The last BQTA post was the unadorned transcript of the conversation between Dan Dennett and Sean Carroll published on 1/6/20, edited for readability and slightly shortened. https://better-questions-than-answers.blog/2020/01/18/daniel-dennett-on-sean-carrolls-podcast-minds-patterns-and-the-scientific-image/
Here we present Part I of an annotated transcript of their conversation with commentary by BQTA contributor Rick Lenon, M.D. (RL) and me (GNS). We add some background and references, and throw in a few questions and comments. We have some discussions of our own. The original verbatim transcript as well as the recording can be found on Carroll’s website, https://www.preposterousuniverse.com/podcast/2020/01/06/78-daniel-dennett-on-minds-patterns-and-the-scientific-image/, but we think our edited text of their conversation we have here is more understandable. This is the longest podcast in the excellent Mindscape series, and Carroll says it’s his favorite.
Part I
Carroll introduces Dennett with a reference to Wilfrid Sellars (1912-1989) whose “effort in philosophy was “to formulate a scientifically oriented, naturalistic realism which would ‘save the appearances’.” https://plato.stanford.edu/entries/sellars/
Sean Carroll: Wilfrid Sellars described the task of philosophy as explaining how things, in the broadest sense of term, hang together, in the broadest sense of the term. (Substitute “exploring” for “explaining” and you’d have a good mission statement for the Mindscape podcast.) Few modern thinkers have pursued this goal more energetically, creatively, and entertainingly than Daniel Dennett. One of the most respected philosophers of our time, Dennett’s work has ranged over topics such as consciousness, artificial intelligence, metaphysics, free will, evolutionary biology, epistemology, and naturalism, always with an eye on our best scientific understanding of the phenomenon in question. His thinking in these areas is exceptionally lucid, and he has the rare ability to express his ideas in ways that non-specialists can find accessible and compelling. We talked about all of them, in a wide-ranging and wonderfully enjoyable conversation.
0:00:00 Sean Carroll: In 2020 we’re going to start off with Daniel Dennett, who is a philosopher who is as well known as any philosopher is in the modern age among the general public, and also once again extraordinarily respected among his professional colleagues. I in particular have enormous respect for what Dan has done.
There’s a framework that ties our conversation together, which is this idea of taking what science teaches us about the world and connecting it to the world of our every day experience.
0:01:34 SC: For whatever science teaches us, it is very often going to be the case that even though it comes ultimately from our experience of the world, the ultimate theories that we end up building might seem very different very surprising even disconcerting. The big bang, cosmology, quantum mechanics, Darwinian evolution are things that you wouldn’t have just guessed just on the basis of your everyday experience without enormous amounts of observation and experimentation into realms that you don’t see in your everyday life.
And therefore the theoretical frameworks you develop don’t sound or feel much like our every day world. This is especially noticeable when it comes to things like consciousness, free will, the nature of human beings.
RL: Imagining the difference between prereflective stance on consciousness, self, and free will vs what happens when we focus our attention on them, especially our scientific attention.
Dan Dennett has devoted his career to taking discoveries from science— whether it’s neuroscience or biology or what have you, computer science, artificial intelligence— and teasing out their philosophical implications. He is one of the world’s leading philosophical naturalists, not a naturalist in the sense of going out into the forest and poking around the trees and the animals, but a naturalist in the sense of not being a super naturalist. An ontology that says there is only the natural world. How do you then explain things like purposes and meanings? And other things that we human beings naturally associate with our lives here in the world?
That’s what Dan has been trying to figure out for the course of his whole career. He has thought very deeply about the nature of existence, the world we live in, the nature of thought, how we conceptualize what’s going on, and questions that are very important to me like emergence and intentionality, how it’s okay to talk about things like purposes and choices in a world that is ultimately governed by the laws of physics. This is probably my favorite podcast interview that I’ve ever done, and I think that you’re going to enjoy it just as much.
0:04:31 SC: Dan Dennett, welcome to the Mindscape podcast.
0:04:33 Daniel Dennett: Delighted to be with you.
0:04:35 SC: Something that you said one time, that I’m sure you’ve said many many times, really struck a cord with me. Talking about Wilfrid Sellars and the manifest image and the scientific image and how you thought of your task as a philosopher to reconcile these. Why don’t we begin setting the stage by telling us what these are?
0:04:58 DD: I’m glad you asked. Wilfrid Sellars, great American philosopher, said the job of philosophy was to explain how things in the broadest sense of the term hang together, in the broadest sense of the term. Well, that sounds sort of trivial. What he had in mind is this: there’s all the things of the everyday world: colors and sounds and haircuts and pains and dollars and home runs. Then in the scientific world, there’s electrons and quarks and fields and molecules. How do we relate the things of our everyday, pre-scientific world to the things that science has discovered? And what 100 years and more have shown is that there’s no simple answer.
0:06:13 DD: When I say 100 years I mean, let’s say, since Einstein. That’s when the world really starts to look weird from a scientific point of view, and you have people saying, “Really it’s all just atoms and the void and there’s no such thing as solidity and there’s no such things as colors. And after all, atoms aren’t colored. And the world’s made of atoms. It’s just atoms and empty space.”
GNS: Phillip Goff recently argued on Sean Carroll’s podcast that the disconnect between common sense and the scientific point of view started with Galileo in the 17th century. https://www.preposterousuniverse.com/podcast/2019/11/04/71-philip-goff-on-consciousness-everywhere/
DD: At one extreme, you have people who have insisted that the scientific image is the gold standard. That’s what sets what’s real. Everything else is illusion. But as a cartoon I like puts it, “The world we live in may be an illusion, but it’s the only place you can get a good cup of coffee.” So it’s not very helpful to be told that not only do dollars and home runs not exist, but colors don’t exist and pain doesn’t exist. Solidity doesn’t exist. So we have to negotiate between the two worlds. Sellars says, that’s what philosophy is for. That’s about as good a definition of philosophy as I can think of.
0:07:54 SC: But you’re adding a little bit, right? One could buy into Sellars’ formulation while still denying that the manifest image is capturing something real.
0:08:05 DD: Oh yeah, Sellar’s image leaves all the options open. It leaves open both the hardcore scientific realist who says everything else is just illusion. Eliminativism, as philosophers say. Or you could go the other extreme and say the electrons and quarks and all that, that’s just a useful fiction. What’s really real is tables and chairs and people and ideas and love and so forth. So those are the two extremes. And then there’s all kinds of positions in the middle.
My view, which might seem to be giving up, especially to philosophers, is to think we have to learn how to get back and forth between these two images, the manifest image and the scientific image. But the way we do that is not by strict definitions that are counter-example proof. The way we do it is with diplomatic and pedagogical ways of easing the passage.
An old example: Voices. Are voices real? What’s it made of? Is it a bodily part? Is it biological material? But you can record a voice. You can recognize a voice. Voices, if you say what category of things they are, they seem to be almost in a category by themselves. We don’t need a voice-throat problem to go with a mind-body problem.
