Contra theologian Roger Trigg on the nature of science

scientismRoger Trigg is a senior theologian and philosopher. His new book, “Beyond Matter”, is soon to be published by the Templeton Press, part of the wealthy Templeton Foundation whose aim is to produce a religion-friendly version of science.

Roger Trigg

An excert from the book promotes a view of science that is common among philosophers. Those of us with a scientistic perspective see it as erroneous, and yet, since Trigg’s account of science is widely accepted, it is instructive to rebut it.

Trigg argues that science rests on metaphysical assumptions:

What then has to be the case for genuine science as such to be possible? This is a question from outside science and is, by definition, a philosophical — even a metaphysical — question. Those who say that science can answer all questions are themselves standing outside science to make that claim. That is why naturalism — the modern version of materialism, seeing reality as defined by what is within reach of the sciences — becomes a metaphysical theory when it strays beyond methodology to talk of what can exist. Denying metaphysics and upholding materialism must itself be a move within metaphysics. It involves standing outside the practice of science and talking of its scope. The assertion that science can explain everything can never come from within science. It is always a statement about science.

This view can be summarised by the “linear” schematic:

sciax1

One can see why theologians like this account of science. If it were really true that science rested on metaphysical assumptions then science would be in big trouble, since no-one has ever proposed a good way of validating metaphysical assumptions.

In Trigg’s account, ideas such as naturalism and materialism are merely matters of personal taste; being preconditions that we impose on science. It is equally valid to start with different metaphysics, such as seeing the universe as a manifestation of the divine. Thus, the theologian argues, atheistic science is willfully blinkered by the unwillingness of atheists to consider the wider picture.

Fortunately the above account is merely metaphysics and therefore, by the very testimony of its advocates (!), is not supported by actual evidence. A much better understanding of science instead sees the process of science as a loop.

sciax2

Science’s method is the progressive iteration of models by adjusting them to better match empirical reality. At each stage one uses everything one knows to predict some aspect of the world, and then compares that against what is actually seen, and then feeds that back into adjustments to improve the model.

The point is that there are no unquestionable axioms. Everything, including working assumptions, including maths and logic, and including the very nature of science itself, is part of and wrapped up with the “scientific world model”. All of those things then get tested and honed as the overall model is compared to reality.

If one wants to question something fundamental, for example the basic logical axiom modus ponens, then all one has to do is produce an alternative model including some negations of modus ponens, and then ask which version of the model best reproduces and predicts features of the empirical world.

This conception of science is often explained using Otto Neurath’s metaphor of being at sea on a floating raft. Any part of the raft can be replaced, just not all of it at once. In the same way, any aspect of the scientific model can be examined and possibly replaced, and thus all aspects of the model are tested by the comparison with empirical reality. That includes any “metaphysical axioms” that might be adopted as working hypotheses, that then get tested by the scientific method’s iterative-loop, in the same way that the “laws of physics” get tested.

Even the scientistic ideas I’m defending here are ones that are products of the scientific method, having been developed as the best explanation of how the world, including human science, actually works.

Thus ideas such as naturalism and materialism are not preconditions of science, not axioms adopted by the blinkered, but rather they are products of the scientific method. The straightforward fact is that naturalistic and materialistic models work better in explaining and predicting the world, and that is why science has come to adopt them. It wasn’t always like that. In the past, invocations of the divine were seen as a normal and necessary part of science; but, as science has progressed, such ideas have fallen by the wayside and are now found to be superfluous, or, more often, making the model much worse.

Thus, Roger Trigg goes wrong when he asks:

Mathematics, though, could be claimed to be merely a tool created by the human mind. Why, then, should we assume that it can express in compressible form the workings of physical reality?

We don’t just “assume” that mathematics can express the workings of physical reality, we test whether it does! And the answer is yes! That is why physicists adopt the language of maths and, indeed, why mathematicians created such language. Historically, maths first arose precisely because it was useful in describing the world.

Trigg quotes Jim Baggot saying that “reality is rational, predictable and accessible to human reason”, but this is not the mere “assumption” that Trigg claims, but, rather, we can test whether it is true. Indeed we do exactly that every time we predict the time and place of a future solar eclipse and then find that our predictions come true.

Indeed, Trigg might be on the verge of realising this when he says:

Alternatively, the reality that we seek to understand may not even be subject to rational understanding. It may be sufficiently chaotic and disordered to be unintelligible. If we are told that this is impossible because science works, we are back with a pragmatic justification rather than a metaphysical one.

Exactly. By a “pragmatic justification” Trigg means a scientific one, since of course science is the pragmatic matter of doing one’s best to figure out the nature of reality based on the actual evidence that we have. Trigg is right that there is no metaphysical justification, but a pragmatic, scientific justification is the much better thing to have.

It may appear convincing, but it is no defense to the worry that we could live in an accidental bay of order on the periphery of a great ocean of disorder.

Well yes, that is true. It is indeed possible that there might be vast swathes of “reality” which lie beyond our ken and for which we have no evidence. And yes, science, being merely pragmatic and limited, will not tell us about things for which we have no evidence. But one can be equally sure that nor will metaphysics and theology. Contra Trigg, scientism does not say that science can answer all questions, merely those questions to which humans can obtain an answer.

Trigg goes further wrong by asserting:

The logical independence of physical reality from mind …

Yet our humans minds and the ideas “created by the human mind” are not independent of the physical world, since our minds are products of that physical world, having evolved by Darwinian natural selection for the very purpose of modelling and predicting the world that we inhabit. Once one accepts that our brains are the product of evolution (not always a given when it comes to theologians!), it follows that our minds must, over evolutionary timescales, have done a pretty good job of modelling and predicting the world around us.

Trigg continues:

Science is then just a human product, rooted in time and place. […] Once the logical independence of reality from science is accepted, the question is why reality has a character that enables it to be understood scientifically. The intelligibility and intrinsic rationality of reality cannot be taken for granted.

But then we don’t just “take for granted” that science works, we test whether it does! We know that we can predict eclipses; we know that we can land probes on comets out in our Solar System; we can predict and then detect the Higgs Boson; and we can even detect molecules in the atmospheres of extrasolar planets that are hundreds of light years away.

Science is not just an arbitrary “human product” that is “independent of reality” — science explains why our brains will have been molded by that reality, and why the brain’s workings will, at a minimum, be a rather good reflection of how reality works. Trigg continues:

Like the way in which mathematics seems to map the intrinsic rational structure of the physical world, this is presupposed within science and cannot be given a scientific explanation. It appears to be a metaphysical fact, and the explanation for which, if there can be one, must come from beyond science.

This is a theologian speaking. He wants there to be an “ultimate reality” that provides such explanations, and he hopes that such explanations don’t come from science, because that would leave no need for the theologian’s god. But science tells us that maths reflects the rational structure of the physical world because it is a product of human brains, and those brains reflect the rational structure of reality because that is what is evolutionarily useful. As usual, Darwin’s Dangerous Idea refutes the theologians’ hopes.

Advertisements

117 thoughts on “Contra theologian Roger Trigg on the nature of science

  1. Ron Murphy

    In “Roger Trigg is a senior theologian and philosopher” I take exception to ‘philosopher’. Same with Plantinga, and WLC. That’s not what these people are. They may learn some philosophy, and may know quite a lot of it – especially the bits that suit their theistic agenda. But they do not engage in ‘philosophy’ in any meaningful sense of the word.

    They are not ‘doing philosophy’ to try to understand the world from a metaphysical perspective with enquiring and open minds. They have decided what the truth is and they are determined to bamboozle with a pretence of philosophy in order to give their fansiful theism some cridibility. It’s a con.

    The pity is that there aren’t that many atheist philosophers prepared to call them out on their BS. I suppose this is understandable when some philosophers are emersed in the subject from a very non-scientific, possibly anti-scientific, direction, through the ‘humanities’.

    The shared ignorance about science is reminiscent of how moderate theists can’t really go into too much detail when criticising religiously inspired terrorists, lest they expose that common failure that is ‘truth through faith’ – what can a moderate theist say when the terrorist claims, “I have faith that this is how god wants me to interpret our texts,” other than to say, “we must agree to disagree”. Is the non-sciency philosopher compelled to back up his theistic commrade?

    In another parallel, I suppose a disclaimer or two is due: ‘NOT ALL MUSLIMS ARE VIOLENT’ and ‘NOT ALL PHILOSOPHERS ARE ANTI-SCIENCE’.

    Reply
    1. Coel Post author

      Good point. Philosophy needs better quality control. Plantinga and WLC are regarded by serious academic philosophers as serious academic philosophers, and Roger Trigg was a Professor of Philosophy at Warwick, and is now part of the philosophy department at Oxford.

    2. ontologicalrealist

      Ron Murphy said:- “They are not ‘doing philosophy’ to try to understand the world from a metaphysical perspective with enquiring and open minds. They have decided what the truth is and ……….”.
      I think that this is true.

    3. Andrew Farmer

      You my dear sir are doing the very thing you accuse Trigg, Craig and Pantinga of doing. You have presupposed what you think the truth is then set out to prove your claim.Instead of letting the evidence lead you to the truth.

  2. Paul Braterman

    if it were not the case that “reality is rational [and] predictable”, there is no way in which our basic biochemistry could function, let alone our brains, or the evolutionary processes that led to them. As to why reality should be “accessible to human reason”, there is no additional mystery there; but perhaps Professor Trigg has not come across Douglas Adams’ parable of the puddle.

    As to whether there is some deeper reason why reality is rational and predictable, that indeed is a metaphysical question, but one that science is not called on to answer. And as with all variants of the argument from design, any purported theological answer leads to an infinite regress, and any attempt to say anything about the Answer leads to a morass of further difficulties, thereby guaranteeing Professor Trigg and other apologists indefinitely continuing employment.

    Reply
  3. richardwein

    Good points, Coel.

    According to Trigg:

    What then has to be the case for genuine science as such to be possible? This is a question from outside science and is, by definition, a philosophical—even a metaphysical—question. Those who say that science can answer all questions are themselves standing outside science to make that claim.

    This way of thinking is confused, but let me start by going along with it for a while, and see where it leads. If speaking about science requires standing outside science, does speaking about philosophy require standing outside philosophy? Does speaking about metaphysics require standing outside metaphysics? In making that point about metaphysics, was I doing meta-metaphysics? Oops, I just did some meta-meta-metaphysics. Oops, there I go again. Please stop me before I drown in metas.

    A metaphysicist might respond by saying that, by definition, metaphysics comes first. So speaking about metaphysics is just more metaphysics. No need for any more metas. How convenient. How useless. When people play the “by definition” card, they are usually saying nothing of any substance. If there was something of substance to be said, it couldn’t be true merely “by definition”.

    If we were interested in substance, and not in empty language, we would actually consider what kind of thinking we should engage in when we discuss the question at hand. And the best way to do that would be to try answering the damn question, instead of pre-categorising it. Once we’ve had a go at answering it, we might observe that the kind of thinking we employed falls neatly into a category. Or we might not.

    So, “what then has to be the case for genuine science as such to be possible”? In brief: the world must be sufficiently regular to be reasonably predictable.

    Now, did I just do metaphysics? It didn’t seem much like what metaphysicists usually do. I certainly wouldn’t call it “metaphysics”, but I won’t ask you to call it “science” either. If it doesn’t fit neatly enough into a category for that category-label to be useful, then don’t categorise it. Sometimes, we don’t need no steenkin’ category.

    I take a very “naturalised” approach to philosophy, and my answer to the question was in that mode. Perhaps Trigg would try to answer the question in a more “metaphysical” way, i.e. take an approach more in keeping with what philosophers have traditionally tended to do when they see themselves as engaged in “metaphysics”. In that case, he might then say: see, a metaphysical approach was the right one, so I was right to call it a metaphysical question. But, of course, to make that determination we would need to make a judgement about whether he had in fact approached the question in the right way. I expect I would find his approach and his answer to have little merit. He and I might disagree on that, but at least we would then be discussing the question on its merits, instead of prejudging the issue on the basis of a lazy categorisation.

    Our language encourages us to reify and categorise. Of course, it’s often useful to categorise. But, if we’re not careful, we may be too quick to assume that things fall into neat and significant categories, and that just by assigning an instance to a category we’ve done something useful. We need to be aware of the fuzziness and context-dependency of language, and use it with care.

    Reply
    1. verbosestoic

      So, “what then has to be the case for genuine science as such to be possible”? In brief: the world must be sufficiently regular to be reasonably predictable.

      Now, did I just do metaphysics? It didn’t seem much like what metaphysicists usually do.

      No, it’s not metaphysics. It’s philosophy of science (which is also outside of science). You’d be doing metaphysics if you argued that reality REALLY IS that regular. Now, given the success of science, that might seem like an obvious statement to make, but you run into Kant here, and thus the argument that maybe the world only SEEMS that regular to use because we impose it mentally as a precondition of any kind of experience at all. As long as acting in that way doesn’t strongly contradict reality — at least not in any way that we can’t rationalize away and order — then we’d be successful, but reality itself wouldn’t have to have those regularities at all. So either reality really is that regular, or our experiences are imposing those regularities so that we can make any sense of anything at all. And given that even our empirical observation shows that our experiences are HEAVILY filtered by our mentally processes before we have them, even science provides evidence for Kant’s notion.

      Then the question is: how do we find out what’s really the case? Well, that’s a metaphysical/philosophical question, and as it concerns the underpinnings OF our empirical observations can’t be settled by that. Even saying that we can never answer the question is giving a metaphysical/philosophical answer to that question, not a scientific one.

    2. Coel Post author

      So either reality really is that regular, or our experiences are imposing those regularities so that we can make any sense of anything at all.

      If our science is anything like correct then we humans could not have evolved in a zero-regularities world. If the order is merely our imagination then we have no explanation of ourselves or of anything. Yet, we experience things. What is the best explanation of those experiences? The two competitors are (1) the standard scientific model including an orderly external world, or (2) “we have no explanation of ourselves or of anything”. Clearly the former does a better job of explaining and predicting our experiences. Therefore, by the standard practice of science, we pick the former as the better model. That doesn’t mean it is flawless or proven correct, it just means it’s the best we have.

      By that application of the scientific method, science arrives at the (provisional) conclusion of an orderly external world. Might there be aspects of reality not including in that? Why sure, there might be, and if we ever get evidence of them then they also will get included in the scientific model.

      By the above process of reasoning, the scientistic stance of ditching metaphysics seems to me to work fine.

    3. verbosestoic

      If our science is anything like correct then we humans could not have evolved in a zero-regularities world. If the order is merely our imagination then we have no explanation of ourselves or of anything.

      Well, while we don’t have to accept that we couldn’t exist in a true reality that had no regularities, let’s start from assuming that. The comment would be that the real world has regularities, but that we don’t know what they are because the regularities we find are too influenced by our own mechanisms, and we need to try to step outside of that. This would mean that our explanations are in some sense wrong, but would they be more wrong than, say, Newtonian physics is? Or would our explanations just work out well enough for our use while still being wrong? If the worst outcome is the possibility that our explanations are just wrong, that hardly seems like something for you to worry about.

      But it gets better. Following Kant, it’s possible that our explanations AREN’T wrong. Kant split things into the Noumenal and the Phenomenal, where the Noumenal is how things and the world are in themselves, while the Phenomenal is the world as it appears to us. Kant, then, pointed out that the Phenomenal world worked precisely as it appeared to us to work, and that therefore all of the scientific data we had and all of those investigations worked perfectly well and were completely and totally accurate when applied to the Phenomenal world, and so we should all use them. Thus, given that, we STILL have explanations for ourselves and everything else. We just can’t use them to make claims about the Noumenal world, or at least not without making some argument to get us to the Noumenal world. Which, it seems to me, is again what Trigg is complaining about.

      And it gets even better. From this, we can see that one quite relevant and reasonable question is “Is the Noumenal world just the same thing as the Phenomenal world?”. If this can be validated, then we’ve solved metaphysics and all of your view works out. Unfortunately, this is easiest to validate if we know that we have direct apperceptions of reality … and unfortunately we know that we DON’T. We have illusions, hallucinations, dreams and even processing — eg the blind spot — that demonstrates that what we experience isn’t necessarily how things are. So, then, we can’t just blindly trust our own perceptions. And we can’t even trust that we can puzzle out where our perceptions are inaccurate because of how complicated and detailed the patch-ups and processing can be (again, see the blind spot). And we can’t do this empirically because that relies on our senses and anything that we could do to try to work this out, well, comes from our senses. Even asking someone else about it doesn’t work because we get those answers from our senses, which is what we’re trying to validate.

      But it gets even better. You can declare, if you follow Kant, that you, personally, simply don’t CARE about the answer to that question, and that as long as acting on the Phenomenal works you’ll just stick with that … and by Kant, it always will. However, this doesn’t let you then say ANYTHING about the Noumenal … or about people who want to find out about it, other than maybe that you don’t think they’ll ever answer that question. But, hey, if that’s what they want, that should be no skin off of your nose. So, again, you can ignore the Noumenal and metaphysics all you want … as long as you don’t make claims that bridge into that gap.

    4. Coel Post author

      Hi verbosestoic,

      However, this doesn’t let you then say ANYTHING about the Noumenal world …

      I agree. In that scenario science could not tell us about the Noumenal world. But science is not making any claims about the Noumenal world, it just gives a silent shrug on the matter. But I then dispute that Trigg or theology or metaphysics or anything else “non science” can then tell us about the Noumenal world.

      … or about people who want to find out about it …

      Well hold on, Trigg and his claims, theology and its claims, metaphysics and its claims, are all in the Phenomenal world! So all of those are fair game! Really, by that account, any statement of theology and metaphysics, and indeed theologians and metaphysicians, and their hopes and utterances, are in the Phenomenal world and thus — by this account — are within the purview of science.

      It’s only the un-knowable Noumenal world that science cannot know about — but then neither science nor scientism is really making any statement about it, other than to say that, if we ever encountered evidence of anything such, then we’d take note of it.

    5. Coel Post author

      A metaphysicist might respond by saying that, by definition, metaphysics comes first. So speaking about metaphysics is just more metaphysics. No need for any more metas. How convenient. How useless. When people play the “by definition” card, they are usually saying nothing of any substance. If there was something of substance to be said, it couldn’t be true merely “by definition”.

      If we were interested in substance, and not in empty language, we would actually consider what kind of thinking we should engage in when we discuss the question at hand. And the best way to do that would be to try answering the damn question, instead of pre-categorising it. Once we’ve had a go at answering it, we might observe that the kind of thinking we employed falls neatly into a category. Or we might not.

      This is all a very good point!

  4. verbosestoic

    In Trigg’s account, ideas such as naturalism and materialism are merely matters of personal taste; being preconditions that we impose on science. It is equally valid to start with different metaphysics, such as seeing the universe as a manifestation of the divine. Thus, the theologian argues, atheistic science is willfully blinkered by the unwillingness of atheists to consider the wider picture.

    The problem here is that you’re taking YOUR view of what metaphysics is and imposing it on HIM, but I’m quite sure that he doesn’t think that metaphysics is just a matter of personal taste, or that it is equally valid to start with any metaphysics you want. I’m pretty sure that he thinks that there is indeed a way to determine what is the right metaphysics, and the right answer to philosophical questions. The charge he makes against atheists/scientismists is that if they argue that all questions that can be answered are questions of science, and then also deny that metaphysics and philosophy are not science, then they leave themselves unable to actually answer those sorts of questions, and therefore unable to actually verify that their presumptions are true. Which does mean, then, that TO THEM they are just axioms, and then it is indeed the case that anyone can pick whatever they want and what works for them.

    Which then means that you can do that as long as you only want to be a methodological naturalist, and don’t want to espouse materialism. Metaphysical naturalism, obviously, takes a position in metaphysics, and materialism states that reality itself is such that nothing immaterial exists. Both of these are statements about reality as it is, and so are metaphysical questions. If you deny that metaphysics is part of science, then you have no way to actually validate these claims. Otherwise, you can lump the relevant metaphysics and philosophy into science, but then have a really hard time convincing anyone that scientism or even the category of science is in any way meaningful.

    If you tried to argue that it is meaningful because all of those questions really are scientific ones, you’d still have to justify that, which is a philosophical question. And you’d run into the problem that the metaphysical and philosophical questions that you’d be trying to answer are ones that we can’t answer empirically. We can’t know that no material objects exist by looking really hard at the world and seeing that we haven’t found any. We can’t know that there are no questions that science can’t answer — or that any questions that science can’t answer can’t be answered — by looking at science answer questions, particularly questions that most people DO expect science to be able to answer. And science has a TERRIBLE track record in answering questions that philosophers have said that it can’t answer; invariably, all of the attempts merely recycle philosophical answers that were already shown to be problematic with a new object stuck in there (see Krauss for a prime example).

    So, either way, you have to acknowledge that metaphysics and philosophy MATTER, and go about things differently than science. Thus, you have to start taking it seriously. If you refuse to, then either you’re limited to statements about models and not about reality itself, per se, or you have unsupported statements that even you have to claim are unjustifiable.

    Reply
    1. Coel Post author

      Hi verbosestoic,

      I’m pretty sure that he thinks that there is indeed a way to determine what is the right metaphysics, and the right answer to philosophical questions.

      Well maybe. But metaphysicians have never been that clear on how they think one can validate metaphysical answers (though, yes, they likely have their own preferred answers to such questions). To a metaphysician, “science” is what there is empirical evidence for, and thus (by definition) metaphysics is stuff for which there is no empirical evidence. Thus, how they think they can arrive at right metaphysical answers, as oppose to wrong ones, is unclear. Indeed, that question is itself considered a live issue in metaphysics!

      Otherwise, you can lump the relevant metaphysics and philosophy into science, but then have a really hard time convincing anyone that scientism or even the category of science is in any way meaningful.

      Scientism is meaningful in the sense that it is a statement about epistemology, and is essentially the statement that we can only know things through empirical evidence.

      If you tried to argue that it is meaningful because all of those questions really are scientific ones, you’d still have to justify that, which is a philosophical question.

      I’ve no problem with it being a “philosophical” question so long as it is also a scientific problem. Thus, the nature of science and how we find things out (epistemology) is itself a part of science.

      And you’d run into the problem that the metaphysical and philosophical questions that you’d be trying to answer are ones that we can’t answer empirically.

      I, as a scientismist, would say that if we can’t answer them empirically then we can’t answer them (I have no problem with there being plenty of questions that we can never answer).

      We can’t know that no material objects exist by looking really hard at the world and seeing that we haven’t found any.

      Which is why science doesn’t say that non-material things don’t exist. Rather, science says that we have no evidence of non-material things and so have no reason to suppose that non-material things exist. Even the doctrine of “materialism” doesn’t say that non-material things do not exist, it says only that materialism is entirely adequate to account for all the evidence that we have.

      Thus materialism is a conclusion of science, not a metaphysical notion, and as with all conclusions of science it is provisional and might be over-turned by new evidence. Many things that would otherwise be problematic in the “philosophy of science” are solved by accepting that science is only making provisional claims.

      We can’t know that there are no questions that science can’t answer …

      We know, because science tells us, that there are almost certainly questions that science can never answer. I listed some here.

      … or that any questions that science can’t answer can’t be answered …

      The evidence that nothing else can answer questions that science cannot answer is that no good and validated answers have ever come from anything other than empirical evidence. Again, this is a provisional claim that could, in principle, be overturned.

      And science has a TERRIBLE track record in answering questions that philosophers have said that it can’t answer …

      But where science has a terrible track record in giving an answer, philosophy and everything else has an equally bad track record!

    2. verbosestoic

      To a metaphysician, “science” is what there is empirical evidence for, and thus (by definition) metaphysics is stuff for which there is no empirical evidence.

      You’re still doing it: translating what they think based on what YOU think about the subject. Metaphysicians do NOT, in general, think that science is what there’s empirical evidence for, and that isn’t how they define metaphysics. Metaphysics is the study of reality as it really is and must be, and philosophy of science says that science is critically empirical. So when philosophers or metaphysicians argue that science can’t do metaphysics, it’s because they have arguments that say that you can’t do metaphysics empirically, and science is critically empirical. You seem to be treating it as mere stipulation, which is absolutely not the case.

      Thus, how they think they can arrive at right metaphysical answers, as oppose to wrong ones, is unclear. Indeed, that question is itself considered a live issue in metaphysics!

      This is one of the strengths of philosophy in general, in my opinion, that it DOESN’T presume a methodology; you not only have to justify your answer, but if the method you’re using CAN provide an answer to that question. What most scientismists miss when they delve into philosophical issues is that justification part; they assume that they can say “This is science!” and expect philosophers to then simply agree that, hey, if it provides an answer then it must be the right one. Philosophy doesn’t accept anything as a simple stipulation, and so will ask if the method can produce a justified answer … and will do so especially in cases where they have reasons to think that that method will not produce an answer.

      Which leads to this: many scientismists seem to think that philosophy rejects empirical data based on either some kind of precommitment or some kind of turf protection. However, philosophy doesn’t reject empirical data AT ALL. If a question can be settled by empirical data, they not only will accept that but will most often pass the question off to science to settle. So when philosophy says that science can’t settle a question, it’s usually because they’ve a) tried it and haven’t had success and b) have separate arguments for why it won’t work. Scientismists rarely acknowledge either when they make their claims of solutions to philosophical questions.

      Scientism is meaningful in the sense that it is a statement about epistemology, and is essentially the statement that we can only know things through empirical evidence.

      Ah, the logical positivism problem again.

      How do you know that we can only know things through empirical evidence? Saying that we HAVE come to know things through empirical evidence doesn’t work, because a) that’s not the statement you’re defending here and b) you’re doing it for questions that CAN be answered empirically. But this question does not seem to be one that you can answer empirically without becoming circular or assuming your conclusion. So how did you manage to empirically come to know that it is true? And if you don’t know that it is true, why should anyone besides you take it seriously?

      Which is why science doesn’t say that non-material things don’t exist. Rather, science says that we have no evidence of non-material things and so have no reason to suppose that non-material things exist.

      This is actually the same problem that Jerry Coyne runs into in “Faith vs Fact”: Presuming that you want to base everything on an evidence-based approach — putting aside whether your definition of evidence is too narrow — what reason do you have for dividing things up into material and non-material and saying that we have no reason to think that non-material things exist? In any particular case, if the material explanation has more evidence in its favour, then it would be the preferred theory, and if the non-material explanation has more evidence in its favour, then it ought to be the preferred one by your own method. So the only time it would matter is when you have the same degree of evidence in favour of a material and a non-material theory … but in that case, saying that we have no evidence that non-material things exist is kinda assuming your conclusion. And, of course, given that if someone is objecting that some particular phenomena — say, consciousness — must be non-material for other reasons, citing this principle isn’t going to in any way address those actual complaints. So, for example, if you say that God can’t exist because he’d have to be immaterial but immaterial things can’t exist, you can imagine how unimpressive the principle you cite here is, as God would BE the thing that provides that evidence if you weren’t dismissing it out of hand.

      Even the doctrine of “materialism” doesn’t say that non-material things do not exist, it says only that materialism is entirely adequate to account for all the evidence that we have.

      You have to be careful with your terms here, because you are kinda equivocating. You CAN have a materialistic view that doesn’t say that immaterial things can’t exist, but this would have to be “weak materialism”, or perhaps “methodological materialism”. There ARE, however, “strong” or “metaphysical” materialisms that say just that, so you can’t say that materialism doesn’t say that. Just like you can’t say that “naturalism” does not say that supernatural things don’t exist just because methodological naturalism doesn’t state that, because metaphysical naturalism clearly does.

      Now, you can indeed hold “weak materialism” and “methodological naturalism” if you want, but if you do you have to be careful not to make claims that extend beyond that. Which means, for example, that if you’re a methodological naturalist you can’t make metaphysical naturalistic claims … and that’s what Trigg is essentially arguing in the article: if you want to stick to methodological naturalism, don’t use it to make metaphysical claims, and if you want to make metaphysical claims, you have to do metaphysics. Scientismists, in my experience, tend to want to do the former without doing the latter.

      The evidence that nothing else can answer questions that science cannot answer is that no good and validated answers have ever come from anything other than empirical evidence. Again, this is a provisional claim that could, in principle, be overturned.

      Again, how do you justify that empirically? What are your standards for a good and validated answer that don’t involve empirical justification?

      But where science has a terrible track record in giving an answer, philosophy and everything else has an equally bad track record!

