Four Features of the Mental
There are four features of the mental which someone who denies the ultimacy of mind maintain must not be found on the rock bottom level of the universe. The first mark of the mental is purpose. If there is purpose in the world, it betokens the existence of a mind that has that purpose. So for anyone who denies the ultimacy of the mind, an explanation in terms of purposes requires a further non-purposive explanation to account for the purpose explanation. The second mark of the mental is intentionality or about-ness. Genuinely non-mental states are not about anything at all. The third mark of the mental is normativity. If there is normativity, there has to be a mind for which something is normative. A normative explanation must be explained further in terms of the non-normative. Finally, the fourth mark of the mental is subjectivity. If there is a perspective from which something is viewed, that means, once again, that a mind is present. A genuinely non-mental account of a state of affairs will leave out of account anything that indicates what it is like to be in that state.
If the mind is not ultimate, then any explanation that is given in terms of any of these four marks must be given a further explanation in which these marks are washed out of the equation.
IV. Minimal Materialism
There seem to be three minimal characteristics of a world-view which affirms that the mind is not ultimate. First, the “basic level” must be mechanistic, and by that I mean that it is free of purpose, free of intentionality, free of normativity, and free of subjectivity. It is not implied here that a naturalistic world must be deterministic. However, whatever is not deterministic in such a world is brute chance and nothing more.
Second, “basic level” must be causally closed. Nothing that exists independently from the physical world can cause anything to occur in the physical world. Second, the level of basic physics must be causally closed. That is, if a physical event has a cause at time t, then it has a physical cause at time t. Even that cause is not a determining cause; there cannot be something non-physical that plays a role in producing a physical event. If you knew everything about the physical level (the laws and the facts) before an event occurred, you could add nothing to your ability to predict where the particles will be in the future by knowing anything about anything outside of basic physics.
Third, whatever is not physical, at least if it is in space and time, must supervene on the physical. Given the physical, everything else is a necessary consequence. In short, what the world is at bottom is a mindless system of events at the level of fundamental particles, and everything else that exists must exist in virtue of what is going on at that basic level. This understanding of a broadly materialist world-view is not a tendentiously defined form of reductionism; it is what most people who would regard themselves as being in the broadly materialist camp would agree with, a sort of “minimal materialism.” Not only that, but I maintain that any world-view that could reasonably be called “naturalistic” is going to have these features, and the difficulties that I will be advancing against a “broadly materialist” world-view thus defined will be a difficulty that will exist for any kind of naturalism that I can think of.
posted by Victor Reppert @ 11:27 PM
73 Comments
73 Comments:
Purposiveness and normativity are oddballs there. I'd chop them off as they can (arguably) be reduced to the two good ones.
Subjectivity and intentionality are certainly kosher. Could it be that the two "aspects" of the mind are really two different processes that don't bear any necessary relation to one another? E.g., could there be a creature with intentional states but not subjective states?
Hooray for semantic victories, I guess.
I don't think that causal closure is a critical issue for naturalism. For example, as a naturalist, I don't dogmatically declare that there are no supernatural causes. My contention is that there are no supernatural explanations, and so if there is supernatural causation, it manifests itself in the form of fundamentally inexplicable events.
So there are really two issues at stake here. The first is dualism, and the other is naturalism, and I don't think they're intrinsically conflicting. Is there any reason (in principle) why basic mental properties could not be natural?
I think the reason we associate dualism with supernaturalism is that dualism is utterly predictionless, just like other supernatural phenomena. A predictive dualism would predict minds that violate physical laws. Yet in every experimental test, dualism is not only bound by physical laws, it looks just like physicalism. Dualism has been reduced to a supernatural claim because it shirks prediction. All dualism can do is (dogmatically) deny that mental phenomena can be fully explained by predictive mechanisms.
If dualism predicts that there just aren't going to be any successful reductions from the mental to the physical, and it turns out after a few centuries that the same "mysteries" that even naturalistic philosophers say are going to remain are still mysteries in, say, 2608, wouldn't that be a successful prediction on the part of the dualists?
Hi Doctor Logic,
Mental phenomena, as mental, seem to me to have a good deal of explanatory success. For example, my decision to have chicken for supper tonight at 6:00 predicts the sort of particles that will enter a lump of matter at a specific time in the future. More often than not, those types of predictions are accurate.
Hi BDK,
Perhaps we have chatted about this before, but could you review for me how normativity can be found in the language of physics alone? I'm with Davidson on this issue, but I am open to an argument which does derive an ought from an is, either with respect to morality or epistemology.
Victor,
What constitutes a successful reduction as far as the dualism question is concerned?
If I create a simulation of a human brain (and body), and the simulation exhibits apparent consciousness, does that falsify dualism?
rino,
For example, my decision to have chicken for supper tonight at 6:00 predicts the sort of particles that will enter a lump of matter at a specific time in the future. More often than not, those types of predictions are accurate.
I don't see what this has to do with dualism. This prediction is in the nature of mental phenomena, whether or not the mental reduces to the physical.
Hi doctor logic,
My example does presuppose several things that go unmentioned. first of all, that mental states such as decisions to eat chicken offer genuine 'mental as mental' causation, which by definition could not be reduced, since that would be mental as physical. secondly, that 'mental as mental' causation is real, in that it successfully predicts the way the world will turn out. and third, that 'mental as mental' causation implies at least a dualism of levels (there are, of course, more like five levels in the traditional layer cake model of the universe), so it is a sort of dualism. thus, if the example i gave is true, then we have an example of how the mental phenomena implicated by dualists has explanatory success. if this much is true, then dualism is not utterly predictionless, in fact 'mental as mental' explanations successfully predict thousands of things per day.
perhaps you will say that this form of dualism is not what you were talking about. that's fair enough. however, from the other points you made in your post, i got the impression that you were sympathetic to some sort of a naturalistic dualism, which is what i was trying to provide.
cheers.
Hi rino,
I guess I still don't get it. If I claim that girls are made of sugar and spice and all things nice, that won't explain that girls give birth, breast feed babies, have higher voices, etc. Those attributes that define girls are prior to the thesis that girls are made of sugar and spice. Likewise, our observation of mental phenomena is prior to the thesis of dualism. So what would be different if there were no dualism?
I would not describe myself as sympathetic to naturalistic dualism. I am a reductionist through and through. However, I do think that there could be a good-faith a naturalistic dualism, or, rather, that one could have existed before we learned what we have learned about cognitive science. I think that to claim that the mind is not physical is akin to claiming that water does not reduce to H2O.
Rino: I commented that it is strange to say that normativity is constitutive of mind, which is a different issue than the one you bring up.
DL is right that it is possible to be a dualist naturalist, as Chalmers actually takes this approach. He's not a substance dualist, though.
BDK
Perhaps "normativity is constitutive" is not the best way to put it. But it is hard to imagine non-mind being sensitive to ethical rightness and wrongness, or correctness or incorrectness of reasoning (broadly defined). To use Victor's example, rocks falling down a hill do not contemplate whether it is right or wrong, correct or incorrect, for them to fall as they do. At least in the case of morality and rationality, minds alone seem to have such sensitivity.
Another way to think of this (at the risk of just repeating the last go-round) is that there is no scientific demonstration that cheating or treachery is morally wrong. There also can be no scientific demonstration of the correctness or incorrectness of reasoning that does not presuppose such correctness/incorrectness. And where there is no possibility of scientific demonstration, we are not talking about natural properties.
Hi Doctor Logic,
I fear I am going to be getting off of Victor's original topic, and I'm hesitant to do so, but if you wish to reply, then I will assume it is fine.
You say: "Likewise, our observation of mental phenomena is prior to the thesis of dualism. So what would be different if there were no dualism?"
I would disagree here. Our observation of mental phenomena implies that there is an 'us', and there is something being 'observed', namely 'mental phenomena'. These things are the very subject matter of dualism, and it is not clear that a reductionist can simply smuggle them in if their worldview does not anticipate their existence.
Perhaps I can ask the question this way: as a good reductionist, you would seem to think that physics is all there is in the end. Is that a fair assumption? If so, then why did the world end up in such a way as to have conscious minds? Alright, for the sake of the argument I will grant that it just happened, and is a mystery. But then, once these minds learned that everything was just physics, why did they continue to talk as though there were minds? The move to reduce not eliminate, thereby keeping the mental, seems very valid to you, of course, but very strange to me. Maybe I can flip the question around so you can see the oddness from my point of view.
Imagine a theist who loses belief in a god, and thinks that there is just the good nature of humans instead. What should this theist do? Obviously, he should say 'i no longer believe in god, i'm an atheist now'. But then, he realizes that his family and friends will get mad at him, and he may not get to go to heaven, so he says 'I do believe in god, but god just is the good nature of humans, but i am still a theist'. Doesn't that seem odd to you? Either just believe in god, or don't. Don't say god exists, but then redefine him in such a pickwickian manner that he is unrecognizable. Obviously such a move smacks of desperation, and is laughable to the good many smart atheists out there.
The issue is the same in the case of the mental. If you believe in genuine mental-as-mental causation then fine, there is mental causation and presumably physical causation, ie, two (dual) causes of some sort. If you really only believe in physical causation, then just believe in physical causation, and no need to complicate things with an additional mental vocabulary. Surely science is flush with enough moral conviction in parsimony (Ockham), elegance, and exclusion to bar such a move.
Thus, the acceptance of genuine mental-as-mental causation does lead to dualism, in fact, it is the dualism. An acceptance of mental phenomena is the acceptance of dualism, ie, that there are mental phenomena and physical phenomena, ie, two (dual) things. If there is a reductive identity, then there is only one thing, and we should stop talking about two things. But you are still talking about mental phenomena, and presumably would talk of physical phenomena as well, so you are talking about two things, so you have accepted a dualism. If there is only one thing, then stop speaking in terms of both physical and mental phenomena.
Anyway, like I said, I'm off topic. But, if you have any comments, please feel free to share. Cheers.
DB: that is fine, I'd accept that, as I think normativity supervenes on intentionality and experience, so doesn't introduce novel metaphysical problems beyond those. The same with purposiveness.
Hi Rino,
But then, once these minds learned that everything was just physics, why did they continue to talk as though there were minds? The move to reduce not eliminate, thereby keeping the mental, seems very valid to you, of course, but very strange to me.