0:10:15 DD: We may not know how to answer the question of what voices are, but we’re not mystified, we’re not puzzled, we’re not baffled. It’s just a curious fact about the way language and our perception of the world— our pre-scientific perception of the world— carves things up. And the otolaryngologists, and the other biologists and the acoustic engineers can tell us all about voices, without ever settling that issue.
0:10:47 SC: I label this view, in my book The Big Picture, poetic naturalism. The motto being that there’s only one world, the natural world, but there are many ways of talking about it. And those ways all can capture some elements of reality and it’s silly to call them illusions just because they’re not the most fundamental thing.
0:11:05 DD: Gee, I haven’t read your book yet.
GNS: It’s a good book. An attempt to put it all together from a scientific perspective. Although it may not ask—much less answer—the toughest questions, it may be the strongest attempt yet by any scientist to understand the world as a whole.
RL: I do need to get to that. Bought it back in March.
0:11:17 SC: The idea was not supposed to be anything original, it’s just a label to help people understand. Because there are people who are eliminativists, right? Who want to say that some of these higher level structures shouldn’t count as real.
0:11:27 DD: Yeah. I’ve been battling against that view for decades.
0:11:39 SC: So therefore you will think of things as consciousness and free will as real? For exactly this kind of reason?
0:11:52 DD: Real, but they’re not what you think they are. That’s my motto. X is real, but it’s not what you think it is.
GNS: This is Dennett’s way of trying to avoid eliminativism. If we say that consciousness is a way of talking about behavior, have we said that consciousness is real, but it’s not what you think it is? Or have we reduced consciousness to behavior? And if mental states are behavior, rather than whatever we think they are, haven’t we eliminated mental states?
RL: I think of behavior as action, something that’s done. Can’t get to consciousness as something that’s done. Talking about it, yes, that’s a behavior. Thinking about it, that too. But how does it get to be something that we can do or not do?
Simon Blackburn: Philosophically the doctrine of behaviourism is that mental states are “logical constructions” out of dispositions to behaviour, or in other words, that describing the mental aspects of a person is a shorthand for describing the various dispositions to behaviour that the person possesses. The most influential work promoting this point of view was The Concept of Mind (1949) by Gilbert Ryle [Dennett’s teacher at Oxford. Dennett wrote the introduction for the 2000 edition of Ryle’s most famous book] which urged behaviourism as the best defence against the Cartesian myth of the ‘*ghost in the machine’. The extent to which Wittgenstein, writing the Philosophical Investigations at the same time, intended to promote a behaviourist doctrine is subject to dispute. Like other reductionist doctrines, behaviourism fell foul of the difficulty of providing workable analyses, notably because of the holism of the mental, or the fact that how a person behaves is not a function of one belief or one desire, but of a whole field or network of beliefs and desires. The modification to take care of this turns behaviourism into its more popular modern successor, functionalism. The Oxford Dictionary of Philosophy
SEP: Analytical or logical behaviorism is a theory within philosophy about the meaning or semantics of mental terms or concepts. It says that the very idea of a mental state or condition is the idea of a behavioral disposition or family of behavioral tendencies, evident in how a person behaves in one situation rather than another. When we attribute a belief, for example, to someone, we are not saying that he or she is in a particular internal state or condition. Instead, we are characterizing the person in terms of what he or she might do in particular situations or environmental interactions. Analytical behaviorism may be found in the work of Gilbert Ryle (1900–76) and the later work of Ludwig Wittgenstein (1889–51) (if perhaps not without controversy in interpretation, in Wittgenstein’s case). More recently, the philosopher-psychologist U. T. Place (1924-2000) advocated a brand of analytical behaviorism restricted to intentional or representational states of mind, such as beliefs, which Place took to constitute a type, although not the only type, of mentality. Arguably, a version of analytical or logical behaviorism may also be found in the work of Daniel Dennett on the ascription of states of consciousness via a method he calls ‘heterophenomenology’ (Dennett, Sweet Dreams: Philosophical Obstacles to a Science of Consciousness, 2005, pp. 25–56). https://plato.stanford.edu/entries/behaviorism/
Wittgenstein in Philosophical Investigations §307: “Aren’t you nevertheless a behaviourist in disguise? Aren’t you nevertheless basically saying that everything except human behaviour is a fiction?” — If I speak of a fiction, then it is of a grammatical fiction.
GNS: I see Dennett as a behaviorist— a behaviorist in disguise— though he would not agree.
0:12:01 SC: You wrote a paper a while ago called “Real Patterns”. [The Journal of Philosophy, Vol. 88, No. 1. (Jan., 1991), pp. 27-51.] I don’t know if you are aware that this has become an important fun topic in quantum mechanics. David Wallace, who is one of the leading theorists of the Everett (or Many Worlds) interpretation, which I’m also a partisan of, leans on your paper and your notion of real patterns very heavily in his book, The Emergent Multiverse. Trying to explain how the classical world— forget about tables and chairs, but even electrons with positions, and atoms, and things like that— are somehow not there in the most fundamental formulation of quantum mechanics. But they describe the pattern and therefore they’re real.
GNS: It’s patterns all the way down? Only patterns are real? That sounds a bit like Max Tegmark’s view that in physics only the math is real. Our Mathematical Universe: My Quest for the Ultimate Nature of Reality (2014) He also recently appeared on Carroll’s podcast: “Max Tegmar has thought a great deal an ambitious and speculative idea: the Mathematical Multiverse, in which we imagine that every mathematical structure is real, and the universe we perceive is just one such mathematical structure.” https://www.preposterousuniverse.com/podcast/2019/12/02/75-max-tegmark-on-reality-simulation-and-the-multiverse/
GNS: Patterns are perfectly real, but they are supposed to be patterns of something: behavior, lines on paper, neuron firings, birds flying, etc. How can there be patterns all by themselves?
RL: Sometimes seems to me as if things only seem genuinely “real” to us if they are constant. An electron can pop into and out of existence; quantum mechanics can prove that to you. Is it as real when it isn’t, as it is when it is? “A quantum state of a particle or system, as characterized by a wave propagating through space, in which the square of the magnitude of the wave at any given point corresponds to the probability of finding the particle at that point.”
GNS: A big part of the scientific project is to figure out what’s real—e.g. electrons— and what is not: heaven and hell, gods and godesses, and much of the rest of what people believe. What’s real is a legitimate question.
0:12:48 SC: Can you give the sales pitch for what your view is there in that “Real Patterns” paper?
0:12:57 DD: The main idea of the paper is to think about information theory. To put it in sort of everyday terms, “How big a file do you need to capture this particular phenomenon?”