      You’re ignoring WHY I said its track record was bad. Every time it tries to do it, it ends up citing as answers things that philosophy ALREADY CAME UP WITH … and found problems with. In short, its wonderful new approach only ended up with it repeating what philosophy had done already without even knowing it. If science is claiming to provide answers that philosophy can’t get by providing the answers that philosophy already came up with and found problems with, that does not bode well for its ability to provide answers.

      Note that, of course, that’s not the only or even the main reason that philosophers think that science might not be able to come up with answers to these questions. We have good reasons and arguments beyond that. But it should be something that scientismists should be concerned about.

    3. Coel Post author

      Hi verbosestoic,

      Metaphysics is the study of reality as it really is and must be … when philosophers or metaphysicians argue that science can’t do metaphysics, it’s because they have arguments that say that you can’t do metaphysics empirically, …

      Can you give any evidence at all that metaphysics *is* actually the study “of reality as it really is”, and can actually supply any answers about “reality as it really is”? (As oppose to metaphysicians and theologians merely claiming to be able to do that?)

      … many scientismists seem to think that philosophy rejects empirical data based on either some kind of precommitment or some kind of turf protection.

      Well no, it’s just that, as I understand it, those who regard metaphysics as a valid discipline tend to regard conclusions-from-empirical-data as “science”. It’s not that they *reject* such evidence, it’s that they regard metaphysics as something different from that.

      So when philosophy says that science can’t settle a question, it’s usually because they’ve a) tried it and haven’t had success and b) have separate arguments for why it won’t work.

      And when they do that, they rarely go on to demonstrate that metaphysics *can* settle the question.

      How do you know that we can only know things through empirical evidence?

      It is an empirical fact. So far, we have zero good evidence of anyone obtaining reliable knowledge any other way. Of course this conclusion (as with all of science) is provisional, and is open to refutation should anyone demonstrate that non-empirical metaphysics is generating reliable knowledge.

      But this question does not seem to be one that you can answer empirically without becoming circular or assuming your conclusion.

      Yes, science is circular! The scientific method continually goes round a loop continually updating and improving its world-model. The idea that “we can only know things through empirical evidence” is simply the best model so far. It is a conclusion from science.

      You would be entirely right that this is not an absolute proof of the concept, as with all science it is a provisional best-fit-so-far claim, and is open to revision.

      … what reason do you have for dividing things up into material and non-material and saying that we have no reason to think that non-material things exist?

      Let’s rephrase that statement to “the only things we have evidence for are things that can be sensibly regarded as “material”.”

      Really, it’s not the scientismists who are dividing things into material v non-material (to scientismists everything seems to be of the same nature, which we can call “material” if we wish), rather it is others who want to create a divide within what exists, a divide between “material” and “non-material”, and thus the burden of proof is on them to supply evidence for that divide.

      … but in that case, saying that we have no evidence that non-material things exist is kinda assuming your conclusion.

      It’s really placing the burden of proof on those making the claim, and saying that they’ve not fulfilled that task.

      So, for example, if you say that God can’t exist because he’d have to be immaterial but immaterial things can’t exist, you can imagine how unimpressive the principle you cite here is …

      Why sure, that would be a feeble argument. But I’ve never heard anyone actually make that argument! The more common stance is to invite those advocating a material–non-material division to supply their evidence.

      Again, how do you justify that empirically? What are your standards for a good and validated answer that don’t involve empirical justification?

      I don’t know! It’s up to anyone arguing for non-empirical knowledge to make their case. I wouldn’t want to pre-judge it, but will evaluate it on its merits. What I’d say is that, so far, no-one has produced good evidence that we can obtain reliable and validated knowledge from non-empirical means.

    4. verbosestoic

      Can you give any evidence at all that metaphysics *is* actually the study “of reality as it really is”, and can actually supply any answers about “reality as it really is”? (As oppose to metaphysicians and theologians merely claiming to be able to do that?)

      That it is the study of that is just the definition, so it’s odd that you’d want to challenge that. Now, whether or not it CAN actually ever answer those questions is indeed something that metaphysics, as a whole, is indeed considering. Claims of SPECIFIC metaphysicians are not yet accepted in metaphysics. About the only ideas that I think are roughly accepted are Kant’s, and they follow from multiple sources, and even that is at a minimum: there are certain concepts that we need in order to have any kind of intelligible experiences, and our perceptions to not necessarily reflect how the world is in and of itself.

      Well no, it’s just that, as I understand it, those who regard metaphysics as a valid discipline tend to regard conclusions-from-empirical-data as “science”. It’s not that they *reject* such evidence, it’s that they regard metaphysics as something different from that.

      Nope. It’s that scientismists like you insist that anything that uses empirical data in any way is by definition science — that’s generally how the broadening of science plays out — and metaphysics has concluded for good reasons that empirical data, without any other kind of justification, can’t work for those questions. Russell even uses science to make that claim: we could take empirical data unadorned if we thought that we had direct perceptual access to reality, but if science is correct we clearly don’t. And any test that we could run to correct our perceptions has to be run through our empirical perceptions, and so runs into the same issue; I can’t use propositions derived from experience to validate experience as a whole. So, empirical data is out.

      Now, you could use this to conclude that it isn’t possible, then, to learn anything about meta-reality because everything ends up filtered through our empirical senses. What could possibly, then, justify that? If we don’t just accept our perceptions as at least roughly right, what else can we use? We can’t really step outside of our perceptions of the world, after all. And that’s a perfectly valid metaphysical argument, and one that, you know, might actually BE an argument when compared to your “You haven’t answered the question yet!” argument. But it isn’t a scientific argument. It’s at least mostly conceptual: if a being depends on something as its access to reality, and we think that might not reflect reality, how could we ever step outside of that to know anything about reality? It’s an argument that metaphysics has indeed considered and indeed takes seriously … but they aren’t willing to give up on it yet. You may be, but that leaves you with the option to just ignore the questions … which ISN’T what you’re doing.

      It is an empirical fact. So far, we have zero good evidence of anyone obtaining reliable knowledge any other way.

      The problem is that people disagree with you about this “empirical fact”. For example, many people think that mathematics does that and don’t buy your argument that mathematics knowledge is really empirical, mostly because I’d argue that you — as you do with metaphysics — mix up “access” with “justification”. 2 + 2 = 4 in base 10 is not JUSTIFIED by taking 2 things and 2 things and counting them empirically, even if we learn about it that way. If someone takes 2 things and 2 things and counts “1, 3, 5, 7” and says that therefore there are 7 things, you aren’t going to appeal to anything empirical to prove that person wrong, but to the concepts of 2 and 4 and + and = in base 10.

      Thus, we do seem to have non-empirical knowledge. The question is if that sort of method is what we should use to study metaphysics — or other philosophical questions — an if they can indeed even work to do that.

      Yes, science is circular! The scientific method continually goes round a loop continually updating and improving its world-model.

      You’re confusing “circular” with “cyclical”. Yes, science starts from a point and feeds back into itself to adjust its models, but that doesn’t mean that it uses circular reasoning to justify its claims. My argument is that you make a circular argument there, which is a logical fallacy. If you want to try to stand on a logical fallacy, be my guest, but it’d be hard for you to include logic in science if you are willing to abandon it just to preserve the supremacy of science …

      Let’s rephrase that statement to “the only things we have evidence for are things that can be sensibly regarded as “material”.”

      Which falters the instant someone points out something that we think exists but that doesn’t seem to be material, which is what everyone you oppose actually does. Take the mind, for example. Descartes took the standard definition of material/physical at the time, pointed out the properties that mind seems to have that don’t seem to fit that definition, and concluded that mind can’t be material from that. Saying, then, that we have no evidence for the immaterial fails there because Descartes can say “Yeah, we do: the mind and these properties”. And this happens over and over in discussions with strong materialists, and they — and you — point to other and more general cases to conclude that THIS case, that has a specific argument for it, ought not count as evidence if they can come up with ANY other explanation … and sometimes even if they can’t. That then pretty much fits into the argument I made below, which is made sometimes explicitly, but very often implicitly, like arguments that God can’t be immaterial because then He wouldn’t be able to cause things, which boils down to “God can’t exist because He’s immaterial”.

      BTW, for an example of someone who DOES do that, Philipse in his book does that twice in the same CHAPTER, once implicitly, and once explicitly. It’s the chapter that has blocked my reading it, because I don’t want to move on until I finish commenting on that chapter, and I have a lot to say there.

      I don’t know! It’s up to anyone arguing for non-empirical knowledge to make their case.

      Do you think you know it? If you think you know it, then you have to know how to justify that empirically. If you don’t know it, then you are just expressing a belief that we can legitimately ignore. And this is an important question for you, because if you claimed that you knew it but couldn’t justify it empirically, we’d have a counter-example to your own principle and so it would be self-defeating. So abandoning the attempt to justify it empirically simply means that your opponents win. And if you can’t say what the standards are for a good and validated answer that doesn’t involve empirical justification, then we can legitimately ask how you know that you’ve never seen one, and what you expect us to do to prove it. If you can’t tell us what it would take to demonstrate your view false, it starts to look like it’s not possible to do that, and that’s NOT a road you want to go down.

    5. Coel Post author

      Hi verbosestoic,

      But it isn’t a scientific argument. It’s at least mostly conceptual

      I don’t accept that being “conceptual” means that an argument is outside science. Science uses concepts just as much as philosophy or anything else. Indeed, concepts are largely arrived at as models of the empirical world.

      2 + 2 = 4 in base 10 is not JUSTIFIED by taking 2 things and 2 things and counting them empirically, …

      That justifies “2 + 2 = 4 in base 10” as a model of reality.

      If someone takes 2 things and 2 things and counts “1, 3, 5, 7” and says that therefore there are 7 things, you aren’t going to appeal to anything empirical to prove that person wrong, but to the concepts of 2 and 4 and + and = in base 10.

      Agreed, but all of those concepts of “2” and “4” and “+” and “=” etc are also concepts adopted as models of reality.

      Which falters the instant someone points out something that we think exists but that doesn’t seem to be material, which is what everyone you oppose actually does.

      Sure, anyone succeeding in doing that would indeed refute materialism.

      Take the mind, for example. … Saying, then, that we have no evidence for the immaterial fails there because Descartes can say “Yeah, we do: the mind and these properties”.

      The “mind” is just a pattern of brain stuff. No-one has ever demonstrated the presence of mind in the absence of the material brain stuff. Therefore Descartes demonstration doesn’t work. You’re right that, in principle, this sort of thing could refute materialism, but so far it hasn’t.

      Do you think you know it? If you think you know it, then you have to know how to justify that empirically.

      Do I think that I know that there can be no non-empirical knowledge? Yes, in a *provisional* sense I do think that I know what. It’s because that is the best model of how things work that we currently have.

      Of course that is open to revision by someone demonstrating validated non-empirical knowledge. For comparison, if you ask me whether I think that I know that there are no unicorns on Earth, then I’d say yes as a provisional matter, owing to the absence of evidence of unicorns. Again, the provisional claim is open to revision given better data. In both of these cases, the answer is an empirical one: it is an answer based on the best model we have of the empirical world.

      If you can’t tell us what it would take to demonstrate your view false, …

      If someone wrote down a list of all the lottery numbers for the next year (claiming they used their sensus divinitus), and if that list were later verified as accurate, then it would be very strong evidence for non-empirical knowledge.

    6. verbosestoic

      Science uses concepts just as much as philosophy or anything else. Indeed, concepts are largely arrived at as models of the empirical world.

      Which just demonstrates how science does NOT conceptualize the way philosophy does, because philosophy DEFINITELY thinks of concepts as more than being models of the world, even though they may use the things in the empirical world to formulate their concepts.

      To understand this , let me differentiate between concepts and instances. Instances are, well, instances of a concept in this world, while the concept describes what critically must the true of all instances. Thus, any instance of a concept in the empirical world must have the conceptual properties — or else it’s not an instance of that concept — but it may have other properties that are not conceptual properties, and it is also possible to have a concept that cannot be instantiated in this world.

      Science conceptualizes only so far as to be able to identify and work with the instances in this world. Science does not care about whether those are truly conceptual properties or are instance properties that just happen to apply to all instances in this universe. Philosophy, on the other hand, wants the conceptual properties, the things that are true of the concept no matter what universe it is in, and cares regardless of whether the concept can be instantiated in this universe.

      For example, we can ask whether or not if a moon was cubical instead of spherical if it would still be a moon. Given the laws of physics, it would be reasonable to say that we can’t have cubical moons. Given that, science — and I think you’ll agree with this — is likely to say that they don’t CARE whether a cubical moon would still be a moon. But then this is simply not caring about a conceptual question, which is definitely valid and something that would interest philosophy (although it is a question that would be lower priority than some other conceptual questions).

      So, no, science does NOT use concepts as much as philosophy does. And there’s nothing wrong with that.

      That justifies “2 + 2 = 4 in base 10” as a model of reality.

      To insert this back into the conceptual/instance divide, it means, potentially, that this universe includes it. But even if this universe didn’t include it, the original conceptual truth of that statement would stand.

      To put it better, it is always true that in Euclidean geometry parallel lines never cross. If we found that in this universe they DO cross, that would NOT mean that we didn’t know it; it would just mean that this universe is non-Euclidean. That’s the distinction here that we need to keep in mind when assessing the relevant propositions.

      Agreed, but all of those concepts of “2” and “4” and “+” and “=” etc are also concepts adopted as models of reality.

      But the TRUTH of the statement “2+2=4” is therefore conceptual, not empirical, and so it must be justified that way. Thus, we don’t come to know those conceptual truths in any way empirically, because the justification is not empirical.

      Sure, anyone succeeding in doing that would indeed refute materialism.

      But if someone, like me, is merely skeptical about materialism, and points out things like that that raise doubts, why would the burden of proof be on me to refute materialism, if you cannot justify materialism?

      The “mind” is just a pattern of brain stuff. No-one has ever demonstrated the presence of mind in the absence of the material brain stuff. Therefore Descartes demonstration doesn’t work. You’re right that, in principle, this sort of thing could refute materialism, but so far it hasn’t.

      So, Descartes took our conceptual truth of what it means to be material, took our only direct access of mind, and points out that the latter gives evidence that mind cannot be material, and you’re saying that it can’t work because of your assertion that the mind is just a pattern of brain stuff and that you haven’t seen enough evidence to convince you yet, without ever actually ENGAGING Descartes’ evidence? Why don’t you have the burden of proof to demonstrate that the mind really is nothing more than patterns in the brain before you can use that fact to dismiss non-material explanations?

      Do I think that I know that there can be no non-empirical knowledge? Yes, in a *provisional* sense I do think that I know what. It’s because that is the best model of how things work that we currently have.

      By what standards do you judge what is the “best model”? How is your model — that presumes that non-material explanations will never work — better than a model that simply leaves the question open, and choose the explanation based on which makes more logical sense and explains more of the evidence in each specific case? In order for you to claim knowledge from your model, it has to be the case that the claim FOLLOWS from your model, given the evidence and logic we have. You can’t argue that it so follows because starting from material explanations gets us to answers faster, because that’s not a justification that would demonstrate the statement is TRUE, and that’s where you need to get. You can’t insist that we’ve always found material explanations before because a) that argument IMMEDIATELY wilts in the light of a reason why THIS time it has to be immaterial and b) that’s committing the inductive fallacy and so your conclusion wouldn’t follow from your model. Either way, you need some kind of justification for saying that you have the best model, and that the conclusion follows from that model. In my opinion, you have neither.

      If someone wrote down a list of all the lottery numbers for the next year (claiming they used their sensus divinitus), and if that list were later verified as accurate, then it would be very strong evidence for non-empirical knowledge

      Considering that that proposition is utterly unrelated to what you and I are talking about, I hope you can see that this is less than helpful [grin]. What would count as evidence? What burden of proof do we have to meet? An overall method/practice to demonstrate it to you is what’s needed here, not specific claims.

      Although:

      1) What specifically would convince you that there is a reality that is different from our experiences?

      2) What specifically would convince you that the mind is immaterial and/or is not brain?

    7. Coel Post author

      1) What specifically would convince you that there is a reality that is different from our experiences?

      Science does this all the time, demonstrating that there are aspects of reality that are not readily perceivable with our senses. Examples include discoveries of things like neutrinos, quarks, Higgs bosons, and dark matter — none of which we can see or hear or taste or touch.

      If the question is about some sort of meta-reality, then the answer is the same, we could be convinced of this by evidence. Specifically, if including such a meta-reality made models of our empirical world work better (in terms of explanatory and predictive power) then we would adopt that meta-reality.

      2) What specifically would convince you that the mind is immaterial and/or is not brain?

      Clear evidence of the mind acting in the absence of a material brain would be a good start. Evidence that the mind is unaffected by all material things (alcohol, LSD, anesthetic, bullets through the head, etc) would be another good start.

    8. verbosestoic

      Specifically, if including such a meta-reality made models of our empirical world work better (in terms of explanatory and predictive power) then we would adopt that meta-reality.

      So, if we sub this back into the “experience vs noumenal” distinction, you’re saying that you’ll be convinced that there is a noumenal reality — or, at least, noumenal QUALITIES, as it isn’t really a separate, disconnected reality — that we don’t have access or easy access to from the world of experience and that doesn’t have a DIRECT effect on our models in the realm of experience when … it’s required to build a model in the realm of experience.

      Yeah, at this point you’re back to “I have no interest in metaphyics and couldn’t care less” [grin].

      Evidence that the mind is unaffected by all material things (alcohol, LSD, anesthetic, bullets through the head, etc) would be another good start.

      Considering that your opposition does not accept that the immaterial mind is UNAFFECTED by all material things, this would be asking them to prove something that they don’t believe.

      Is there no way to demonstrate that the properties of the material are not compatible with the properties that the mind has? And note that in this case YOU would have to define what it means for something to be material.

      The first one is actually good, although problematic because, again, interactionist dualists will think that those cases are few and far between. But I suspect, then, that if we proved that ghosts existed, for example, you’d concede the point?

    9. Coel Post author

      So, if we sub this back into the “experience vs noumenal” distinction, …

      If your question amounts to “what would convince you of the existence of something for which, ex hypothesi, we can never have any evidence of at all, then my answer is “nothing”, since the only thing that could convince me of the existence of something is evidence.

      That is not the same as declaring no interest in metaphysics, it is declaring that it is up to the metaphysicians to prove any claims they are making.

      Is there no way to demonstrate that the properties of the material are not compatible with the properties that the mind has?

      My stance is, again, that it is up to those arguing that mind is independent of matter to make their case. I suggested two lines of evidence that would start to persuade me. If metaphysicians want to put forward other lines of evidence then I’m happy to consider them.

    10. verbosestoic

      If your question amounts to “what would convince you of the existence of something for which, ex hypothesi, we can never have any evidence of at all, then my answer is “nothing”, since the only thing that could convince me of the existence of something is evidence.

      In this case, though, you’d have to be subbing “empirical evidence” in for “evidence”. It’s reasonable to say that you won’t accept something that you have NO evidence for, but that’s not what you mean here because the argument is that we can have evidence for it, but not empirical evidence that would justify it.

      My stance is, again, that it is up to those arguing that mind is independent of matter to make their case.

      But you dismissed Descartes’ attempt to do that out of hand by insisting that mind was physical, even as his argument purported to demonstrate that the properties we observe of mind can’t be physical or material by the accepted definitions of the terms. You didn’t argue AGAINST his interpretation, you just dismissed it as not being relevant. THAT’S what that question was aimed at: is there any way for an argument like Descartes’ to work for you? Replying that you’d need to see an argument doesn’t work when people are PRESENTING them, even if they might be wrong, especially if you refuse to actually engage them.

    11. Coel Post author

      … the argument is that we can have evidence for it, but not empirical evidence that would justify it.

      To which I reply that it is up to those arguing for non-empirical evidence to make their case.

      But you dismissed Descartes’ attempt to do that out of hand by insisting that mind was physical, even as his argument purported to demonstrate that the properties we observe of mind can’t be physical or material by the accepted definitions of the terms. You didn’t argue AGAINST his interpretation, you just dismissed it as not being relevant.

      As I see it, I wasn’t dismissing the argument out of hand, I was simply saying that it was in no way convincing that mind was not physical.

      … is there any way for an argument like Descartes’ to work for you?

      In principle, yes, but I’ve yet to see any argument that even begins to convince.

    12. verbosestoic

      To which I reply that it is up to those arguing for non-empirical evidence to make their case.

      Which means that you have to know and be clear about what counts as evidence in a way that doesn’t simply define any sort of evidence that isn’t empirical as not being evidence. Or, rather, that you can’t use the terms “evidence” and “empirical evidence” as identical. Which is what you did here, when you answered my comment about subbing in “noumenal” and “phenomenal” with “Well, if we can’t have any evidence at all …”. Either you were doing that, or you don’t understand the discussion.

      As I see it, I wasn’t dismissing the argument out of hand, I was simply saying that it was in no way convincing that mind was not physical.

      In order to be doing that, you’d have had to give some indication of why the argument wasn’t convincing. Instead, you simply said that mind is just brain stuff and so Descartes was wrong.

      In principle, yes, but I’ve yet to see any argument that even begins to convince.

      And I find the arguments that the mind is just brain stuff unconvincing as well. So what? Do you KNOW that you’re right? If so, how? And how does that knowledge stand up to the counter-arguments, like that of Descartes?

      You need to distinguish yourself from someone who is convinced of their position and is just ignoring any and all evidence and arguments to the contrary.

    13. Coel Post author

      Which means that you have to know and be clear about what counts as evidence …

      Well no, *I* don’t need to be clear about what counts as evidence. If people wish to make the case for non-empirical evidence then it is up to them to make the case, and it is up to them to point to whatever evidence they are using, and it is up to them to justify the evidence and to explain why we know it is reliable. I do not need to have preconceptions about that, I can simply listen to and then evaluate their case.

      In order to be doing that, you’d have had to give some indication of why the argument wasn’t convincing. Instead, you simply said that mind is just brain stuff and so Descartes was wrong.

      Can you outline what about Descartes argument you find convincing that mind is independent of matter?

    14. verbosestoic

      Well no, *I* don’t need to be clear about what counts as evidence.

      If, when people are pointing out what they think is evidence for their claim, you continue to insist that there’s no evidence for their claim, you DEFINITELY need to say what counts as evidence or else all you’re doing is denying their arguments without engaging them, or more importantly all your doing is moving the goalposts and redefining your position to avoid it being refuted. And specific but unlikely “proofs” don’t work; at the very least, you’d have to use something that the people you’re arguing with at least somewhat expect to or could happen.

      Can you outline what about Descartes argument you find convincing that mind is independent of matter?

      His arguments are less convincing now, not because he was proven wrong, but because the definition of “physical” has changed to include odder properties. But this only highlights why clear definitions are required: Descartes’ arguments were never refuted, and the properties he cites are still considered to be properties of mine (like the fact that when we introspect, mental things don’t seem to be in space, which Descartes used to argue that they can’t be material as all material things are in space). If Descartes was defeated, it was by people redefining what it meant for something to be physical or material. This has two main problems:

      1) If you’re arguing that you think you know that immaterial things don’t exist because you’ve never seen one and all of our other explanations weeded them out, this counter becomes disingenuous if the solution in those cases was to redefine material things to include things that likely ought to be considered immaterial once you found out they exist.

      2) This doesn’t actually settle the argument, because Descartes, for example, was REALLY after the argument that it can’t be body, and I’m after the argument that it can’t be brain. One argument for that was that if its immaterial it can’t be brain. Redefining material to include anything that exists that might be considered immaterial gets around that side argument, but the properties it has are still there … and weren’t known properties of the brain, so you still have to explain how the BRAIN can do that.

    15. Coel Post author

      If, when people are pointing out what they think is evidence for their claim, you continue to insist that there’s no evidence for their claim, you DEFINITELY need to say what counts as evidence or else all you’re doing is denying their arguments without engaging them …

      Yes, if such people claim evidence, then I have to engage with their claim and evaluate whether it really is evidence. But, I don’t need pre-conceptions before hand about what sort of evidence I will accept; I can keep an open mind and be willing to consider what they propose.

      this counter becomes disingenuous if the solution in those cases was to redefine material things to include things that likely ought to be considered immaterial once you found out they exist.

      In considering that matter we should really consider whether a material/immaterial divide improves the explanatory and predictive power of the model or not. As I see it, the model in which mind is a function of the material brain has the most explanatory and predictive power. That is not ruling out non-material by fiat, but is saying that the evidence for it is unconvincing.

    16. verbosestoic

      Yes, if such people claim evidence, then I have to engage with their claim and evaluate whether it really is evidence. But, I don’t need pre-conceptions before hand about what sort of evidence I will accept; I can keep an open mind and be willing to consider what they propose.

      The thing is, you have to be careful that your “open mind” isn’t hiding a “there is no evidence that will ever convince me” sort of line, where because you don’t have any set standard for what you’d accept — beyond some specific examples that you don’t really expect to happen because the original claims don’t say that will happen — you can always rationalize the claims away as being “unconvincing”, as some theists do. If you really had an open mind towards it, you wouldn’t be claiming to KNOW that it was the case and then demanding that people prove you wrong; you’d be simply saying that you’re unconvinced and neutral on the topic. If you aren’t neutral on the topic, then your reasons for not being neutral are equally open to scrutiny … and, as you know, I find your reasons unconvincing [grin].

      As I see it, the model in which mind is a function of the material brain has the most explanatory and predictive power

      And I disagree. I disagree because that model has, in fact, failed to explain in any way what is most indicative of mind — at least as related to consciousness — which is qualia. That model cannot explain why we have it or what it actually does, and doesn’t even predict that we WOULD have it. It can’t say whether or not a sufficiently advanced computer would have it, and its model is built to assume that it OUGHT to be able to answer those questions if it works, and it obviously can’t. The only thing that it explains and predicts is the interaction between mind and brain state … and dualists knew about that one long, long ago. Sure, they can’t predict what impacts changing the brain will have on consciousness, but then neither can your model. So this model seems to be rather inferior, unless you can give me some good examples of what it really does do well.

      Also, my whole argument is that any sort of material/immaterial divide is problematic because materialists keep redefining “material” as soon as they discover something that didn’t fit. Thus, I’m willing to redefine everything as material as long as you understand that this doesn’t mean that you then get to say that mind being brain is more likely because “materialism won”. Materialism — and naturalism — are either meaningless — because by definition everything that exists is material and natural — or false. If we put that debate aside, we still have to settle all of the other debates that led to the false debates over then in the first place.

    17. Coel Post author

      The thing is, you have to be careful that your “open mind” isn’t hiding a “there is no evidence that will ever convince me” sort of line, …

      Agreed.

      If you really had an open mind towards it, you wouldn’t be claiming to KNOW that it was the case and then demanding that people prove you wrong; you’d be simply saying that you’re unconvinced and neutral on the topic.

      But science doesn’t work like that! Science works by adopting the model that best fits the evidence we currently have. Given current evidence, the best explanation is “monist”, without any divide between “material” and “non-material”. Now, if new evidence came along pointing to such a divide, then we’d need to evaluate it. However, that doesn’t stop us adopting the current best model.

      I disagree because that model has, in fact, failed to explain in any way what is most indicative of mind — at least as related to consciousness — which is qualia.

      Well, fair point. However, non-material explanations do not do better in explaining qualia. Just saying that something is non-material is not an explanation. Again, science goes with the simplest explanation that best explains the evidence. Since adding in a material/non-material divide does nothing to improve the explanation, we don’t add in any such divide.

      So this model seems to be rather inferior, …

      It would only be inferior if the non-material explanation were significantly better in some regard. But I don’t see any way in which it is better; saying it is non-material just raises more questions than it answers, it doesn’t actually improve the explanation. (And, unless it does improve the explanation, then Occam’s razor disposes of it.)

    18. verbosestoic

      But science doesn’t work like that! Science works by adopting the model that best fits the evidence we currently have

      But when science says it has knowledge, it doesn’t determine that by simply taking the model that currently happens to fit the evidence best. The model needs to be confirmed first. Even more importantly, when science wants to extend that model to cover other phenomena it requires an argument that the model does apply there and confirmation that it still works. Science can PROVISIONALLY accept models because they happen to work without declaring that therefore they KNOW that the models are right. And, just to remind you, this isn’t about certainty, but about justification. In science, you are not allowed to say that you know a model is correct until you have confirmed it to a great degree. As Larry Moran put it. science is critically rational, empirical, and SKEPTICAL. I argue that when you use induction as you do to justify your models, you drop the “skeptical” part, and so only end up with beliefs, not knowledge.

      Which isn’t a problem as long as you don’t treat it as knowledge, but as we saw when I brought up Descartes, you really DO seem to treat it as knowledge.