Reduction does not eliminate collective effects. Knowing that the Pacific Ocean is a collective effect of atoms doesn't make the ocean disappear. Nor is it the case that waves no longer have causal effects. The reduction simply means that oceans aren't basic or fundamental.
Similarly, if minds are collective effects of atoms, etc., minds won't disappear, and thoughts won't cease to be causally effective.
What instance of reduction makes you think otherwise?
Perhaps I can ask the question this way: as a good reductionist, you would seem to think that physics is all there is in the end. Is that a fair assumption?
Well, I'm not so committed to reductionism and physicalism that they are my starting point, but I would say that even non-material explanations would require a "physics" of their own. So let's call it a fair assumption for now.
If so, then why did the world end up in such a way as to have conscious minds?
I think that almost any replicating life form needs a sense of self, e.g., an ability to distinguish its body from the environment or food. I also think that abstract thinking is an extremely powerful tool for survival (and it is certainly so in social groups). If you combine the two, you get an animal that can do abstract, reflective thinking about itself.
BDK
>>I'd accept that, as I think normativity supervenes on intentionality and experience, so doesn't introduce novel metaphysical problems beyond those.<<
If intentionality consists of physical relationships between objects/states, and if the supervenient properties of these relationships are natural, then I cannot see why propositions concerning normative concepts would not be, potentially at least, scientifically demonstrable or confirmable. Working backward from that, if such propositions are in principle not scientifically confirmable, then something is wrong with that kind of naturalizing explanation.
DL
Rino's point may have to do with holding the mental within the same causal context as the physical. For example, gravitational fields are distinguishable from physical forces/effects generally in the way that any subset is distinguishable from the set of which it is a part. But gravitational fields can be placed in the same causal context as other physical forces/effects, showing that it is indeed a subclass not an entirely separate class from the physical broadly conceived. This is not obviously true of purpose, meaning, moral rightness/wrongness, rational correctness/incorrectness, etc.
Hi Doctor Logic,
You say:
"Reduction does not eliminate collective effects. Knowing that the Pacific Ocean is a collective effect of atoms doesn't make the ocean disappear. Nor is it the case that waves no longer have causal effects. The reduction simply means that oceans aren't basic or fundamental.
Ok, I will use your example here, though it is an unhappy way to put my point: would you say that the 'ocean as ocean' causes things, or the 'ocean as microparticles' causes things? Would you say the 'wave as waves' cause things, or would you say that the 'wave as microparticles' causes things? Presumably you would say that the wave as microparticles causes things, since there is no 'wave' over and above the microparticles. So, my point is that we could tell a complete causal story of everything that the ocean ever does without ever talking about waves or water, but only by talking about H2O, (or actually by talking about something more fundamental than H2O). So, why keep talking about waves, or oceans? Only microparticles exist. Parsimony, and exclusion, and elegance, should compel us to stop talking about waves and other such wholes that are no more than their parts.
I am trying to question the internal consistency of reductionism here. Reverting to my previous example, if our theist friend says 'god does exist, he just is reduced to the good nature of humans', is that person still a believer in god? If he says 'god still has causal power, since he is the collective effect of all the good that humanity has ever achieved, and humanity has achieved lots of good', does that logic make sense to you?
If we reduce, then we are left with the conclusion that 'god as god' does not cause, but 'god as the good of humans' does. 'waves as waves' does not cause, but 'waves as a cloud of particles' does. The mental as mental' does not cause, but 'the mental as a cloud of particles' does cause. I'm trying to say, using the ethics of most scientists, that we should abandon the dual explanations if the bottom level explanation will suffice. This amounts to the elimination of 'waves', 'the mental' and anything else that is reduced. This logic is actually quite persuasive, in my opinion, and has been often used in getting rid of such things as the concept of phlogiston, the four humors, witchcraft, ether, etc... The only problem is that physicalists see the problem in eliminating the mental in such a way as well, so they do all this logical gymnastics to preserve it somehow, but it doesn't work. Or rather, if it can work, then we can preserve the existence of phlogiston, humors, witchcraft, and other eliminated theories as well.
Anyway, I know this is a big topic. If we want to get into it more deeply, we would have to sort out the differences between ontological reduction and reductive explanation, for it is hard to stay clear on the issues without being constantly mindful of the distinction.
Hi Rino,
Reducing the ocean to microparticles does not eliminate the waves or the need to talk about them. I still need the concept of a wave, even in terms of microparticles because I need to talk about the collective motion of the microparticles when I see a wave.
As a scientist, I find it very odd that you think reduction eliminates anything. Consider the reduction of atoms to protons, neutrons and electrons. In theory, we don't need to talk about elements like Helium and Sodium, we can just speak of atoms with two protons and two neutrons, or of atoms with 11 protons and 12 neutrons, respectively. But why shouldn't we give names to stable collections of the microparticles that make up atoms? And why shouldn't we call these stable states by the corresponding chemicals element name? After all, that's precisely what they are. Nuclear physicists don't stop using the word Sodium because Sodium is what the proper name is for that configuration.
Indeed, this goes beyond naming. Sodium is 11 protons and 12 neutrons. The reduction does not eliminate Sodium, it tells you what Sodium is.
You give phlogiston, the four humors, witchcraft, and the ether as examples of things that were eliminated by reduction. However, they weren't eliminated by reduction. It's not as if phlogiston exists, and we reduced it to chemical bonds. Phlogiston doesn't exist, period. The phlogiston theory was completely wrong. The obsolete things you cite were theories that were eliminated because they failed to be explanatory or because they made the wrong predictions or because better explanations came along.
I'll just throw in one more example. Suppose something is breaking into your garage and nibbling at insulation, knocking down paint cans, etc. Suppose you theorize that the guilty party is a gang of creatures belonging to a species that is half snake, half duck. However, you later learn that the guilty party is just a local squirrel. In this case, you have not reduced the gang of snakeducks to a squirrel. You have replaced the snakeduck gang with a squirrel. The snakeducks do not exist. They don't reduce because they don't exist.
Hi Doctor Logic,
Thanks for the comments. I would say you have missed something important. For example if I say H2O, then I refer to two hydogen and one oxygen in whatever structural relationship I want, no structural relationship is given, it has been eliminated, or at least not mentioned. One of the Hydrogen molecules could be a mile away from another, whilst the oxygen molecule is ten feet away. If I say 'water', then I am referring to H2O in a certain structural organization, and this structural organization cannot be lost without it no longer being water. So, to say H2O is water is not accurate. Counterfactually, there could be H2O without it being water. To use another example, a car on the factory floor has all the parts of the finished car, but it has not been put together. If we reduce to the parts alone, but pay no heed to the important structural properties, then we are stuck with saying that it is a car when all the parts are randomly distributed on the factory floor. Obviously, all the parts are there, but the organization is not there. And it is the organization that gives it various causal powers that the parts alone do not have. My point is that you seem to be eliminating the structural properties whenever you reduce.
I would argue this leads to a non-reductive picture: You can't reduce structural properties to the parts. If you do try to reduce structural properties, you have eliminated them, and why talk about the higher level anymore which involves the parts and a certain structure of the parts?
Hi Rino,
For example if I say H2O, then I refer to two hydrogen and one oxygen in whatever structural relationship I want, no structural relationship is given, it has been eliminated, or at least not mentioned.
You may say this, but no one in the sciences ever does. H2O refers to a specific molecule with a specific structure. This is the same reason we refer to O2 and N2 molecules instead of just oxygen and nitrogen respectively. If the designations referred only to proportions, we would not say O2 or N2. We do not say the Earth's atmosphere is made of CO21N78, but rather we say it is 21% O2, 78% N2, some tiny percentage of O3 (ozone), and so on.
Now, suppose we had not yet discovered atoms and molecules such as H2O, but were aware that water decomposed into two parts hydrogen and one part oxygen. In that case, we would still know that the hydrogen and oxygen has to be mixed or fused in some homogeneous manner in order for the reduction to work. After this reduction, no one ever thought the left side of a glass of water might be all oxygen. So even before atoms and molecules, structure was part of the reduction.
There's not one example of reduction in the sciences that omits structure. So I think you are attacking a straw-man version of reductionism.
On the issue of mind, when dualists criticize reductionism, they're not denying the reduction on the grounds that the reduction omits the structure. Everyone in science of mind knows that the structure is highly complex. But just like we knew water was some complex of hydrogen and oxygen, we know that the mind is some complex of chemical compounds. They're saying that even if reduction includes mechanical structures, it still won't account for mental functions.
Hi Doctor Logic,
Thanks for the comments. You say:
"There's not one example of reduction in the sciences that omits structure. So I think you are attacking a straw-man version of reductionism"
So long as you are willing to acknowledge that 'water is H2O in a specific structure' is more accurate than 'water is H2O', I will be happy. I think you are willing to acknowledge this. You even seem to think that this perhaps an obvious observation. I think it is not. There are numerous reasons we need to go with the first. Not only for the reason that H2O without a specific structure could mean one hydrogen molecule in Africa, one is Argentina, another oxygen molecule in Finland. Under this structure, various causal properties of H2O with its specific structure found in water would not have emerged. Not only this, but it also stops reduction, in my opinion. Since structural properties are relations between parts, you cannot look only at parts without eliminating the relationship between parts. So, may I just clarify that you accept the view that water is more properly defined as 'H2O in a specific chemical bonding relationship'.
As for the further issue of mind, you are definitely correct that dualists want more than just an admission of structure. However, in my opinion this is a building block. After the structure is established, we need to posit intelligible properties which supervene on the structure of the whole.
Here is an example. Take a typical biological property, perhaps the fact that a mouse is attempting to survive (I take it that Darwinism is not controversial with you). Typically the reductionist will say 'look for the mechanism that realizes the survival instinct, and we have reduced survival instinct to that mechanism'. To this we can ask 'so is the mouse really surviving, or are we just reading into that'. If we say yes he is surviving, then we are saying that the survival instinct is a property of the mouse, it is really there. If we say no, then we are just talking fictitiously. Although this is a path frequently taken, I would not recommend it. I'll assume you take path (1), until I hear otherwise.
So, assuming the first path is taken, the property of having a survival instinct is really a property of the mouse. We say 'where do we see that'? And the answer is we see it nowhere. But we can understand what the mouse is doing. Perhaps we say that survival instinct is an intelligible property of the mouse, a property which can be understood by an intellect. We can look at a certain structure of chemicals and say 'aha, I understand what those chemicals are doing together, that is quite smart, they are surviving'. On the contrary, if the same group of chemicals were arranged in wildly different ways, they would not be surviving, they would be doing nothing intelligible, they would just be a lump of soup on the floor.