If you have a checker board which has got just 64 squares and some are black and some are white, it’s pretty easy to give a very limited description of that pattern, and write it on the back of an envelope. If you got a color picture of confetti, and you have to describe it in detail, you’ve got a much bigger file. That’s why some pictures on your phone are bigger, have used more megabytes than other pictures. It all depends on how much complexity there is in the picture. And if there’s no pattern in the picture at all, if it’s just random, oddly enough that’s the one that takes the most information to record because you have recorded every pixel. You can’t say, “Well there’s a region of deep-blue over here and there’s a region of red over here.” Those are nice concise ways of taking advantage of the pattern in the phenomenon. The idea of real patterns is, take any phenomenon and are there patterns in it? A pattern is a summary, something that permits you to generalize so that you’re better than a coin flip about what the next little bit of it is. If you’ve got any predictive edge at all on the data set that you’re looking at, you got a pattern.
0:15:00 SC: Right. We should be happily surprised when there are such patterns. What the patterns enable you to do is to ignore certain pieces of information.
0:15:14 DD: Absolutely. Evolution— natural selection— has designed organisms to be ruthless pattern finders, to ignore almost all the information that’s officially available at their surfaces and just focus on what matters to them. If they can latch onto those patterns, they can feed themselves and avoid getting eaten; live long happy lives and mate and all the rest. So, the idea of a pattern is I think a very useful and deep idea. It can be given a nice clear mathematical formulation. And it’s the key. What science does is find patterns, but it’s also what the manifest image does. We take for granted all the patterns that we see. In fact, we do more than that. We over-interpret them. That is, if we see two things that look the same shade of green to us, we think, “Well, deep-deep down, they are the same.” No. They might be green for entirely different reasons. They might be so-called metamers and they only look the same color to us because we’re green-green color blind.
[advertisement]
0:17:55 SC: I did a podcast with Melanie Mitchell, who is a computer scientist, about the struggle of artificial intelligence to capture common sense. Would it be off-base to think that some of this struggle has to do with the fact that even AI’s very advanced Deep Learning networks are not as good at finding the patterns as a human intelligence is?
RL: A lot of what seems like “common sense,” stuff that we don’t even have to think about, is hardwired in us. Stuff we know without ever having to have learned it, and would struggle to explain. All the way up to Chomsky’s language machine. It’s code all the way down; it’s just that we didn’t write it. Can’t even read it.
0:18:25 DD: I think that’s true except for the fact that if you crank your deep learning system long enough, it’ll find patterns where there aren’t any patterns. Deep learning systems— algorithms— are very good at squeezing pattern out of apparently random data. That’s how the neuroscientists categorize on this FMRI data about what’s going on in people’s heads. They discover they can make a prediction about what a person is going to do 10 seconds later. Yeah, they can. And that shows that it’s a real pattern. But a lot of the patterns they find by these methods aren’t real. That is, they don’t predict the thing.
SC: What AI — the Deep Learning networks— are so good at is manipulating huge amounts of data. They don’t need to be as tricky as human beings are to find the patterns.
DD: We pay a price for that. It’s not a miracle. The price we pay is that we have a lot of false positives. We see a lot more pattern in the world than it is really there. We see similarities that are only similar in that they have the same effect on us, but they are otherwise as different as can be.
0:20:00 SC: Does the word real in the phrase “Real patterns” have the same meaning as the word “real” when we were just talking about baseballs being real?
DD: Well, that was the idea. I wanted to say, if we have the concept of a pattern, we do have some pretty good tests as to whether it’s real. That is, to put it bluntly, can you make money betting on it? If you can, it’s predictive, it’s real. And that’s a touchstone of reality that seems to hold up very well. Let’s say maybe patterns are the thing that’s most obviously where we can make a real vs non-real distinction. And every other distinction between real and unreal— real and fictional, real and bogus— is somehow dependent on that.
0:21:05 SC: So if there’s a room and there’s this huge number of atoms in the room, and if you were infinitely smart and I gave you the location of all the atoms and their velocities, then you could predict anything. You could be Laplace’s Demon. But the patterns— the other structures— are the idea that I could give you much less information than that. I could say there’s a baseball and it’s headed toward a window. And then you could infer an enormous amount from that. And therefore, baseballs are real in some sense.
Laplace: We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes.—Pierre Simon Laplace, A Philosophical Essay on Probabilities 1814
0:21:37 DD: That’s right. Years ago I concocted an example to show the power of this. We have a visiting Martian, who’s a sort of Laplacian Demon. He’s in somebody’s house and the phone rings and the lady picks up the phone and says, “Yes dear, you’re bringing the boss home for dinner? Do get a bottle of wine on your way home. See you in half an hour.” Hangs up, okay. So now, both the woman and the Martian predict that within 30 minutes two people are going to walk in the door, one of them holding a glass bottle filled with an alcoholic beverage. But the Laplacian Demon has had to trace out the whole trajectory— the stop signs, and the lights. and the paying of the wine… Every photon, and to the Laplacian, this is a miraculous prediction. How did she do this without all that information? Well, very simple. She understood what was being said.
GNS: Of course, people’s plans sometimes go awry. What’s said is not always what happens. Laplace’s Demon has done the extra work of tracking every particle. His prediction— maybe that hubby ends up having dinner with the glamorous boss alone—does not go awry, because human behavior is determined by the laws of physics. Predictions based on the correlations between words and actions can never be more than a rough guess.
0:22:45 SC: So let’s take this point of view— patterns at the higher level capture some influence, some predictability of the world— and apply it to the difficult cases where we have things like people and agents. Another phrase that you popularized way back in the day is “the intentional stance”. We might ask, are these real? Or do they have some special status? Things like intentions, reasons why, about-ness, why a certain painting is about something. So how do those boundary-contentious words fit into this picture?
0:23:30 DD: Oh they fit in beautifully. The intentional stance patterns are just one particular set of patterns. They’re the set of patterns that have to do with living agents and non living agents— that living agents have made. And what’s the simplest one? When I first started writing about the intentional stance, I chose a thermostat. Because you can consider the thermostat as a little agent that can be instructed to keep a certain temperature. It senses the temperature and when the temperature falls below the set line, it has a desire to raise the temperature. Treat a thermostat as an agent surrogate. You could have a person standing there and throwing logs on the fire, but you can replace it with this dead simple thing. You can explain it to a child without going into the mechanics. There are 100 different ways you can make a thermostat.
GNS: For Dennett, mental events are nothing but higher level patterns of behavior. Those patterns are real, but we think of the mind as something other than behavior, however we describe it.
0:24:50 SC: In other words, you can explain it in terms of its purpose, rather than its atoms.
DD: Exactly. Considered as a little homunculus, a little agent, and it has one desire only. And that is to maintain the temperature, but it has a way of sensing the temperature and responding to changes by making an appropriate move.
SC: By the way, this is the way that one gets taught about transistors in physics class. As if there’s a little man in there, transistor man, who decides how much current to let through.
0:25:20 DD: Well, it turns out that this tactic— this strategy— of adopting the intentional stance works throughout biology. It works not just for brains and for higher organisms. It works for bacteria. It works for archaea. It works for single celled organisms. The question is, does it work for things smaller and simpler than that? Well I like to say, we’re robots made of robots made of robots made of robots made of robots… And once you get down to sub-cellular, you get down to the canasins, the motor proteins, and tubulin, and things like that. Think of ribosomes, fantastic little machines. You can treat them from the intentional stance.