      Given current evidence, the best explanation is “monist”, without any divide between “material” and “non-material”.

      You need to be careful with your terms here, unless you’re claiming to be a neutral monist:

      Neutral monism is a monistic metaphysics. It holds that ultimate reality is all of one kind. To this extent neutral monism is in agreement with idealism and materialism. What distinguishes neutral monism from its better known monistic rivals is the claim that the intrinsic nature of ultimate reality is neither mental nor physical. This negative claim also captures the idea of neutrality: being intrinsically neither mental nor physical in nature ultimate reality is said to be neutral between the two.

      You seem to be a materialist, despite your discussions here. Also, taking this position doesn’t really help you, because we’d still need to determine whether the brain itself can indeed just be the thing that does the work, and you couldn’t rely on what you call the physical properties to define what the mental is. Also, neutral monism isn’t a popular view, so you’d need to justify that with more than simply definition.

      Ultimately, I argue that the distinction between material and non-material is meaningless because of redefinitions of terms like “material” and “physical”, so we might agree on that. But materialism about mind also dovetails with ideas like “It’s all the brain” and “We can figure out everything about the mental from the third-person view” (see Dennett) and models of that. The model I’m talking about is THAT one, and that’s the one that doesn’t work out very well for a number of reasons related to qualia. THAT’S the one we have to comparing here.

      However, non-material explanations do not do better in explaining qualia. Just saying that something is non-material is not an explanation.

      But dualism doesn’t actually require that. SUBSTANCE dualism does, but they do so, as I said, because the qualities of qualia don’t seem to work by the definition of material that we traditionally went with. For example, material things seem to be in space, but mental things don’t. But the key debate here is over the sort of materialism vs dualism question that talks about whether the mental can be reduced to or explained by the brain, or from any third person view, with the key question being: can we learn everything interesting about the mental or about consciousness from the third-person view? So let me run down some issues:

      1) What’s the hallmark of the mental? Materialists: brain state. Functionalists: observable behaviour. Dualists: Inner experience.

      2) Could computers be conscious? Materialists: ? Functionalists: If it acts sufficiently conscious. Dualists: Maybe.

      3) Can we tell if something is conscious or not from the third person? Materialists: Yes, by looking at the brain state. Functionalists: Yes, by looking at behaviour. Dualists: No. We can only make assumptions.

      Now, I’m playing a bit fast and loose with the terms here, but these are some of the fundamental distinctions. I explain in more detail the problems with the third-person approach here: https://verbosestoic.wordpress.com/nailing-the-third-person-science-of-consciousness-to-the-wall/

      The dualist position is superior in my opinion because it makes qualia the defining attribute of consciousness, takes the fact that qualia is fundamentally a first-person phenomena seriously, and avoids epiphenomenalism: there’s always something for the specific experience to actually cause. There are numerous other issues as well, but these are the big ones. The worst problems with that model are explaining mental causation — which, if you drop a strict physicalist view, isn’t a problem anyway — and explaining how changing the brain impacts mental states, which all interactionist dualist positions have already done.

      If we need an actual material/immaterial divide will be determined once those who claim that everything is material figure out what it means for something to be material. That they keep changing the definitions is not a problem for those who took them at their word and built their position on that understanding, and you can’t refute their positions by accepting all the phenomena they used as evidence into your own position and then saying they were wrong all along.

    19. Coel Post author

      But when science says it has knowledge, it doesn’t determine that by simply taking the model that currently happens to fit the evidence best.

      Well, yes it does! Though predictive power is important, in additional to explanatory power, and also picking the most parsimonious model.

      The model needs to be confirmed first.

      But we can never fully “confirm” a model, we can only say that new data that comes along is also in line with the model.

      Even more importantly, when science wants to extend that model to cover other phenomena it requires an argument that the model does apply there and confirmation that it still works.

      Science would make the default assumption that the model does work for new phenomena, until observations show that it doesn’t. That’s all it can do. If we validate a physical model by observations in Australia, Canada and Belgium, then we might presume it applies also to Mongolia and Peru. We don’t first need to validate it in Mongolia and Peru before presuming that it does work there also.

      I argue that when you use induction as you do to justify your models, you drop the “skeptical” part, and so only end up with beliefs, not knowledge.

      Would you be happy applying induction to extrapolate from Australia, Canada and Belgium to Mongolia and Peru? Skepticism is important, but can be overdone. If you had to test absolutely everything first then you could never do anything. For example you could never build a bridge or trust an airplane, since no amount of testing of bridges and airplanes will guarantee that your new bridge will hold, or that the place will still fly tomorrow.

      Which isn’t a problem as long as you don’t treat it as knowledge, but as we saw when I brought up Descartes, you really DO seem to treat it as knowledge.

      In the same way that I “know” that Airbus 330s fly and pigs don’t! 🙂 Both claims depend on induction.

    20. verbosestoic

      But we can never fully “confirm” a model, we can only say that new data that comes along is also in line with the model.

      I didn’t say “Fully confirmed”. I said that it had to be confirmed to some degree first. In science, you can’t just take the evidence, come up with a theory that fits the evidence, and say “Well, now we know what it is!”, even if that’s the only model available (and so by definition the best one). Science is indeed more skeptical than that.

      Science would make the default assumption that the model does work for new phenomena, until observations show that it doesn’t.

      No, it would presume that it does work for new phenomena as long as the model’s properties themselves suggest that it ought to work for that new phenomena. As soon as someone suggests that it MIGHT not cover the new phenomena — particularly if they have a reason for saying that — then science would have to go and test to see if it does.

      Would you be happy applying induction to extrapolate from Australia, Canada and Belgium to Mongolia and Peru?

      It depends on the model/question. For psychology, I absolutely would not; we actually KNOW that a lot of psychological phenomena are culturally dependent. For physics, I probably would, because the models suggest and have confirmed that typically there aren’t such differences. Thus, skeptically speaking, I have to evaluate each case on its own merits before deciding if the inductive generalization ought to apply or not, which means advancing reasons why it ought to work.

      For example you could never build a bridge or trust an airplane, since no amount of testing of bridges and airplanes will guarantee that your new bridge will hold, or that the place will still fly tomorrow.

      But we have reasons in the models that we have confirmed for thinking that those things should still work. Without those reasons, you’d be committing a fallacy. We don’t need certainty, but we need reasons. ESPECIALLY in cases where people are pointing out that the model may not apply in those cases. For example, extrapolating from the determinism of non-thinking objects to the purported fact that all MENTAL events must also be deterministic doesn’t work because of the whole free will debate; we have reasons to think that mental events MIGHT NOT be deterministic. Now, if you demonstrated that mental events really are just brain events, THEN the induction works, as we would have reason to think that they are the same and so the trait can be so generalized.

    21. Coel Post author

      As soon as someone suggests that it MIGHT not cover the new phenomena — particularly if they have a reason for saying that — then science would have to go and test to see if it does.

      Well, no, they need a *good* reason for saying that. For reasons of necessary pragmatism, science assumes that the current model is the best, until given good evidence that it is inadequate. If a nobody crank claims that science doesn’t apply to such-and-such then science can ignore that claim, until good evidence is provided.

      For example, extrapolating from the determinism of non-thinking objects to the purported fact that all MENTAL events must also be deterministic doesn’t work because of the whole free will debate; we have reasons to think that mental events MIGHT NOT be deterministic.

      Well no, actually, we don’t have good reasons for supposing that mental events might not be deterministic (except for the issue of quantum indeterminacy, which isn’t all that relevant for “free will”). All we have is human dislike of the idea that mental events are the playing out of physics.

      This one is again dealt with by Occam’s razor. Until such time that we’re given good evidence for adding in a non-material component (that is, until such models actually explain or predict a lot more) then we can discount the suggestion.

    22. verbosestoic

      Well, no, they need a *good* reason for saying that.

      You rely on the vague term “good” a lot, and are indeed vague about what it means … particularly on how strong that has to be. When you use it in our discussions of religion, it’s pretty weak, but here it seems like it would have to be pretty strong, given your comments.

      Let’s stop talking about that. Let’s look at the examples that I talked about.

      So, for any psychological theory, if we gathered our initial data in Canada, we’d have to test in other cultures to see if it held cross-culturally, because we know that culture can greatly impact psychological traits. If we had separate reason to think that this was cross-culturally valid, then we might suspend that. So if, for example, we noted that this trait went AGAINST typically Canadian cultural traits, we MIGHT argue that this is probably cross-cultural … but we should test it anyway before committing to knowing that.

      We don’t do this for physics and chemistry and biology, but that’s only because they’ve ALREADY done the confirmation: the basic laws of those fields apply everywhere based on all of the tests that we’ve already done. Given that MASSIVE success and testing, we’re pretty confident that they still apply, and so you’d need a strong reason to say that in this case, it doesn’t. This might not apply to other galaxies; we assume that they are, but any oddity that might be explained by a change in the laws will trump that assertion and force extra testing.

      This is what I mean by “reasons”. As far as I can see, we NEVER take a simple “Everything has turned out that way before” case and claim that justifies knowledge. Even the physics cases include confirmed theory, not just the results of experiments.

      Well no, actually, we don’t have good reasons for supposing that mental events might not be deterministic (except for the issue of quantum indeterminacy, which isn’t all that relevant for “free will”). All we have is human dislike of the idea that mental events are the playing out of physics.

      No, the debate ITSELF provides good reasons because we have far more than that. There are clear differences between actions taken from conscious deliberation and those that aren’t. To see this, take the case of someone throwing a rock at a window, and breaking it. We can easily assign ultimate responsibility to the thrower because they took the conscious action there, and the rock had no impact on that. Now imagine that we throw YOU at the window. You’re a conscious agent, unlike a rock, so it’s POSSIBLE to assign you the responsibility here … but we know that in that instance you are acting a lot more like the rock than you are like the thrower. So the behaviour is different there, and is different in terms of PRECISELY the sort of behaviour that we’re talking about (the mental). And we note that the determinism presumption is based on cases like rocks, not like conscious deliberators. Even the experiments of Libet are based on things that we don’t reason about, but take instinctively and never thought were really the prime examples of free will. So conscious deliberation and the examples you’re using to get to “determinism” are not only different, but differ in the critical aspects that we are considering. Thus, you can’t validly inductively generalize from the rock-type cases to the deliberator-type cases; we need to actually prove that deliberator-type cases are actually the same sort of case.

      If all material things are deterministic, and turns out that deliberator-type things are not, then we need a non-material component. So Occam’s Razor can’t slice that away. And it can’t slice away the “deliberator-type things are not deterministic” because of the debate that we’re having. So neither induction nor Occam save you here.

      Note that I am not claiming that I have proven that we have free will, or that mind is not deterministic, or that mind is immaterial. This line works as long as we don’t KNOW the outcome, and demonstrates, again, how you can’t use induction on this because you’d end up presuming the very thing that’s under debate, not demonstrating it.

    23. Coel Post author

      We don’t do this for physics and chemistry and biology, but that’s only because they’ve ALREADY done the confirmation: the basic laws of those fields apply everywhere based on all of the tests that we’ve already done.

      Which means you’re accepting this sort of extrapolation in principle. So the only discussion is whether it is sensible in a particular case.

      So the behaviour is different there, and is different in terms of PRECISELY the sort of behaviour that we’re talking about (the mental).

      Agreed. But the difference there is one of social convention and social utility. The difference is not about physics, not about materialism vs dualism, and not about determinism vs non-deterministic explanations. One can produce just as good an account of your scenario from the point of view of materialism and determinism.

      So conscious deliberation and the examples you’re using to get to “determinism” are not only different, but differ in the critical aspects that we are considering.

      We treat them differently socially, yes, but that’s not saying there is any difference in the physics, nor anything non-deterministic.

      So neither induction nor Occam save you here.

      I can still invoke Occam and say that, since the deterministic account works fine here, there is no reason to go for a non-deterministic account, given the slew of problems that would throw up.

    24. verbosestoic

      Which means you’re accepting this sort of extrapolation in principle. So the only discussion is whether it is sensible in a particular case.

      And somehow my explicit and repeated statements that induction can produce knowledge but that the case you were relying on WASN’T one of those cases confused you on that point? Again, they don’t just say “I haven’t seen it fail”. They actually TESTED the basis principles in all of those cases, and so if the new principles follow on from those, then we can be confident that it holds for them as well unless we have reason to think otherwise.

      Agreed. But the difference there is one of social convention and social utility.

      No, it’s experiential: when I’m flying at the window, I am in a completely different mental state vis a vis that situation than I am when I throw a rock at the window, and it is from that experience — and experiences of intention — that I conclude that I am responsible for the breaking of the window in the latter case in a different way than I am in the former case, and that that case has a bearing on social and moral matters. Thus, to reduce it to social convention is to ignore actual experience.

      The difference is not about physics, not about materialism vs dualism, and not about determinism vs non-deterministic explanations. One can produce just as good an account of your scenario from the point of view of materialism and determinism.

      But the main issue has been this: if the material is deterministic, and the rock’s interactions are deterministic, and my “intentional actions” are deterministic IN THE SAME WAY, then at BEST the perceived difference is illusory. But I have no reason to think that the difference IS illusory. Thus, if materialism implies that AT MINIMUM, then materialism about mental things must be false. Or, in short, if materialism implies determinism, and determinism implies that this distinction is illusory, then determinism seems false, and so seems materialism about mind.

      That determinism implies that the difference is illusory is agreed upon by libertarians and hard determinists. Compatibilists dispute that, but are having a devil of time preserving what is important about both views, without reducing me to the rock, or on the other hand overly weakening determinism.

      I can still invoke Occam and say that, since the deterministic account works fine here, there is no reason to go for a non-deterministic account, given the slew of problems that would throw up.

      As soon as you appeal to problems, you abandon Occam; you are talking about evidence, not simplicity. And you also have to ignore all of the problems that the DETERMINISTIC approach has. And again, induction doesn’t help you because mental phenomena are too different from what we’d call “strictly physical phenomena” to make the generalization valid.

    25. Coel Post author

      if the material is deterministic, and the rock’s interactions are deterministic, and my “intentional actions” are deterministic IN THE SAME WAY, then at BEST the perceived difference is illusory.

      I beg to differ. The brain would be in a particular state (a real, physical state). That state we call the “intention”. The rock would not be in that state. That difference is material, physical and real. In the same way, a chess-playing computer about to play a particular move has a particular pattern in its innards, a pattern that isn’t there if the computer is idling.

    26. verbosestoic

      I beg to differ. The brain would be in a particular state (a real, physical state). That state we call the “intention”. The rock would not be in that state.

      Yes, as I said, I’m aware that you’re a compatibilist [grin].

      The issue here is that you end up saying that the intentional actions are NOT deterministic “in the same way” as those of the rock. They’re different, but not in a way that means that they aren’t deterministic. That’s why you can keep the distinction between rock actions and intentional actions, which hard determinists (who claim they are no different) and libertarians (who say they are different in a way that precludes them being deterministic) both deny. The problem for compatibilists is preserving the determinism while also preserving that the decision-making process — and, particularly — the experiences of that — matters. But if the experience of decision-making is produced by the neural firings, and the neural firings are causally closed, and the neural firings are produced only by past experiences and by genetic and physical structure, and the changes to the brain based on that are ALSO causally closed,then there is no room for the experiences to do anything without having overdetermination. At which point, you lose the experiences again, as they appear to matter but it seems like they don’t or can’t.

      Hard determinists like Jerry Coyne use these sorts of results to try to prove their cases (Libet experiments, for example).

      In the same way, a chess-playing computer about to play a particular move has a particular pattern in its innards, a pattern that isn’t there if the computer is idling.

      So let’s do the comparison there, then. Take a bunch of chess books, and sit down and play chess just by analyzing the current position of the board and choosing the best option. Now play chess yourself. There’s a sharp difference in your decision-making there, and in fact the first case doesn’t seem like you making decisions at all. Thus, there is a critical difference between the two cases, so again you can’t analogize from one to the other.

      Note that this isn’t an appeal like the Chinese Room, because while there it isn’t clear if the person in the room is really in the role of the computer, in this case you really are doing only what the computer is doing wrt decision-making, and so they really are directly comparable (and so that might even make the Chinese Room clearer).

    27. Coel Post author

      The issue here is that you end up saying that the intentional actions are NOT deterministic “in the same way” as those of the rock.

      I’d say — as a compatibilist — that they are indeed deterministic in the same way as those of the rock. The term “intention” is a useful one about a particular sort of pattern, one that the rock does not have, but everything is thoroughly deterministic in the same way.

      The problem for compatibilists is preserving the determinism while also preserving that the decision-making process — and, particularly — the experiences of that — matters.

      But we can build deterministic decision-making processes. Good examples are a chess-playing computer or an aircraft autopilot or a Google self-driving car. Those demonstrate that compatibilism is a self-consistent account. Our human brains make decision just as those robotic devices do.

      But if the experience of decision-making is produced by the neural firings … then there is no room for the experiences to do anything without having overdetermination.

      If one sees the “experience of decision-making” as simply an integral part of the neural-firing decision making, then there is no problem to resolve. It is dualists who see the “experience” as something different from the physical hardware, the compatibilists do not.

      Take a bunch of chess books, and sit down and play chess just by analyzing the current position of the board and choosing the best option. Now play chess yourself. There’s a sharp difference in your decision-making there, and in fact the first case doesn’t seem like you making decisions at all.

      Sorry, I don’t understand your distinction. Once you’re into the middle-game, the chess books won’t tell you which move to make. They might help you understand a position, but it’d still be you analysing the position and choosing the move. I don’t see any real difference between these two cases.

    28. verbosestoic

      I’d say — as a compatibilist — that they are indeed deterministic in the same way as those of the rock.

      One of the key things you learn doing philosophy — and, I think, one of the reasons why people not versed in it do it so badly — is that you have to read and respond to whole sentences and paragraphs, and that the words used are really, really important. Here’s what I said:

      The issue here is that you end up saying that the intentional actions are NOT deterministic “in the same way” as those of the rock. They’re different, but not in a way that means that they aren’t deterministic.

      So, yes, in a sense they are just as deterministic, as I stated. But they are also different in a critical way, and in a way that implies control, and intentional control. In short, that intentional pattern absolutely MATTERS and characterizes the interaction.

      But we can build deterministic decision-making processes. Good examples are a chess-playing computer or an aircraft autopilot or a Google self-driving car. Those demonstrate that compatibilism is a self-consistent account. Our human brains make decision just as those robotic devices do.

      But are they making decisions, or just doing pattern-matching and stimulus-response? Because we know in ourselves that pattern-matching and stimulus-response are NOT making decisions; the experiences are notably different.

      If one sees the “experience of decision-making” as simply an integral part of the neural-firing decision making, then there is no problem to resolve.

      Um, actually, it’s seeing it as that that CAUSES the problem, as I pointed out, because it implies that it is possible for the same physical reactions to happen even if the experiences weren’t there. The electrical transfer between the neurons is sufficient cause to move from the stimulus to the action in the world, and so that specific experiences are associated with that can add no more causal power to the story. This implies that if we replaced the neurons with things that produced no experiences but created the same sorts of electrical activations, we’d see the same external behaviour. We would create, in essence, a zombie a la Chalmers, except that it would have a different physical structure. This would mean, essentially, eliminating what our experience, then, is telling us about the process. You, as we’ve established, aren’t really willing to go there, at least not yet.

      Once you’re into the middle-game, the chess books won’t tell you which move to make. They might help you understand a position, but it’d still be you analysing the position and choosing the move.

      Okay, let’s back up with your example. What do you think a chess computer is doing, in detail, when it analyzes a position and chooses one to play? If you’re using it as your example, you have to have some idea as to the details, right, so that we can see if we can emulate it ourselves and see what set of experiences it maps to in us: rock or thrower?

    29. Coel Post author

      But are they making decisions, or just doing pattern-matching and stimulus-response?

      Both. Pattern-matching, stimulus response, and other forms of information processing, is exactly what “making decisions” is.

      Because we know in ourselves that pattern-matching and stimulus-response are NOT making decisions; the experiences are notably different.

      We actually have very little awareness of how we make decisions. None of us are aware of most of what is going on in our brains.

      … it implies that it is possible for the same physical reactions to happen even if the experiences weren’t there.

      That sees that experiences as something separate to the physical interactions. The materialist would see the experiences as an integral part of the physical interactions.

      This implies that if we replaced the neurons with things that produced no experiences but created the same sorts of electrical activations, we’d see the same external behaviour. We would create, in essence, a zombie a la Chalmers, …

      This thought experiment is based on the idea that a p-zombie is actually possible. That is based on the idea that conscious experiences are an optional extra, rather than being an integral part of the process. I don’t see any reason to grant that.

      Okay, let’s back up with your example. What do you think a chess computer is doing, in detail, when it analyzes a position and chooses one to play?

      It’s doing a whole lot of information processing and evaluation of options.

    30. verbosestoic

      The key here is that I claimed that your theory was problematic because it denied our experiences and didn’t have sufficient evidence to do so. Here, you move away from and start denying our experiences to preserve your theory:

      Both. Pattern-matching, stimulus response, and other forms of information processing, is exactly what “making decisions” is.

      Except that we can do that sort of strict, rote information processing and note that it is different from what we experience when we actually make decisions. If we were the person inside the Chinese Room, we CLEARLY wouldn’t be understanding the language or deciding what to say. If a computer or a system can only do that, then it doesn’t seem like it makes decisions.

      We actually have very little awareness of how we make decisions. None of us are aware of most of what is going on in our brains.

      Sure, but if our experiences are accurate, we shouldn’t need to. And if they aren’t accurate, you had better have good reasons and evidence to demonstrate that.

      That sees that experiences as something separate to the physical interactions. The materialist would see the experiences as an integral part of the physical interactions.

      Which gets massively complicated, because you’d have to demonstrate what it is that the EXPERIENCES are doing that impacts the neural firings. This means that the experiences can’t be the RESULT of the neural firings, because then that causation wouldn’t be there. So you need to find a way that the mental experiences themselves can have a direct causal impact on the state above and beyond what the neuron firings themselves do. If you can’t, then you have overcausation and the thought experiments follow, but it’s hard to see how you can and still reduce mind meaningfully to neural firings.

      The materialist view, to be meaningful, has to imply that the causal chain in the brain is closed and follows the neural firings solely. Any additional causation added there means that dualism is the simpler explanation. But without additional causation, you don’t have the experiences actually having causal power.

      This thought experiment is based on the idea that a p-zombie is actually possible.

      Actually, no, as this is roughly my thought experiment from https://verbosestoic.wordpress.com/nailing-the-third-person-science-of-consciousness-to-the-wall/ , and I don’t think that p-zombies are possible. The reason is that I think that phenomenal experiences are an input to the system, and so if you take out experiences you might — and likely are able — to get the behaviour, but the brain would have to act differently to replace the missing inputs (more on that here: https://verbosestoic.wordpress.com/phenomenal-experience-and-cognitive-function/ ). The thought experiment follows from the materialist/neurological response to dualism: all of the causation is explained by the interactions between the neurons, and so there is no room for anything else to have causal power. Bien sur. But then I can replace each individual neuron with something that fires off in the same way in reaction to the same responses, and get the same behaviour. Even if the experiences are integral to neuron firings, they wouldn’t be with those replacements, but the visible behaviour would be the same. So if we take the tack that experiences are fundamental to the operation of neurons AND that there is no causal room for dualism, then as long as I replace the causal chain that the neurons were doing, I can build something that acts exactly like a conscious person with conscious experiences but who actually has NO conscious experiences. One of your materialist principles has to be false to avoid this.

      It’s doing a whole lot of information processing and evaluation of options.

      That word, detail, I do not think it means what you think it means [grin].

      Let me say why detail is important here: what I want is the outlined and direct procedure, so that I can simulate it, not on a computer, but myself, and see that if I do that does it feel like decision-making. In short, if I play chess the way the computer plays chess, does it feel the same way as if _I_ was playing chess? This is tying it back to experience itself; if the experiences are different, then your standard inductive argument doesn’t work, because a) they aren’t the same thing and so might not be similar enough to make the comparison and b) I actually have good reason to think that they ARE different. What you say here, obviously, isn’t good enough to do that.

    31. Coel Post author

      Hi verbosestoic,

      The key here is that I claimed that your theory was problematic because it denied our experiences and didn’t have sufficient evidence to do so.

      Well what actually are our “experiences” of our decision making? We “experience” making a decision, and we’re not really sure how we make them, because we are not conscious of most of what is going on in our brains. We are often aware of conscious deliberations revolving about that decision, but that does not imply that our consciousness is making the decision, it only means that the only aspects of it that we’re consciously aware of are the conscious ones.

      Thus I don’t see that my theory denies what we experience, it merely denies some possible interpretations of what we experience.

      If we were the person inside the Chinese Room, we CLEARLY wouldn’t be understanding the language or deciding what to say. If a computer or a system can only do that, then it doesn’t seem like it makes decisions.

      You seem to be asserting that because we experience consciousness along with decision-making, therefore consciousness is necessary for decision making. I don’t see that I need to accept that. (More on the Chinese Room in an imminent post.)

      Which gets massively complicated, because you’d have to demonstrate what it is that the EXPERIENCES are doing that impacts the neural firings.

      This whole question presumes that the conscious experiences are something “other than” the neural firings. Again, I don’t see that I need to grant that. (Essentially, I don’t grant the possibility of a p-zombie, a set of neural firings that are functionally identical, but not conscious.)

      I’ll need to read your posts on this, but for now:

      The thought experiment follows from the materialist/neurological response to dualism: all of the causation is explained by the interactions between the neurons, and so there is no room for anything else to have causal power. Bien sur. But then I can replace each individual neuron with something that fires off in the same way in reaction to the same responses, and get the same behaviour. Even if the experiences are integral to neuron firings, they wouldn’t be with those replacements, but the visible behaviour would be the same.

      How do you know that the experiences would then be different? Again you’re asking for a p-zombine: identical function but no experiences. I can assert that, if the replacements were functionally equivalent, then they would generate experiences indistinguishable from the original ones.

      Let me say why detail is important here: what I want is the outlined and direct procedure, so that I can simulate it, not on a computer, but myself, and see that if I do that does it feel like decision-making. In short, if I play chess the way the computer plays chess, does it feel the same way as if _I_ was playing chess?

      You could not experience being a chess-playing computer, since you are very different from that chess computer, and have very different experiences and consciousness. So no, it wouldn’t feel the same way if you simulated it: for starters you’d be doing a conscious first-person reflection on the proceedings whereas the computer would not be doing so. That, though, does not demonstrate that the first-person reflection is necessary for the decisions to count as “decisions”. That is, unless you want to define “decision” as requiring that, which I would not.

    32. verbosestoic

      Thus I don’t see that my theory denies what we experience, it merely denies some possible interpretations of what we experience.

      If you are denying, as you seem to here, that the most obvious interpretations of our experiences are wrong simply because it clashes with your preferred theory, then it’s pretty safe to say that you’re denying our experiences. If there is an alternative theory that can explain all of the evidence you’re citing in favour of your theory and doesn’t have to argue that our experiences aren’t really indicating what’s going on, that theory is to be preferred, even if that explanation has to be convoluted. In general, a response of:

      We “experience” making a decision, and we’re not really sure how we make them, because we are not conscious of most of what is going on in our brains.

      without strong evidence that our experiences really ARE misleading up or missing critical information can be taken as denying our experiences: our experiences suggest X, but because we don’t really know what’s happening in the brain it might not be X. If you don’t have really good evidence to suggest that our experiences are misleading us about X, there’s no reason to think that the brain really is doing something different.

      In these examples, I’ve been pretty careful about what I was using experiences to imply. In general, I was taking myself through two methods and noting that my experiences using both methods were different, therefore it was reasonable to think that the two methods weren’t the same, and especially weren’t the same wrt consciousness and conscious deliberation which is what, I pointed out, we were trying to explain. To retreat to “Well, maybe the experiences aren’t reflecting what’s REALLY going on”, as you did, is to deny what our experiences are telling us without good reason.

      You seem to be asserting that because we experience consciousness along with decision-making, therefore consciousness is necessary for decision making. I don’t see that I need to accept that.

      No, what I’m saying is that the deliberation and decision-making that I do when I consciously deliberate is accompanied by certain experiences , and those experiences are ABSENT in the rote cases that I’ve talked about. Thus, doing those rote cases is NOT doing the same thing as the rote cases to, or at least we have very good reason to think that there’s some difference somewhere. Thus, you have no reason to conclude or even think it REASONABLE that those rote cases really are what we do when we perform the CONSCIOUS versions of those actions. You need much better evidence than you’ve provided to get there.