Anyway, I know you will not accept this, but I'm trying to offer an alternate way of looking at things, just for something to think about, if nothing else. Reductionists tend to take for granted the structure and the intelligibility that arises out of the structure. I understand their reasons, but for the life of me I don't understand what the horror is of granting that intelligibility (things understoood but not seen) into their ontology. It is as is intelligibility is a boogy man, better not say anything is actually intelligible, that would be some crime. Anyway, please correct me if you gladly accept such properties.
Nice talking.
Rino said:
If intentionality consists of physical relationships between objects/states, and if the supervenient properties of these relationships are natural, then I cannot see why propositions concerning normative concepts would not be, potentially at least, scientifically demonstrable or confirmable.
We don't even have to discuss naturalism with this: my claim is that normativity supervenes on intentionality and experience. So I can be neutral wrt the metaphysics of intentionality and experience. Assume I am a dualist about intentional contents as well as qualia. It doesn't technically matter for the argument I'm making against normativity or purposiveness being constitutive of the mental or irreducible. I think they are reducible to the other two main mental properties. Or at the very least we'd need an argument for the odd claim that normativity and purposiveness are not explicable in terms of intentions and qualia.
But since you've mentioned it, since I am a naturalist, I think that there should be a naturalistic take on normativity (e.g., the study of the emergence of normative judgments in children and other animals). This is different from a science of first-order ethical judgments. E.g., a science that can tell you if abortion within a week of pregnancy is something that should be legal. I am not sure such a thing is possible. It would be like a science of humor telling you what, objectively, is humorous. That would be a nonsensical thing since what is humorous is dependent upon our minds: there is no objective fact of the matter about what is, or isn't, humorous.
Yes, it is a form of antirealism, and no I don't feel like entering that debate right now.
BDK
I made the statement you quoted, not Rino.
>>Or at the very least we'd need an argument for the odd claim that normativity and purposiveness are not explicable in terms of intentions and qualia.<<
I recognize that your point does not by itself settle the issue of naturalism. At the same time, however, can't we still turn this around the other way? Presumably you believe you can construct an argument that intentionality entails rational normativity. But wouldn't that argument in fact presuppose rational normativity? I have put this in terms of a root norm or perhaps a "metanorm" that there are correct as opposed to incorrect ways of arriving at conclusions. Doesn't any derivation of normativity from intentionality assume such correctness/incorrectness?
Hello Victor-
I am a Christian, a reader of your blogs (the 2 I know of), and fairly new to studying Naturalism in my Apologetics stuff. I have a question for you.
I am debating a Naturalist who does not believe that naything exists which cannot be explained apart from natural causes (so...he's a naturalist) I am frustrated trying to show him how silly this seems to me. Can you give me an example or 2 of things to pose to him which cannot be explained by naturalism. I have tried talking about First Principles, Logic in general, etc. But he keeps calling these "conventions, defined by man", so we just go around and around.
Any help would be appreciated. I have e-mail as well: philosophiatheos@inbox.com
Thank you for your time. Naturalism is so new to me.
--D Brown
Sorry for the misattribution DB.
We've hashed this out here a bit, ad nauseum, and I don't have much new to say. It is an interesting question, but I don't think we have to assume such things, but fine-tune our argument practices over historical time. Perhaps there is a kernel of an "urge for truth" that motivates it all. And I don't mean anything particularly sexy or transcendental or a priori about that urge.
This reminds me of people who say "Well, how can we ever hope to understand brains, given that we have to use our brains to understand them?" There isn't an intrinsic barrier to using X to understand or reason about X.
Rino,
So, may I just clarify that you accept the view that water is more properly defined as 'H2O in a specific chemical bonding relationship'.
It's not just me. :) It's every chemist on the planet. H2O is a water molecule. It doesn't mean structureless hydrogen and oxygen, and it never has.
Reductionism is about showing that an entity is a mechanism made up of more fundamental components. I can reduce an old pocket watch to a collection of gears and springs, and when I do this, there's no suggestion that having these components unstructured in a bag is equivalent to a pocket watch. Reduction says that the watch (and all its intelligible properties) are the mechanism, its structure and components. The reduction may also indicate that the intelligible properties of the watch need not be made out of the same stuff. I can make a time-teller out of aluminum or steel or perhaps even a simulations of those things.
I also think that reduction is sometimes an exercise in formal abstraction. By reducing human bodies to molecular machines, we learn how to accomplish the same macroscopic intelligible functions (e.g., locomotion, blood circulation, etc) in alternative ways. We know precisely what the intelligible function is, and how to implement it (often in more than one way).
Reductionists tend to take for granted the structure and the intelligibility that arises out of the structure. I understand their reasons, but for the life of me I don't understand what the horror is of granting that intelligibility (things understoood but not seen) into their ontology.
I'm afraid I did not really understand this claim. I don't think there's much taking for granted going on.
Reductionists tend to be able to identify the components of an entity before they can identify the mechanism that makes the components work in the entity. For example, it's easy to crack open a watch and see a bunch of broken cogs and springs fall out. At that point, we can make a very good inference that watches are made of cogs and springs. However, figuring out how those parts go together to make the watch tell the time is a more challenging task.
In my opinion, this is a good analogy for the mind. We have cracked the watch open and found cogs and springs (neurons). We know how to create some mechanisms with cogs, how they connect, how they implement ratios (neural networks, learning, prediction). However, we don't yet know how the watch works in its entirety (stream of consciousness). The question is whether there's any reason to continue to believe that the watch is driven by some sort of non-physical watch ghost (soul).
Nevertheless, I don't see any reductionists saying structure is not important, or saying that intelligible properties of the entity don't really exist.
Sorry if I'm not understanding your question.
Getting to DL's point, I know in neuroscience we take structure very seriously: knowing what types of, and how many, neurons there are in the brain isn't enough to understand brain function. We also have to know how they are hooked together, how they interact with one another. Seems like a straw man to say that anyone would think structure/organization isn't important.
D Brown:
There is no knock-down argument against naturalism, but here are some things that are certainly difficult for naturalists to deal with:
1. Mathematical truths and logical truths. Claims like 'The dog is in the yard' are true in virtue of facts about dogs and yards. In virtue of what types of facts are mathematical truths true? There are no obvious physical facts that correspond to '1+1=2.'
2. Consciousness. Have him explain the experience of a red sunset. If he gives a bunch of facts about the brain, ask him where, in that description of the causes and functions in the brain, the experience of redness lies. Couldn't we imagine the same whirring machine doing just fine even in the absence of the experience of redness?
3. Morality. Does he think it is wrong to torture babies? Why? What is the ultimate justification for such moral claims? If he resorts to facts about psychology, happiness, and the like, you can tell him those don't tell us that it is wrong to torture babies, but only some facts about such behavior. God is one ultimate grounding force for moral claims.
Note I am a naturalist, but if the above didn't trouble me I'd be an idiot. If he just acts all confident, then he is no better than a head-in-the-sand creationist type in the face of uncertainty. (For the record, I think problem 2 is the hardest one of all).
Victor, of course, has his argument from reason. I think the parts of that argument that are worth listening to are special cases of 1 and 2 above. He might even say 3 too, based on the topic of this post.
BDK: LOL, you are the real deal. I knew I liked you for a reason.
BDK,
Could it be that the two "aspects" of the mind are really two different processes that don't bear any necessary relation to one another? E.g., could there be a creature with intentional states but not subjective states?
Hmmm. I see why you ask, because if so that would presumably be a first step towards naturalising intentionality. It raises Victor's "is/ought" argument, though. So I don't think so.
The reverse certainly seems possible: my experience of pain is not obviously "about" anything.
God is one ultimate grounding force for moral claims.
Have I just encountered a naturalist who doesn't put stock in the Euthypro dilemma?
(Off-topic, I know).
Hi Doctor Logic,
I should say that I'm happy you find these points obvious, so do I. Also, I should note that at the beginning you asked for a naturalistic dualism, so the argument I am attempting to construct is not some Cartesian form of substance dualism, so it is neither as satisfying nor as difficult to accept for a naturalist. It does not address the problem of consciousness, though I think it does address other properties of the mental, such as rationality.
In any event, let me ask this question:
Imagine someone describes an elephant in the following way: 'this object is grey'. Then another person describes the same elephant is the following way: 'this object is ten feet tall'. Then another describes the elephant in the following way: 'this object is hungry'.
We have no problem with accepting all these descriptions of the same object (the elephant), but we tend to say the elephant has all of these three different properties. Now suppose someone says: 'the greyness of the elephant is the hunger of the elephant, which is the tallness of the elephant'. Surely we would say 'no, this is not possible, different properties cannot be identified (by Leibniz)'.
In the same way, we can say that the same brain has properties of mass, structural properties, and intelligible properties. But, we cannot say that the intellibigble properties just are the neurochemical properties, they are just different things. So, a form of property dualism is generated. There is one brain, but it has properties that are not reducible to one another (again, I don't think this is controversial). So, if we imagine that rationality supervenes on the structural properties/intelligible properties (ie, when a structure is doing something we consider quite brilliant, we call it rational, when it is just chaotically bouncing around without rhyme or reason, we call it arational), then we have a non-reductive form of naturalism. There is one brain, but mental properties, such as rationality/intelligibility cannot be reduced to neurochemical properties.
Again, this doesn't solve the mystery of consciousness, nor does it imply the existence of ghosts in machines. Those are separate issues. My goal is much simpler, to provide a form of naturalistic dualism, as you wondered about. If you grant all of my points, I don't see how you remain a full blown reductionist. Perhaps you could explain it to me. Perhaps you think that different properties can be identified with one another? That would still give you full blown reduction.
Cheers.
BDK
Yes, I know we've gone round about this, but I can't let you off quite that easily.
>>but I don't think we have to assume such things, but fine-tune our argument practices over historical time.<<
Compare your statement that we "fine-tune" our reasoning practices over time with the statement that our reasoning practices simply change or drift over time. To me it is clear that the difference is the assumption (in "fine-tuning") of a time-independent value of correctness/incorrectness. But I do agree that ultimately someone either sees this or they don't.
DB: compare this to the statement that our wheel-building practices simply drift over time. They don't, which suggests there is a time-independent value of wheeledness.
