinguish the feeling of anxiety (you need drugs for that) but it can help you understand it.


--- The End

Even if there is life after death, if there is an end to sensation there is an end to us, and what follows doesn't really matter. I think we're all aware if this to some degree, and as Becker points out, we spend the better part of our existence in denial of this.

"If only we could go on living forever," we think, forgetting (or not realizing) that 'forever' is itself a fabrication of our imagination, no more comprehensible than necessity or universals or absolute nothingness.

"Maybe there is a thing that lives forever," we think, perhaps a gene, a meme, a society, a culture, an artifact, an identity, a soul. But we forget that none of these is us, because our conscious existence is, and will always be, only our sensation.

There is solace, perhaps, to be found in the idea that there is something greater than ourselves, that we may be a part of, and have an impact on, something greater than ourselves, and even if this is the beginning and end of our existence, the feeling of that existence will be satisfactory. Some people find their consolation in philosophy, some people find their solace in fame.

For me, the lesson I draw, is that life is everything. This applies to my own life, of course, but I also extend the same feeling to those around me, whose existence I can only comprehend through science and culture, but whose dilemma I assume is the same as mine, that is, that it is the quality (and duration, and variety, and related properties) of their experience that overwhelms everything else.



Stephen Downes Stephen Downes, Casselman, Canada
stephen@downes.ca

Copyright 2024
Last Updated: Oct 11, 2024 11:43 a.m.

Canadian Flag Creative Commons License.

Force:yes