      This whole question presumes that the conscious experiences are something “other than” the neural firings. Again, I don’t see that I need to grant that. (Essentially, I don’t grant the possibility of a p-zombie, a set of neural firings that are functionally identical, but not conscious.)

      I’m not sure what you mean when you argue that the phenomenal experiences are nothing more than the neural firings, at least in any way that matters for my discussion, so perhaps you can expand on that a bit.

      Let me outline what the issue is, since I’m not sure you’re getting it:

      The process for, say, seeing a red stop sign and stopping, at the neural level, is essentially this: light hits the retina, this spawns an initial neural firing that triggers a number of other neurons to fire, cascading through the brain until the neuron firings trigger nerves that then result in specific bodily actions, that result in my stopping the car at the stop sign. To avoid dualism, this has to be causally closed; once that light hits the retina, it is one single uninterrupted causal chain. Now, experiences also happen at this point. I’m not going to talk about HOW they happen, but they happen in some way (either they are just inherent to the actions or they are caused by them or whatever). But, again, this is causally closed. So the question is: can the nature of those experiences impact the neural firings we have? Well, it looks like they CAN’T, because that’s all causally closed. So, in order to refute p-zombies, you have to demonstrate that if it happened to be the case that neural firings were identical in all ways to the ones we see here but whatever mechanism it was that produced the experiences wasn’t present that we’d see different behaviours and outcomes. And based on the standard materialist theory, we can’t; as long as the firing happens to activate the next neuron down the line, the nerves will be activated and the car will be stopped.

      You can try to argue that the only way to get different experiences — or to have none — is to have different neural firings, but this doesn’t help you, because all you’d be saying is that different neural firings produce different experiences — however they do it — and thus potentially different behaviours. That’s “the experiences are different because the neural firings are different”. To avoid the issues I’ve raised here, you need to be able to say that you have different experiences, and BECAUSE you have different experiences the neural firings are different, and that therefore the properties of the experience CAUSE the neural firings to be different. But, again, by the standard materialist theory the neural firings are causally closed, and so that can’t happen.

      In short, you need to explain mental causation by pointing out the causal role the MENTAL properties, in and of themselves, have. If you can’t, then it looks like there’s no room for those properties, in and of themselves, to have causal power, and thus they can be removed and the behaviour is still the same.

      I deny p-zombies as well, but it’s only because I argue that the experiences and mental properties THEMSELVES have to have causal power (for me, experiences are the input to the system that we react to). But standard materialist theories aren’t open to that move.

      How do you know that the experiences would then be different? Again you’re asking for a p-zombine: identical function but no experiences.

      A p-zombie is something that is PHYSICALLY identical to us but has no experiences. My thought experiment is not physically identical, as it doesn’t have neurons, and there is no reason to think that the small processors that replace each neuron produce or contain any kind of experiences. In addition, we KNOW that we can have identical functional responses to having experiences when we don’t have that experience at all (actors faking pain, for example). So we know that simple functionality isn’t sufficient to show that the experience is there, and we have no reason to think that the small processors produce experiences in any way, so here if the standard materialist theory is true, it definitely does seem like we’d produce the observable behaviour without having experiences, because we would again have a causally closed chain from the light striking the eye to the nerve activations that we have no reason to think produces phenomenal experiences.

      If you think you have a way out of that, I’m all ears [grin]!

      BTW, you probably shouldn’t focus so much on brain arguments, because that doesn’t seem to be your position. Ironically, you seem to be a strong FUNCTIONALIST when it comes to consciousness and mind; consciousness is the functions it performs that are externally observable. But functionalism doesn’t care about implementation; you can build it out of anything and as long as it produces the same observable behaviour it’s equally conscious. The problem with functionalism, for me, is that the phenomenal experiences AREN’T externally observable, and so we can generate the BEHAVIOUR of phenomenal experiences without HAVING phenomenal experiences. Thus, functionalism’s definitions ignore phenomenal experience — which is what it means to be conscious, to me — or alternatively it is provably false that you can look at the external behaviour and know what the phenomenal experience is. Or you could include the phenomenal experience in the functionality and functional definitions, but this just gets us right back to where we were before trying to be functionalist about it, so it doesn’t help much.

      You could not experience being a chess-playing computer, since you are very different from that chess computer, and have very different experiences and consciousness.

      If this is true, then it is possible that it has NO experiences at all, and you’ve conceded the possibility that my thought experiment works; our collection of processors is a lot different from the set of neurons that make up our brain.

      That, though, does not demonstrate that the first-person reflection is necessary for the decisions to count as “decisions”. That is, unless you want to define “decision” as requiring that, which I would not.

      Here I’m defining “decision”, in this context, as doing what we do when we typically make decisions, like when I deliberate and decide where I’m going for lunch, or what move to make in a chess game. Those sorts of decisions are accompanied by certain experiences. If you are going to claim that we are doing just exactly what the chess computer does when we decide what move to make, I should be able to do its procedure and have roughly similar experiences. But I don’t. In fact, it doesn’t seem like I’m making a decision at all. This is strong empirical evidence that they AREN’T the same procedure, as long as you don’t deny what our experiences are saying … which leads back to the start of the comment.

      You can define the word “decision” to be broader and include what the computer is doing, and it may not be a bad definition. However, even if you do that, you’ll still run into the issue that human deliberative and conscious decisions don’t seem to be just the same thing as computer decisions, because I can make both human deliberative AND computer decisions and note that the experiences are radically different.

    33. Coel Post author

      … and doesn’t have to argue that our experiences aren’t really indicating what’s going on …

      What do you think that our experiences are telling us about what is going on when we make a decision?

      Personally I think that our experiences are telling us rather little about how we make decisions.

      But, again, this is causally closed. So the question is: can the nature of those experiences impact the neural firings we have?

      This question again presumes that the experiences are not already entirely specified by the neural-firings account, but are something different. Again, it is presuming that that one could have a totally identical set of neural firings but without the conscious experiences (a p-zombie), and again I don’t grant that possibility.

      Once you’ve specified the neural-firing account I assert that everything about the conscious experiences is also fully specified, and that entails the effect the conscious experiences are having on the neural firings.

      Let me give a comparison. One could analyse a physical situation by, say, a particle-by-particle account of it. Or one could apply the concept of conservation of energy and describe it that way, totally ignoring individual particles. But, if you did give the particle-by-particle account then everything about the second account would be entirely entailed by it.

      So, in order to refute p-zombies, you have to demonstrate that if it happened to be the case that neural firings were identical in all ways to the ones we see here but whatever mechanism it was that produced the experiences wasn’t present …

      But, if one were to grant that that was possible, then that alone would *validate* the concept of the p-zombie. It is exactly that possibility that I’m disputing.

      If there is a neural-firing account of a consciousness generating situation, then, I assert, there is no *identical* neural-firing account that lacks the consciousness.

      That is like asserting that, if a particle-by-particle account conserves energy, then there could be an entirely identical particle-by-particle account that does not conserve energy. That, to me, is incoherent.

      … and that therefore the properties of the experience CAUSE the neural firings to be different.

      Or that the two are linked together, such that they are effectively just different aspects of the same thing.

      My thought experiment is not physically identical, as it doesn’t have neurons, and there is no reason to think that the small processors that replace each neuron produce or contain any kind of experiences.

      The small processors, each by themselves, would not be conscious or have experiences, just as each neuron will not be conscious in its own right. But, the pattern of your processors could be conscious in the same way that the pattern of neurons is conscious. I don’t see how you know that it wouldn’t be. If you asked it whether it was conscious it would say yes!

      The problem with functionalism, for me, is that the phenomenal experiences AREN’T externally observable, and so we can generate the BEHAVIOUR of phenomenal experiences without HAVING phenomenal experiences.

      How do we know that? Do we really know that we could, in principle, build a functionally equivalent simulation of a human that did not have phenomenal experiences?

      If this is true, then it is possible that it has NO experiences at all, and you’ve conceded the possibility that my thought experiment works; our collection of processors is a lot different from the set of neurons that make up our brain.

      I certainly concede that a chess-playing computer (likely) has no experiences, but then it’s a vastly more limited machine than a human brain. The interesting question is whether, if you built in all the much greater capabilities of the human brain, that you would be building in consciousness whether you liked it or not!

      Here I’m defining “decision”, in this context, as doing what we do when we typically make decisions, like when I deliberate and decide where I’m going for lunch, or what move to make in a chess game. Those sorts of decisions are accompanied by certain experiences.

      OK, so you’re defining “decision making” in such a way that it needs consciousness to be part of the process, or it doesn’t count. OK, noted (philosophers seem to *always* do this, saying that “understanding”, and “purpose” and all such similar things require consciousness 🙂 ).

      If you are going to claim that we are doing just exactly what the chess computer does when we decide what move to make, …

      No, I wouldn’t claim that. In saying that a chess-playing computer makes “decisions” I’m making a much more limited interpretation of the concept “decision making” which does not require any consciousness.

      What I would suggest is that human brains are doing the sort of thing that chess-playing computers are doing, *along* *with* an array of other things and much-more-general capabilities, and that the *ensemble* is conscious.

    34. verbosestoic

      What do you think that our experiences are telling us about what is going on when we make a decision?
      Personally I think that our experiences are telling us rather little about how we make decisions.

      Well, remember, what I REALLY said here was this:

      If there is an alternative theory that can explain all of the evidence you’re citing in favour of your theory and doesn’t have to argue that our experiences aren’t really indicating what’s going on, that theory is to be preferred, even if that explanation has to be convoluted.

      You’re not answering that here, at all. You seem to be insisting that our experiences aren’t completely accurate, but this does nothing to refute the idea that if someone can produce a theory that explains the physical/neural evidence AND maintains that our experiences are accurate, then that’s the theory that we ought to prefer. That, I think, is what must be accepted AT A MINIMUM. The only question, then, is what we have to do if your theory explains the physical better but leaves out the experiences. My first blush answer is that the burden of proof is on you, because our experiences are the foundations of all possible knowledge, so we ought never abandon them without EXCEPTIONALLY good reason.

      This question again presumes that the experiences are not already entirely specified by the neural-firings account, but are something different.

      The experiences don’t APPEAR in the neural firing account, nor do we derive the experiences FROM the neural firings account as a higher level summary/description of what’s going on. The PRIMITIVE, lowest level point here are the experiences, and the neural firings account is supposed to EXPLAIN those experiences and their role in causation. Thus, the issue: what is the relation between the neural firings and the experiences? My thought experiment points out that under the neural firings model it seems like as long as the electrical circuit is made — even with things that don’t seem to have experiences, or that at least that we have no reason to THINK have experiences — the same behaviour will result. We also know that mere behaviour isn’t sufficient since we know that we can act as if we are having a specific phenomenal experience even if we aren’t (for example, we can read about a red car without seeing it and know, therefore, what seat covers would look good there without having a phenomenal experience of it). Given that, it looks like there’s no room for the experience itself — and its details — to have causal power. If you think that there IS still room there, you need to do more than talk about how the neural firings account MIGHT entirely specify the experiences, but instead show HOW it does that, so we can see what impact that might have on my thought experiment and the general issue of epiphenomenalism. For the most part, you’ve been quite vague on how precisely this is all supposed to work.

      And remember, I think that you CAN’T have a physically identical zombie who has no phenomenal experiences, both as a consequence of your theory AND in general. But the general consequence of your theory, it seems to me, proves epiphenomenalism, and so what I need is for you to show me how you expect to escape the trap that I think my other posts, comments, and thought experiments show that you have to get into.

      But, if one were to grant that that was possible, then that alone would *validate* the concept of the p-zombie. It is exactly that possibility that I’m disputing.

      Well, no, although I might not be being entirely clear on this point. Let me be more precise: I argue that on your account I can preserve the “electrical circuit” properties of the neurons but not the “produces phenomenal experiences” properties of the neurons, and still produce the same behaviour. That’s not a p-zombie, because I concede that we’d have to make a physical change, but it IS saying that the electrical circuit properties don’t have to be — and we have no reason to think ARE — the properties that result in us having phenomenal experiences. Thus, the question to you is to demonstrate that AT A MINIMUM the electrical circuit properties DEPEND on the phenomenal experiences properties, without merely assuming it,since we have no reason to think that electrical circuit properties produce experiences in anything.

      But, the pattern of your processors could be conscious in the same way that the pattern of neurons is conscious.

      But we have no reason to think that it’s the PATTERN that matters when it comes to phenomenal experiences. If you want to assert that this is the case, you had better have good reason to think so, and accept a number of rather odd conclusions (like that it is possible that a neural net RIGHT NOW has phenomenal experiences even though it gives no sign of it. We have phenomenal experiences that we give no sign of, too, except internally).

      How do we know that?

      We do it ourselves, as in a number of cases we act as if we are having phenomenal experiences that we aren’t having. We can even do it by rote. Actors do it constantly, for example. So, yes, it is obvious that you can’t simply look at the external behaviour and conclude that a particular phenomenal experience is happening. And from there, we can conclude that it is possible to constantly act as if a phenomenal experience is happening without having any AT ALL. Only consciousness-about-consciousness MIGHT be immune to this … but my thought experiment about phenomenal knaves certainly raises doubts about THAT. So, yes, we know this, from personal and consistent experience.

      Do we really know that we could, in principle, build a functionally equivalent simulation of a human that did not have phenomenal experiences?

      Do we know that we can build one that DOES? And if you accept this, then there’s nothing special about neurons, and thus being conscious does not mean having neurons in a certain state and producing certain behaviours, at which point the neural firings account seems to be less definitional and more descriptive, and so you’d still need a definition of conscious beyond that. In short, you’d reduce neurons to being a way to implement consciousness, at which point, again, you should stop talking about neurons and just admit that you’re really a functionalist about mind, so that at least we can have a consistent theory to work with [grin].

      The interesting question is whether, if you built in all the much greater capabilities of the human brain, that you would be building in consciousness whether you liked it or not!

      Which, to me, as I’ve said, is whether or not that thing is actually HAVING experiences or not. But I will say that I am confident that adding more code is NOT going to get us there, because code doesn’t seem to be the sort of thing that can produce phenomenal experiences. Thus, if you think that adding more processors that DON’T produce experiences in the chess playing computer and/or a bigger look-up table when look-up tables don’t produce experiences in the chess playing computer but that somehow, if we add enough of them we will magically get there, I think that it is clear that you are making the extraordinary claim and so the burden of proof is on you here.

      OK, so you’re defining “decision making” in such a way that it needs consciousness to be part of the process, or it doesn’t count.

      No, I’m defining CONSCIOUS decision-making that way, which is what I’ve been talking about from the beginning. Whether or not the computer makes decisions doesn’t matter to me unless those are also conscious, and as I pointed out when I simulate it on my own hardware the computer process maps not to CONSCIOUS decision-making, but to UNCONSCIOUS decision-making. It also maps to cases where I DON’T understand what’s going on or involved, and so don’t have understanding. And so on. So if you are going to assert — as you did — that the methods are the same, I have EXCEPTIONALLY good evidence that they aren’t. Again, the burden is on you to overturn that.

      What I would suggest is that human brains are doing the sort of thing that chess-playing computers are doing, *along* *with* an array of other things and much-more-general capabilities, and that the *ensemble* is conscious.

      First, for me the only extra capability relevant is “Has phenomenal experiences”. If you disagree with that, then we aren’t talking about the same definition of consciousness.

      Second, given what you said, it is perfectly reasonable to think that some of those things and capabilities are beyond the capability of a machine. It’s especially reasonable to think that those things are more than simply symbolic processing — which is what Searle was talking about in the Chinese Room — ESPECIALLY since we note that when WE do simple symbolic processing we don’t have the consciousness stuff that we have in the conscious deliberation. Your suggestion, then, gives the game away to your opposition; you admit that they may well be right, and that you don’t really know that they’re wrong. Until you can show what those other things are and how to get a machine to do them, you can’t say that the make conscious decisions, even if we agree that they make decisions of a sort. The debate, though, has always been about conscious, semantic decisions, and so your move here REALLY risks equivocation … as your argument with me seems to have been the entire time.

    35. Coel Post author

      You seem to be insisting that our experiences aren’t completely accurate, …

      No, I think that our experiences are accurate. What is it about our experiences that you think is inconsistent with my model of how this works? Of course some interpretations of our experiences say something different, but one can equally make another interpretation that is just as much in line with our actual experiences that is fully in accord with my model.

      The experiences don’t APPEAR in the neural firing account, nor do we derive the experiences FROM the neural firings account …

      If one take a particle-by-particle account of gas particles in a metal box, the concepts “entropy” and “pressure” would not be part of that particle-by-particle account, but they would be fully entailed by that particle-by-particle account. As I see it, the conscious experiences are not something separate from the neural-firings account, they are entailed by it.

      My thought experiment points out that under the neural firings model it seems like as long as the electrical circuit is made — even with things that don’t seem to have experiences, or that at least that we have no reason to THINK have experiences — the same behaviour will result.

      You are entitled to postulate a totally identical neural-firing account that does not have conscious experiences, but I am entitled to deny that possibility. To me it is akin to arguing for an identical particle-by-particle account of a gas that has a different pressure. That is self-contradictory.

      … you need to do more than talk about how the neural firings account MIGHT entirely specify the experiences, but instead show HOW it does that, …

      But neither of us can produce an actual account of how experiences are generated. So I’m no worse off than you on that point.

      For the most part, you’ve been quite vague on how precisely this is all supposed to work.

      Yes, I admit that. That’s because I don’t really know. But your account is no better in that regard, and thus there is no basis for preferring your account.

      That’s not a p-zombie, because I concede that we’d have to make a physical change, but it IS saying that the electrical circuit properties don’t have to be — and we have no reason to think ARE — the properties that result in us having phenomenal experiences.

      So, we agree that *some* physical properties would have to be changed in order to have no phenomenal experiences. Whether those properties are “electrical circuit” properties or other physical properties we can argue about. I’ll readily admit that I’m not sure what physical properties are actually necessary for conscious experiences. But, so long as they are entailed by the overall physical nature of the brain, my stance is entirely defendable. In causal terms, whichever physical properties are necessary for the experiences are the ones causing the experiences; they can also physically influence the flows of electric charge along neurons. Thus there is no “overdetermination” problem here.

      But we have no reason to think that it’s the PATTERN that matters when it comes to phenomenal experiences.

      Well we do: phenomenal experiences seem to be strongly correlated with the highly patterned objects we call brains.

      Actors do it constantly, for example. So, yes, it is obvious that you can’t simply look at the external behaviour and conclude that a particular phenomenal experience is happening.

      But, in the case of the actor, there are still phenomenal experiences occurring. So, at most this proves that more than one type of phenomenal experience can map to the same external behaviour. It doesn’t prove that one can have the same external behaviour but have no phenomenal experiences at all (and thus be a p-zombie).

      But I will say that I am confident that adding more code is NOT going to get us there, because code doesn’t seem to be the sort of thing that can produce phenomenal experiences.

      I’d agree with you as regards abstract “code”. But if that code is physically instantiated in the form of a pattern of physical stuff, then that is exactly the sort of thing that (it seems to me) produces phenomenal experiences. That seems to be the case because phenomenal experiences seem to exist only given those highly complex physical patterns called brains.

      Thus, if you think that adding more processors that DON’T produce experiences in the chess playing computer … but that somehow, if we add enough of them we will magically get there, …

      That’s exactly what I’m claiming, yes. The low-level units in themselves are not conscious. A limited array of a limited number of them is not conscious, but if you add enough of them together in the right sort of pattern then — yes — that generates consciousness. How else might this whole thing work (unless one wants to go for dualism?)?

      Second, given what you said, it is perfectly reasonable to think that some of those things and capabilities are beyond the capability of a machine.

      Is it? (If we interpret that as “beyond the capability of any conceivable machine”.)

      It’s especially reasonable to think that those things are more than simply symbolic processing — which is what Searle was talking about in the Chinese Room …

      But I don’t accept that there is any such thing as “simply symbolic processing” that has no semantic content.

      ESPECIALLY since we note that when WE do simple symbolic processing we don’t have the consciousness stuff that we have in the conscious deliberation.

      Sorry, this one I don’t understand. If we were doing conscious symbolic processing we would indeed be conscious of it. Of course our brains are also doing a lot that we are not conscious of. As I see it, consciousness is an extra layer of processing and monitoring and decision-making that might or might not accompany and affect other processes.

      Your suggestion, then, gives the game away to your opposition; you admit that they may well be right, and that you don’t really know that they’re wrong.

      Yes, granted, since none of us actually understand consciousness. But in the same way you can’t know that I’m wrong about this.

    36. verbosestoic

      What is it about our experiences that you think is inconsistent with my model of how this works?

      I’ve been doing that throughout the comment thread, and your response has been to dismiss that. It’s a little late now to be asking me what it was I said throughout the entire comment thread [grin]. Although it will come up a bit later in this comment.

      If one take a particle-by-particle account of gas particles in a metal box, the concepts “entropy” and “pressure” would not be part of that particle-by-particle account, but they would be fully entailed by that particle-by-particle account. As I see it, the conscious experiences are not something separate from the neural-firings account, they are entailed by it.

      But we can show how to reduce entropy and pressure TO the specific behavior of the particles, and so in some sense how we can DEFINE those things in terms of the particles. You can’t do that for neurons or neural patterns. At all. This, at least for now, is more than merely something that we can figure out later; again, experiences and neural patterns really do look like ENTIRELY different things, and no such reduction or definition seems forthcoming.

      If you actually have an interpretation that could even possibly accomplish this task, I’m all ears. But you haven’t presented anything even close yet.

      You are entitled to postulate a totally identical neural-firing account that does not have conscious experiences, but I am entitled to deny that possibility.

      I’m not postulating it. I’m arguing for it. You aren’t free to deny the conclusions of arguments that you never address.

      Look, let me take a step back here and talk more about the neurological account. By the biology — or, at least, the biology required to be a threat to dualism, and remember you asserted that the biology refuted or at least called into question dualism — outside of the neurons themselves we have light hitting the retinas — say — and at the end of the chain nerves getting activated to produce the behaviour. As long as those nerves get activated in the right way as a response to the light hitting the retina, the right behaviour will be produced. Thus, by the biology, the main purpose of the brain is to translate that retina activation into the right nerve activation points, but even if there was no brain there at all with nothing like the right neural patterns, as long as the electrical charge flowed, the right behaviour would be produced.

      So, now we can look at the brain. And from the perspective of the electrical pattern, all the neurons do is charge up and then activate the next one down the line, according to the pattern established by neurons that activate together. The challenge this presents to dualism is that this looks to be causally closed; if we could map out all of these electrical connections, we’d know exactly how everything activates and have no need for any other causal factors. So all we need to do is create the right electrical pattern, and we can activate the nerves in the right way and produce that behaviour.

      But NONE of this relies on particular neural patterns necessarily. Again, as long as the electricity flows the right way, the behaviour will be produced.

      At the same time, the neurons have to be producing the phenomenal experiences somehow. But there’s no room for that process to impact the causal chain because according to our own best biological science, only the electrical activations play a causal role in the story, and there is no room for anything else to play a part. So, at best, when a neuron or neural pattern activates it PRODUCES an experience AND generates the right pattern to produce the behaviour. But this makes phenomenal experiences epiphenomenal. But there is no room for any other causal interaction or else dualism is brought back into play.

      So that’s why I argue that the sort of situation I argue for is possible, because there is no reason to conclude that ANYTHING that can produce that electrical pattern must produce phenomenal experiences or give us any indication that it does, and that electrical pattern is in and of itself sufficient to produce the behaviour.

      But neither of us can produce an actual account of how experiences are generated. So I’m no worse off than you on that point.

      But YOU’RE arguing that the neural account entails the phenomenal experiences and that you can reduce one to the other. I’m not. So you are missing the very heart of your explanation, and the explanation that you need in order to make your claim that my thought experiment doesn’t work, and that it isn’t possible. You don’t have anything similar against my account, as far as I can see, but if you think you do please outline it so that I can address it.

      In causal terms, whichever physical properties are necessary for the experiences are the ones causing the experiences; they can also physically influence the flows of electric charge along neurons. Thus there is no “overdetermination” problem here.

      Except that we have no evidence of any missing physical properties of neurons that we need there, and no evidence of any other properties AT ALL having any causal impact. Or, rather, if we DID have the latter then dualism would be the preferred theory, given that we don’t see any physical properties of neurons or neural patterns that we are in theory missing and so no evidence that the extra causation would be internal to the brain and not external to it. So, if you take this line, you actually make dualism MORE credible, and the only reason you are taking this line is to preserve the idea that phenomenal experiences cause behaviour in an interesting way. The standard materialist theory doesn’t in any way assert that, and in fact asserts the opposite. Thus, in some sense, you are a mind-brain dualist when compared to straight materialists, which should not make you comfortable [grin].

      Well we do: phenomenal experiences seem to be strongly correlated with the highly patterned objects we call brains.

      There are many other highly patterned objects that don’t seem to do phenomenal experiences, and to argue about the pattern you’d have to show that it isn’t a property of neurons or the brain itself that’s doing the work, not the pattern itself. Thus, the only reason you appeal to the pattern is that you need to preserve the ability for things that aren’t brains to produce phenomenal experiences, but can’t show anything about neurons themselves that means that they should produce phenomenal experiences. Dualism avoids this by putting the phenomenal experiences into another object that hooks up causally to the brain, and so has two-way causal interaction. If non-brain things can hook up to those sorts of objects, then they too can have phenomenal experiences; if only brains can do that, then that’s a property of the mind, not a property of the brain (necessarily). Those options aren’t open to you, and so you argue for the pattern for no other reason that to avoid the consequences of my thought experiment … but you have no real reason to do that beyond that desire to avoid the consequences of my thought experiment. And, of course, it still doesn’t even solve your problem.

      But, in the case of the actor, there are still phenomenal experiences occurring. So, at most this proves that more than one type of phenomenal experience can map to the same external behaviour.

      Except that the phenomenal experience isn’t actually there AT ALL. You can’t map the lack of the phenomenal experience to the behaviour and claim that that’s a valid mapping. So what this thought experiment does is show that what it means to be a certain type of phenomenal experience is NOT the behaviour it produces, because we can clearly conceptually have the experience without the behaviour and have the behaviour without the experience. That’s all I need the actor example to do. So you’d need to show me how I’m wrong there WITHOUT assuming your conclusion and just defining that to be the case.

      I’d agree with you as regards abstract “code”. But if that code is physically instantiated in the form of a pattern of physical stuff, then that is exactly the sort of thing that (it seems to me) produces phenomenal experiences.

      1) Abstract code and its physical instantiations are all we have.

      2) There are lots of patterns in the world that don’t produce phenomenal experiences, and we’ve also simulated the same patterns of the brain and haven’t really had a lot of success in any way at reproducing anything that looks like consciousness. So this seems … unevidenced, to say the least.

      But I don’t accept that there is any such thing as “simply symbolic processing” that has no semantic content.

      Unless you think that computers are conscious now — and you don’t — we have a system that does that: most computers. Heck, do you think that a pocket calculator does semantic processing? Yet it definitely does symbolic processing, as it processes the 2 and + and = symbols to get 4. Do you think it reasonable to say that a pocket calculator really understands math?

      Sorry, this one I don’t understand. If we were doing conscious symbolic processing we would indeed be conscious of it.

      We’re aware that we’re processing symbols, but not necessarily aware of what they MEAN. It’s understanding meaning that is the key to semantic processing, and what distinguishes it from symbolic processing. I can process symbols that I don’t understand like the person in the Chinese Room, and that is clearly different from processing symbols that I DO understand. Thus, there’s a difference, and we can compare methods to see which ones impart understanding and which ones don’t.

      Yes, granted, since none of us actually understand consciousness. But in the same way you can’t know that I’m wrong about this.

      Look back at the discussion chain. You argued that chess computers don’t do everything we do, and so that one, at least, doesn’t have consciousness like we have it. Thus, you grant that machines DON’T HAVE to have consciousness like we do, and so need an argument to suggest that they can EVER get that. And your best options aren’t going to work, because external behaviour and phenomenal experiences don’t always map — as you conceded yourself — and if you retreat to the brain since machines don’t have that and may not instantiate that pattern to produce the right behaviour, you might eliminate any chance of machines being conscious at all. This is not a good situation for you to be in, methinks [grin].

    37. Coel Post author

      But we can show how to reduce entropy and pressure TO the specific behavior of the particles, and so in some sense how we can DEFINE those things in terms of the particles. You can’t do that for neurons or neural patterns. At all.