0:26:20 SC: They have a job to do.
GNS: Of course, none of these things— from thermostats to transistors to motor proteins really have intentions. So is Dennett’s point that humans do not have intentions either? Intentions are merely patterns of behavior?
RL: This seems more like language problem than an ontological one. When we say “have intentions” we are imagining an agent, and generally a conscious self to do the having part. For instance, you have a knee jerk reflex. It’s hardwired to participate in maintaining posture. Any sudden stretching of that tendon that connects your quadriceps to the tibial head via the kneecap will be automatically countered by tensing of the quad. You and I would not call that tensing intentional. But DD has defended speaking of evolution as having “designed” this or that to serve the function that it does. In that sense, the knee jerk reflex is “intended” to help maintain posture. Just to make it a little more complicated, the knee jerk reflex is moderated/inhibited by your brain. Interrupt the spinal cord and you get an exaggerated knee jerk. But that inhibition from your brain is not conscious either; its “intent” is to smooth out the reflex action, moderate it. Worse yet, often happens that if the patient is expecting the knee jerk, it won’t happen; it gets inhibited by whatever accompanies the conscious awareness in the moment. Is it intent yet?
GNS: Reflex reactions are a classic example of unintentional actions. Yes, we say we “have intentions” all the time. “My intention today is to do some writing.” Is that kind of talk indulging in sheer mythology, because it requires imagining an agent or a conscious self, which do not exist? Maybe, but just try to give it up.
RL: So in that sense a thermostat does have intent, it just doesn’t belong to the thermostat. It’s more than the atoms it’s made of. It’s the pattern in which those atoms are arranged, and it’s the predictable behavior it will exhibit. And it manifests the intent of the people who designed and built it. As DD put it above, “…you can consider the thermostat as a little agent that can be instructed to keep a certain temperature.” It manifests intent, but it doesn’t have a self or consciousness. But the fact that it does stuff is real, so it’s not just atoms and moving electrons and heat and the lack thereof.
GNS: The thermostat “manifests” the intent of its builders and users. It does not have intentions, mental states or consciousness of its own.
RL: Or let’s pretend that we’re all taken out by a super neutron bomb. Is the thermostat still a thermostat? Does it have intent/purpose, meaning?
GNS: We can discover the purposes and intentions of the builders of the pyramids. Piles of rocks do not have purposes or intentions of their own.
DD: They have a job to do and they know how to do it. One of the things that I particularly like about motor proteins is that — it now turns out — basically they’re sailing. They’re using the storm of the water molecules inside the sail, and they have sort of ratchets in their feet, so they’re actually selectively using the energy in the random bombardment of the water molecules as a source of power. It reminds me of Ricky Skaggs, great line, “I can’t control the wind, but I can trim the sails.” And that’s your basic agent, and it’s just a protein.
0:27:10 SC: Right, right. It’s a little Maxwell’s Demon. In the last just 15 or 20 years, this has become another hot topic in physics, understanding these non-equilibrium fluctuations and very very tiny things. It’s probably still under-appreciated in my personal world how much this transition from the world of individual particles— where it would make no sense to adopt the intentional stance— to the macroscopic world is driven by entropy and the arrow of time.
RL: Per wikipedia, “Maxwell’s demon is a thought experiment created by the physicist James Clerk Maxwell in 1867 in which he suggested how thesecond law of thermodynamics (the total entropy of an isolated system can never decrease over time) might hypothetically be violated. In the thought experiment, a demon controls a small door between two chambers of gas. As individual gas molecules approach the door, the demon quickly opens and shuts the door so that only fast molecules are passed into one of the chambers, while only slow molecules are passed into the other. Because faster molecules are hotter, the demon’s behavior causes one chamber to warm up and the other to cool down, thereby decreasing entropy and violating the second law of thermodynamics. This thought experiment has provoked debate and theoretical work on the relation between thermodynamics and information theory extending to the present day, with a number of scientists arguing that theoretical considerations rule out any practical device violating the second law in this way.” https://en.wikipedia.org/wiki/Maxwell%27s_demon
0:27:40 DD: I’m not a 100% sure of this, but I think the key element to being an agent is having a history. A history that makes a difference. That is, something can happen to it that changes it, and that changes it again, and it has a sort of memory.
GNS: Particularly good points here.
RL: I’m having a similar problem with this. More about what the word “agent” means to us, than about what exists or does’t, is real or not. A mouse trap has a history that determines what will happen. Is it set or not? Baited?
GNS: DD and SC take a very broad view of agency, which has an extensive philosophy, of course.
SEP: In very general terms, an agent is a being with the capacity to act, and ‘agency’ denotes the exercise or manifestation of this capacity. The philosophy of action provides us with a standard conception and a standard theory of action. The former construes action in terms of intentionality, the latter explains the intentionality of action in terms of causation by the agent’s mental states and events… it seems that genuine agency can be exhibited by beings that are not capable of intentional action, and it has been argued that agency can and should be explained without reference to causally efficacious mental states and events. https://plato.stanford.edu/entries/agency/
0:28:15 DD: An interesting thing about electrons, they don’t pick up scars or dirt or anything. An electron over a billion years doesn’t change at all, and that’s a huge difference. We had a wonderful argument at the Santa Fe Institute with David Wolpert and his colleagues, and one of our big issues was whether tornadoes count as agents.
GNS: Great point about electrons not having scars.
SC: I’m trying to predict which side you are on. I can’t do it.
0:28:50 DD: I was against it, because I didn’t think that the tornado could actually explain information.
SC: It’s a complex system but ascribing agency or intention to it doesn’t seem to help as much.
DD: But it was a very illuminating discussion. So, if you want to look at the boundaries, you want to look at things like tornados or motor proteins. In the living world, everything bigger than a motor protein is a designed thing and it has purposes. It’s got parts that have jobs to do.
SC: I think participating in the arrow of time is probably a necessary pre-condition for being an agent in this sense. The thing about an electron is, as you said, they don’t have scars. They don’t change over time. More complicated things have different access to the past versus the future. They have memories of the past and they can a little bit all they can do is predict the future. And that’s when it becomes, that’s when purposes and things like that might become necessary. So, I presume what you’re going to say is that ascribing intentionality or purposes to things… there’s a reason why we do that. It makes sense for ribosomes. It makes sense in exactly the same way for human beings, not in a different way.
0:30:20 DD: Absolutely. It’s interesting to think about the history of this. Before there was language, you didn’t have any agents that were comparing notes, that were arguing, that were explaining. Language brought into the world— onto our earth— something that Wilfrid Sellars called, “the Space of Reasons”, and this is where human persuasion and explanation and querying and challenging happens. The whys and the becauses. The arrival, the emergence, of the Space of Reasons has to have an evolutionary history too. And this only in one species. So, that’s why I’m interested so much in the evolution of language and in evolution of human minds which are profoundly different from even chimpanzee minds or dolphin minds or whale minds. Human minds are really different, and they’re different precisely because they are obliged to articulate reasons.