The "urge for truth" is analagous to the "urge for a better wheel". Once the end is fixed, there are objective features of the universe that will constrain what we will find--if you want to build a good wheel, don't try to make it out of olive oil. If you want inference rules that don't lead to incorrect claims, don't assert ~A using A as a premise.
Having a goal of knowing the truth, and the ancilliary goal of finding sentence patterns that preserve this property of truth, is the guiding normative factor here, not an assumption about any particular rule of inference. The desire to find rules of inference has been met with the happy result that we have found some, empirically tuned over time, that seem to work pretty well in most contexts.
And now we really are talking about the exact same topic as we did in the post I linked to above, which once again, people can see here, where I still agree with what I said, and I won't post here again unless I wouldn't just be repeating myself from this or that previous thread.
WOW! Thanks Blue Devil! That was some great stuff. Pretty brave of you! LOL
One quick question on your point #1--You said "There are no obvious physical facts that correspond to '1+1=2."
Couldn't a naturalist just say that we have placed conventional terms like "1" and "2" as labels on a quantity of objects? I mean, couldn't a Naturalist just say "Well, we used the word 'ONE" for a single object, 'TWO' for the next higher quantity, etc...then we made up word for combining objects together (addition), or grouping them and making simpler ways to add (multiplication), fair distribution of objects (division), etc..."?
I hope my question isn't too dumb...
The other 2 were awesome I will think more about those.
Thanks again,
--D Brown
BDK
I won't go over in detail what I said before either, other than to remind other readers that I was not talking about a particular rule of inference but about the presupposition that there are such rules, i.e., that there are correct versus incorrect ways of arriving at conclusions. Just as believing that nature must be lawlike does not entail the assumption of specific laws of nature.
However, to comment on what you said about consciousness/qualia in a previous post, there is an intriguing similarity between consciousness and reasoning. To analyze sight you do not need to see; to analyze hearing you do not need to hear; someone can analyze walking even if they are wheelchair-bound. But to analyze consciousness a person has to be conscious, and to anaylze the reasoning process a person must be reasoning. There is inability in principle to establish objective epistemic distance in both cases. To some of us this is just an oddity, while to others (like me) it is a clue about the bifurcated nature of reality.
Hi Rino,
I'll try to explain my position some more.
Now suppose someone says: 'the greyness of the elephant is the hunger of the elephant, which is the tallness of the elephant'.
I don't think this is analogous to reduction. Reduction does not eliminate phenomena, or say that one macroscopic function is the same as another.
Let's use the example of a watch. The watch tells the time. The watch ticks. The hands of the watch move. The watch contains movable springs and gears. The watch has a glass cover and a part that connects it to a strap or watch-chain.
The reduction says that the ticking and the time-telling and the hand movements are caused by the springs and gears working as a mechanism. The reductionist does not say that the hand movements and the time-telling are identical to the springs and gears in the absence of a mechanism.
When the reductionist theory is first put forward, the mechanism is unseen. So we're not saying that something we have seen is something else we have seen. We are saying that something we have yet to see (the mechanism) causes the seen phenomena.
So, in this case, what we have observed to date about the watch does not contradict the reduction. The ticking, the gears and springs, the time-telling are all still present after the reduction, and they don't disappear after the reduction. The reduction does not eliminate the motion of the hands or the telling of the time. It explains them.
Now, let's backtrack to a time before we decided knew the watch reduced.
A priori (before the reduction), we might think that the watch properties cannot be reduced. We might think that the external properties of the watch are the fundamental ones. For example, we might think time-telling is in the watch as a fundamental property of the whole and cannot be broken down to a mechanism of more fundamental parts.
When we propose irreducibility like this, there will generally be parts in the watch that we can remove without breaking the time-telling. For example, we can remove the glass cover or the parts that connect the watch to a chain or strap. Take off these other parts and the watch continues to tell the time.
When we open the watch and see the springs and gears moving in sync with the hand movements, we have to ask whether they are all fundamental or whether one is more fundamental than the other. For example, maybe it is the intrinsic time-telling that moves the hands, and the hands drive the gears. If that were the case, then the gears and springs are at least as likely to be unnecessary as necessary. Maybe the watch is intrinsically a time-teller, and the gears serve only to cause the ticking.
On what basis do we successfully and completely reduce the watch to a mechanism of gears and springs? Well, the watch reduction is totally successful once we understand the mechanism, and once we can build watches or show that the gears and their mechanism predict the ticking and the time telling.
However, we can make partial reductions without understanding the whole watch. If we find that the watch cannot tell time without the gears, there's (a priori) 2:1 odds that the gears cause the the time-telling (and that time-telling is not fundamental to the watch). This is because if time-telling is fundamental, the time-telling is compatible with gears moving or not moving, but if gears and mechanism are fundamental, the gears must move.
Similarly, if we find that tweaking the gears causes the time-telling to speed up or slow down, then we have another 2:1 (now 4:1) statistical advantage for the reductionist gears theory. If time-telling is fundamental, then it may or may not be possible (2 possibilities) to change the hand movements by tweaking the gears. On the other hand, if the gears and mechanism are fundamental, then it will certainly be possible to change the time-telling rate by tweaking the gears.
In cognitive science, we have performed many partial reductions. If mind or its aspects are fundamental, then we don't need physical brains, physical memories or any of the other things comparable to the gears in the watch. Yet we have them. If minds are fundamental, then we don't need circuits in our brains that function as memory, that recognize, that predict, and that emote. Yet we have all those things. It turns out that just about every function of mind can be tweaked physically or chemically. So while we don't know the whole mechanism, we ought to be well over 99% certain that the mind is not fundamental, and that the neurochemical mechanisms are fundamental to mind.
DL
>>If mind or its aspects are fundamental, then we don't need physical brains, physical memories or any of the other things comparable to the gears in the watch.<<
Wouldn't it be correct instead to say that if mind and its aspects are fundamental, then we need more than physical brains and physical memory storage mechanisms to explain all aspects of the mental? In order to reduce successfully, we have to be able to understand all aspects of the mental as mechanical (i.e., biochemical) functions. At least, we have to show that there is no aspect of the mental that in principle cannot be understood as such a function.
Hi Doctor Logic,
Thanks for your thorough example. This helps to clear some things up, I think. First of all, you respond by saying this:
"Reduction does not eliminate phenomena, or say that one macroscopic function is the same as another."
Yes, my illustration was saying that one cannot identify macroscopic properties of one object, but I was trying to say more as well. Specifically, I was trying to say you cannot identify any property of an object with another property of the same object. Hence, it is also not possible to identify the macroscopic property of 'hunger or the elephant' with any microscopic property that is conceptually non-equivalent, for example 'chemicals in the elephant's brain'. This is what I am suggesting that you are doing. You are identifying intelligible properties of the whole, with microphysical properties. I am saying this is not possible, by Leibniz.
Moving to your helpful analogy of the watch, you say: "The reductionist does not say that the hand movements and the time-telling are identical to the springs and gears in the absence of a mechanism."
Fair enough. But the reductionist often does say that 'time telling' is identical to the springs and gears and appropriate mechanism. I am questioning this.
'Time telling' as an intelligible property cannot be identified with the mechanism by Leibniz' law, and also because you could have one without the other. I.e., you could have the mechanism, but it not be a time teller. For example, if humans stopped existing, but a watch ticked on, would it still be a time-teller? Or, if after humans stopped existing, the wind wisked together the same mechanism out of the assorted gadgets, would it be a time-teller? I would argue no. I will await to see if you agree with this or not. If it is not still a time-teller, then 'time-telling' cannot be identified with the mechanism (under the assumption that identities are necessary, Kripke, etc...).
In the same way, 'reproduction' or 'survival' cannot be identified with the underlying mechanism/realizer. It can supervene on the mechanism, thus we still have naturalism/physicalism, but it cannot be identified with the mechanism, thus we have non-reduction.
Again, I'm not suggesting that there is a 'time-telling' elan vital or soul in the clock which is causing the tick tocking. For the sake of this argument, I am agreeing with you on that point. But that does not mean that 'time-telling' can be reduced to the mechanism.
Regards.
Darek,
Wouldn't it be correct instead to say that if mind and its aspects are fundamental, then we need more than physical brains and physical memory storage mechanisms to explain all aspects of the mental?
Well, we can go a step further and say that we either need more or we need less. I would go back to the watch. The watch dualist could argue that one needs more than just the physical aspects of the watch to tell the time. But it would be at least as valid to say that the watch can tell time without some of the physical aspects.
I see very little difference between mind dualism and water dualism. The water dualism is a better example because we have never simulated waterfalls as large (molar) quantities of H2O molecules. We only have approximations and molecular results that are compatible with water being H2O molecules and a mechanism. Yet we do not doubt the reduction of water to H2O interactions just because the simulation lacks the subjective properties of water.
In order to reduce successfully, we have to be able to understand all aspects of the mental as mechanical (i.e., biochemical) functions. At least, we have to show that there is no aspect of the mental that in principle cannot be understood as such a function.
I disagree here. If you have a proof that the reduction is impossible, that's one thing, but the mere possibility that we have overlooked such a proof is not evidence against the reduction.
Of course, dualists claim they have proofs, but I don't think they're proofs as much as statements of the unknown. For example, citing the fact that non-conscious objects (e.g., contemporary PCs) aren't "about" other non-conscious objects isn't a proof that conscious minds cannot be mechanical. The argument is rather like claiming that a mechanical doorbell isn't a time-keeper, so a mechanical watch must be impossible. It's really just a statement that we can't yet imagine what a mechanism of consciousness would be.
Also, I think many dualists argue that a molecular simulation of a person that behaved exactly like a person would not be conscious. But to me this is like arguing that if I build a watch and it tells the time, it's not a real watch, but a mere robotic simulation of a watch.
Rino,
You have a point about time-telling as an abstraction as opposed to time-telling of the watch.
You can take "time-telling" in my previous posts to mean "time-telling ability of this particular watch".
This abstraction issue is an important one.
Suppose I see a rabbit for the first time. I know it is small, brown, has four legs, is furry and has big ears. In recognizing the bunny, I have created a filter in my brain - a rabbit recognizer. Anything that is recognized by this filter I will call a "rabbit". This filter is an abstraction because it will recognize any rabbit, including rabbits I have yet to see.