      I agree that we can’t do it. But that does not mean that it can’t be done, it only means that we don’t know how to do it.

      … experiences and neural patterns really do look like ENTIRELY different things, and no such reduction or definition seems forthcoming.

      A neural network of 100,000,000,000,000 connections is so vastly beyond our competence to analyse and understand that we would not expect to be able to give an answer. Thus the absence of an answer is not a problem for the stance I’m taking.

      As long as those nerves get activated in the right way as a response to the light hitting the retina, the right behaviour will be produced.

      Yes.

      And from the perspective of the electrical pattern, all the neurons do is charge up and then activate the next one down the line, according to the pattern established by neurons that activate together.

      Yes.

      But NONE of this relies on particular neural patterns necessarily. Again, as long as the electricity flows the right way, the behaviour will be produced.

      But only the right neural patterns would produce the right pattern of electricity flowing. There may indeed be multiple different right neural patterns that would produce the same behaviour, but it does need to be one of those patterns.

      So, at best, when a neuron or neural pattern activates it PRODUCES an experience AND generates the right pattern to produce the behaviour.

      Yes.

      … because there is no reason to conclude that ANYTHING that can produce that electrical pattern must produce phenomenal experiences or give us any indication that it does, …

      Here I disagree. Since only a very particular set of neural patterns would produce the right electricity flow, I can assert that all members of that set of patterns would produce phenomenal experiences.

      You don’t have anything similar against my account, as far as I can see, but if you think you do please outline it so that I can address it.

      You have given no account of what phenomenal experiences are or of how they are generated. Thus, even if there are things about my account that I cannot answer (which I fully admit to), that does not mean I’m worse off than you.

      Except that we have no evidence of any missing physical properties of neurons that we need there, and no evidence of any other properties AT ALL having any causal impact.

      I’m not asking for anything extra. I’m saying that the neural-network pattern does generate phenomenal experiences — in a similar way that, analogously, the behaviour of gas particles generates the properties “pressure” and “entropy”.

      There are many other highly patterned objects that don’t seem to do phenomenal experiences, …

      Agreed.

      … and to argue about the pattern you’d have to show that it isn’t a property of neurons or the brain itself that’s doing the work, not the pattern itself.

      I’m arguing that it is the particular pattern of neurons, not any old pattern of neurons.

      but can’t show anything about neurons themselves that means that they should produce phenomenal experiences.

      Correct. It may be that a very similar pattern implemented in silicon might also be conscious.

      Except that the phenomenal experience isn’t actually there AT ALL.

      But, in the case of the actor, other phenomenal experiences are there. All I need to allow is that more than one phenomenal experience can map to the same behaviour.

      2) There are lots of patterns in the world that don’t produce phenomenal experiences, …

      Sure, but so what? There are lots of material patterns that are not plant-like life forms. That doesn’t negate the claim that plants *are* particular material patterns.

      … and we’ve also simulated the same patterns of the brain and haven’t really had a lot of success in any way at reproducing anything that looks like consciousness

      No, not at all, we are vast numbers of orders of magnitude away from simulating a brain, on anything like the scale of the human brain.

      … Unless you think that computers are conscious now — and you don’t — we have a system that does that: most computers. Heck, do you think that a pocket calculator does semantic processing?

      I don’t accept that consciousness is required for semantic processing. Yes I do think that pocket calculators are doing semantic processing (in a rudimentary and low-level way).

      [Of course you could define “semantic processing” in a way that requires self-reflective consciousness, but in that case we’re using the term “semantics” differently.]

      Do you think it reasonable to say that a pocket calculator really understands math?

      In a very limited degree, yes. But, essential to that answer is the fact that “understanding” would be a continuum, not a binary on/off quantity, and thus that it is possible to have small amounts of it.

      Thus, you grant that machines DON’T HAVE to have consciousness like we do, …

      Yes.

      … and so need an argument to suggest that they can EVER get that.

      My argument is: we’re glorified machines; we’re conscious; ergo …

      And your best options aren’t going to work, because external behaviour and phenomenal experiences don’t always map — as you conceded yourself …

      I conceded only that they don’t have a one-to-one mapping. But I don’t see that I *need* a one-to-one mapping. A many-to-one mapping from experiences to behaviour would work fine.

  5. richardwein

    Coel wrote:

    If one wants to question something fundamental, for example the basic logical axiom modus ponens, then all one has to do is produce an alternative model including some negations of modus ponens, and then ask which version of the model best reproduces and predicts features of the empirical world.

    Have you actually imagined what would be involved in adopting the negation of modus ponens? I don’t think we could do it, even on a trial basis. Modus ponens is not just a statement. It’s a cognitive habit, an inclination to proceed in a certain way. For example:

    A: Do we need anything from the shops?
    B: If your friend is coming to dinner we’ll need some wine.
    A goes to the shops and buys wine.

    A’s response to B’s conditional statement is in accordance with modus ponens, even though A has probably not said to himself any such thing as “If my friend is coming to dinner we’ll need some wine. My friend is coming to dinner. Therefore we’ll need some wine.” Still less likely is it that A will have recalled some generalised statement of modus ponens, of the sort you might call an “axiom”.

    Similarly, scientists don’t normally refer to an axiom of modus ponens; they don’t include any such axiom in their models. They simply respond to conditional statements in the normal, appropriate way. Adopting the negation of modus ponens would mean adopting the contrary habit, which would be a habit of drawing conclusions that are inconsistent with prior beliefs and evidence. I don’t think they could consistently do that if they tried. Nor is there any point in trying. It’s simply obvious that that way lies madness. To question modus ponens is to so fundamentally question our cognitive capabilities as to be self-defeating. Asking for a justification for modus ponens is like asking for a justification for induction, which leads to the misguided “problem of induction”.

    None of this is to say that we need metaphysics to justify or “ground” modus ponens. Formal statements of modus ponens just reflect our existing cognitive habits. We’ve developed those habits because they’ve been effective in the past. In that sense, you could say that they’re empirically grounded. The habit of modus ponens is part of the habit of using conditional language. It’s not something invented by philosophers.

    Reply
    1. Coel Post author

      Hi Richard,

      Have you actually imagined what would be involved in adopting the negation of modus ponens?

      I imagine that it would make models nonsensical, with inconsistent or gibberish outputs. For example, if we tried to predict future eclipse times using a model with neg-MP, then the model would likely output nonsense, or predict any and all eclipse times, or state that an eclipse will both happen and not happen at a given time, or something equally useless.

      The one thing it would not do is produce a clear prediction of an eclipse time, with no eclipses at other times. Thus the model would fail badly, in explanatory and predictive power, in comparison with a model invoking MP.

      To question modus ponens is to so fundamentally question our cognitive capabilities as to be self-defeating.

      Yes, in the sense that MP is one of the things that our brains have evolved to use, in order for those brains to be in any way useful. But that still means that MP is adopted in order to model the world, which is effectively what I was claiming.

      Asking for a justification for modus ponens is like asking for a justification for induction, which leads to the misguided “problem of induction”.

      My attempt at that one is here. 🙂

  6. richardwein

    Verbosestoic wrote:

    Metaphysics is the study of reality as it really is and must be…

    Science and everyday observation tell us how the world is, and adding in the word “really” doesn’t make any difference. I recognise that what science and everyday observation give us are models, or abstractions if you like. But there’s nothing more to give. You can’t give me the world; you can only give me models of it.

    As for “must be”, that presupposes that in some sense the world couldn’t have been any other way. Metaphysicists might aspire to discover such supposed facts, but it’s a bit rich to say that metaphysics is the study of such things, when it has yet to show that it is successfully studying anything at all. It’s like saying that astrology is the study of astrological relationships. It may aspire to be that, but aspiration doesn’t make it so. Get back to me when you’ve made some progress. I won’t be holding my breath, since all the metaphysics I’ve looked at has been conceptually confused emptiness. Quite a few philosophers have thought that too.

    Reply
    1. richardwein

      P.S. You can’t get any facts about the world without using some evidence from the world. The only other thing you can get is purely abstract relationships, as in mathematics. I think this is basically the same dichotomy that Hume invoked.

      One kind of facts about the world are facts about our use of language. One thing philosophers can do is study how we use language. In response to a “metaphysical” question, like “What is time?”, we can study our discourse using the word “time”, and see what function this discourse serves. Of course, that’s not the goal that traditional metaphysicists have set themselves. They think there are special “metaphysical facts” to be discovered. But their concept of a metaphysical fact is incoherent.

    2. verbosestoic

      Science and everyday observation tell us how the world is …

      See, this is a metaphysical position, and it might be true … but it might not be. So how are you going to justify that position? Any attempt to do so will be metaphysics, and it’s not like this position isn’t one that metaphysics has tried to adopt, usually hopefully. They WANTED the answer to be this simple, but it turns out that it’s not, for both philosophical and empirical reasons.

      As for “must be”, that presupposes that in some sense the world couldn’t have been any other way.

      You’re reading too much into that. What it means is to get to what reality really is like independently of what we think about it, and our own perceptions of it … at least as far as we possibly can.

      Metaphysicists might aspire to discover such supposed facts, but it’s a bit rich to say that metaphysics is the study of such things, when it has yet to show that it is successfully studying anything at all. It’s like saying that astrology is the study of astrological relationships. It may aspire to be that, but aspiration doesn’t make it so.

      Your example relies on, essentially, relying on the field of astrology being nothing more than those who think that there ARE those relations, which doesn’t work for metaphysics, because metaphysics accepts as possible answers both “What our senses are telling us about the world” AND “We have no way to actually tell”. The thing is, the first answer has a number of challenges to overcome and no one has been able to prove that we should just give up trying yet. Thus, there may indeed either be no need to study anything or there may be nothing there to study, but it cannot be denied that what metaphysics is TRYING to study is that, and that that is an interesting philosophical question, regardless of whether it has any direct practical implications.

      Get back to me when you’ve made some progress. I won’t be holding my breath, since all the metaphysics I’ve looked at has been conceptually confused emptiness. Quite a few philosophers have thought that too.

      Two points:

      1) There’s a lot of muddled crap in science, too, and in, say, things like psychology, but that doesn’t mean that we can dismiss the entire field.

      2) Are you sure that the problem with the metaphysics you’ve looked at isn’t more with you than with the metaphysics? I can’t really say without knowing what you HAVE looked at, but I do think there is reasonable metaphysics that just doesn’t quite get as far as it needs to.

      P.S. You can’t get any facts about the world without using some evidence from the world.

      And your proof of that is … ? You can start by telling me how you define “facts about the world” and “some evidence from the world”, and also how much evidence counts as “some”.

      n response to a “metaphysical” question, like “What is time?”, we can study our discourse using the word “time”, and see what function this discourse serves.

      That’s potentially a useful study — although it may be more relevant to linguistics than metaphysics — but it won’t tell us what the concept “time” really is or has to be. In short, it won’t give us the conceptual properties, which are the things that would allow us to tell whether time really exists in this world or not, since people can, obviously, conceptualize incorrectly, and trying to conceptualize to what seems to be the most useful runs into the issue that just because an answer is useful doesn’t mean it’s true, as well as issues about whether a different concept is more useful in different contexts or not.

      Of course, that’s not the goal that traditional metaphysicists have set themselves. They think there are special “metaphysical facts” to be discovered. But their concept of a metaphysical fact is incoherent.

      Can you outline what you think the concept of a metaphysical fact is, so that I can see if a) it is what most metaphysicians rely on and b) that it is incoherent?

  7. richardwein

    Can you outline what you think the concept of a metaphysical fact is, so that I can see if a) it is what most metaphysicians rely on and b) that it is incoherent?

    I refer you to your own concept of metaphysical facts. Namely, facts about how things “really” are, on top of the familiar ordinary facts about how things are. There’s no difference between saying how things are (like saying the earth is round) and saying how things “really” are. The earth really is round (approximately). Do you deny that the earth is round?

    Reply
    1. verbosestoic

      That specific question isn’t a question of metaphysics, nor something that metaphysics in and of itself is concerned with. The idea wrt metaphysics is to note that our experiences and science tell us that the Earth is round … but how can we know if it REALLY is? In short, how do we justify that what that tells us reflects reality at all? Thus, it ties into questions like “How could we know whether or not we’re in the Matrix?” and “How do we know that we’re not dreaming right now?”. It’s essentially about justifying claims about reality itself. It doesn’t deny that relying on our experiences works or anything like that, but just asks if we can know that our perceptions are giving us more than appearances.

      As I said above to Coel, you can decide that since acting on the appearances works, you don’t really care about justifying that they are more than mere appearances. But if you do that, you don’t get to say that people who do care about that are wrong, or make claims that make metaphysical claims.

    2. Coel Post author

      Hi verbosestoic,

      As I said above to Coel, you can decide that since acting on the appearances works, you don’t really care about justifying that they are more than mere appearances.

      What science tries to do is make sense of our experiences. The aim is to produce a model that provides the best explanatory and predictive power possible about those experiences. We can demonstrate that science can do this, because we can make predictions about experiences that then come true.

      Your question is whether such experiences are just part of a meta-reality that has some other nature. Well, yes, they might well be. And, if that meta-reality produces no empirical evidence about itself in our experiences, then indeed science would have no way of knowing about that meta-reality.

      In that sense you are entirely right, there could be some meta-reality to which we are entirely oblivious. Scientism does not deny that! What scientism does say is that, if science cannot tell us about that meta-reality, owing to the lack of empirical evidence, then nor can any other human method.

      Thus, science would be useless about this un-evidenced meta-reality. But so would metaphysics! And so would theology! And so would tarot cards. Et cetera. Metaphysicians and theologians are fond of the conceit that, where science cannot provide answers, they can. Well, there is absolutely no reason to suppose that metaphysics can do that. Yes, one can scheme up scenarios and make wild guesses, but in terms of supplying actually reliable answers about such meta-realities, metaphysics simply can’t to it.

      That last statement, by the way, is not in itself metaphysics! The evidence for that statement, rather, is scientific. Namely, we conclude that metaphysics is useless at supplying reliable answers about such meta-realities because, as an empirical fact, it has never done so.

      That claim can be refuted by putting forward reliably-known statements about such meta-realities that have come from metaphysics, and have not or could not have come from science.

    3. verbosestoic

      What science tries to do is make sense of our experiences. The aim is to produce a model that provides the best explanatory and predictive power possible about those experiences. We can demonstrate that science can do this, because we can make predictions about experiences that then come true.

      Great. We’ve gotten to Kant at this point: science is about the world of experiences (he called it appearances), works for that, and produces true statements about that. And as I said, if that’s all you want, then you can stop there and no one will think any less of you. But then note that, by definition, any question OUTSIDE of that is not a scientific question. Thus, ANY question about the noumenal is not a scientific question, which INCLUDES whether or not there is one, or if even if there was one we could know anything about it, or whatever. Note that this means, if you’re going to be consistent with your own definition, that even IF we can demonstrate things about the noumenal world with empirical data, it still wouldn’t be SCIENCE. And the question of whether or not we can do that empirically is in itself a question that is outside of science.

      For convenience, let’s call the field that tries to figure out things about the noumenal “metaphysics”. Then these are all, by definition, metaphysical questions, no matter what method we try to use to to figure out these things. Thus, we could even do a NATURALIZED metaphysics — where we try to apply the scientific method to these questions — and we’d STILL be doing metaphysics and not science. Starting from your own, accepted definition.

      In that sense you are entirely right, there could be some meta-reality to which we are entirely oblivious. Scientism does not deny that! What scientism does say is that, if science cannot tell us about that meta-reality, owing to the lack of empirical evidence, then nor can any other human method.
      Thus, science would be useless about this un-evidenced meta-reality. But so would metaphysics! And so would theology! And so would tarot cards. Et cetera.

      But why would science being useless to discover things about phenomena that it explicitly, by definition, is not structured to study mean that nothing else could possibly do that? Either you’re saying that we could never know anything outside of science — which is an epistemological question, and one that’s almost certainly wrong unless you broaden the definition of “science” to such a degree that it becomes meaningless — or you have to insist that the question it raises simply cannot be answered. At which point, I’d have to ask you to at least demonstrate how you know that. If you don’t know that, then I really don’t care about your statement here, and if you do know it, then you have to be able to justify it. And this attempt doesn’t do it:

      Namely, we conclude that metaphysics is useless at supplying reliable answers about such meta-realities because, as an empirical fact, it has never done so.

      While I think you underestimate the progress that philosophy has made in a number of fields without using science, even accepting that it’s not enough to justify a claim that you know the question cannot be answered. Just because we don’t have an answer YET doesn’t mean that the question can’t be answered, and that’s the proposition that you need to justify. The most you can try to do is appeal to success rates for science vs the other options … but science hasn’t been able to answer these questions EITHER and, by your own definition CAN’T even answer them. Thus, at most you are expressing your own personal beliefs, but I fail to see how you get to knowledge for that, and fail to see how a reply of “These are really hard questions” is not only acceptable, but the more reasonable belief to hold.

      Again, if you don’t want to do metaphysics, don’t. But as soon as you try to claim that metaphysical questions can’t be answered, you are stepping onto the metaphysical playing field, and so you do have to in at least some way play by its rules.

    4. Coel Post author

      Hi verbose stoic,

      science is about the world of experiences (he called it appearances), works for that, and produces true statements about that.

      It phrase it that science is an *attempt* to understand and know about everything, starting with our experiences. It may then be that there are things that science cannot succeed with, such as the “noumenal”, but such a limitation is not a pre-condition of science, but would rather be a conclusion.

      But then note that, by definition, any question OUTSIDE of that is not a scientific question.

      But I do not define science that way, I don’t place anything outside science a priori. If something turns out to be outside science then that would be a conclusion. To me the idea that we can know about things only from empirical evidence is a conclusion of science, not a prior commitment.

      And the question of whether or not we can do that empirically is in itself a question that is outside of science.

      To me, all such questions are part of science.

      … IF we can demonstrate things about the noumenal world with empirical data, it still wouldn’t be SCIENCE.

      If we could get empirical data about the noumenal world, then, since the empirical data would be part of our world of experiences, then studying that data and its implications would indeed be science.

      But why would science being useless to discover things about phenomena that it explicitly, by definition, is not structured to study mean that nothing else could possibly do that?

      It wouldn’t. We do not leap from “science cannot do it” to “therefore nothing else can do it”. Rather, we start with “has anything else told us about such things, or given us good reason to suppose that it can tell us about such things?”. The answer to that is “no”. From that we reach the *provisional* conclusion that these other things cannot do it.

      It is basically a burden-of-proof assertion, that if metaphysics wants to claim that it supplies validated knowledge of the noumenal, then it is welcome to make the case that it has done so. Absent that, we reject the claim.

      But as soon as you try to claim that metaphysical questions can’t be answered, …

      As a *provisional* conclusion, as best we can tell, metaphysics cannot answer such questions. This can be overturned by someone demonstrating that metaphysics can answer such questions.

    5. verbosestoic

      It phrase it that science is an *attempt* to understand and know about everything, starting with our experiences. It may then be that there are things that science cannot succeed with, such as the “noumenal”, but such a limitation is not a pre-condition of science, but would rather be a conclusion.

      At which point you’ve included philosophy in science — although philosophy might be better described as attempting to answer all questions that can be answered — which means that either you are including all of the philosophical and metaphysical questions and arguments in science — making them as scientific as your claims — or are invalidly dropping them without sufficient justification. Additionally, this ignores the history of science, which started as a subset of philosophy — natural philosophy — and was split off to work on empirical examinations as those questions were of less interest to philosophy as a whole. So philosophy DEFINITELY gets first claim on that definition, and if you include “by any means” you have to address the idea that we cannot therefore give ANY preference to empirical evidence or methods when answering the questions.

      If we could get empirical data about the noumenal world, then, since the empirical data would be part of our world of experiences, then studying that data and its implications would indeed be science.

      This presumes that using ANY empirical data makes something science, which I strongly disagree with. What is your justification for saying that while at the same time not making science critically empirical?

      It is basically a burden-of-proof assertion, that if metaphysics wants to claim that it supplies validated knowledge of the noumenal, then it is welcome to make the case that it has done so. Absent that, we reject the claim.

      Which is a clear case of the inductive fallacy, and also of shifting the burden of proof. What you have to realize is that my reply, at least, is not of the form “Metaphysics can produce knowledge about reality”. My only claim is that if anything can do it, metaphysics can, because a) it is the field dedicated to doing so and b) it isn’t limited to any specific method, meaning that it can use science, empirical data, pure logic, or anything else that might get it there. Given that, I think I can make a strong knowledge claim for my statement. Your claim, on the other hand, doesn’t seem so well-supported. The fact that metaphysics hasn’t come up with an answer yet does not mean that it never will, any more than the fact that you might never have seen a black swan doesn’t mean that they don’t exist. You need further arguments if you want me to accept your claim … and if you want to claim it as knowledge, then you pretty much have to want that, or else we can ask how it is that you know that that’s true when you can’t present me with your evidence and have me thus ALSO know it. And calling it “provisional” doesn’t do as much work here as you seem to think it does, because I’m not demanding certainty or, I dare say, an excessive amount of justification for your claim. I’m merely asking that you not base it on logical fallacies and on things that apply in your worldview but not necessarily on anyone else’s. That seems pretty reasonable to me [grin].

    6. Coel Post author

      What you have to realize is that my reply, at least, is not of the form “Metaphysics can produce knowledge about reality”. My only claim is that if anything can do it, metaphysics can, because a) it is the field dedicated to doing so and b) it isn’t limited to any specific method, meaning that it can use science, empirical data, pure logic, or anything else that might get it there.

      OK, if you’re including any and all methods in “metaphysics” then, agreed, if anything can, metaphysics can. However, most people define “metaphysics” as something distinct from “science” and not including science. From that point of view I deny that metaphysics can do anything useful in a way that is distinct from science.

    7. verbosestoic

      OK, if you’re including any and all methods in “metaphysics” then, agreed, if anything can, metaphysics can.

      You also have to include the “And that’s the goal of the field” in there, because, say, philosophical moral studies also can use any possible method, but are not likely to solve metaphysical problems because they aren’t trying to solve them.

      However, most people define “metaphysics” as something distinct from “science” and not including science.

      I’ve commented on this before. While metaphysics is not science, that doesn’t mean that it a priori says that it can’t use scientific results or even methods, just as metaphysics is also not epistemology but it can and indeed relies on epistemology for a number of its issues. The field as a whole does not reject science a priori. As with any philosophical field, there have been a number of attempts to use either science directly or the methods of science to answer those questions. When most metaphysicians say that science doesn’t work for those questions, it’s not a rejection by definition, but simply a recognition that we’ve tried and it hasn’t worked at all.

      Metaphysics includes more than science, and the usefulness of science to the endeavour has been argued over.

      From that point of view I deny that metaphysics can do anything useful in a way that is distinct from science.

      I think you need to clarify the point of view you’re referring to and how it gets to this conclusion.

    8. Coel Post author

      I think you need to clarify the point of view you’re referring to and how it gets to this conclusion.

      My view is that: we can only learn things through empirical evidence. As far as we can tell, there are no other ways of learning things.

      We can and do, of course, build models resulting from empirical evidence, and we can examine those models to learn something, but we can only validate those models using empirical evidence and so, effectively, this is still empirically based enquiry.

    9. verbosestoic

      I think you conflate “justification” and “learning” here. As I’ve said previously, we may learn mathematical terms empirically — ie by people talking to us — but that’s not how they’re justified. Here you start with learning, move to validation, and ignore that your opponents are merely saying that some truths aren’t validated using empirical evidence, and can’t be. That all truths can only be validated empirically, for example, is one of those statements that you can’t justify that way.

    10. Coel Post author

      As I’ve said previously, we may learn mathematical terms empirically — ie by people talking to us — but that’s not how they’re justified.

      Then what is your method of justifying mathematics? If you want to claim that one reasons it from axioms, what is the proof that the reasoning is correctly done and that the whole system is self-consistent?

      That all truths can only be validated empirically, for example, is one of those statements that you can’t justify that way.

      I disagree. The justification for the idea that only empirical evidence can validates truths is that, as a matter of empirical fact, there are no truths that we know to be validated, for which the validation was done by non-empirical means.

    11. verbosestoic

      If you want to claim that one reasons it from axioms, what is the proof that the reasoning is correctly done and that the whole system is self-consistent?

      Okay, I’m a bit confused by this, so forgive me if I get things wrong or am inadvertently insulting.

      First, we don’t need absolutely indubitable knowledge, so the possibility that someone might have made a mistake in their derivation doesn’t mean that we don’t know it and that it isn’t justified. Otherwise, science doesn’t produce knowledge either.

      Second, what we normally do is check it over following the rules of logic — and, in mathematics, those can actually BE non-standard logics — and get others to do so as well.

      From this, either you don’t know what the standard way to check reasoning from axioms is — which I find unlikely — or you are demanding perfect knowledge — which has a bad consequence for science — or else you’re trying to claim that that method is really empirical. If you do the latter, then you’re pulling the standard scientismist fallacy of defining a term so broadly that it encompasses all possible things that are relevant, which works to some extent but doesn’t actually settle the argument. If doing this is empirical, then everything is empirical … including that metaphysics you deride. That method becomes just as empirical as anything else … and therefore just as valid. And then we still have to decide what will work for the relevant questions, and what actually justifies it.

      In short, if you don’t use “empirical” in the same way as people who claim that mathematics is not justified empirically use it, you aren’t actually debating them when you say they’re wrong.

      The justification for the idea that only empirical evidence can validates truths is that, as a matter of empirical fact, there are no truths that we know to be validated, for which the validation was done by non-empirical means.

      1) I say this is false for mathematical and conceptual truths.

      2) This is also the inductive fallacy. Which is not to say that induction can’t produce knowledge, but that a claim of “I have never seen an instance of X, so I know that we can never have an instance of X” is NEVER justified. This holds even if we aren’t pushing for indubitable knowledge, as you can never say “I have not seen X” and use that to get to “I know X cannot exist”. It’s not a matter of doubt, but a matter of logic.

      Science itself does not use induction that way. It simply says that if X is true in this area, I can assume that it is also true in all areas unless I have good reason to think it isn’t. So, rather, “I have seen X here, therefore I know a) that X exists here and b) X exists in all relevantly similar areas”. Then it builds rules that build in the conditions of that existence, updated when it finds out what those relevantly similar areas are.

      So, no, your justification doesn’t hold. I submit that you do not know that, and cannot justify it the way you claim you can justify it.

    12. Coel Post author

      Two main replies:

      First, on reasoning from axioms (aka mathematics). Yes, we do indeed arrive at truths by mathematical methods within the system of maths, without any empirical input in that process, once we have that system. However, my argument is that “the system” is itself a product of empiricism. The axioms of maths, the rules of logic, the process of reasoning, are all the results of attempting to model the empirical world. Thus, if we *then* reason within the system, that knowledge is still, ultimately, the product of empirical enquiry.

      Let’s make a comparison. Suppose we want to know the temperature at the centre of the Earth. We have no direct way of measuring it. All we can do is turn to a theoretical model of how the Earth works, and use that to calculate the temperature at the centre of the Earth. That process is very akin to maths. But, in order for it to work, the model we use has to itself have been the product of empirical enquiry.

      Further, the same applies to *all* empirical evidence, since empirical evidence is *always* entwined with theoretical models.

      Thus, when it comes down to it, physics and maths are basically the same. In physics we often do calculate from theory to find out things we want to know, but the overall models are themselves empirical. There is no way of making a clean distinction between physics and maths.

      … that a claim of “I have never seen an instance of X, so I know that we can never have an instance of X” is NEVER justified.

      If often is justified to quite an extent, it is just never *completely* and infallibly justified. Suppose we take the claim “pigs do not fly”. Ultimately, the justification for that is the fact that we’ve never seen any pig flying. Logicians will then complain about induction and point out that it doesn’t prevent us seeing a flying pig tomorrow. They are right, but the extrapolation of empirical evidence to the idea that pigs do not fly is pretty good.

      My claim that no non-empirical methods lead to validated truths is of the same sort — it’s simply that we’ve never encountered such non-empirically-validated truths.

    13. verbosestoic

      Let me start here first:

      There is no way of making a clean distinction between physics and maths.

      Sure there is: we can point to mathematical and logical systems that do not apply to this physical world and almost certainly will NEVER apply to this physical world. These are still perfectly valid logics and mathematics, and still perfectly justified for the purposes of logic and mathematics. In short, we KNOW them in the relevant sense for mathematics and logic, even though we can’t have empirical evidence to justify them. As another example, I can know what it means for something to be a vampire even though, empirically, they don’t exist, and I don’t even need to justify it historically in a sense of “This is what everyone says they are”, and if we discovered a basis for the legends I don’t need to say that vampires “really are” whatever that basis is, but can instead say that those things weren’t really vampires.