0:32:17 DD: They learn how to do this, and it’s an imperfect business, and some are better than others, but it’s the fundamental basis for morality. If you are responsible, it’s because you respond to reason. You can’t argue a bear out of what it’s doing, but a human being is supposed to be persuadable.
RL: I don’t buy that. Two bears having an encounter over a deer carcass. The bears are exchanging information, but it’s not articulated. Information they both have includes degree of hunger, a combat risk assessment, and a status factor, if future encounters are likely. Behavior that occurs will be determined by the processing of that information. All of that can be weighed without consciousness, or language. The two bears are probably better at choosing the optimal behavior in that situation than you would be (if you were a bear).
0:32:34 SC: So, you can train or teach a dolphin or a dog or whatever, but you don’t give it a reason why. It’s pure stimulus and response, right?
RL: Dogs do seem to get a sense for what we want, and try to do it.
0:32:55 DD: Recently, in my work I’ve had lots of examples of what I call free-floating rationales. This is where the reasons are clear, but they’re not the reasons of the organisms involved. So, the stotting or pronking gazelles are throwing these great extravagant leaps while they’re running away from the lions. It’s a tremendous waste of energy and it makes it dangerous. What they’re doing is they’re showing off. They are signaling to the lions, “Don’t bother try to catch me. I can throw these big expensive dangerous leaps and still outrun you. Go after my cousin over there, he can’t do it.” The lions believe them and the evidence for this is pretty clear.
0:33:50 DD: This is a testable and tested hypotheses and sure enough, the lions discriminate and they don’t go for the ones that are stunting. Now, I’ve given you the rational explanation. The lions are, in effect, wise to take this information that’s being offered to them. It benefits both the speedy gazelle and the lion who doesn’t have to work as hard to get his supper. And there’s lots and lots of cases of this. But don’t think that the lion understands this or that the gazelle understands this. This is a rationale that has been uncovered by natural selection. The gazelle just doesn’t know why it wants to make those leaps if it can. The lion doesn’t know why it doesn’t care for those jumpy ones. They don’t have to know. So they are the beneficiaries of a rational system that they don’t have to understand.
GNS: Animals behave rationally in the sense that they do what they need to do to survive, as dictated by evolution. What we think of as their reasons are the causes of their behavior. Also, we think that nature makes sense. That’s the basis for science.
0:35:00 SC: And that’s more or less exactly the same sense in which AlphaGo doesn’t know why it puts a certain token on the Go board in some way. It knows what to do. It couldn’t tell you why.
RL: There is a reason why it did it though. What’s not there is a conscious entity that will explain it to you. But could a really good go player really tell you why she made the choice that she did.
RL: Not saying it never happens that humans will make a decision that’s entirely determined by articulated reasons, considered carefully before a decision is reached. Just saying that more often than not, the reason/defense gets formulated after the fact. That the information that gets processed en route to the decision is generally not fully articulated before the deciding begins. Before it’s over, sometimes. Jonathan Haidt’s rider on the elephant and all. The Righteous Mind (2012).
RL: Saw it someplace recently that as humans in any proximity to this arena get older, they are increasingly likely to fall into making statements about how we are different from other animals. To date, everyone who has tried has been embarrassed. The luckier ones, postmortem.
0:35:11 DD: So do the ribosomes in every one of your cells. There’s a rationale for every part of the job. If you look at the machinery, elegant, elegant, engineering. But the ribosome doesn’t know, and in fact, no agent figured that out in advance. The Nobel-winning molecular biologists, the chemists, they worked it out for the first time what the rationale is, but the rationale is secure as anything.
RL: I’m not so sure that the model the molecular biologists created in their heads was comprehensively figured out in advance, or for that matter, entirely accessible consciously. They ask themselves, what would this thing do if this happened? And then it tells them.
0:35:50 SC: And in some sense— because we’re among philosophers here— the fact that we human beings can attach reasons to this has to do with some sort of counter-factual thought experiment. If the gazelles were not leaping in that way, then we know that the lions would chase them, even if the gazelles don’t know that.
0:36:10 DD: The intentional stance is like an instinct. It’s a Baldwin effect. [In evolutionary biology, the Baldwin effect describes the effect of learned behavior on evolution. wikipedia] It first came on the scene in its articulate form with human beings discovering they could talk about the reasons why things were happening. We’re very, very good at it. And in fact, if you wanna see it as an instinct, you can go back and see the early animations of simple triangles and circles moving around on the screen. Everybody looks at it and says, “Oh, the big circle is trying to catch the little circle.” Everybody instantly sees intentionality and purpose in these cases. Infants, quite young infants, are puzzled by violations of the apparent agency in very, very simple displays.
GNS: Dennett is refering to movies made by Fritz Heider and Marianne Simmel. Heider, F. and M. Simmel, 1944, “An Experimental Study of Apparent Behavior”, American Journal of Psychology, 57: 243–259.
GNS: Per SEP, We have a pervasive tendency to interpret and explain behavior in terms of intentional mental states. We tend, even, to interpret the interaction between animated objects in terms of desires, beliefs, and intentions (Heider and Simmel 1944). This raises the question of when it is appropriate to attribute mental states in the explanation of behavior. According to an instrumentalist stance (Dennett The Intentional Stance 1987: Ch. 2), the question of when it is appropriate to ascribe mental states cannot be separated from the question of when it is appropriate to ascribe agency, and both questions are to be answered in terms of predictive success: it is appropriate to attribute mental states in the explanation of agency when doing so supports successful predictions of behavior. https://plato.stanford.edu/entries/agency/
0:37:30 SC: But in circles and triangles, isn’t that an edge case here, because we’re saying that the ascribing of intentions to human beings or to the behavior of the gazelles is real and true, whereas presumably it’s not real and true in the case of the circles and triangles?
DD: Whose experiments am I thinking of? A German psychologist. [Heider and Simmel 1944] When he made the films to show to people, he deliberately set out to create these intentional patterns.
SC: I see. There was an intention working behind the scenes.
0:38:15 DD: In fact, he tested this by having randomly moving circles and triangles, and people did not attribute intentionality to those. That was just noise. In the same way, natural selection has enforced the patterns that we see in the jumping gazelles and the lions. That’s not just random. So we can see intention where there’s none. We’re very good at that. It’s called paranoia.
SC: Thomas Pynchon has some novels about this. And you can see it emerging, I guess. You don’t use the word emergence that much, but are you happy with the word?
0:39:15 DD: Emergence is a word that I don’t use much, because it has a sordid history in philosophy, where emergence comes to mean woo woo, inexplicable.
SC: Physicists use it all the time, but I’m warned by my philosophy colleagues I shouldn’t.