So when I say "rabbits have long ears" I really mean that "my rabbit filter is triggered by (among other things) long ears."
This exposes an important fact about language. It means that when we speak in terms of abstractions, we don't have to be referring to some Platonic ideal or some floating universal. We can be referring to our own faculties, and what would trigger those faculties to recognition.
So when I recognize what the watch is doing when it is keeping time, I automatically create an abstraction filter for time-keeping. I can speak of time-keeping mechanisms in the abstract because I refer to the filter in my mind that recognizes such things. And I can say that the time-keeping ability of this particular watch is due to the mechanical mechanism inside of it.
So when you ask "Is a watch a time-keeper in the absence of minds?" you have to decide what you mean by the question. Time-keeper could mean that I presently see and recognize and use the device as a time-keeper. Or time-keeper could mean that, if I had such a device here and now, I would recognize it as a time-keeper. You would have to take time-keeper only in the strict, former sense of the word to say that time-keepers would not exist without us. However, taking the word "time-keeper" in this sense is misleading. If I used that strict definition, then any watch not in my presence would not be a time-keeper. (And any rabbit yet to be born would not be a rabbit, etc.) No one takes language to mean this. The language is taken such that a device is a time-keeper if it would be recognized as such by a mind, if a mind were present.
DL
>>If you have a proof that the reduction is impossible, that's one thing.<<
That's the idea. Let me illustrate. The brain works chemically, but suppose the statement "There are conditions under which atoms of different elements combine, and conditions under which they do not," were incapable of being confirmed by observation. That would be a big problem for the scientific status of our knowledge of chemical processes. Fortunately for science, the statement can be observationally confirmed.
I earlier proposed the norm, or better, metanorm, "There are
correct and incorrect ways of arriving at conclusions." And I pointed out that this metanorm is in principle impossible to confirm, because any confirmation would in fact presuppose the very thing to be confirmed. Since what I propose as the metanorm of rationality cannot be confirmed, it cannot be identified with or reduced to a statement that can be confirmed, such as "Some biochemical processes in the brain generate more adaptive behavior than do others."
It might be argued that we are fundamentally misuderstanding the
metanorm, except that our conception of the brain as a physical organ is valid only to the extent that we do, in fact, understand the proposition that there are correct versus
incorrect ways of arriving at conclusions. To the extent that we are mistaken as to what the metanorm means, our accumulated knowledge is impugned.
Therefore it is indeed impossible in principle to reduce some aspects of the mental to biochemical functions. It should be obvious that this is not an argument from gaps in our knowledge of neurochemistry.
In addition, I have made the case before that logical necessity cannot be confirmed by observation. Again, if the physical properties of the brain are "physical" to the extent that they can be confirmed by observation, then the sensitivity of the mind to logical necessity cannot be identified with physical properties of the brain.
Finally, the comparison of the ticking of a watch with consciousness is problematic. The ticking might be compared to behavior, but consciousness is distinguishable from behavior. You can gather evidence to confirm that you are walking, talking, breathing, or otherwise physically active. But you cannot gather evidence to confirm that you are conscious, since the evidence so gathered would be of no more value than your intuition that you are conscious. The argument, "I am aware of evidence that I am in a state of awareness" pulls no weight as a confirmation. This fact distinguishes behavior from consciousness.
Good post, DB. I'm learning a lot from you geeks.
Darek,
The problem with your argument is that you conflate the abstraction of what minds do with instances of the abstraction (i.e., specific minds).
It's analogous to conflating the abstraction of time-telling machines with an instance of a watch. Reducing the watch does not reduce the abstraction of all time-telling machines, but it reduces one or more instances of the abstraction.
I agree that I cannot rationally prove that the rules of rationality are correct, but I can most certainly show that an instance of a machine implements those rules.
In other words, the purpose of the reduction of the human mind to a mechanism is not to prove that the abstract rules of rationality are correct. It is not, as you suggest, reducing the abstraction to something scientifically confirmed, but scientifically confirming that the physical mechanism is an instance of the abstraction.
DL
>>I agree that I cannot rationally prove that the rules of rationality are correct<<
Well, I did not precisely challenge you to prove that the rules of rationality are correct. I said that we cannot confirm by observation that there are correct versus incorrect ways of arriving at conclusions.
The comparison I made was to the statement that there are conditions under which atoms of different elements combine and conditions under which they do not. Consider that statement as an abstraction if you like, nevertheless we can confirm it by observing specific instances of molecular bonds forming and not forming.
I also gave the example of the statement that there are biochemical brain processes that generate more or less adaptive behavior. Again, take that proposition as an abstraction. Nevertheless it is at least potentially confirmable through observation of specific instances.
Abstractions or not, obviously there is a difference between the confirmability of the three propositions--concerning correct ways of arriving at conclusions, molecular bonding, and chemical generation of adaptive behavior.
You have demanded demonstration that there are aspects of the mental that cannot be reduced to scientific facts about the brain, that is, that cannot be reduced to properties confirmable by observation. At the very least a critical distinction exists here that is inadequately addressed by invoking abstraction versus instantiation.
>> but I can most certainly show that an instance of a machine implements those rules<<
That would not be a confirmation of the metanorm of rationality. It would presuppose it.
Darek,
I dispute the claim that the "metanorm of rationality" is fundamental. I think it's just an abstraction, like "the metanorm of shiny rocks" = "the set of all things we would recognize as shiny". The set of all things we would recognize as rational is an important set for us humans, but it's not fundamental to the universe.
But even if I granted that your metanorm was fundamental, that has nothing to do with the reduction of human minds to a physical mechanism. The mechanism doesn't have to prove the metanorm. If it did, then you would have to say that the watch does not reduce to a mechanism because the "metanorm/abstraction of time-keeper" cannot be reduced in its general form to something physical.
You could certainly argue that any mechanistic model of mind must account for our ability to appreciate, assume, and recognize the metanorm of rationality. However, even this requirement is not the same thing as proving that the metanorm is correct (which is rationally impossible).
DL
>>I think it's just an abstraction, like "the metanorm of shiny rocks" = "the set of all things we would recognize as shiny".<<
Because we learn that there are shiny rocks by observation, we can confirm by observation that there are such things as shiny rocks. No so with the metanorm of rationality.
>>You could certainly argue that any mechanistic model of mind must account for our ability to appreciate, assume, and recognize the metanorm of rationality. However, even this requirement is not the same thing as proving that the metanorm is correct (which is rationally impossible).<<
Confirming is not exactly the same as proving. But the character of the metanorm of rationality is inextricably entangled with the question of whether it can play a natural role in brain processes. For example, if it is a fact confirmable by observation that a clock keeps the correct time to within a certain margin of error, it must likewise confirmable that there is correct versus incorrect timekeeping.
I admit that a naturalist can answer concerning the metanorm of rationality, "Oh, that. Well, that's so obvious that it doesn't count!" But what naturalism demands is a distinction so clear-cut and obvious that it is unarguable. It makes no sense after that to reject the distinction for the very reason that it is clear-cut and unarguable.
C. S. Lewis, whose book Miracles inspired this blog (or inspired the book that inspired this blog) said that the difficutly with seeing the transnatural (my word) character of reason was not that it was so arcane and subtle, but that it was so immediate and pervasive that we tend to look right through it.
If the mental has transnatural aspects, those aspects are necessarily transnatural and are liable to be looked past or discounted because of that.
Darek,
For example, if it is a fact confirmable by observation that a clock keeps the correct time to within a certain margin of error, it must likewise confirmable that there is correct versus incorrect timekeeping.
I disagree. I am not confirming that the clock keeps correct time. I am confirming that the clock keeps time the way I think is normative. I cannot prove or confirm what correct time-keeping is, but I do assume it is something I can recognize. So I am proving that the the clock keeps time the way I think it ought to keep time.
Similarly, by showing that a material mind thinks the way I say it ought to, I do not prove that my oughts are correct. It only shows that it functions the way I think it ought to. And it can be scientifically confirmable that that it does so. Now, I certainly assume that there is a correct way of thinking, but that is not what is being proven by the reduction.
Because we learn that there are shiny rocks by observation, we can confirm by observation that there are such things as shiny rocks. No so with the metanorm of rationality.
I can observe that certain linguistic string transformations work better than others for preserving truth. I think DL has nailed it.
DL
>>I disagree. I am not confirming that the clock keeps correct time. I am confirming that the clock keeps time the way I think is normative. I cannot prove or confirm what correct time-keeping is<<
You seem to be saying that while you cannot confirm that clocks keep correct time, you can confirm that they keep what you recognize as correct time.
I doubt that we must assume there is such a thing as accuracy of time-keeping in order to confirm that there is. But I chose the example because you cited time pieces as illustrations of naturalistic-mechanical
reduction. Your implication is that since it is matter of human convention how timepieces "ought" to work, we cannot confirm accuracy of time-keeping. Like saying that we cannot confirm that bishops ought to move only on their same color squares on the chess board, because it is us who decide how they "ought" to move in the first place.
I see. If brains can be compared with clocks, and if there is no confirmable fact about how clocks ought to keep time, then there are no confirmable facts about how brains "ought" to work. And it naturally follows from this that there is not a confirmable fact that there are correct versus incorrect ways to reach conclusions. Because we decide how clocks ought to work, and we decide how correct conclusions are arrived at. It is all a matter of our subjective if not arbitrary choice. That seems to be your point.
Of course, humans make clocks and invent games like chess. But humans didn't invent the human brain. It is the naturalist contention that blind nature did that, a nature consisting of unthinking forces that adopt no conventions and make no decisions but simply grind forward.
In any case, if the metanorm of rationality is a human convention, something we invented, then we in effect invented all of its fruitage--which includes scientific knowledge. There is no objective fact about whether the ideas of Einstein about space, matter and motion are truer than those of Aristotle, because we have simply invented the standard of correctness that recommends one set of ideas over the other. Anti-realism about science lies at the end of this road. Do you, of all people, want to take it?
Every stitch of your current argument presupposes the metanorm of rationality as objective fact, not mere human convention. Otherwise neither of us has an objective basis for appealing to the other or to anyone else who happens to be reading this.
If our scientific picture of the world depends upon the metanorm of rationality, and if that scientific picture reflects reality precisely because of this dependency, how can we avoid seeing the metanorm as fundamental? If it does not qualify as fundamental, nothing can.