      So the problem with your empirical evidence above is that it’s answering a different question. Let me use the example of Euclidean geometry here to make this clear. There are two questions we can ask here:

      1) What does it mean to be Euclidean geometry?

      2) Is Euclidean geometry a good way to model the empirical world?

      The first question cannot be settled by empirical investigation, but we can know it with the method I outlined and you accepted, and we need to be able to answer it before we can answer the second question. The second question is what we use your empirical investigation for, obviously, but is just ABOUT that link to the empirical world. But I can imagine many, many geometries where I can answer 1) and know that answer to be true and yet have to answer 2) as “No”, or even as “I don’t know”. Thus, you seem here to be insisting that we can’t justify an answer to 1) without having an answer to 2), which seems false.

      Also, that we started these sorts of conceptions in an attempt to find useful models for the empirical world does not mean that that is what they really are at the heart of it all, and as I’ve outlined it seems that they HAVE gotten more abstract from there. Whether you find those abstract ruminations useful or not is up to you, but that wouldn’t mean you get to insist that the only things useful to those fields are those that interest you (and coincidentally are empirical [grin]).

      If often is justified to quite an extent, it is just never *completely* and infallibly justified.

      I’m claiming that it is insufficiently justified to count as knowledge, because it relies on an explicit logical fallacy. I am not demanding certainty; I am demanding that you do more than simply say “I haven’t seen it” when I ask you to justify how you know something is true.

      Suppose we take the claim “pigs do not fly”. Ultimately, the justification for that is the fact that we’ve never seen any pig flying. Logicians will then complain about induction and point out that it doesn’t prevent us seeing a flying pig tomorrow. They are right, but the extrapolation of empirical evidence to the idea that pigs do not fly is pretty good.

      Except this is precisely the reasoning used to conclude that there were no black swans, which is the typical example of the inductive fallacy.

      What you want to get at is this, I think, from the wiki:

      The opposite, Slothful induction, is the fallacy of denying the logical conclusion of an inductive argument, dismissing an effect as “just a coincidence” when it is very likely not to be.

      In short, you want to claim that I and others are engaging in THIS fallacy. My counter is that in order to make this claim, you have to give me a SEPARATE argument for why you think you’ve sampled enough cases and enough properly representative ones to justify an inductive knowledge claim. Otherwise, it seems to me that you are engaging in THIS fallacy:

      Hasty generalization is the fallacy of examining just one or very few examples or studying a single case, and generalizing that to be representative of the whole class of objects or phenomena.

      So you either need to argue that your experience is representative, or give another reason that says that this is unlikely to be the case, that you also know. Taking the pigs can fly case, if you just say that you haven’t seen one fly, that’s weak because your experiences might not be — and likely aren’t — representative enough. But if you counter that the genetic structure of pigs is such that the likely can’t develop any organs that could allow for flight, that’s a stronger argument and does give us a lower probability.

      If you’re tempted at this point to say that even if you aren’t really justified in generalizing from experience that way, if all of the pigs in your experience don’t fly it’s reasonable for you to ACT as if they don’t fly until you need to adjust that, we’d be in agreement. I’d just call that a belief and not knowledge, mostly because you couldn’t use the same reasoning to convince me that I had to come to know that pigs couldn’t fly if I doubted you, or had differing experiences.

      Also, all of these reasonings die out as soon as someone points at example at you or provides an argument about why pigs might be able to fly. Which is what this whole debate has been about, as people have been firing examples at you all over the place.

    14. Coel Post author

      Sure there is: we can point to mathematical and logical systems that do not apply to this physical world and almost certainly will NEVER apply to this physical world.

      But science is much the same. Physics *also* considers the conceptual implications of models. For example, cosmologists spend time considering alternative universes that could be constructed from the same basic model. This is doing much the same as maths, asking about “what if?” scenarios and implications of models.

      Whether the model one is working with is logical, mathematical or physical makes little difference, since all of them were ultimately arrived at as being models of the empirical world.

      There are two questions we can ask here: 1) What does it mean to be Euclidean geometry? … The first question cannot be settled by empirical investigation, …

      But you can ask exactly the same sorts of questions about science’s models. Thus you cannot distinguish between maths and science this way.

      For example, one can ask, “in such and such a conceptual universe, would there be a Big Bang?”. That’s a question cosmologists would ask. The answer would come from mathematical reasoning about a physical model. But, the physical model comes from modelling the empirical universe, and thus the whole enterprise is rooted in empiricism.

      I’m claiming that it is insufficiently justified to count as knowledge, because it relies on an explicit logical fallacy.

      Then nothing can ever count as knowledge, since all of knowledge is provisional and nothing can ever be validated for certain ex nihilo.

      Note that I’m not claiming that we know for sure that empiricism is the only root to knowledge, nor am I claiming a known-down proof of this, I am merely asserting that it is the best model so far; I am justifying it in the same way that the claim “pigs don’t fly” is justified. You are right that neither claim is certain.

      But if you counter that the genetic structure of pigs is such that the likely can’t develop any organs that could allow for flight, that’s a stronger argument and does give us a lower probability.

      All of that will also involve inductive reasoning. It would indeed give you a lower probability, but never a zero probability.

    15. verbosestoic

      For example, cosmologists spend time considering alternative universes that could be constructed from the same basic model.

      I don’t deny that science does some limited conceptual analysis. The difference is in the purpose for doing so. So, let’s imagine that we are considering two models, one of which works well in this universe but is awkward in those alternative universes that we don’t think exist, and one of which works well in the alternative universes but not so well in this one:

      Physics — and science in general — will prefer the former, because they do conceptual analysis to better understand the instances in this world, and so that it’s awkward in other universes that don’t exist doesn’t really matter to them.

      Philosophy, in general, prefers the latter, because it wants to get to the concepts themselves, and a good concept has to work reasonably well in all universes, not just the one we happen to have.

      For mathematics and logic, the difference is meaningless, because they don’t care about modelling any world at all (as mathematics and logic). Both models are valid, useful, and possibly interesting, and so they have no reason to prefer one over the other. It’s only when science and philosophy get involved that some maths and logics are preferred, but otherwise just teasing out the implications is enough for those fields.

      That’s the difference, here, and why simply appealing to “We use thought experiments too!” doesn’t in any way impact my argument.

      But you can ask exactly the same sorts of questions about science’s models. Thus you cannot distinguish between maths and science this way.
      For example, one can ask, “in such and such a conceptual universe, would there be a Big Bang?”.

      Um, why do you think that your question is equivalent to my question? From my perspective, they aren’t the same question at all. MY question is more akin to “If moons were cubical, would they still be moons?” than “If the universe was like this, would a Big Bang happen?”.

      Then nothing can ever count as knowledge, since all of knowledge is provisional and nothing can ever be validated for certain ex nihilo.

      You ignore that I said, IMMEDIATELY THEREAFTER:

      I am not demanding certainty; I am demanding that you do more than simply say “I haven’t seen it” when I ask you to justify how you know something is true.

      I also then went into a detailed analysis of the inductive fallacy that you REALLY need to address before this can proceed any further. Even your comment about the “pigs can fly” example relies on insisting that I couldn’t have certainty, which I have said on numerous occasions and in great detail that I am not demanding.

    16. Coel Post author

      Philosophy, in general, prefers the latter, because it wants to get to the concepts themselves, and a good concept has to work reasonably well in all universes, not just the one we happen to have.

      That’s true. But scientists will take the same attitude! They really are also interested in the concepts themselves. Indeed theoretical physicists often spend their lives more interested in the concepts than in the real-world applications. 🙂 If a concept worked well in other possible universes but not in ours, then that would be an interesting topic for theoreticians to study.

      MY question is more akin to “If moons were cubical, would they still be moons?”

      But why do you think that scientists would be uninterested in that question? Surely scientists would be the ones most interested in such a question. In that same way, scientists have recently been discussing how to define a “planet”, and whether it is spherical is indeed one of the major criteria.

      I also then went into a detailed analysis of the inductive fallacy that you REALLY need to address before this can proceed any further.

      I’ve defended the scientific use of induction here. But, more generally:

      I am demanding that you do more than simply say “I haven’t seen it” when I ask you to justify how you know something is true.

      What more do you want? Scientific models are, at root, simply generalisations of empirical experience, codified into the simplest models that will fit the observations and give explanatory and predictive power.

      “Pigs don’t fly” is the simplest model that fits the data on the topic of pigs and flying. On the topic of monism versus dualism, monism fits the data fine and is simpler. Saying “we haven’t seen it”, or rather that we have no data which suggest that we need to abandon the simpler model (monism) and include a dualistic divide in our models, is exactly why we don’t include dualism in scientific models.

      You seem to be wanting more justification for materialism than that. As I see it, there isn’t any more justification than that, and yet that justification is sufficient for the (provisional) adoption of materialism.

    17. verbosestoic

      Indeed theoretical physicists often spend their lives more interested in the concepts than in the real-world applications. 🙂

      Physics is the only formal science with that degree of conceptual focus, and even then theoretical physicists will often get criticized if their theories are too “out there” when compared with the empirical data. Moreover, pretty much any theoretical physicist’s theory isn’t accept in physics until it is empirically confirmed, which is not true for philosophy. Thus, physics definitely ties itself to THIS world much more strongly than philosophy does, and as we’ve seen when people like Hawking and Krauss try to drift into the more philosophical topics they are very bad at it.

      So, essentially, physics is rather unique in its conceptual focus, and still has less of a conceptual focus than philosophy does. At best, then, it is unclear where theoretical physicists stop doing physics and where they start doing philosophy, but there is a distinction to be made there.

      If a concept worked well in other possible universes but not in ours, then that would be an interesting topic for theoreticians to study.

      I think you’re treating “possible universes” here as the multiverse model, while I’m thinking of it in terms of modal logic and “possible worlds”, which is different. I don’t see that much interest in philosophical “possible worlds” in any science and among most scientists. Even thought experiments are woefully primitive in physics and the other sciences, mostly because, as I said, it’s more interesting to just go and look at the world in physics than to speculate about it. So they don’t use them as evidence, but as illustration, while for philosophy that’s the only evidence you can really have.

      But why do you think that scientists would be uninterested in that question? Surely scientists would be the ones most interested in such a question.

      Why would they be interested in that question if they’ve never seen a cubical moon and don’t think that one is possible? Most scientists I’ve seen really DON’T care about those questions; it’s one of the reasons that science isn’t natural philosophy any more, as it adopted a far stronger focus on only explaining what was necessary to figure out the empirical world than philosophy had.

      What more do you want? Scientific models are, at root, simply generalisations of empirical experience, codified into the simplest models that will fit the observations and give explanatory and predictive power.

      But in order to count as knowledge, they have to have MORE than that, because as I think Quine pointed out you can have multiple models that explain the empirical data — and even predict it — just by changing some of your starting presumptions. So you need a justification to say that this model is the one that is RIGHT, not merely one that works for you. And you can’t use a justification of “I haven’t seen an instance of the thing you’re talking about” to justify that, especially if you are dealing with someone who claims that they have or that it’s necessary.

      In short, the SPECIFIC inductive justification you’re using here fails the instant someone proposes that a different model is the right one. And Occam’s Razor will not save you when you have two models that differ in what they explain and in their fundamental details. As I pointed out — and you ignored — you are using induction in the way that is fallacious and so cannot produce knowledge. This is not me saying that induction can’t produce knowledge, nor is it me saying that we need certainty. It is me saying that in that case, you aren’t justified in claiming knowledge. You generalizing to far too different phenomena and far too widely without providing sufficient justification to think that the generalization holds. Thus, you can’t use induction there to claim knowledge. And if you can’t do that, then I don’t need to accept your model — even provisionally.

    18. Coel Post author

      Thus, physics definitely ties itself to THIS world much more strongly than philosophy does …

      I’m not sure I agree. The reasons that both theoretical physicists and philosophers are interested in “concepts”, including concepts that might apply to other possible worlds, is that the more we understand the concepts the more we understand about our world. It’s really, at root, the same attitude from both.

      … and as we’ve seen when people like Hawking and Krauss try to drift into the more philosophical topics they are very bad at it.

      One can dispute that! A lot of philosophers are worse than Krauss and Hawking at philosophy!

      Why would they be interested in that question if they’ve never seen a cubical moon and don’t think that one is possible?

      Because, as above, the more they understand about concepts, the more they understand about our world. Understanding why our world is the way it is requires considering it as one possibility among the realm of conceptual worlds.

      as I think Quine pointed out you can have multiple models that explain the empirical data — and even predict it — just by changing some of your starting presumptions. So you need a justification to say that this model is the one that is RIGHT, not merely one that works for you.

      You then use Occam’s razor to pick the simplest model. If you pick the model that has the most explanatory and predictive power, and also the fewest ad-hoc features, you’re then pretty much down to one (or at least very few) models. If there genuinely is ambiguity between two models that fare as well on these tests, then science would not decide between the two, but would ask for further investigation and evidence.

    19. verbosestoic

      The reasons that both theoretical physicists and philosophers are interested in “concepts”, including concepts that might apply to other possible worlds, is that the more we understand the concepts the more we understand about our world. It’s really, at root, the same attitude from both.

      You didn’t address my comments on what “possible worlds” means, and the different sorts, which is important because it is my understanding that theoretical physics cares little if at all for modal logic, and the other sciences care for it not at all, while it’s VERY important in philosophy. Let’s look at Putnam’s Twin Earth experiment:

      https://en.wikipedia.org/wiki/Twin_Earth_thought_experiment

      My understanding is that philosophers find it fascinating, theoretical physicists might find it mildly interesting at best, and all other sciences find it useless. What’s your take on that?

      One can dispute that! A lot of philosophers are worse than Krauss and Hawking at philosophy!

      It’s not a defense of them to say “Hey, there are some bad philosophers too!” [grin].

      But they are bad in a typical way because they fail to understand how to do true conceptual analysis. Krauss gets confused in thinking that the word expresses the concept, and so takes a meaning of the word — one that we find in his field — declares THAT to be what the problem is, and then declares it solved, never realizing that he’s not talking about the same concept of “Nothing” there. Hawking, on the other hand, doesn’t delve deeply enough into the concepts of “law” and “thing”, and so again doesn’t realize what the concept “Nothing” used there really entails. When scientists jump into philosophy, they typically treat instances as if they were the concepts and argue based on that, and when philosophers jump into science they typically think that when they’ve come up with a consistent concept that all the work is done. This is how their methods differ, and it follows strictly from what the focus of the fields actually is.

      Because, as above, the more they understand about concepts, the more they understand about our world.

      Unless those concepts don’t apply to this world, which is not a concern for philosophy. Science cares about the properties of objects in this world, and so only cares about the conceptual properties if we have instances in this world. Philosophy isn’t that concerned about that, which is why it can deal with things that are purely conceptual (like normative concerns).

      You then use Occam’s razor to pick the simplest model. If you pick the model that has the most explanatory and predictive power, and also the fewest ad-hoc features, you’re then pretty much down to one (or at least very few) models.

      The problem is that what features count as “ad-hoc” depends on the theory itself; it may be crucial for that theory to work but not exist in others. Is that “ad-hoc”? Well, if it’s strict theoretical entities, maybe, but this still only works if the compared theories all explain all the same things in the same way. And, as pointed out, if it’s just a matter of a difference in starting assumptions, Occam’s Razor will not save you, because it can’t slice away STARTING assumptions (which would be on the level of “My senses reflect a real world”), only differences in the theories.

      If there genuinely is ambiguity between two models that fare as well on these tests, then science would not decide between the two, but would ask for further investigation and evidence.

      Which means that it would claim to not know that, right?

      I say that when it comes to material/immaterial and natural/supernatural, that ambiguity exists. I’d actually argue that that must be the case because in the specific cases you’re talking about, you can only use “In all of the other cases, it ended up material!” as an argument, suggesting that you can’t DEMONSTRATE that empirically, which would be ambiguous evidence pretty much by definition.

    20. Coel Post author

      Let’s look at Putnam’s Twin Earth experiment: … What’s your take on that?

      I’m in the “mildly interesting” but easily-resolved camp. That resolution is simply that “water” means “sufficiently water-like”. Thus H2O and XYZ are both “water”. Indeed, we already know of several types of “water” with different compositions. As well as H2O there is HDO and water with O18 instead of O16. All of these are sufficiently water-like and so are “water”. Of course the qualifier “sufficiently” depends on the context. If we’re putting a fire out, sea-water counts as “water”, if we were drinking it then it wouldn’t.

      So, overall, twin-earth has a fairly mundane resolution and I can’t share philosophical fascination with it.

      But they are bad in a typical way because they fail to understand how to do true conceptual analysis.

      I do not agree that physicists are any worse, in general, than philosophers at analysing concepts. Indeed, if there were an analysing-concepts Olympics, I’d tend to pick physicists for my team rather than philosophers!

      Krauss gets confused in thinking that the word expresses the concept, and so takes a meaning of the word — one that we find in his field — declares THAT to be what the problem is, and then declares it solved, never realizing that he’s not talking about the same concept of “Nothing” there.

      There has been a large amount of utter tosh spoken about Krauss and his book (by David Albert and others). If one reads the book it is blatantly obvious that Krauss is aware of all the different interpretations of the concept “nothing”. To me he showed more philosophical awareness than the replies by Albert and most critics do.

      Science cares about the properties of objects in this world, and so only cares about the conceptual properties if we have instances in this world.

      That’s not true. You can only properly understand the concepts if you also think about how such concepts might vary, regardless of whether there are instantiations of those variations. Scientists think about concepts in such ways just as much as philosophers. Indeed, often it’s the physicists having wilder flights of fancy about conceptual worlds than philosophers.

      Occam’s Razor will not save you, because it can’t slice away STARTING assumptions …

      Yes it can. All “starting assumptions” can then by questioned and assessed and subjected to Occam and other tests. That’s the whole point of the Quinean-web understanding of scientific models, in which no statements are primary and unquestionable.

      Which means that it would claim to not know that, right?

      Yes.

      I say that when it comes to material/immaterial and natural/supernatural, that ambiguity exists.

      Whereas I’d argue that in no case does the “immaterial” or the “supernatural” explanation actually improve the overall explanation, and thus they are disposed of by Occam.

    21. verbosestoic

      I’m in the “mildly interesting” but easily-resolved camp.

      1) The point of the question was not to get your opinion on whether you found it interesting or not, but gauge your opinion on whether SCIENCE found it interesting or not. Given your reaction, you seem to have proven my point: philosophy finds it interesting, science … not so much. In fact, your recent post on fitness is the same thing: philosophers find it interesting, and you even explicitly state that scientists go on doing what they do regardless of those issues. So there is a difference in the sorts of conceptual analysis that the fields do.

      2) You’ve hit one of the main issues I have with people who wander in and try to do philosophy. You say that you think this is easily resolved, and give your preferred resolution (I’ll talk about it in a minute). Do you think that philosophers are incapable of thinking of that obvious and easy resolution? If you accept that they are, then you probably should accept that they probably have thought of it … and that if they don’t consider that to be the answer, that maybe there are issues that you haven’t considered as to WHY it’s not taken as the obvious answer?

      And, it turns out, there are:

      That resolution is simply that “water” means “sufficiently water-like”.

      So … what does it mean to be “sufficiently water-like”? We’d want to distinguish water from, say, iced-tea, or even rocks, so what sort of properties can we use, if we aren’t going to use physical composition. Well, it seems that you try to take a strong functional view of this — water is what water does/what we use it for — from this part:

      Of course the qualifier “sufficiently” depends on the context. If we’re putting a fire out, sea-water counts as “water”, if we were drinking it then it wouldn’t.

      The problem is that this confuses the word with the concept. If I ask for a glass of water, and you bring me a glass of salt water, technically you’ve fulfilled my request. The reason I’d be upset is NOT that you didn’t bring me “water”, it’s that you didn’t fulfill my request, which included the context that I wanted it to drink, and you ought to have known that I can’t drink salt water. It is in general considered absurd to suggest that salt water is EVERY not water; it is water, it’s just water that it’s useless to try to drink. So that strong a functional stance leads to immediate absurdities.

      But we CAN take a less strong functional stance, and argue that what it means for something to be water is that it has certain functional properties … in fact, all of the ones that hold true on Twin Earth. But what this means is that discovering that water on this Earth is H2O — or even the other forms you mentioned — doesn’t really help us in understanding what “water” is, because it has no link to its meaning. So if we can say that H20 and XYZ are both equally “water”, then whether water is H20 or XYZ is IRRELEVANT to our understanding of what water is.

      But science and everyone who respects it seems to think that we really did come to understand WATER better when we discovered that. It also strikes against the idea of water being a natural kind if its composition is irrelevant to what it really means for something to be water. (I’ll talk, in a minute, about a way to resolve this, but you likely won’t like it).

      There’s another problem. We think that there’s a clear difference between gold and fool’s gold, but the reason it’s CALLED fool’s gold is that for most purposes it behaves pretty much like gold. So it is possible, then, that fool’s gold is “sufficiently gold-like” to count as gold. And yet we definitely consider it to be different than gold. While there are some functional differences, we’d need a way to determine what gold “really is” to make use of them, and we can’t just do that in certain contexts because of what I said above. So if you’re going to take this route, you need to have a way to determine what “gold” really is.

      Ultimately, what this tack does is take the most definitive property we have — physical composition — and make it irrelevant to the meaning of the term. Thus, the physical composition becomes an ACCIDENTAL property; it’s what it happens to have, but in no way defines what it is. For you, it even being MATERIAL would have to fall in this category as well, which would be a problem. But at least science thinks that the physical composition of the object is a key piece of knowledge, and is essentially what it strives to find most of the time … even though it doesn’t in any way help us to understand what it really means for something to be what it is. This … is not a good outcome.

      We can save this by pointing out that what it really means for something to be what it is is a conceptual question, and how it is instantiated in this world is an instance question, and so science is interested in the instance properties and other fields in the conceptual properties. If you make this move, you end up agreeing with me on the very question that your disagreement with me ended up spawning this debate.

      Personally, I think that the physical composition of an object is indeed an instance property and so not what really makes that thing an instance of that concept, but I can see the issues when you make that move. They’re just not issues I find problematic. You, on the other hand … [grin]

      I do not agree that physicists are any worse, in general, than philosophers at analysing concepts. Indeed, if there were an analysing-concepts Olympics, I’d tend to pick physicists for my team rather than philosophers!

      But you aren’t a philosopher, so how do you know that you’re really doing conceptual analysis [grin]?

      Anyway, while this is just a rhetorical trick, if you really think that, then could you explain what you think conceptual analysis is and why you think that physicists do it better? What standards do you think the “Olympics” would judge it by?

      To me he showed more philosophical awareness than the replies by Albert and most critics do.

      Krauss’ question is at least traditionally a philosophical one, and has been criticized by philosophers who have definitely read the book. Why do you think that your amateur understanding trumps theirs?

      At any rate, if Krauss tries to answer the question by using the quantum vacuum “Nothing”, then I don’t even need to read the book — and I admit I haven’t — to know that he’s gone horribly wrong, because the QV is actually a “Something” in terms of the original question, answering that it came from a necessary and constant “Something” IS a valid answer to the question, and so all he ends up doing is equivocating by answering a different question … and philosophy already thought about whether we could just answer another question and discovered, oddly enough, that doing so just left that question unanswered, and we don’t have any reason to think that the question, in principle, can’t be answered. In terms of conceptual analysis, Krauss dropped talking about the conceptual issues that he can’t handle to talk about the ones he can, and then pretends that either he’s answered the original question or at least that the original question isn’t worth answering. He’d be badly wrong on both counts.

      You can only properly understand the concepts if you also think about how such concepts might vary, regardless of whether there are instantiations of those variations.

      Your comments here and on the evolution post prove that while they may be of idle interest, such things are not going to be the heart of your research program, while for philosophers, they are … which is rather my point.

      Indeed, often it’s the physicists having wilder flights of fancy about conceptual worlds than philosophers.

      That you think talking about “wilder flights of fancy” says anything about the quality of the conceptual analysis simply proves that you don’t do conceptual analysis well. The conceptual worlds examined in philosophy are not about being weird, but are about being critical to the concept being examined. They have to raise interesting conundrums, not simply be cool.

      Yes it can. All “starting assumptions” can then by questioned and assessed and subjected to Occam and other tests.

      They can’t, because either Occam and those other tests are starting assumptions as well, or else they are justified BY other starting assumptions. As an example, you know that I don’t find Occam’s Razor to be useful at all. If I reject its use, then, based on different starting assumptions than you, what agreed upon principle — meaning one that I CANNOT rationally disagree with — will you appeal to to justify?

      That’s the whole point of the Quinean-web understanding of scientific models, in which no statements are primary and unquestionable.

      Actually, in Quine some of them are going to be more primary than others, because they will form the bedrock upon which all of the other beliefs and assumptions rest; they will be justified by them. They will also be difficult to dislodge, because they would require giving up a lot of beliefs that are working, or at least finding a new way to justify them. As I do think it was Quine who talked about how you can always make a theory work on multiple presumptions simply by deciding what to keep and what to abandon, this doesn’t help you much [grin].

      Whereas I’d argue that in no case does the “immaterial” or the “supernatural” explanation actually improve the overall explanation …

      You’d have to know what constitutes one of those explanations first, hence the ambiguity.

    22. Coel Post author

      but gauge your opinion on whether SCIENCE found it interesting or not …

      I think “science” finds it only semi-interesting because there is a straightforward and easy resolution of the “issue”.

      In fact, your recent post on fitness is the same thing: philosophers find it interesting, and you even explicitly state that scientists go on doing what they do regardless of those issues.

      Again, I think the reasons why scientists don’t find that issue interesting is that they’ve already resolved it.

      Do you think that philosophers are incapable of thinking of that obvious and easy resolution?

      If we’re talking about Twin-Earth “water” then the resolution I gave has indeed already been given by philosophers.

      … and that if they don’t consider that to be the answer, that maybe there are issues that you haven’t considered as to WHY it’s not taken as the obvious answer?

      Possibly, yes, that might be the case!

      If I ask for a glass of water, and you bring me a glass of salt water, technically you’ve fulfilled my request. The reason I’d be upset is NOT that you didn’t bring me “water”, it’s that you didn’t fulfill my request, which included the context that I wanted it to drink, and you ought to have known that I can’t drink salt water.

      Agreed. In that context the term “water” means potable water, and yes I would define that in functional terms.

      But what this means is that discovering that water on this Earth is H2O — or even the other forms you mentioned — doesn’t really help us in understanding what “water” is, because it has no link to its meaning.

      Well “understanding” is a whole can of worms, and could shortly lead into Chinese Rooms and such. But: there are a whole lot of contexts in which we might talk about “water” and thus there are a whole lot of “meanings” of the term “water”. For some of these contexts, the composition “XYZ” will be relevant, and for other contexts it will be irrelevant.

      What discovering the composition implies is an enlargement of the “linkages” between different “meanings” of the term “water”. If we imagine a whole web of linkages between concepts, learning that composition adds new linkages, and so adds to the web.

      Whether that is pertinent to the “meaning” of the term in one particular context is another issue. It is certainly pertinent to the meaning of the term in the set of all possible contexts.

      So if we can say that H20 and XYZ are both equally “water”, then whether water is H20 or XYZ is IRRELEVANT to our understanding of what water is.

      No, I disagree. As above it extends the web of linkages.

      So it is possible, then, that fool’s gold is “sufficiently gold-like” to count as gold.

      In some contexts, yes.

      And yet we definitely consider it to be different than gold.

      Yes, in the same way that a metal chair is different from a wooden chair, and yet — in some contexts — they are both a “chair”.

      we’d need a way to determine what gold “really is” to make use of them, and we can’t just do that in certain contexts because of what I said above.

      But I don’t agree with your analysis above; I think that a functional account works fine.

      Ultimately, what this tack does is take the most definitive property we have — physical composition — and make it irrelevant to the meaning of the term.

      Not at all. In some contexts the physical composition is entirely pertinent. For example: “I need to make some water from its components: do I need some X, Y and Z atoms, or some H and O atoms?”. In that context, the composition is entirely pertinent to the meaning of the term.

      What the Twin Earth does is to narrow down the number of contexts to those for which the composition is irrelevant, and then declare that the composition is irrelevant to the meaning. But, if one broadens out the contexts to include those where the composition is relevant, and the issue is resolved.