GNS: Does “emergence” mean much at all? At the Big Bang, no consciousness. Later, there is consciousness. So it emerged. The phrase does exclude divine intervention, which is useful. Everything emerged from the nature world; nothing has a supernatural source. Granting that, not much else has been explained by saying “consciousness emerged”.
RL: I agree the term doesn’t seem to tell us much. It is hard for me to imagine how consciousness might have “emerged” gradually. It does sort of feel all or none, implausible as that seems. When it’s degraded, as by alcohol, seems like quality of content suffers more than consciousness itself. I am aware of interruptions of consciousness, fading, as happen when nodding out. I have noticed that I’m less aware of falling asleep, waking than I used to be. Sometimes I’ll find myself awake, look at the clock, and there’s an hour or two missing. Got from there to the Hume, where he could see content but not consciousness itself.
But I can imagine consciousness emerging relatively suddenly as a result of gradual increases in the complexity of our hardware, much as we’ve seen with game playing and translation by AI. That comes no where near explaining the actual mechanism, but it semi-justifies the use of the term “emergence.”
RL: At 0:42:26 Dennett talks more about this. It does seem that we are bound to imagine consciousness as all or none. But it’s not plausible that it first appeared in its present full glory. When we degrade it, with alcohol for instance, it seems like the diminishing quality of the content is apparent before the impairment of consciousness itself. There’s the nodding in and out effect, even without actual nodding.
GNS: Consciousness cannot be “all or nothing”. Who imagines that? There must be degrees of consciousness, as you point out.
RL: Here I go back to my insistence that consciousness itself doesn’t choose its own content, because it has to be assembled before you can be conscious of it. Can’t be conscious of something that isn’t there yet.
GNS: You don’t mean that we become conscious of brain activity. You mean that consciousness is produced by—Carroll might say “emerges from”— brain activity. So the brain has to be active before consciousness can emerge. No mental change with a physical change is the slogan. Dennett would disagree, I think. He refers at 0:54:27 to this kind of theory as “up through the various cortical areas, and then ta-da, consciousness happens.” For him, patterns of brain activity and the resulting behavior is all there is. There would be, for Dennett, no other sort of “consciousness” that gets produced by brains.
RL: Agree that consciousness has to be a product of brain activity. You and I have struggled over whether consciousness IS brain activity, or a product of brain activity. Both, I’m thinking. Conscious experience happens.. most philosophers will admit that, which does not mean that we understand what it is. One problem is that when we start talking about it, consciousness seems inextricably bound to a recipient “self,” and that thing is a model, not an experience. That conscious experience occurs is a fact; what it is or does is not a fact.
RL: This does remind me of Hume’s inability to perceive consciousness itself, independent of content. Conscious vs unconscious, we can do; but not consciousness without content. Seems like maybe it’s a goal for meditation, but I’ve never seen anyone suggesting that we try for it directly. It’s always breathing, or a word, or a thing that you concentrate on, excluding all else. Consciousness itself, independent of content, is very difficult to observe.
GNS: I do not see how we can say both 1) conscious IS brain activity, which is a version of materialism, and 2) consciousness is produced by brain activity, which would lead to dualism.
RL: Maybe we could just say that consciousness is brain activity when viewed from the third-person, and a product of brain activity when viewed from the first-person. If we had a machine that could see and discriminate all of the specifically consciousness-supporting brain activity, we’d be looking at consciousness. But we wouldn’t be having the first-person experience.
0:39:40 DD: In fact, when John Holland wrote his book Emergence, I said, “John, you’ve got to put a foreword in where you say what you don’t mean.” And I completely approve of John Holland’s work on emergence, because it does not mean that this is an inexplicable pattern, precisely not. In fact, I like to illustrate emergence with John Horton Conway’s Life World and the amazing patterns that emerge there, and say, look, that’s emergence, and that’s completely explainable and predictable. There’s no question mark anywhere in that system, but it creates stunning emergent effects.
The Guardian: Conway “invented the Game of Life, a cellular automaton that to this day retains cult status. It is not a game proper; Conway calls it a “no-player never-ending” game. It is played on a grid, like tic-tac-toe and, according to three simple rules of Conway’s devising, the cells placed on the grid proliferate, resembling skittering micro-organisms viewed under a microscope. A cellular automaton is in essence a little machine with groups of cells that evolve from iteration to iteration in discrete rather than continuous time – in seconds, say, each tick of the clock advances the next iteration, and then over time, behaving a bit like a transformer or a shape-shifter, the cells evolve into something, anything, everything else. As such, the Game of Life demonstrates how simplicity generates complexity, providing an analogy for all of mathematics, and the entire universe.” https://www.theguardian.com/science/2015/jul/23/john-horton-conway-the-most-charismatic-mathematician-in-the-world
A single Gosper’s Glider Gun creating “gliders” in Conway’s Game of Life.
0:40:40 SC: Yeah. And you can talk this higher-level vocabulary and capture some of the real.
DD: That’s what emergence gets you.
GNS: That patterns “emerge” out of mathematics or the motions of particles is all well and good, but it’s still a long way from understanding how consciousness, morality or value emerge. Maybe those things do not exist. Dennett would say, they exist, but they are not what you think they are.
[advertisement]
0:42:00 SC: So with that definition of emergence on the table, it seems to make sense that patterns are real, if we’re going to ascribe reality to these higher-level things that give us some way of capturing what’s going on. Purposes and intentions are in that bucket. They serve a purpose. So now we get to consciousness. Maybe I’ll just let you fit it in.
0:42:26 DD: And it emerges in this innocent sense. The idea that it’s one thing, that everything in the universe is either conscious or not, that it’s the light is on or the light is off— that is a fundamental error. But it’s very widespread. It’s just amazing how many really deep and clever thinkers can’t get out of their heads that consciousness is all or nothing. And I think, no, it’s emergent. What that means is that the search for the simplest form of consciousness, that’s a snipe hunt, it’s a wild goose chase. Because it emerges. And yes, starfish have some of the aspects of consciousness, so do trees and bacteria and as you do.
GNS: Dennett means that patterns of behavior emerge in more complex animals. Nobody thinks consciousness is “all or nothing” in the sense that there cannot be degrees of consciousness, in the way RL has described. But many “really deep and clever thinkers” (plus me) do think that everything in the universe either has at least a teeny bit of consciousness or none at all. “An organism has conscious mental states if and only if there is something that it is like to be that organism—something it is like for the organism. (Tom Nagel “What Is It Like To Be a Bat?” 1974. Dennett “while sharply disagreeing on some points, acknowledged Nagel’s paper in Consciousness Explained as “the most widely cited and influential thought experiment about consciousness.” wikipedia. Dennnett is everywhere) Bats? Yes. Bacteria? No. Trees and rocks? No. Laptops? No. Parrots? Yes. Octopus? Yes. See how easy it is! Bumblebees? Flatworms? Uh, not sure.
0:43:50 SC: But not electrons?
DD: But not electrons. And we can argue about motor proteins.