BDK
>>I can observe that certain linguistic string transformations work better than others for preserving truth.<<
Are you saying that this is the same as confirming the metanorm of rationality by observation? Or are you saying that you make inferences from observation that presuppose the metanorm?
Darek,
Every stitch of your current argument presupposes the metanorm of rationality as objective fact, not mere human convention. Otherwise neither of us has an objective basis for appealing to the other or to anyone else who happens to be reading this.
But we're both human. :)
But seriously, I'm not sure in what sense you are using the term objective here. What is the test of objectivity? There are several meanings of the word, and I would say that each of those meanings presupposes the metanorm. So that means that the metanorm is probably objective in a trivial way.
However, I think you are missing my point with this objectivity issue.
For whatever reason (or perhaps for unfathomable reasons), we have a sense of what constitutes correct reasoning. We find this intuition to be incorrigible because we can never reach an alternative conclusion. At the same time, this does not constitute a formal, rational proof that the intuition is correct (due to circularity). Thus, the metanorm is an intuition that we all share, and that we all presuppose whenever we use language propositions.
When it comes to the reduction of human minds to physical mechanisms, the question is not whether our intuitions about reason are formally correct. The question is whether material minds reason in accordance with our intuitions (unprovable as they may be).
Now, if I were arguing that a working model of a material mind would physically prove how we ought to think, then I would side with you. I would supposedly be proving an ought from an is, which is not possible. However, it is not the intent of the reductionist to prove how one ought to think. The goal is to prove that a material mind thinks the way we think it ought to, i.e., according to the metanorm.
Finally, I want to say one more thing about the objectivity of the metanorm. The metanorm is presupposed in every argument. So it acts much like an axiom (or perhaps a meta-axiom). Now, axioms are assumed true. And for the purposes of a logical argument, axioms are objectively true. This is why I think the objective status of the metanorm is confusing. The axioms of Euclidean geometry are objectively true within the scope of Euclidean geometry, but they are not objectively true in the broader sense. It's like an algebra problem where we are given x=5. In the scope of the problem, it is objectively true that x=5. But it is not objectively true that x=5 in all problems or all contexts.
In the case of the metanorm, it is objectively true in any instance of reasoning because it is an axiom of reasoning. That doesn't make the metanorm objectively true outside of this context. It's not like the objectivity of the fact that the fine structure constant is 1/127. The latter is objectively true while not being an axiom. I guess I would say that all scientific truths are meta-theorems. So the scientific truth that we reduce to physical mechanisms can only be a theorem of our founding assumptions, one of which is the metanorm of rationality.
DB: Huh?
I'm much more prosaic than all this 'metanorm of rationality' business.
I'm just observing some things. E.g., I see we have an ability to detect true/false sentences, an ability to see or formulate patterns amongst the truth values of sentences. So, for people interested in truth, these are useful properties to know about.
I leave it to anyone interested in the language of metanorms to determine whether I'm discussing them. I don't care what we call it.
I've seen the identical mental allergy to normative irrealism in the case of epistemic norms as we see in the case of moral norms.
Just as the theist insists he *must* be dragged, kicking and screaming by the ineluctable power of some mysterious force called a "transcendent moral objective truth", into helping little old ladies across the street, so he insists that without some "objective rational metanorm" he can see no reason to go about believing true things.
(this is Rino)
Hi Doctor Logic
Perhaps it would be best to make the point in the second way that I mentioned: no human exists and the wind whips up the exact same mechanism that we today call a watch. Is it a time-teller in that case? I would argue no. No one intended it to tell time, no one is reading time off of it, it is not a time-teller. However, it has the same mechanism. Therefore, since identities are like mary's little lamb ("everywhere that Mary went the lamb was sure to go"), then these things cannot be identified, for one exists without the other. Thus, we cannot identify 'time-telling' as a macroproperty of the mechanism with the underlying realizers. This conclusion brings about a non-reductive picture, but it should not be threatening to you as a naturalist. There are plenty of non-reductive physicalists. And that is one step in the right direction :)
Incidentally, I agree with you on the rabbit example, and your endorsement of content internalism.
I agree that we can be abstracting 'time-teller' about the watch, and that abstraction is a product of our own minds.
However, I think we have to add one thing. For the sake of honesty, we have to assume that our abstraction really corresponds to the way the world actually is. For example, when we say 'that is a time teller', we cannot then say 'but that object has no 'time-telling properties', I'm just talking as if it is a time teller, but it is not. Dennett, with his intentional stance, is bad at doing this. If we go down that route, we end up endorsing massive amounts of fictitious story telling about the world, that no good scientist should embrace, I would argue.
So, we have to also assume that the watch has the intelligible property of 'time-telling'. I think you have agreed on this point, so I won't belabour the point. One thing I will add: it is fair to ask where intelligible properties come from. Ie, when we say 'the clock is really telling time', how did the clock become a time-teller? Obviously, someone had to make it with that purpose. Otherwise, there would be no reason to suppose it was really a time teller.
Anyway...carry on.
DL
>>So that means that the metanorm is probably objective in a trivial way.<<
Yes, to the extent that triviality here is, as you go on to say, the same as incorrigibility. Not only do we find that we cannot help believe that the metanorm is true, but that it is true for all minds at all times. Whatever we start to build, in terms of the products of intellect, we find the metanorm there at the bottom holding everything up while it itself seems to be bottomless (resonates with your term "unfathomable"). Calling it "trivial" can be a backhanded way of acknowledging it as fundamental.
As Victor has pointed out, if there is a fundamental constituent of matter/energy--say the vibrating strings of M Theory--these constituents are the way they are because they are that way, not because of further sub-constituents. They are physically "trivial," as it were. At the bottom level of material reality are these strings, but the strings are, so to speak, bottomless. But that's exactly what we find with the metanorm of rationality except that it is mentalistic rather than material.
>>At the same time, this does not constitute a formal, rational proof that the intuition is correct (due to circularity).<<
Exactly. It's scarry to find so much agreement between us.
>>Now, if I were arguing that a working model of a material mind would physically prove how we ought to think, then I would side with you.<<
This has come up before. I have tried to say that the question is not so much that reduction must "prove" the metanorm. It is that in order to interpret the metanorm in terms of some physical parameter of the mechanism in its environment, we must turn the metanorm into something confirmable--which it simply cannot be.
So how do we give the metanorm the causal role to play in a mechanism that it plays in our own mental life? We just cannot do so, except that in the case of a computer the metanorm influences how we design the mechanism and interpret its outputs. Our minds seem to be the medium through which the metanorm influences the computer's function. If mechanism as such cannot have everything that our rational mental life does, then our rational mental life is something more than mechanism.
C. S. Lewis put it as follows. He said that whenever we mentally try to put rational thought into a purely naturalistic, mechanical picture of reality, we find that the thought we have put into the picture is different from the kind of thought we were engaged in as we put it in. Here we see one reason why--because the naturalistic model cannot, in virtue of its purely natural character, make causal room for the metanorm of rationality.
The AfR draws out an implication: Naturalism cannot be a complete description of reality; reality has an irreducibly mentalistic aspect. Note that this is not an attempt to justify the norm, but to allow room for the norm to play the causal role in our thought processes that it self-evidently does.
So, what philosophical points of view are compatible with an irreducibly mental aspect of reality? Theism--the belief that an all-governing mind is the bedrock of everything that exists--is not the only contender, but it is a strong one.
>>In the case of the metanorm, it is objectively true in any instance of reasoning because it is an axiom of reasoning. That doesn't make the metanorm objectively true outside of this context.<<
Well, I would just direct you to your earlier comment that you couldn't think of a definition of "objectivity" that does not presuppose the metanorm. The metanorm of rationality is the axiom that lies below all reasoning about the role of axioms. It is the context outside of which nothing can be conceived of as "objectively true," including scientific inferences.
BDK
I'm fairly prosaic myself, to tell the truth.
>>I see we have an ability to detect true/false sentences, an ability to see or formulate patterns amongst the truth values of sentences.<<
Would I be way out of line here to take your word "see" to mean "understand" or "conclude"? Would it be wild to think that you could just as well have said, "I conclude based on observation that we have the ability . . ." Now, is this ability that you conclude we have "to detect true/false sentences" effectively the ability to arrive at sound conclusions (at least sometimes) about whether sentences are true or false? If it is, then we can recast your statement as follows:
"I conclude (soundly, I think) that we have the ability to reach sound conclusions."
You must see the problem. The first part of the sentence already assumes our ability (at least occasionally) to reach sound conclusions. If you assume it in order to conclude it, you haven't really concluded it. This illustrates what DL has called the incorrigibility of the proposition that there are correct and incorrect ways of arriving at conclusions--what I call the metanorm of rationality.
Hi Rino,
Perhaps it would be best to make the point in the second way that I mentioned: no human exists and the wind whips up the exact same mechanism that we today call a watch. Is it a time-teller in that case? I would argue no.
How about days and years? If I live on a planet in a solar system, I can use the planetary rotation and the annual orbit to define the notion of years and days.
It seems to me that you would say that solar systems don't have sidereal days or years unless someone is there to see them.
So I strongly disagree with you on this point. A watch is a time-keeper because time-keeper refers to what we would call it if we were there. Alternatively, it refers to what it could possibly be used for, if we were there to make use of it.
You are making a distinction about the intent of an object, not what it is. You are saying that the watch whipped-up by the wind was not the product of design or intention, or was not actually and intentionally used for time-telling. That distinction is all well and good, but it doesn't prove that intent and design cannot be reduced. Indeed, I think that intent and design are reducible.
Darek,
So how do we give the metanorm the causal role to play in a mechanism that it plays in our own mental life?
What role does the metanorm play? Well, most of the time, it plays no role at all. We simply can't help ourselves acting rationally.
We do what we do by instinct, or by social convention. People did this for a long time before the Greeks showed up and began to tell us about the metanorms of rationality. Indeed, a lot of inner city folk are quite oblivious to the metanorms, and yet they apply reason.
The metanorm is an abstraction from our intuitions about thinking.
So I would say that the reduction of a mind would have to show that there were intuitions and instincts for rational processes, and rather imperfect ones at that. The mind mechanism would also have to be able to make abstractions about mental processes and experiences in order to be able to comprehend the metanorm as we do.
I have to reject the comparison with M theory. M theory is a theory of causation, and even if the theory is inexplicable if it is fundamental, it causes everything else we see. The metanorm is not a force that compels rational things to happen in our minds, at least not directly. It is an abstraction of our best instincts for reaching conclusions.