      Twin-Earth seems to me one of a number of “problems” where the philosophers have essentially created a problem for themselves by a false analysis of the situation.

      But you aren’t a philosopher, so how do you know that you’re really doing conceptual analysis [grin]?

      Because all scientific models are concepts, and working with them is what scientists do!

      then could you explain what you think conceptual analysis is and why you think that physicists do it better?

      I think physicists do it better because philosophers tend to tie themselves in knots by creating problems (Twin Earth, Chinese Room, p-zombie, and others) which are actually pretty easy to resolve.

      … if Krauss tries to answer the question by using the quantum vacuum “Nothing”, then I don’t even need to read the book — and I admit I haven’t — to know that he’s gone horribly wrong, because the QV is actually a “Something” …

      Why sure it is! And do you *really* think that Krauss is too stupid to have thought of that?

      In that book Krauss adopted a good pedagogical technique. He realised that one can take lot of interpretations of the term “nothing”. [ E.g. “what have you got in your pocket?”, “Nothing”. The answer is correct *in* *context*, even though there are air molecules in the pocket.]

      As Krauss proceeds through the book, he gradually strips down the concept of “nothing” to more and more nothing-er versions of nothing. He dispenses with material, showing how it can be created by quantum fields. But then, realising that quantum fields are not “nothing”, he asks whether we can dispense with them, and get to a nothing-er nothing.

      But then David Albert points to half-way along that process and says “look, the silly physicist thinks that counts as nothing!”. And then lots of philosophers join the chorus. If you actually read the book, you’ll see that Krauss was well aware of the “somethingness” of quantum fields, and explicitly discussed that issue. The complaints by Albert et al were just unfair.

      Krauss dropped talking about the conceptual issues that he can’t handle to talk about the ones he can, and then pretends that either he’s answered the original question or at least that the original question isn’t worth answering.

      How do you know that if you haven’t read the book? Don’t just go on Albert’s utterly unfair review!

      If I reject its use, then, based on different starting assumptions than you, what agreed upon principle — meaning one that I CANNOT rationally disagree with — will you appeal to to justify?

      Science does not work based on “starting assumptions”, it works on a Quinean-style web, where everything is justified by other parts of the web. One can then justify Occam’s razor (and similar things) that way.

      Actually, in Quine some of them are going to be more primary than others, because they will form the bedrock upon which all of the other beliefs and assumptions rest; they will be justified by them.

      The Quine-style web does not work that way, with “bedrock” assumptions. It works by justifying all assumptions in terms of other parts of the web. The web overall is then justified against empirical reality.

    23. verbosestoic

      Again, I think the reasons why scientists don’t find that issue interesting is that they’ve already resolved it.

      Okay, I realize that I took the wrong tack here. I was pointing out the potential issues with simply taking the functional view, but that’s not really an issue for you given your epistemological view, as if you don’t have a better solution you’ll stick with what you have, even given its problems, and so that won’t, in and of itself, cause you to rethink your “obvious” solution. Fortunately, the real interesting thing about the experiment is not that we don’t have any “obvious” answers, but that instead we have two completely incompatible yet equally “obvious” answers, or perhaps to use your terms better that we have two equally “easy” resolutions:

      1) The functional one that you favour: if it looks like water and behaves like water, then it’s clearly water, no matter what its physical composition. Thus, when we say “water” in this Earth and they say “water” in Twin Earth, we are talking about the same thing.

      2) That what we mean by water in THIS Earth is really “H20” (or sufficiently similar derivations, as determined by science), and so since “water” in Twin Earth means “XYZ” which is clearly NOT a sufficiently similar derivation, when we talk about water on this Earth and when they talk about water on Twin Earth we are actually talking about two completely different things, despite their similar functionality.

      It seems to me that the resolution that science would prefer is 2), not 1), as that fits into what they really wanted to know, and is what they usually considered to be what water really is. As another example, take the reclassification of “Pluto” to being a dwarf planet and so not really a planet; this fits more into a “what it means to be a thing is what its physical properties are” as opposed to “what it means to be a thing is how it can be used”, which to me seems to be more of an “everyday reasoning” approach.

      At any rate, we have two perfectly good easy resolutions here. Both fit the evidence, and both work. There are good reasons for either, and reasons against taking that tack. I can’t think of a way to use Occam’s Razor here without being self-serving (ie picking the one you prefer). And yet, they both can’t be right. And we have good reason to think that there IS a right answer here, and that this isn’t just a matter of personal preference.

      Hence the interest: there’s a right answer here, but depending on how you look at it two completely opposite answers both look to be obviously right. The problem is to resolve that conflict, either by proving one answer true or providing an alternative that captures what seems right about both.

      This has proven … difficult.

      If we’re talking about Twin-Earth “water” then the resolution I gave has indeed already been given by philosophers.

      Given that, why, in your opinion, has it not simply become the standard answer in philosophy? What are philosophers missing that you have seen?

      Agreed. In that context the term “water” means potable water, and yes I would define that in functional terms

      I think you’re confusing the contextual theory of language — which I personally do favour — with the idea of concepts. Because our terms are ambiguous, we often have do — and often do — use context to determine what a sentence “means”, ie what concepts and referrents the speaker is trying to express to us. But this doesn’t change what the concept itself actually is, or mean that the concept is defined by the context. What we are trying to do is determine what concepts are being expressed, not define them.

      Let me use a clearer example of this. Take the word “bank”, and the phrase “I’m going to the bank today”. Now, the English word “bank” can refer either to the place where you get money, or to the side of a river. Clearly, these are different concepts; you can never say that the place where you get money or the side of a river are the same concept, and we determine by context what that is. But the word is ambiguous, so we need to use context to determine which concept is being referred to here … and, given the sentence, what physical think is being referred to here. So in order to figure out what that sentence and what that word “means”, we need to disambiguate the usage here, using context. But what it means to be “the sort of bank where you get money” is NOT determined here, in context. That is determined by conceptual analysis.

      The same thing applies to your water case. Conceptually, both salt water and potable water always count as water. But in the sentence you gave, you were using the WORD “water”, and expecting the person using the context to understand that you were referring to a sub-concept of water, that of potable water. Thus, this is again a clear case of disambiguation; we’re trying to figure out what concept the term there is supposed to be referring to. If the person gets it wrong, it’s because they were assuming that you meant the parent and more general concept, when you were aiming at the more specific one. But, just as in the bank case, this does not change the concept of water, or make the concept context dependent. All it does, again, is disambiguate the word, and has no impact on the concept itself … or on how we go about figuring out what the concept is.

      Yes, in the same way that a metal chair is different from a wooden chair, and yet — in some contexts — they are both a “chair”.

      In ALL contexts they are both chairs. Sometimes, they aren’t the droids … er, CHAIRS you’re trying to refer to.

      To get back to discussions of what conceptual analysis really is, a really interesting question is if you can call something a chair if it can’t possibly be sat on. In general, we tend to think that it can be when faced with something that looks like a chair but can’t be sat on, but when we do conceptual analysis it seems that one of the defining primary qualities of chairs is that they can be sat on, creating a contradiction. Examples like this are used in introductory philosophy classes to show the issues, but aren’t focused on because we have far more interesting conceptual problems to short out, and if we figure out concepts we’ll probably have the answer to that anyway.

      But I don’t agree with your analysis above; I think that a functional account works fine.

      If you have to count fool’s gold as actually being gold in ANY circumstance, you have a major issue because NO ONE accepts that; they are definitely considered as different things as the two concepts of “bank” that I pointed out above, and have no relation like “potable water” and “salt water” have to each other conceptually to make substitution possible. So you’d need a lot more to justify your functional approach than that, especially considering all of the other theories that work quite well WITHOUT running into that issue.

      Not at all. In some contexts the physical composition is entirely pertinent. For example: “I need to make some water from its components: do I need some X, Y and Z atoms, or some H and O atoms?”. In that context, the composition is entirely pertinent to the meaning of the term.

      Well, as it turns out, it’s provably NOT pertinent to the meaning of the term. You start from asking “I want to make some water”. Then you go about figuring out how to make water. On Earth, you build some H20. On Twin Earth, you build some XYZ. But on the functional answer, as stated above, it’s still water no matter WHAT this procedure is; the procedure — and therefore, the physical composition — is IRRELEVANT to whether or not it is water. It’s like trying to make iced tea, and in my case looking around my kitchen I grab the instant mix, while someone else takes real tea and ice and whatever. At the end of the day, it’s still iced tea no matter HOW I have to build it, and the same thing applies to the Twin Earth case: on the functional approach, it’s still water regardless of its physical composition, so its physical composition is irrelevant to what it means to be water.

      And the “linkages” move doesn’t work because that really looks like disambiguation, not conceptualization. The links determine when you’d be referring to “Twin Earth Water” instead of “Earth Water”, but it’s still water. You could argue that the physical composition is relevant to the concepts of “Earth Water” and “Twin Earth Water”, as that’s the difference between them … but then, again, you end up moving from the functional approach to the physical composition approach, as that means that the person on Earth and the person on Twin Earth are actually referring to different things when they use the term “water”, and that the difference in meaning and in concept is based entirely in the physical composition.

      Twin-Earth seems to me one of a number of “problems” where the philosophers have essentially created a problem for themselves by a false analysis of the situation.

      And so, lots of intelligent people thinking about this a lot — this is a standard problem taught in philosophy classes — have failed to come up with an answer that you consider obvious, and it is thus clear that THEY missed something as opposed to YOU missing something? Especially since you admit that they already DID think of this answer? Maybe the problem is that you don’t understand the question … or don’t understand concepts.

      Because all scientific models are concepts, and working with them is what scientists do!

      Yep, it’s that you don’t understand concepts. A model of something is a representation, not a concept. A scientific model — like, say, a theory — is a mix of concepts and observations that represent something and provides a descriptive — not normative — view of that thing. A model car, for example, is not a concept of car, but is a representation of a car. A weather model is not the concept of weather, but is instead a representation of present and future weather, using concepts and observations and other models and theories. Because science deals primarily with models, dealing with CONCEPTS is always secondary; they conceptualize only enough to build a good model of what they are trying to represent. Anything that makes the model cluttered or problematic is put aside. Since strong conceptual analysis — as you’ve noted — tends to clutter models by considering things not directly relevant to it, science tends to drop it out as soon as it does so. Thus, philosophy does more conceptual analysis than science does.

      This is not a criticism of science. If you tried to do science using philosophy, you’d chase a lot of irrelevant questions a lot of the time, as you yourself note. You’d take a LONG time to get any productive models. But for strong conceptual analysis, that’s what you have to do. That this isn’t what you’re interested in doesn’t make philosophy pointless, or mean that scientists who don’t actually do that are better at it because they do science better and you think that’s all you need to do to get good concepts.

      I think physicists do it better because philosophers tend to tie themselves in knots by creating problems (Twin Earth, Chinese Room, p-zombie, and others) which are actually pretty easy to resolve.

      Again, many intelligent people have spent lifetimes studying this and come up with the same “easy resolutions” as you, but don’t consider the questions resolved. Isn’t it a little arrogant to declare YOURSELF to be better at THEIR FIELDS than they are based on that?

      Why sure it is! And do you *really* think that Krauss is too stupid to have thought of that?

      No, what I think is that Krauss is arrogant and ignorant enough — based on what I’ve read of his own responses to the issue — to think that he can sub “QV nothing” into the “Nothing” part of the “Can you get something from nothing” question and have that be a BETTER question than the one that was asked.

      But let’s look at that technique that he used. First, let’s look at the question. There are two answers to the “Can you get something from nothing?” question:

      1) No, so we need some kind of fundamental something.

      2) Yes, and here’s how.

      By looking in detail at the things we call “nothing”, Krauss looks like he’s trying to give an answer of form 2). But it has always been the case that that nothing is known to not be things like “What’s in my pocket?” but has in fact been absolute non-existence. So to give an answer of form 2), he can only start with absolute non-existence; for anything else, it will be an answer of form 1). So since the QV is indeed not absolute non-existence, he can’t use it to give an answer of form 2). He is not, therefore, showing how we can get something from nothing.

      So, let’s assume he knows this, and is really giving an answer of form 1). Then what is the purpose of examining the things that we CALL nothing? If he’s trying to establish a fundamental something, then he unreasonably limits himself talking about things that we happened to call “nothing”. That’s only relevant if he wants to give an answer of form 2). If he’s trying to give an answer of form 1), then what he’s doing is the equivalent of saying “What am I thinking about? Nothing! But as I can think about nothing, then thought is the fundamental something and I’ve solved the problem!”. It might be the case that thought is the fundamental something — not your position, I know — but that argument in no way supports that; that we call it “nothing” does not support it being the fundamental something.

      So either way, he’s gone wrong. Either he’s trying to give an answer of form 2) with something that is clearly not nothing, or he’s been misled by nothing into picking a particular something at least in part because it happens to be called “nothing”. More likely, he’s trying to demonstrate that all reasonable concepts of nothing are really somethings, which doesn’t work for the nothing we’re talking about. So, no, without even reading it I can tell from his even trying to use a QV “nothing” that he’s gone horribly wrong somewhere.

      The Quine-style web does not work that way, with “bedrock” assumptions. It works by justifying all assumptions in terms of other parts of the web. The web overall is then justified against empirical reality.

      Which leads precisely to the problem I already noted: that you can adjust the web to make it fit the evidence in a great many if not infinite ways. If you have no starting propositions, that gets even more malleable. So it doesn’t help you much here.

    24. Coel Post author

      2) That what we mean by water in THIS Earth is really “H20” (or sufficiently similar derivations, as determined by science), …

      I don’t think that this stance is viable. First, what we “mean” by water is clearly not one specific thing (such as the composition), but a whole range of things depending on context (ie., I go for the functionalism option). E.g. scientists before the discovery of oxygen could not have meant “H2O” by “water”.

      Secondly, the “really” H2O bit privileges one particular account of “water” over others, with no good reason for doing so. (E.g., one cannot defend it on the grounds that is it the “basic” composition, since one can de-compose both H and O into smaller components.)

      Third, the “sufficiently similar derivations” gives the game away, since it reverts to functionalism. The fact that we can substitute deuterium for hydrogen and O-18 for O-16 destroys any “Platonic essentialist” account of what water “really is”, and leaves only the functionalist account.

      … and so since “water” in Twin Earth means “XYZ” which is clearly NOT a sufficiently similar derivation, …

      Why is it not sufficiently similar? If “X” were just another label for deuterium and “Z” for O-18 (as oppose to O-16) then would it be “sufficiently similar”? This whole argument seems to elevate mere words, mere labels, to some sort of status that they don’t actually merit.

      Let’s consider “crystal” instead of “water”. Clearly, “crystal” can be made up of all sorts of different compositions, and thus the “meaning” of “crystal” is functional, rather than in terms of the composition.

      when we talk about water on this Earth and when they talk about water on Twin Earth we are actually talking about two completely different things, despite their similar functionality.

      But, by the set-up of the experiment, they are not “completely” different, they are different in one way and similar in lots of other ways. I don’t find this any more puzzling than the idea of “crystals” that are composed of different stuff.

      Let’s suppose that twin-earth lacked an element, say silicon, that is present on Earth. Thus an Earth “rock” might have a composition that no twin-earth “rock” has. Maybe, where we have quartzite rock, they have limestone. So what? Why would that cause a problem for the concept “rock”?

      One might counter that we consider that water is different in having only one possible composition. But whether a concept is singly realisable (water) or multiply realisable (rock, crystal, etc) is simply a fact we learn about the world. What twin-earth does is simply tell us what “water” is multiply realisable.

      (By the way, what happens if you get a space ship and take a bottle of XYZ from twin-earth to earth? Does it act like water on earth? Does the “thought experiment” tell you that?)

      It seems to me that the resolution that science would prefer is 2), not 1), as that fits into what they really wanted to know, and is what they usually considered to be what water really is.

      This — I suggest — is where the philosophers are going wrong, with a misconceived view of science. They are trying for a (Platonic essentialist?) view of what water “really” is, and of what account science “would prefer”, whereas science is pretty much pragmatic and functionalist. Even the description H2O is functionalist, as are all other descriptions in science.

      Given that, why, in your opinion, has it not simply become the standard answer in philosophy? What are philosophers missing that you have seen?

      Well, I don’t really know, and am not that familiar with the culture of philosophy. I’m aware that *some* philosophers take the line I do. As for others, well, maybe they are lacking in appreciation of science? (I do think that a lot of “philosophy of science” suffers from being too far divorced from science.) I also think that philosophy lacks a good mechanism for quality control.

      If you have to count fool’s gold as actually being gold in ANY circumstance, you have a major issue because NO ONE accepts that.

      How about if I’m putting on a Christmas play, with three kings carrying incense, myrrh and gold. Wouldn’t a nugget of fool’s gold count as “gold” in that context? It seems to me that an eight-yr-old at a nativity play would by happy with it!

      Well, as it turns out, it’s provably NOT pertinent to the meaning of the term.

      Yes it is pertinent. If I was on Earth, and wanted to make water, I’d need H and O. Having C and N wouldn’t work. Thus the composition is relevant in that context.

      Similarly, if I was on twin-earth, I’d need X, Y and Z. P, Q and R would not work. Again, the composition is relevant. So, again, the functional account is entirely satisfactory.

      And so, lots of intelligent people thinking about this a lot — this is a standard problem taught in philosophy classes — have failed to come up with an answer that you consider obvious, and it is thus clear that THEY missed something as opposed to YOU missing something?

      It’s not so much that they’ve failed to resolve the question. Many philosophers do think they have a resolution, it’s just that the others don’t agree with them. 🙂 Again, the issue is quality control and dispute-resolution in philosophy.

      A model of something is a representation, not a concept.

      I don’t really agree (at least as the term is used in science). Something like an “inverse square law” is a concept, not a representation. Something like an electron’s “charge” or its “spin” or indeed its “mass” are all concepts.

      No, what I think is that Krauss is arrogant and ignorant enough — based on what I’ve read of his own responses to the issue — to think that he can sub “QV nothing” into the “Nothing” part of the “Can you get something from nothing” question and have that be a BETTER question than the one that was asked.

      These critiques of Krauss fail to realise that that is not what he did. He did not just consider one conception of “nothing” (ie “QV nothing”) and declare that he had solved the problem.

      He very explicitly considered the ENTIRELY OBVIOUS point that “QV nothing” is not “nothing”. He pointed out that starting from “QV nothing” was obviously not “nothing nothing” and so went on to consider whether one could dispense even with the quantum vacuum and so start from a nothinger nothing.

      David Albert was simply too busy sneering to notice what Krauss was actually saying.

    25. verbosestoic

      First, what we “mean” by water is clearly not one specific thing (such as the composition), but a whole range of things depending on context (ie., I go for the functionalism option).

      Well, you’d have to demonstrate that that’s “clearly” the case, which means that you’d have to address my claim that you’re confusing disambiguation of words with conceptual analysis.

      E.g. scientists before the discovery of oxygen could not have meant “H2O” by “water”.

      Yes, this is one of the criticisms of that other approach, where saying “We didn’t know what it meant when we said ‘water’ before we discovered it was H20” seems impossible. However, if that is rephrased as “We didn’t really know what water WAS until we discovered that it was H20”, this seems to be more reasonable, and seems to be pretty much in line with the stance that science takes towards the issue. It ALSO seems to be more related to a conceptual question than your EXTREME functionalism is, which again seems far more about disambiguating the word, deciding what concept the sentence is meant to refer to rather than what concept that actually is.

      Secondly, the “really” H2O bit privileges one particular account of “water” over others, with no good reason for doing so. (E.g., one cannot defend it on the grounds that is it the “basic” composition, since one can de-compose both H and O into smaller components.)

      I’m afraid I don’t understand this argument at all, unless it is merely insisting that we can’t say that water on this Earth is really “water” if it could be different on Twin Earth, which would be assuming the conclusion. So what is your argument here?

      Third, the “sufficiently similar derivations” gives the game away, since it reverts to functionalism. The fact that we can substitute deuterium for hydrogen and O-18 for O-16 destroys any “Platonic essentialist” account of what water “really is”, and leaves only the functionalist account.

      1) Um, we are still substituting what science considers to be hydrogen and oxygen in there, so that doesn’t destroy anything. The differences are not conceptual ones, but are instead differences in accidental properties, at least as they relate to oxygen, say. That’s pretty much what “sufficiently similar derivations” MEANS there; it’s not merely that they can be used in the same way.

      2) “Platonic essentialism” here means “conceptual properties”. Platontic Forms weren’t rejected because they weren’t a good way to talk about concepts, but were instead rejected because they relied on the Forms themselves being separately and independently real. And even then, we might need that to make some things work.

      Why is it not sufficiently similar? If “X” were just another label for deuterium and “Z” for O-18 (as oppose to O-16) then would it be “sufficiently similar”?

      Or, to put it another way, if X stood for hydrogen, Y stood for hydrogen, and Z stood for oxygen, would they be sufficiently similar then? Well, yes, but that’s not what the thought experiment is saying, and so it’s not comparable. Your argument here is the one elevating words or symbols, not the counter-argument. The counter-argument says that there’s a real physical difference there, and that physical difference is enough to justify saying that they’re talking about two different things.

      Let’s consider “crystal” instead of “water”. Clearly, “crystal” can be made up of all sorts of different compositions, and thus the “meaning” of “crystal” is functional, rather than in terms of the composition.

      Sure. But, then, we never say that “crystal is physical composition X”, whereas we DO say “Water is H20”. So you have two options here: either say that that phrasing for water is at best misleading and we SHOULD be talking about it like we talk about crystal, or say that the difference is a meaningful one, at which point we end up back at the option you don’t like in the Twin Earth experiment.

      Let’s suppose that twin-earth lacked an element, say silicon, that is present on Earth. Thus an Earth “rock” might have a composition that no twin-earth “rock” has. Maybe, where we have quartzite rock, they have limestone. So what? Why would that cause a problem for the concept “rock”?

      Well, what do you think it means to say that something is a "rock"? We've NEVER thought that it had to mean "contains silicon", so of COURSE that won't matter. But, for example, if they had limestone where ever we have quartzite, it wouldn't make much sense to say that therefore limestone really is quartzite there, even if in terms of language use they still called it quartzite, would it?

      (By the way, what happens if you get a space ship and take a bottle of XYZ from twin-earth to earth? Does it act like water on earth? Does the “thought experiment” tell you that?)

      Um, you don’t understand thought experiments. They’re there to test concepts, not instances. So, in this case, we’d ask what might happen if we brought XYZ back to Earth, and there are two options. One is that, of course, it acts nothing like water. This isn’t all that interesting to the thought experiment, and if that happened it would even break your functionalist account (the other option works great). It’s also what would probably actually happen, given the sharp difference in physical composition. On the other hand, maybe it WOULD act just like water. This could be seen as an issue for the physical composition view — of water specifically, mind you; it’s not a generic view which is where I think you do wrong — except we can ask what we’d do if we found something that acted just like water but wasn’t anything AT ALL like H20. This would likely devolve back to the “fool’s gold” example; we wouldn’t call it water, although you could use it like water in most if not all circumstances. At the very least, this is not an unreasonable stance, which again highlights the “two easy resolutions, both of which generally work and both of which have oddities”.

      This — I suggest — is where the philosophers are going wrong, with a misconceived view of science. They are trying for a (Platonic essentialist?) view of what water “really” is, and of what account science “would prefer”, whereas science is pretty much pragmatic and functionalist. Even the description H2O is functionalist, as are all other descriptions in science.

      Science is the one that introduced the phrase and description “Water is H20”. Also, they tend to deal in natural kinds, which are a lot closer to Platonic Forms than they are to functionalist descriptions. For example, why is it that science NEVER describes whales as “fish”? Functionally, they are a lot more like fish than they are like elephants, but science considers them far more akin to elephants than to other fish.

      I also think that philosophy lacks a good mechanism for quality control.

      But you, as far as I know, haven’t actually worked in formal philosophy. What makes you qualified to discuss their standards of quality control? And since by that you’d have to find the discussions of those who agree with you equally likely, your only tack is to insist that you, a non-philosopher, know how to do philosophy and the details of philosophical discussions better than philosophers do. This … is not very likely.

      How about if I’m putting on a Christmas play, with three kings carrying incense, myrrh and gold. Wouldn’t a nugget of fool’s gold count as “gold” in that context? It seems to me that an eight-yr-old at a nativity play would by happy with it!

      You’re missing representations again. The fool’s gold REPRESENTS gold, there, in the same way as a rock painted yellow would. That doesn’t make it actually gold, even to the 8 year old.

      Yes it is pertinent. If I was on Earth, and wanted to make water, I’d need H and O. Having C and N wouldn’t work. Thus the composition is relevant in that context.

      By your standard, it’s not relevant to whether IT’S REALLY WATER OR NOT, which was my point; at the end of the day, you end up with something, by your definition, that’s still water. So whether it is H20 or XYZ is irrelevant to whether or not it’s water, which means that saying “Water is H20” is misleading at best and false at worst.

      Many philosophers do think they have a resolution, it’s just that the others don’t agree with them. 🙂

      But they disagree for good reasons, raising what do seem to be real problems with the view. You are dismissing that far too lightly.

      I don’t really agree (at least as the term is used in science). Something like an “inverse square law” is a concept, not a representation. Something like an electron’s “charge” or its “spin” or indeed its “mass” are all concepts.

      The inverse square law is a representation of a set of physical behaviours. “Charge” and “Mass” are representations of physical states. There are concepts associated with this, but the former is clearly not a concept in the philosophical sense that we’re talking about, and the latter MIGHT be depending on how YOU use them. But the theory of evolution is not A concept, and so you haven’t done anything to refute what I’ve said.

      These critiques of Krauss fail to realise that that is not what he did. He did not just consider one conception of “nothing” (ie “QV nothing”) and declare that he had solved the problem.

      1) My summary above is based on what I recall him saying in his own defense, although it could be wrong.

      2) As I went through that entire procedure you talked about, I’M certainly not accusing him of that, nor, in fact, is Albert. We’re criticizing his endpoint; how he got there isn’t all that interesting since that endpoint is, I presume you’ll agree, his purported answer to the problem.

      He very explicitly considered the ENTIRELY OBVIOUS point that “QV nothing” is not “nothing”. He pointed out that starting from “QV nothing” was obviously not “nothing nothing” and so went on to consider whether one could dispense even with the quantum vacuum and so start from a nothinger nothing.

      This … doesn’t help, because as described it sounds like he tunneled down to the last nothing that could be a something and stopped there, and then said that we can get something from that. But nothing in the question as ALWAYS meant “absolute nothing”, so even doing THAT ends up with him equivocating rather badly.

      As I pointed out, either he’s answering based on form 2 — I’ll find a nothing that can produce somethings — or form 1. If the former, then he needs to start from absolute nothing, not any nothing that remains a something . If the latter, then that we call it a nothing is irrelevant to justifying that that is the fundamental something. So either his answer is equivocating and so wrong, or else his procedure is wrong. Take your pick.

    26. Coel Post author

      I’m afraid I don’t understand this argument at all …

      What I’m saying is that saying that “water is really H2O” privileges one particular account of “water” over others. Why would “water is H2O” be privileged over “water is wet stuff that we drink when thirsty, that falls as rain, and that fish swim in”? Knowing that water is H2O is useful and important, but all the other statements about what water is are just as much what water “really is”.

      1) Um, we are still substituting what science considers to be hydrogen and oxygen in there, so that doesn’t destroy anything.

      Yes, but the only reason science considers them to still be hydrogen and oxygen is the *functionalist* stance that they are sufficiently like hydrogen and oxygen that we can call them that. Again, my argument here is that even the claim “water is really H2O” only makes sense in functionalist terms.

      The differences are not conceptual ones, but are instead differences in accidental properties, at least as they relate to oxygen, say. That’s pretty much what “sufficiently similar derivations” MEANS there; it’s not merely that they can be used in the same way.

      Again, I’ll have to disagree. Whether the differences in properties are “accidental” or “conceptual” is again dependent on context, and can again only be resolved from a functionalist perspective.

      The counter-argument says that there’s a real physical difference there, and that physical difference is enough to justify saying that they’re talking about two different things.

      This is a real physical difference between O-16 and O-18, and that physical difference is enough to justify saying that we’re talking about two different things. Now, whether they are *sufficiently* similar for a particular purpose can only be settled from a functionalist perspective.

      But, then, we never say that “crystal is physical composition X”, whereas we DO say “Water is H20”.