0:44:00 SC: But once you admit that, it’s nothing mystical. Something that builds up, then you can write it.
DD: And the question where do you draw the line is an ill motivated question. That’s like where do you draw the line between night and day?
0:44:10 SC: Do you have a simple definition of what consciousness is that you prefer?
DD: No.
GNS: We love Dennett, but he is a master of evasion. He got it from Gilbert Ryle, who got it from Wittgenstein. For Dennett, consciousness is a form of behavior, but he avoids saying “behaviorism”, because that would be bad PR.
SC: You did write a book called, Consciousness Explained, so this might be an okay question.
GNS: His snarky critics call it Consciousness Explained Away.
0:42:04 DD: I did, but I think that’s the way science proceeds too. Scientists don’t sit around wasting hours and hours and hours trying to define time or energy. They get on with the theory and once they’ve got a really good theory, it will be obvious what time or energy is. And I think that’s the same as consciousness.
GNS: What time, energy and consciousness are is not obvious.
0:45:00 SC: Okay. But still, you must have something in mind.
RL: We assume that there has to be a physical basis for consciousness. In any case, it depends on whether we’re looking at it first person or third. Pretend we have a scanner that can isolate all the physical events in your brain that track with consciousness. Just consciousness itself, not its content. From third person, we’d be looking at consciousness. But the subject of this study, looking at the same activity in real time, what he was sees on the screen will be very different from the experience of looking at that screen. On the other hand, what our subject is looking at on the screen is an entirely sufficient explanation for what he is experiencing. No mental change without a physical change.
What does conscious do or not do? From my feeble introspection, does consciousness actually produces its own content? I just don’t see it happening. I can’t be conscious of something that isn’t there yet in the neurons.
People are often wrong about the sources of their thoughts, not just in the academic sense of errors in recollection or reasoning, but in identifying factors that influenced their conclusions.
GNS: You are never conscious of the subpersonal neuronal processing necessary for consciousness. What you mean is that you cannot be conscious of anything before the appropriate brain activity has occurred, and the specific contents of a mental state are a function of the underlying neuronal patterns.
DD: Let’s talk about human consciousness. My view is an embattled one, but I’m pretty sure of it. Human consciousness is much different from the consciousness of any other species. There are many reasons why this is hard to see. One is that consciousness has a moral dimension. We want to be kind to animals, so when we say, “Well yes dogs are conscious but not the way we are”, people immediately get their backs up, “He’s about ready to talk about mistreating animals.” No, no, no. I think the properties of human consciousness that we share with dogs and mammals and birds, to some degree with reptiles and fish, have moral significance. So let’s see if we can take moral significance as itself a graded notion. Under British law, Octopus Vulgaris is protected. It’s an honorary vertebrate. It’s against the law to throw a live octopus on a hot grill. It’s not all cephalopods. It probably should be. Squid, you want to throw a squid on a live grill, you can.
0:47:15 SC: But the point is that you are allowed to boil a lobster. I think cut-offs are okay in general for these tricky questions. People say, “Well if you draw a line here, then they argue about either side.” But you gotta draw it somewhere, especially for legal purposes.
0:47:45 DD: What we should recognize is the law draws lines that are reasonable to the vast majority of people, and we can talk about exceptions. And this is an interesting case. The wonderfulness of octopus, the amount of convergent evolution between octopus and human beings is enough to push them over the line. And I think I approve. Before we put the moral issue behind us, we should know that almost nobody wants to hold any non-human species morally responsible for their behavior. That’s key. They may be moral patients, but they’re not moral agents. A bear that kills a tourist has not committed murder. Because they don’t have the mental wherewithal, they don’t have the kind of free will that we have.
0:49:00 SC: We couldn’t have offered them a reason not to do that.
DD: That’s right. We can’t expect them to appreciate the societal norms that we’ve set up and so forth. So don’t look in the bear’s brain and a human brain for the fact that one of them is indeterministic and the other one is deterministic. Determinism has nothing to do with the issue. It has to do with information. It has to do with self control and with degrees of freedom. And degrees of freedom is a term that I’ve been using more and more recently, and really seeing it come more out of engineering than out of physics, and thinking a degree of freedom is an opportunity for control. You can clamp a degree of freedom and then you don’t have to control it. You can just lock it down in one way or another.
0:50:10 DD: How many degrees of freedom do we have? Millions. Billions, because if we can think about so many things we have orders of magnitude more degrees of freedom than a bear does.
SC: With roughly the same number of cells and so forth, but the complexity is much higher.
0:50:30 DD: Yeah. It means that the options that the bear has are a vanishing subset of the options that we have. Learning to control our perusal of those options, that’s not a science, it’s an art. We try to train our kids so that when we launch them, and they are no longer in our control, they will be able to control themselves in ways that will lead them to have happy and productive lives. If they can’t, they are going to get in trouble.
0:51:30 SC: You’ve used the word agent a few times, and I use it all the time, but we haven’t yet described what that word means. So it’s clearly a relationship between agency, responsibility, consciousness. Is there a simple definition of “agent” if not of “consciousness”?
0:51:40 DD: Agents come in all sizes and shapes too. A few minutes ago we were talking about bacteria as agents, viruses as agents. So that’s not the sense of “agent”. We want a moral agent. We want to talk about a moral agent as not just a locus of self control with purposes and an ability to fend for itself, and prolong its existence and enhance its circumstances. That’s a pretty good definition of an agent: something that can fend off the second law of thermodynamics, fend off dissolution. Mountains aren’t agents because of erosion. They can’t protect themselves, or move or anything.
0:52:44 SC: But you can see why tornados are an interesting edge case.
0:52:47 DD: Exactly and that’s why tornadoes are an edge case. But at the most sophisticated, as we climb that ladder, a pretty good scale would be: how many degrees of freedom are available for control? When it gets up into the billions, as it does for even young children, now we’re talking about potential moral agents. A moral agent is simply a human being— at the moment we don’t have any other, we could but we don’t— a human being that is mature enough to control the degrees of freedom that matter when they matter, and to be able to foresee and understand the outcomes of possible actions, and act accordingly. I call it a member of the moral agents club.
0:54:18 SC: Good. I think that does make sense to me, but it deviated us from our task, which was explaining the salient features of human consciousness.
0:54:27 DD: One of the curious features of the way the science has proceeded here is that of the many theories of consciousness, only a tenth have a theory. This is the inbound path or the upward path, and we get from the photons striking the retina, and the sound waves and up and up through the nervous system, up through the various cortical areas, and then ta-da, consciousness happens. That’s the end of the theory. Wait a minute, I want to ask what I call the hard question: And then what happens? What makes whatever you say amount to consciousness? What does it enable? What does becoming conscious of this or that enable the agent whose consciousness it is to do? Or disable that agent from doing? What effects does it have on those multiple degrees of freedom? And the answer is, almost anything can happen, but we need to have the neuro-scientific theory of how that can be true, and how the various sequelae, the various outcomes, can spell themselves out.