DL
Very well, let's call the metanorm of rationality an intuition. Are you really going to press the claim that your intuition that there are correct versus incorrect ways of arriving at conclusions plays no causal role in your thought processes? Really?
Remember how I said that whenever we construct an intellectual product we find the metanorm of rationality underneath holding it all up. Your last post is such a product. Do you recommend it as consisting of conclusions correctly arrived at? If so, you make my case for me.
Hi Darek,
I should have been clearer. Suppose I have an intuition for deduction and inductive inference. And my past experiences of success and failure lead me to apply them in particular contexts.
There are two possible ways of looking at this. One way is to say that this behavior constitutes un/subconscious awareness of the metanorm. The way that a bird has unconscious awareness of the metanorm of flying. In that case, we have a straightforward way of reducing human minds and illustrating metanorm awareness. The metanorm instinct is certainly causal in such a model because metanorm has been identified with a causal instinct, and an instinct is a mechanism.
(BTW, contemporary software does not typically implement any such instinct. There are some research instances, I'm sure, but most PC's don't have the instinct because they lack awareness. Yet I think that creating such an awareness would be quite trivial.)
The other way of looking at the definition is to say that only a conscious, reflective appreciation for the metanorm as an abstraction counts as having a causal metanorm. In that case, the metanorm is rarely a causal factor in our thinking.
Which do you prefer?
Hi Doctor Logic,
You use an interesting example, but I would press you on it.
You say:
"It seems to me that you would say that solar systems don't have sidereal days or years unless someone is there to see them"
In the absense of any intelligent life forms, I would argue that whatever exists would not be cut up into 'days' and 'years' or by 'solar system' or 'star'. These are human concepts. There is something out there, but it is not cut into the joints we use. (I.e., the 'big dipper' is not really in the sky, humans have cut the sky up that way, though the matter is still all there)
So, I would say that the time teller example is still valid. In a world devoid of intelligence, there would be no concepts. Thus, the concept of 'time teller' would not be there, though the exact mechanism would be. In the same way, the concept of 'day' and 'planet' would not be there, although all the matter would be the same.
Cheers.
DL
>>One way is to say that this behavior constitutes un/subconscious awareness of the metanorm.<<
How can "behavior" constitute "awareness"? Behavior can be an evidence of awareness. Perhaps you mean the metanorm can exert influence on a subject who is oblivious to the fact that there are correct versus incorrect ways of arriving at conclusions. The metanorm is a driver of behavior in such a case. In naturalism, all drivers of behavior are confirmable. The metanorm is inherently unconfirmable.
>>The other way of looking at the definition is to say that only a conscious, reflective appreciation for the metanorm as an abstraction counts as having a causal metanorm.<<
Rational people can assert that some conclusions "just make sense" given certain grounds or antecedents while other conclusions "just don't." They may not use the precise words of my formulation of the metanorm, but they use words that substantially convey the idea. We judge the degree of a person's rationality, in fact, by the extent to which they appreciate the distinction between good and bad reasoning. But we don't require people to use the glossary of formal logic or specialized terms before we attribute rationality to them.
With that in mind, the ability to distinguish in some measure between sound and fallacious reasoning plays a causal role frequently in the lives of human beings. I am surprised that I actually have to argue for this point.
BTW, in case it is not clear, I use "metanorm" to distinguish my proposition about correctness/incorrectness of ways of arriving at conclusions from putative "norms" of reason such as the laws of thought and the rules of inference. The metanorm is primary to these, offering a ground from which we can explore and evaluate what are commonly considered to be rational norms.
This comment has been removed by the author.
DB: I don't have to infer that your claim "The TV is on the table" is true, especially when we are in the same room. You keep trying to insert inferences into my mind that I am saying we aren't making. This is why I brought up observation judgments. We can continue this on Victor's post on my comment if you want, as I push this line of thought there.
Hi Darek,
I think you have the arrow of causation backwards in your argument. The metanorm is a rational abstraction of our best reasoning instincts. It is not that the metanorm causes us to think that there are correct methods of inference. It is our instinct for individual methods of inference that leads us to create the abstraction we refer to as the metanorm. Or, to put it another way, the metanorm is the name we give to the instincts (in the abstract).
Since the instinct is the cause of our correct thinking, it is easily reduced to a mechanism because instincts are mechanisms.
DL
>>It is our instinct for individual methods of inference that leads us to create the abstraction we refer to as the metanorm.<<
I suppose that it is individual cases of gravitational interaction that caused Einstein to create the General Theory of Relativity. So we would have the arrow backwards equally if we answered questions of causation regarding gravitational interaction by invoking the General Theory. This cuts across all of science, then.
The General Theory, to the extent that it is true, captures the causal reality of gravitational interactions. I suppose that particular formulations of it in symbolic terms humans can understand is a causal result rather than a driver, but whatever it is that can be represented to the human mind in no other way but by such formulations is a driver of gravitational effects, not a result of them. Otherwise no cause can ever be contemplated as such by the human mind.
Some kind of expression of the metanorm may be, necessarily, the means by which the mind grasps discrimination principle at work in rationality. But that which is grasped by means of the expressions cannot be denied a causal role simply because it is by means of expressions that it is grasped.
Darek,
There's a huge difference between GR and the metanorm.
GR is a predictive theory and everything in the universe follows its rules (as far as we know, and per the theory). In contrast, the metanorm is a moral preference which predicts nothing. And certainly not everything operates in accordance with the metanorm.
The metanorm of rationality is what something ought to do before we label it as 'rational'. It is not the cause of that rationality.
An example. In order for us to call something a cat, it ought to meet certain conditions (having four legs, tail, fur, fangs, inability to digest sugar, or some substantial portion of the above). But this "metanorm of felinity" (the norm for us to describe something as a cat) does not cause embryos to grow up to be cats, nor did it cause cats to evolve. It's not causal. The only thing it causes is us to label something as a cat.
Likewise, the metanorm of rationality causes us to label a mind as rational. That's it.
The thing that causes the rationality is a physical process, an instinct.
DL
>>GR is a predictive theory and everything in the universe follows its rules (as far as we know, and per the theory).<<
You did not engage my point. GR refers to a formulation or description that captures an aspect of reality. As a formulation, it was caused by the reality it describes. At the same time, the reality that caused the formulation is such that it can only be captured by that formulation. The same can be said about the metanorm of rationality.
To continue with your latest objection, how can you know that the GR is successfully predictive unless there actually are correct as opposed to incorrect ways of arriving at conclusions? The formulation of GR is only worth something insofar as it is a conclusion correctly arrived at, and for it to be that kind of conclusion there must be correct ways of reaching it. Apart from the metanorm nothing meaningful can even be thought about GR, much less claimed for it.
>>And certainly not everything operates in accordance with the metanorm.<<
It does to the following extent: Everything in nature apart from what is immediately present to the senses can only be apprehended in the form of conclusions correctly arrived at. In that sense much of nature (and what lies beyond nature) must accord with the metanorm.
>>An example. In order for us to call something a cat, it ought to meet certain conditions.<<
Turn this around. We have the category "cat" because of the way certain animals actually are. That is, our idea of what distinguishes a cat, to the extent that it is realistic, is determined by the assemblage of flesh-and-blood features that actually sets cats apart from other mammals. We did not invent the label "cat" on a lark and then go shopping through the zoo to find some creature we could pin it on like a prize ribbon.
>>nor did it cause cats to evolve.<<
Cats presumably evolved because of selection pressure. But according to your argument, "selection pressure" is just a set of parameters we define. There is a kind of "metanorm of selection pressure" that causes us to label certain physical processes as "selective." That's it!?!?
Again, this just type of objection relies on the ambiguity between an existent and the description or expression by which we identify that existent.
>>Likewise, the metanorm of rationality causes us to label a mind as rational. That's it.<<
The metanorm is a description or expression of a reality. Realities can only be referred to by descriptions. (How basic is this?)
>>The thing that causes the rationality is a physical process, an instinct.<<
If so, it is an instinct or a process that distinguishes between correct and incorrect ways of arriving at conclusions. And for there to be such a process there must be an actual, real, genuine dinstinction between correct and incorrect ways, right? That's all the metanorm says--that there is such a distinction.
But if this process or instinct were physical, it would be confirmable. That's part of what it means to call it physical! But correctness/incorrectness of ways of arriving at conclusions is not confirmable, remember? So this process or instinct cannot be physical after all.
Darek,
The formulation of GR is only worth something insofar as it is a conclusion correctly arrived at, and for it to be that kind of conclusion there must be correct ways of reaching it.
Notice the language. "Worth something." It's worth something to a rational mind. The causal relations described by GR are not caused by the metanorm. The mechanism described by GR works whether or not a rational being is there to see it. What you are talking about is our apprehension of the mechanism of GR.
Apart from the metanorm nothing meaningful can even be thought about GR, much less claimed for it.
I don't see why this is relevant. The mechanism of GR doesn't care about rationality. Again, the metanorm has nothing to do with the mechanism of GR, it only has to do with our apprehension of and inferences about that mechanism.
It does to the following extent: Everything in nature apart from what is immediately present to the senses can only be apprehended in the form of conclusions correctly arrived at.
In other words, everything apprehended through rational means can only be apprehended according to conclusions correctly arrived at. I think this is tautological, and not a deep statement. Those minds we would recognize as rational obey the metanorm.
In that sense much of nature (and what lies beyond nature) must accord with the metanorm.
What must accord with the metanorm is our apprehension of "much of nature", not "much of nature" itself.
You discussed the cat analogy by saying that there's something about instances of cats that is distinguishable, and that this something is what we're labeling with metanorm of felinity. Fair enough.
By analogy, we would both say that there's some real perceivable difference between instances of proper reasoning and instances of improper reasoning. I totally agree that we can perceive those differences. We can perceive these differences formally. Yet the reality we are perceiving is nothing but consistency, and consistency is something that all natural systems have (because they are lawful).
In other words, let's remember that just as we talk about instances of "cat" that lead to the abstraction, there are instances of "reason" that lead to the abstraction. When we examine instances of reason over the same domain, the ones we don't like are the ones that lead to contradictions. And since contradictions are part of the model, it is easily captured by naturalism.
I'll put things another way. Assuming our instincts are correct, we can correlate good versus bad inference by the presence of contradictions. We can devise a formal model of what our instincts are telling us. Once we do this, we have a scientific way to confirm that an instance of reasoning was correct (according to our intuitions). We can't confirm the intuitions to be correct in an absolute sense, but we can confirm that the model or entity under study thinks in accordance with those intuitions.
But if this process or instinct were physical, it would be confirmable.
The thing being confirmed is that the mechanism of a naturalistic mind reaches the (substantially) correct conclusions per our instincts. That is, the formal mechanism agrees with our rational intuition. It does not have to prove the intuition itself to be correct.
What you need to make your case is a contradiction between rationality and naturalism, and I'm not seeing one at all.
Just to try to sum up the system as I see it...
1) Assume that our instinct for inference is correct.
2) Assume nature is lawful.
3) Devise a naturalistic model of rational inference and contradiction.
4) Devise a naturalistic model of human cognition.
5) Show that the model makes correct inferences (per #3), and that the model has the same instincts for correct inferences (per #1, #3).
Just because (1) cannot be rationally proven without being circular does not in any way invalidate the basic chain of inferences in the reduction. If we demonstrate (5), we demonstrate that the model makes decisions that we think are correct. It doesn't have to show that the inferences are correct (i.e., it need not prove #1).
DL
A technical note I should make before going on:
I have chosen to state the metanorm in terms of correct ways of arriving at conclusions because it is a bit more specific than "reasoning," "logical thought," and some other optional terms, yet it is readily understood by ordinary readers.
I use "correct" instead of "consistent" because dialetheists are being taken seriously in academic discussions of logic. And there are always those who see quantum theory as challenging the strict consistency of classical logical. I am not saying that I incline toward questioning the Law of Non-Contradiction. I just feel that the AfR does not need to defeat the claims of dialetheists to go through. Even dialetheists depend upon an assumption of correctness.
Now, to your points . . . I was not claiming that GR depended upon the metanorm, but that our knowledge of it did. This goes back to your giving the metanorm of rationality trivial status compared with theories like GR. I am just observing that if the metanorm is something without which we could have no knowledge of theories like GR, it hardly deserves to be treated as of slight importance. Might even deserve to be treated as fundamental.
>>What must accord with the metanorm is our apprehension of "much of nature", not "much of nature" itself.<<
Nature itself must have the character of being understandable by rational minds in order for us to know it. But that is a side issue.
>>I'll put things another way. Assuming our instincts are correct<<
This is the issue right here. If you call our conviction about the metanorm an "instinct," it's curious that it is not an instinct we can confirm. It is an instinct that we have no choice but to trust.
Take other instincts and turn them into propositions. We can turn the instinctive fear of falling into the proposition that it is a dangerous thing to fall. We can confirm that generally speaking, falling is indeed dangerous. Turn thirst in to a proposition: Our bodies need water. We can confirm biologically that, yes, our bodies do need water. And so on.
The metanorm can be stated as a proposition, too: "There are correct as opposed to incorrect ways to arrive at conclusions." But this one we cannot confirm from observation.
The puzzle grows because we are talking about brain function. If reasoning is purely a biochemical function of the brain, then its critical parameters ought to be confirmable just as with any other biochemical process.
It would be odd, wouldn't it, if we had to say "assuming the pulmonary system maintains organ-critical oxygen levels in the blood, then we can say" and go on to describe oxygen delivery to tissues.
In that case, a bright med student might ask, "Why do we just assume that the pulmonary system does that? Can't we set up an experiment to confirm it?" And then if we had to say, "Well, it's impossible to confirm from experiment and observation that the pulmonary system can maintain critical oxygen levels, but we instinctively feel that it does." Wouldn't the med student be justified in wondering whether pulmonary oxygenation of blood fell within a naturalistic framework of medicine?
Darek,
Take other instincts and turn them into propositions. We can turn the instinctive fear of falling into the proposition that it is a dangerous thing to fall. We can confirm that generally speaking, falling is indeed dangerous. Turn thirst in to a proposition: Our bodies need water. We can confirm biologically that, yes, our bodies do need water. And so on.
Whether these claims are confirmable depends on whether they include "oughts".
What is confirmable is that if we fall, we will die or be injured. That does not prove that we ought not fall. We only ought not fall if we think we ought not die or be injured. But those oughts are not confirmable either.
Same thing with drinking. We know that if we don't drink, we will dehydrate and die. But that by itself doesn't imply we ought to drink. It is only the case that we ought to drink if we want to avoid death or dehydration.
Likewise, we can show that a mechanism that reasons according to our instincts makes better inferences, predicts the world better, and avoids inconsistency. We cannot prove that a mechanism absolutely ought to do this, but we can show what a mechanism gains by doing so.
We happen to think we ought to do all of the above, but that ought is not fundamental. It is contingent. A being that fails at these things is easily imagined (and some exist), and such a being tends not to live long. That doesn't make the morality of survival or reason a fundamental force.
Thus, the proper analogy to the pulmonary system would involve the med student correctly observing that what we know about the heart and blood flow explains what the heart does and how it helps move oxygenated blood around the body. Yet, the med student will also observe that what we can confirm fails to explain in any absolute way why we ought to aid or impede blood flow. Just because we will die if we kill our heart muscles does not mean we ought not do it. But if we can't say why we ought to have a healthy pulmonary system, why should that cause us to think that the pulmonary system was supernatural? Does that mean pulmonary systems don't reduce?
Science tells us what is, not what ought to be. And that's the situation with the metanorm. The metanorm is not about the results of the assumption of rational axioms. The metanorm is about whether we ought to accept those axioms. That's why, if we reduce minds to mechanisms, we can show what it means to think rationally, and what the implications of thinking rationally/irrationally will be, but we cannot show that we ought to be rational.
DL: What does your last post say about truth?
Hi Normajean,
Truth has meaning only after we assume axioms of rationality, and I think it would be impossible for a being to have propositional attitudes if it didn't (at least implicitly) assume the axioms of rationality. But is it the case that we objectively ought to have propositional attitudes or value truth? Well, we can show that we do or don't value truth, or that a mechanism will or will not value truth, but we can never show that it ought or ought not. We happen to feel (and cannot help ourselves feeling) that we ought to value truth, but that's subjective.
For example, I can show that a rock cannot think rationally, but I cannot show that it ought or ought not think rationally.
DL
Admittedly, instincts such as fear and hunger have a "felt" component that cannot be captured in a proposition. Calling the rational sense an "instinct" is questionable for the same reason. Your last post to me, however, implies that the instinctive fear of falling has a moral aspect. It doesn't. If I start to fall from a height, the instinctive twinge that I feel is panic, not guilt. Fear of falling is not a conviction that falling is unethical.
For that matter, all through that post you seem to bring in the normativity of morality as if it were the same as that of rationality. They are distinct. A person can reason impeccably about the best way to cheat investors or carry out a terrorist plot. On the other hand, a person with limited capacity for reasoning can show moral sensitivity. If we find we have made a mistake in reasoning, we tend to feel embarrassment or frustration but not guilt. The normativity of rationality is better captured by the term "correct" than the term "ought," although "correct" can be used in a moral context and "ought" in a rational one.
I, too, am confident that we cannot confirm moral propositions from observation, but conflating moral "oughts" with rational "oughts" confuses the issue at hand.
What I introduced near the start of this discussion was a proposition (I will spare everyone yet another restatement) that I call the metanorm of rationality. It is just that, a proposition, and therefore capable of taking a clear "true" or "false" qualifier. The proposition makes an irreducible distinction, but one that cannot be subsumed under rationalism precisely because it cannot be confirmed from observation.
After all this verbiage, I am still not sure whether you even think the proposition is true, much less whether the distinction it makes is reducible. At times you seem to acknowledge that it cannot be confirmed, but then claim that it can be projected onto mechanism, leaving it in a no-man's-land between reducibility and irreducibility. It seems lately that you think the irreducible aspect is moral and therefore unprovable the way "It's wrong to steal" is unprovable.
I doubt that you grasp the self-reference and circularity that beset all attempts to find an explanation for rationality in terms of its adaptive benefits.
I am willing to leave the subject there for the moment...until we pick it up again down the line, that is. As usual, we must agree to disagree. And if I have misunderstood your position, you are welcome to correct me and close out the exchange.
Hi Darek,
I'll try to sum up my POV on this thread...
There are technical or factual assumptions about rationality, and then there are moral assumptions.
Suppose I give you a math problem. x=5, x+y=11, y=?
Obviously, x=5 is an axiom in this context. Ought you assume x=5? What if you don't feel like assuming x=5?
I think you will agree that how you feel about setting x=5 is irrelevant to the question. The "ought" of the assumption does not change the contingent facts of the assumption. That's what makes y=6 the objective answer to the math problem. When we say y=6 is objectively the answer, we mean that, if you assume the axioms (both facts and procedures), the result follows.
Similarly, if we assume the facts and procedures we regard as inherent to rationality, we can rationally and objectively show what rational behavior consists of, and what the advantages of such behavior will be. We can also prove that a particular mechanism behaves rationally.
Now your criticism of reductionism is that the existence of a rational machine (one which is compelled to think rationally) cannot prove that we "ought" to think rationally. Your claim is true, but irrelevant. The reductionist model doesn't have to prove that.
Is it true that my heart ought to beat? No, and no amount of analysis of what my heart is will show that it "ought" to beat. But analysis can show that my heart is compelled to beat. Your metanorm argument is equivalent to saying that hearts cannot be reduced to matter because that would only show that a heart was compelled to beat, not that it ought to.
The metanorm (as I understand your definition) does not say that we cannot create a rational AI. You are saying that any such machine will not prove that a computer ought to be rational.
I suspect some dualists' might say that we programmers put the "ought" of the metanorm into the AI. However, the ought in the machine takes the form of a compulsion, and evolution is quite capable of putting compulsions into life forms. There's no conservation law of "oughts".
Finally, I do think that the metanorm is an example of morality in general. The morality of the metanorm is distinct from the morality of the Golden Rule, but that doesn't mean the metanorm's morality is not part of morality in general. The use of the word "correctness" doesn't escape this, in my view. Correctness has objective meaning in the context of the axioms of rationality. The metanorm is about is whether we "ought" to accept the axioms, not whether we can label theorems as correct or incorrect once we have assumed the axioms.
Post a Comment
Subscribe to Post Comments [Atom]
<< Home