0:56:15 DD: Some people’s theories of consciousness are a little bit like somebody who mounts a closed-circuit TV camera on the hood of his car and puts a receiver under the hood so the car can see where it’s going. No! What’s going to consume that information? Ruth Millikan talks about the consumers of representations. And in scientific theories of consciousness, there has been a systematic neglect of the consumers.
0:57:07 SC: Sorry, the consumers are? Didn’t understand that.
DD: The consumers are, ultimately, neural structures that respond to representations spread all over the brain in waves that give rise to the ability of people to report and reflect on and remember. There’s a tremendous difference between sensing something and noticing that you’re sensing something. And noticing that you’re noticing that you’re sensing something.
RL: Much of this seems contaminated by an assumed homunculus. He has inputs, and behaviors, moral or otherwise, and a conscious agent in between. He’s assuming that the consciousness is essential to moral decisions, and also that there is a self doing the deciding, rather than some unconscious mechanism. Consciousness is better at seeing what we’ll use for an excuse, than the actual reasons why we did what we did.
GNS: 1/18/20 For what it’s worth, Dennett has been been denying— rather than assuming— the “homunculus” for 50 years. Sometimes he discusses mental states in an “as if” sort of way. His discussion of “consumers” could be considered metaphorical.
RL: Agree. Fair. It’s more that when he talks about understanding, and foreseeing, and moral agents clubs, the language seems to imply a conscious agent. He does seem to reject the “consumer” idea outright.
GNS: 1/22/20 Sometimes Dennett seems to make very free use concepts and terminology he otherwise argues are incoherent or meaningless.
RL: The way I imagine it, Darwin equipped you with a modeling machine, and it will try to model anything placed before it. We don’t generally pay a lot of attention to noticing that we’re noticing that we’re noticing, unless/until some philosopher plants that question in front of us.
End of Part I
Bonus coverage on Consciousness and Free Will
GNS: Your view is that we cannot be conscious of anything before the appropriate brain activity has occurred. And the brain patterns determine the content.
RL: I’m saying that all consciousness can do is display what’s given to it. All of what’s given to it is constructed by subpersonal processing, which we can’t see. Even when we are going through the steps of some learned algorithm, recognition of its applicability, the retrieval of the steps, and the processing that tells us the conditions are met, and that the conclusions are valid… all of that is done by subpersonal processing. If there is doubt at any stage, that’s attached by subpersonal processing. More compactly, consciousness never shows us process, only product. When it shows us steps along the way, it’s again showing us those as product of invisible underlying processing.
GNS: Consciousness is caused by brain activity. The specific contents of consciousness— a tomato, a thought, a headache— must be due to some specific patterns of neuron firings. That’s the basis for neuroscience’s search for the NCC, the neural correlates of consciousness. The neuron firings are subpersonal, which means that we are never “conscious of” them. We are only conscious of the resulting content.
Since we are not conscious of the subpersonal processing, we cannot control it. That means that we cannot control the contents of our consciousness. So everything we think and feel is beyond our control. We are not responsible for our thoughts, much less our actions.
This is dramatically at odds with common sense. It’s one of the major ways the scientific conception diverges from what Wilfrid Sellars calls the “manifest image”.
SEP: That’s what Sellars sees as the major problem confronting philosophy today. This is the “clash” between “the ‘manifest’ image of man-in-the-world” and “the scientific image.” These two ‘images’ are idealizations of distinct conceptual frameworks in terms of which humans conceive of the world and their place in it. Sellars characterizes the manifest image as “the framework in terms of which man came to be aware of himself as man-in-the-world”, but it is, more broadly, the framework in terms of which we ordinarily observe and explain our world. The fundamental objects of the manifest image are persons and things, with emphasis on persons, which puts normativity and reason at center stage. According to the manifest image, people think and they do things for reasons, and both of these “can occur only within a framework of conceptual thinking in terms of which [they] can be criticized, supported, refuted, in short, evaluated”. In the manifest image persons are very different from mere things; how and why normative assessments apply to things is an important and contentious question within the framework. https://plato.stanford.edu/entries/sellars/#5
GNS: If we are not responsible for our thoughts, much less our actions, in what meaningful way can we be “criticized, supported, refuted, in short, evaluated”? There would be no normative morality. Only the facts about what people do or don’t like, only the facts about social practices. People will do what they do, but they would have no right to do one thing rather than another.
Can normativity can co-exist with science?
SEP: Sellars thinks that science’s a-perspectival description of the empirical world is the measure of reality, but he is also committed to the indispensability of the concepts built into the first-person perspective that makes agency possible. If those concepts were not involved in the regulation of our behavior, we would not be persons and could not engage in such activities as moral behavior or scientific research. This is why Sellars calls for a stereoscopic vision in which the descriptive resources of the sciences are united with the language of individual and community intentions and the dualism of the manifest and scientific images is transcended. https://plato.stanford.edu/entries/sellars/#5
GNS: I’m not so sure.
RL: In the discussion coming in Part II, the next BQTA post, you say that Dennett says it’s still free will even if it’s biochemically determined and could not have come out otherwise. Just can’t be externally determined. That’s certainly not how the culture thinks of it. Folk philosophy has it that it’s your self that decides, and that the decision takes place in the conscious arena.
GNS: Your right that Dennett’s compatibilism is not the way we think of free will. I do not understand how we can have any kind of free will if our every thought and action is caused solely by biochemical processes we cannot become conscious of or control. And I do not see how that biochemical determinism can be compatible with normative morality. So does that mean that normative morality is purely mythological, like religion? Folk philosophy (usually called “folk psychology”) seems to require immaterial entities like the Self which cannot exist. And it requires thinking about reasons for action, where the thinking is product of immaterial consciousness, not subpersonal processes.
RL: We haven’t speculated on what consciousness is actually for, from a design standpoint, for a while now. I still think Darwin put it there for a reason.
GNS: We always end up going around in the same circle. First, you insist that consciousness and its contents are beyond our control: “All of what’s given to it is constructed by subpersonal processing, which we can’t see.” So if what’s given to consciousness is always constructed by subpersonal processing, consciousness cannot itself do anything. That seems right.
Subpersonal processing is the cause of consciousness. The effect cannot come before the cause. That’s what “I can’t be conscious of something that isn’t there yet” is supposed to mean. The causation flows only one way— from processing to consciousness. For the causation to flow in the opposite direction—from consciousness to processing—consciousness would need to be operating independently of processing. That’s not possible. The brain is the source of bodily movements. So to affect movement, consciousness would need to cause changes in the brain’s subpersonal processing. That would require you to have a conscious thought before the subpersonal processing occurs. That’s not possible. So consciousness cannot possibly do anything. Consciousness cannot have a function. With no possible function, consciousness cannot be adaptive, and so is invisible to Darwin. Consciousness cannot have evolved, except as incidental to the subpersonal processes that are adaptive. Darwin did not put it there, because there is no conceivable reason for it.
Better Questions Than Answers 