      But that’s only because of the fact that we know of only one way of making water, whereas we know of many ways of making crystal.

      or say that the difference is a meaningful one, …

      Yes, the difference is meaningful, but that meaning is only as above: that in one case there are multiple ways of making the stuff, and in the other case only one. All that Twin Earth does is move water from the latter category to the former. Which is of rather limited interest; it really does not have big philosophical consequences, when viewed from a functionalist perspective.

      why is it that science NEVER describes whales as “fish”?

      It’s just a matter of what concepts and categories are most useful, considering all the different contexts.

      But you, as far as I know, haven’t actually worked in formal philosophy. What makes you qualified to discuss their standards of quality control?

      Given the way philosophers feel free to comment on all sorts of areas, such as science, such questions are a little ironic! 🙂 But, there is sufficient overlap between areas of interest to philosophy and to science, that one can gain some evaluation of philosophy without having had formal training in philosophy. To pick one obvious example, the fact that theology is still considered an academic subject that is part of philosophy gives me license to doubt the quality control in philosophy.

      You’re missing representations again. The fool’s gold REPRESENTS gold, there, in the same way as a rock painted yellow would. That doesn’t make it actually gold, even to the 8 year old.

      But from a functionalist perspective, it *is* gold in that context. (Though that narrow context alone.)

      We’re criticizing his [Krauss’s] endpoint; how he got there isn’t all that interesting since that endpoint is, I presume you’ll agree, his purported answer to the problem.

      Krauss’s endpoint was open-ended. He didn’t claim to have answered the question, he merely presented how far down that road we had got. He fully accepted that he had not fully answered the question. (And, again, if you’re going mainly on Albert’s review, that review was very unfair in failing to note that Krauss had *explicitly* raised and discussed many of the issues that Albert claimed he was oblivious to.)

      This … doesn’t help, because as described it sounds like he tunneled down to the last nothing that could be a something and stopped there, and then said that we can get something from that.

      No, his discussion was much more open-ended than that.

      But nothing in the question as ALWAYS meant “absolute nothing”, so even doing THAT ends up with him equivocating rather badly.

      As he fairly points out, the concept of “nothing” can have a range of interpretations.

    27. verbosestoic

      What I’m saying is that saying that “water is really H2O” privileges one particular account of “water” over others. Why would “water is H2O” be privileged over “water is wet stuff that we drink when thirsty, that falls as rain, and that fish swim in”?

      What would that matter if that is what water ACTUALLY is? Unless you want to argue that there is no right concept of water and water is just whatever we consider water to be, then that we are privileging one interpretation over another isn’t a problem. And if you argue that, then you aren’t conceptualizing at all; you’re assigning words to things, which is not the same thing, and is a far too relativistic an idea to survive, or be compatible with science (you have to think that there are right answers to at least SOME questions [grin]).

      Again, I’ll have to disagree. Whether the differences in properties are “accidental” or “conceptual” is again dependent on context, and can again only be resolved from a functionalist perspective.

      I don’t think you understand what I mean when I talk about “conceptual” and “accidental” properties. Let me use the example of a computer. A conceptual property might be that it can, well, do computing. An accidental quality is that it is brown. So conceptual properties are the properties that the thing has to have to be an instance of the concept it is instantiating, and accidental properties are properties that it just happens to have, but that if they were different it would still be an instance of the same concept.

      According to science, O-16 and O-18 are still instance of the same concept. This would hold even if, on Twin Earth, they breathed in carbon and called it oxygen. If science is as good at conceptual analysis as you think it is, we definitely have to accept that they are reasonably right about that distinction.

      This is a real physical difference between O-16 and O-18, and that physical difference is enough to justify saying that we’re talking about two different things. Now, whether they are *sufficiently* similar for a particular purpose can only be settled from a functionalist perspective.

      There is actually a significant functional difference between O-18 and O-16, because of the extra neutrons and what you can do with them. The reason we consider them the same is that the number of neutrons in a nucleus is irrelevant to the concept of what makes a particular atom an instance of the atomic type that it is instantiating. This is a CONCEPTUAL distinction, not a functional one; what type of atom you have is determined by the protons in the nucleus. Unless you want to show me what functional argument you can make here …

      But that’s only because of the fact that we know of only one way of making water, whereas we know of many ways of making crystal.

      As I pointed out, you’d have to make an argument that what water is is more akin to crystals than it is to, say, oxygen or limestone. You haven’t done that.

      Crystal is defined as being a particular lattice-type structure, not as a specific physical composition. Specific TYPES of crystal, however, ARE defined that way. Water is akin to something like a specific type of crystal, or an acorn, or a specific type of rock, it seems.

      Yes, the difference is meaningful, but that meaning is only as above: that in one case there are multiple ways of making the stuff, and in the other case only one.

      You keep dodging my point that no matter how you make it, it still ends up being water, and so what it’s made of cannot be a conceptual property of water, if your view is correct.

      But we also know from salt water that if you can’t drink it, it’s still water, so that doesn’t work either. So you’re kinda stuck with the radical functionalist position that has some serious problems …

      But from a functionalist perspective, it *is* gold in that context. (Though that narrow context alone.)

      … and so, jumping ahead a little, this leads you to what is considered an absurdity: that fool’s gold, or the painted rock, is gold in that context. To make things even worse, with your radical functionalist view, anything that can be used in the same way is really the same thing. So just as that yellow painted rock is gold here, in any case where you want something reasonably hard that is yellow in colour, then gold is yellow painted rock. To return to water, if I want to quench my thirst, then not only would iced tea count as water, but water would count as iced tea. If I wanted to pound in a nail, a bottle and a shoe are hammers … but a hammer is also a bottle and a shoe. And you can’t add any OTHER properties to distinguish them there without my being able to say that in the Twin Earth case we ultimately need to include the “is H20” case to make them the same, and so they aren’t the same sort of substance.

      What makes philosophical questions interesting, I think, is that we often have obvious answers, but all of them lead to absurdities. In the Twin Earth case, the non-radical functionalist case seems obvious, but it leads to absurdities because we know that we can have many things that we can use for similar purposes, and so just that two things look alike and behave alike doesn’t necessarily mean that they are the same thing. But if we take the physical composition approach, then we hit the absurdity that these two things that act absolutely identically really AREN’T the same thing at all. The interesting philosophical question is to get to a point where we have an answer that doesn’t lead to any absurdities. This is … generally harder than you might think [grin].

      Given the way philosophers feel free to comment on all sorts of areas, such as science, such questions are a little ironic! 🙂

      Which makes me the only one in this conversation qualified to comment, because I have formal training in both [grin].

      But this raises a key point: you clearly seem to be frustrated when philosophers drift into science without having or respecting the background and training of scientists in that regard, so why do you not see how philosophers might reasonably feel the same way when the opposite occurs?

      To pick one obvious example, the fact that theology is still considered an academic subject that is part of philosophy gives me license to doubt the quality control in philosophy.

      On what grounds, other than that you don’t care for theology and are an atheist?

      Let me put it another way: what is the Problem of Evil? It has to, at a minimum, be philosophy of religion. It’s also critical to many atheistic arguments. What is Euthyphro? Why are these counters respectable, but arguments in favour of theism out of vogue? And, BTW, there are philosophical issues with both of those arguments.

      So, no, you need a better example than that [grin].

      He didn’t claim to have answered the question, he merely presented how far down that road we had got. He fully accepted that he had not fully answered the question.

      But then why, when philosophers accused him of not answering the question, was his reply not “I know! I wasn’t trying to!” instead of “You’re just turf protecting!”?

      At any rate, you need to address my actual argument against Krauss: if he was after a nothing that can produce something, his last point was a something, and if he was after a fundamental something, then chasing conceptualizations of nothing was meaningless. You’ve completely dodged that for a number of comments now.

      As he fairly points out, the concept of “nothing” can have a range of interpretations.

      But, again, we knew quite well what conception of nothing we were going for there. His approach, then, would be like us asking how we would be able to survive for two weeks on the ocean since we couldn’t drink sea water, and saying that we can conceive of water in many ways, including many that are drinkable. Sure, but since we were dealing with SALT WATER, none of them are relevant, are they? The same here: he can talk about “nothing” in a lot of ways, but unless he’s talking about “absolute nothing” none of them are actually relevant to the question.

    28. Coel Post author

      What would that matter if that is what water ACTUALLY is?

      I’m not sure what meaning attaches to the phrase “actually is”, as used there, unless it’s an unwarranted privileging of one particular description of water.

      Unless you want to argue that there is no right concept of water and water is just whatever we consider water to be …

      Of course “water” is what we consider water to be. In essence, “water” is a concept that models a feature of reality. That concept has lots of aspects (wet, drinkable, H2O, falls from sky, etc). Such concepts/models are adopted for utility, and are honed and adapted for utility. The “falls from sky” and the “wet and drinkable” aspects of the model are just as integral to the concept as the “H2O”.

      Privileging of one aspect as being what water “actually is” smells of Platonic essentialism or some similar weird philosophical concept that has never had any traction in science (with science being pragmatic and functionalist).

      According to science, O-16 and O-18 are still instance of the same concept.

      Agreed, but that against is a functionalist stance. They are instances of the same concept “oxygen” since, for many purposes, O-18 looks like O-16, waddles like O-16 and quacks like O-16.

      Once we introduce functionalism the twin-earth issue is easily resolved. If we don’t introduce functionalism but hold to some form of essentialism, then O-18 and O-16 are distinct.

      There is actually a significant functional difference between O-18 and O-16, because of the extra neutrons and what you can do with them.

      In some contexts, yes, and in other contexts, no. Functionalism is always context dependent.

      the number of neutrons in a nucleus is irrelevant to the concept of what makes a particular atom an instance of the atomic type that it is instantiating. This is a CONCEPTUAL distinction, not a functional one;

      But the whole concept of “atomic number” is functional. The whole reason that it depends on the number of protons but not on neutrons is that protons have an electric charge and thus the number of protons affects the chemical properties whereas the number of neutrons does not.

      Absolutely every concept in science is functional. Science is about pragmatically modelling data, and the test of such models is whether they function.

      … and so, jumping ahead a little, this leads you to what is considered an absurdity: that fool’s gold, or the painted rock, is gold in that context.

      Yes! (Err, why is that absurd?)

      To make things even worse, with your radical functionalist view, anything that can be used in the same way is really the same thing.

      I’m not sure what you mean by “really” the same thing. It’s that thing *in* *that* *context*, but only in that context. There will be lots of other contexts.

      And you can’t add any OTHER properties to distinguish them there …

      I’m allowed to introduce other properties if we’re considering other contexts.

      The interesting philosophical question is to get to a point where we have an answer that doesn’t lead to any absurdities.

      I’m sorry, but I don’t see anything absurd about the thorough-going functionalist perspective.

      But this raises a key point: you clearly seem to be frustrated when philosophers drift into science without having or respecting the background and training of scientists in that regard, so why do you not see how philosophers might reasonably feel the same way when the opposite occurs?

      Why sure, of course they’ll feel that way! But that doesn’t mean their expertise trumps that of scientists in areas of mutual interest.

      At any rate, you need to address my actual argument against Krauss: if he was after a nothing that can produce something, his last point was a something, and if he was after a fundamental something, then chasing conceptualizations of nothing was meaningless.

      Two points about Krauss.

      First, regarding starting with a vacuum which contains quantum fields. His point was that in a quantum-*gravity* scenario, the vacuum, the space and time itself, and thus the quantum fields themselves, are *also* part of the quantum-gravity fluctuation. Thus one might not need any pre-existing quantum vacuum. The whole fluctuation could be self-contained in the sense of not arising from anything pre-existing (not even quantum fields). In that sense of not being dependent on *anything* else, it would be a universe “from nothing”.

      Note that Albert didn’t even attempt to deal with this part of Krauss’s argument.

      Secondly, Krauss made no claim to have proven or resolved the issue, partly since we don’t have a working theory of quantum gravity and so the above suggestion is speculative. He was thus, in the end, setting out where we had got to and setting out a programme for going further.

      Albert’s response was complete unfair. The trouble with philosophers commenting on physicists is that they tend to go instantly into sneer mode. Well, Krauss’s book was a lot more considered, thoughtful and perceptive than Albert’s response to it.

    29. verbosestoic

      I’m not sure what meaning attaches to the phrase “actually is”, as used there, unless it’s an unwarranted privileging of one particular description of water.

      Okay, let’s go back to Twin Earth:

      If a person X from Earth wakes up on Twin Earth, and talks about “water”, are they saying the same thing as the person Y who was born on Twin Earth?

      You claim that you have a perfectly good answer to this which is why you don’t care about the experiment, but your answer to this question is “I don’t know”, which doesn’t seem like all that enlightening an answer [grin]. Even worse, your REAL answer is actually something like “If they’re talking about the physical composition, no, but if they’re talking about the functional properties, then yes”, which is a lot closer to the compositional view than it is to the functionalist view, as you concede that there IS an important difference between the two substances and thus the two usages. The only difference is that you don’t think it determinative, but they do … and you don’t really have any good reason to think that it isn’t determinative.

      Thus, the question: what does whatever it was that I questioned have to do with what properties determine what it means when we say “water”?

      In essence, “water” is a concept that models a feature of reality. That concept has lots of aspects (wet, drinkable, H2O, falls from sky, etc). Such concepts/models are adopted for utility, and are honed and adapted for utility.

      So … we started down this path because I accused science of not being as conceptual as philosophy because it adopts them strictly as models of reality and for utility and so doesn’t really care about purely conceptual issues, and raised the Twin Earth experiment as an example of a purely conceptual problem that science wouldn’t care about for those very reasons. Your defense of your answer to that here is to insist that for science concepts are adopted as models of reality and for strictly utilitarian reasons. So if we agree, why are we arguing [grin]? The only objection you could have is that I denigrate that sort of process, except that I specifically said that that was perfectly fine, but not what philosophy was interested in. You could also argue, I guess, that philosophy doesn’t do conceptualizing right, but without an agreed upon purpose to that we probably aren’t going to get anywhere.

      Philosophical conceptualizing will impede scientific work. Scientific conceptualizing typically impedes philosophical work. The only problem is when you try to take one approach and directly apply it to the other area. But even from your statement, science cares less about concepts than philosophy does, because philosophy cares about them as ends in themselves, while science treats them solely as means to another end, understanding empirical reality. But as long as science doesn’t insist on saying that they’ve discovered what the concept “really”, if they want to stick with the rough-and-ready concepts, that’s fine.

      Agreed, but that against is a functionalist stance. They are instances of the same concept “oxygen” since, for many purposes, O-18 looks like O-16, waddles like O-16 and quacks like O-16.

      And in many contexts they DON’T. But they’re still called “oxygen” by science in all of those cases. At most, you have to specify a specific TYPE of oxygen, but that’s not enough to claim that they don’t still consider it “oxygen”. So how do they decide what functions to choose to determine whether it should be considered “oxygen” or not?

      Once we introduce functionalism the twin-earth issue is easily resolved. If we don’t introduce functionalism but hold to some form of essentialism, then O-18 and O-16 are distinct.

      But they ARE distinct. Even your functionalism has to accept that in some way at some point. Essentialism says that they are both oxygen because they contain all the conceptual properties of oxygen and only differ in properties that are accidental to the concept oxygen, but O-16 and O-18 also are instances of different sub-concepts of oxygen because some of those accidental properties wrt oxygen are NOT accidental when it comes to O-16 and O-18, and are not individual instance properties (ie merely the properties of an individual atom).

      Do you know anything about object-oriented programming? I’ve always thought of that as a good example of the relationship, at least as I see it.

      Yes! (Err, why is that absurd?)

      1) Ask the eight-year-old and they’ll say that, yes, it’s absurd.

      2) As I pointed out and you didn’t address, it also means that real gold would be fool’s gold in that context, which makes this meaningless.

      I’m not sure what you mean by “really” the same thing. It’s that thing *in* *that* *context*, but only in that context. There will be lots of other contexts.

      Take this thought experiment:

      A person stumbles into your camp in the desert, dying of thirst. They are hallucinating that your camp is a lemonade stand, and desperate to avoid dying of thirst, they croak, “Lemonade!”. Now, not being a lemonade stand, you don’t have any, so you bring them some water. By your functionalist view:

      1) It is reasonable to say that they asked for water, since all they wanted was something to stop them from dying of thirst, and water and lemonade are indistinguishable in that context.

      2) It is also reasonable to say that you brought them lemonade, for the same reasons.

      This seems absurd, but I can’t for the life of me figure out how you avoid it given your starting point.

      I’m allowed to introduce other properties if we’re considering other contexts.

      But we weren’t. In THAT context, a bottle and a hammer were the same thing and so I could call the bottle a hammer and the hammer a bottle and remain perfectly reasonable. But if I can’t do that, then you have NO answer to the Twin Earth problem. So, again, an absurdity.

      First, regarding starting with a vacuum which contains quantum fields. His point was that in a quantum-*gravity* scenario, the vacuum, the space and time itself, and thus the quantum fields themselves, are *also* part of the quantum-gravity fluctuation. Thus one might not need any pre-existing quantum vacuum. The whole fluctuation could be self-contained in the sense of not arising from anything pre-existing (not even quantum fields). In that sense of not being dependent on *anything* else, it would be a universe “from nothing”

      No, it wouldn’t. It would be a universe that didn’t have a specific cause, or that caused itself. The QV itself has to already exist for a fluctuation to happen. Also, this very much looks like a kludge, since the QV creates the universe and so the universe is also caused. You have to get to a very specific notion of cause to eliminate a cause just because time doesn’t exist yet.

      So, still not relevant to the initial question: Krauss is still assuming a pre-existing something, but limits himself to talking about ones that we call “nothings”.

      Secondly, Krauss made no claim to have proven or resolved the issue, partly since we don’t have a working theory of quantum gravity and so the above suggestion is speculative. He was thus, in the end, setting out where we had got to and setting out a programme for going further.

      The objection is not that his answer is speculative. The objection is that his answer ISN’T AN ANSWER. If the QV is a something, then he didn’t get something from nothing, no matter what we call it. Thus, again, he EQUIVOCATES. And Albert was quite right to point out that just because we thought that the vacuum of space is a nothing doesn’t mean that he can take that nothing and say that we were wrong to say that nothing can’t produce things because we once thought that the vacuum was nothing and, look, it produces things! You’ve done nothing to defend Krauss’ approach here.

    30. Coel Post author

      Hi verbosestoic,
      You asked me what water “actually is”, and this led to your question:

      If a person X from Earth wakes up on Twin Earth, and talks about “water”, are they saying the same thing as the person Y who was born on Twin Earth?

      Let me answer by asking a related question:

      If a 5-yr-old boy on Earth talks about “water”, is he saying the same thing as an Earthly adult who talks about “water”? My answer: the word has a related but not identical meaning in the two cases.

      The 5-yr-old means “wet liquid stuff I drink” (plus other connotations), but does not know about the H2O composition so does not mean that. The adult means all the things the child does, but also means extra things, since he knows much more about water, including the composition.

      My answer to your question would be along the same lines: their meanings would be mostly the same, but not identical. In none of these cases is it sensible to ask what water “actually is”; what it is “is” is a whole set of concepts (including “wet”, “drinkable” etc) that will differ from person to person.

      Even worse, your REAL answer is actually something like “If they’re talking about the physical composition, no, but if they’re talking about the functional properties, then yes”, which is a lot closer to the compositional view than it is to the functionalist view, …

      Or, rather, I’d say that even the issue of the composition can only be understood from a functionalist stance, as my point about O-18 v O-16 demonstrates.

      But they’re still called “oxygen” by science in all of those cases. At most, you have to specify a specific TYPE of oxygen, but that’s not enough to claim that they don’t still consider it “oxygen”. So how do they decide what functions to choose to determine whether it should be considered “oxygen” or not?

      The labels are chosen because they are useful, and that means from a functionalist perspective. The reason we have a bigger name change when adding a proton (with an electric charge) than we do when adding a neutron, is that the change in proton number has a bigger effect on the chemical properties (ie. what is important to us most of the time) than the neutron number. Thus the whole label “oxygen” is functional.

      But they ARE distinct. Even your functionalism has to accept that in some way at some point.

      Sure, functionalism says that for some functions they are distinct.

      Essentialism says that they are both oxygen because they contain all the conceptual properties of oxygen and only differ in properties that are accidental to the concept oxygen, …

      Yuck, yuck yuck! What a horrible concept! The distinction between “conceptual properties” and “accidental properties” is totally ad hoc and only makes sense from a functionalist perspective (ie. “conceptual” properties change the function more than “accidental” ones).

      1) Ask the eight-year-old and they’ll say that, yes, it’s absurd.

      That’s because even the eight-yr-old knows about all the other contexts. I still don’t see what is absurd about saying that strictly limited to that context fool’s gold is gold.

      By your functionalist view: 1) It is reasonable to say that they asked for water, since all they wanted was something to stop them from dying of thirst, and water and lemonade are indistinguishable in that context.

      Yes.

      2) It is also reasonable to say that you brought them lemonade, for the same reasons.

      If we narrow down the context enough then yes! Of course the reason that seems absurd is that we’re all aware of the other contexts.

      The QV itself has to already exist for a fluctuation to happen.

      We don’t know that. It may be that, for a quantum *gravity* fluctuation, we do not need a pre-existing QV. We don’t know since we don’t have a working theory of quantum gravity.

      The objection is that his answer ISN’T AN ANSWER. If the QV is a something, then he didn’t get something from nothing, …

      And, again, his suggestion is that, in a quantum-*gravity* fluctuation, we do not even need a pre-existing QV.

    31. verbosestoic

      If a 5-yr-old boy on Earth talks about “water”, is he saying the same thing as an Earthly adult who talks about “water”?

      Yes, but they don’t really know what water is, because they don’t know that it’s H20.

      My answer to your question would be along the same lines: their meanings would be mostly the same, but not identical. In none of these cases is it sensible to ask what water “actually is”; what it is “is” is a whole set of concepts (including “wet”, “drinkable” etc) that will differ from person to person.

      If this is the case, then there is no meaning for the term “water”, because in theory every single person will have a different “meaning” for it in their heads. This … is not what we wanted, because it implies that when you use the word water and _I_ use the word water we DON’T mean the same thing, because surely the meaning of the term “water” in my head is the sum of all of my associations with it, as you yourself imply. If we have different associations, we have different meanings.

      The labels are chosen because they are useful, and that means from a functionalist perspective.

      Following on from this and from the above, if a certain label is less useful to me and a different one is more useful, then I ought to be able to personally adopt the different labeling, otherwise you’d be privileging one set of functions, which is what you accused ME of doing as if that was a negative. Thus, for example, if I am someone mostly interested in harvesting food from the sea, then classifying a whale as a “fish” is more useful to me than classifying it as a “mammal”, and so I should then conclude, with perfect validity, that whales are really fish. Also, if the differences between O16 and O18 are critical to my work, it should be perfectly valid for me to say that O16 and O18 are not, in fact, both still “oxygen”, despite the cases where they DO function the same. At this point, there is no such thing as a natural kind, and all assignments of concepts are PERSONALLY subjective, and based entirely on what is important to me. This would definitely get confusing … and clearly ISN’T what we do.

      Yuck, yuck yuck! What a horrible concept! The distinction between “conceptual properties” and “accidental properties” is totally ad hoc …

      I don’t think you understand what “conceptual properties” and “accidental properties” are, because if there is one thing they are not it is “ad hoc”.

      Okay, let’s take the concept “Car”. The conceptual properties are the properties that we use to identify whether something IS a car at all, and the accidental properties are the properties that a car might happen to have but that can be different — or potentially even missing — while the instance remains a car. So, let’s take “is red”. Cars, of course, can be red, but if I come across an instance of a car that is red, it being red is just what colour is happens to be; it’s not the case that if a car wasn’t red, it wouldn’t be a car anymore. Now, we can sub-concept Car to Red Car, and for that concept it being red is now a CONCEPTUAL property, because all instances of Red Car will have to be red, or else they aren’t red cars. (This is why I asked about OO computing, because this is REALLY clear in OO). And so on. We might get into issues with properties that identify specific instances — ie MY Car — but that’s good enough for now.

      So the properties are not at all ad hoc. The conceptual properties are properties that all instances of that concept both will and MUST have to count as instances of that concept, while the accidental properties are properties that either only some instances of the concept will have, or else are ones that the instances could be missing entirely — even if in general they are always present — and the instance would still be an instance of the concept. In theory, we can objectively derive all of the conceptual and accidental properties for any concept using this method.

      Thus, it is not at all ad hoc. By contrast, your functional model is entirely ad hoc, because it relies on determining what functionality is most “useful”, which is not objective, even by your own arguments. Thus, that distinction is entirely ad hoc in a way that the conceptual/accidental distinction isn’t.

      The problem with this distinction is not that it is ad hoc, but that it requires us to find the RIGHT distinctions, which, as you know, can be rather difficult.

      That’s because even the eight-yr-old knows about all the other contexts.

      Actually, they need not know about the other specific contexts, but only know that the names are different. But since that’s how we learn everything, that will apply to all other cases, too, including the Twin Earth case, and so if it is reasonable for them to say that fool’s gold is not gold, then it is reasonable to them to say that Earth water is not Twin Earth water, either. So we need the ability to be precise and have one answer. So, in the context of the play, you would have to say that the 8 year old OUGHT to call the fool’s gold gold, and consider fool’s gold and gold to be the same thing. But no one actually ever does that; even if the teacher says “Get your gold” everyone knows that what she means is to get the fool’s gold that is representing the gold. So, for your stance to work, you have to essentially claim that we really mean what we consciously seem to KNOW that we don’t mean. That, then, is an absurdity.

      If we narrow down the context enough then yes! Of course the reason that seems absurd is that we’re all aware of the other contexts.

      The reason it seems absurd is two-fold:

      1) It’s not what we do; in those cases, we would know that the person didn’t ask for water (although they should have) AND know that we didn’t bring them lemonade.

      2) We know that there are a number of what we would call different objects that are functionally equivalent in some contexts, and so don’t think that functional equivalence really means that they are the same thing. Functional equivalence, in short, is not IDENTITY wrt concepts or even things.

      In short, it is absurd because it is not how we, in fact, view those things. Thus, if you want to insist that that is really the case, you need some extraordinary evidence to demonstrate that you are right and (almost) everyone else is wrong.

      We don’t know that. It may be that, for a quantum *gravity* fluctuation, we do not need a pre-existing QV. We don’t know since we don’t have a working theory of quantum gravity.

      Thus, his answer is blind speculation and no better than the “It’s God!” answers of the theists, except that they KNOW that they are dealing with a something, and roughly what something they are proposing.

      And, again, his suggestion is that, in a quantum-*gravity* fluctuation, we do not even need a pre-existing QV.

      Do we still need something that would count as a “something”? ‘Cause if he could prove we didn’t (even necessarily), he’d actually HAVE an answer to the question, and if he can’t then he’s still positing a something that does all of this, but can’t even give us a GUESS at what it might be . Again, I fail to see how he gets any further than where we are with this move.

    32. Coel Post author

      Yes, but they don’t really know what water is, because they don’t know that it’s H20.

      Which, again, privileges one particular account of what water “is” over the others.

      Question: would you say that people in the 15th century didn’t know what water was? If the answer is yes, would you then say that scientists in the year 1900 didn’t know what water was, because, despite knowing that it was H2O, they didn’t know what oxygen was made of? How about scientists in the year 1950, who knew that water was H2O and knew what hydrogen and oxygen were in terms of protons and neutrons, but didn’t know what protons and neutrons were made of?

      How about us? Do we know what water is? We know what protons and neutrons are in terms of quarks and gluons, but there’s plenty we don’t know about what quarks are. Are they string-theory “strings”? We don’t know. Anyhow, how have you decided which of these accounts to privilege?

      If this is the case, then there is no meaning for the term “water”, because in theory every single person will have a different “meaning” for it in their heads.

      But those different “meanings” will have sufficient overlap between people that the concept will work and be useful. You’ve accepted that 5-yr-olds and adults mean different things by it, and yet 5-yr-olds and adults manage to talk to each other.

      because surely the meaning of the term “water” in my head is the sum of all of my associations with it, as you yourself imply. If we have different associations, we have different meanings.

      Yes. Is that a problem, provided the meanings overlap sufficiently to be useful?

      if a certain label is less useful to me and a different one is more useful, then I ought to be able to personally adopt the different labeling, …

      Sure! And science and other fields often adopt a technical meaning of a term that is different from how it used outside the field. Further, sub-fields might adopt their own even more specialised meanings. For example, what an astronomer calls a “metal” differs from what the rest of physics means by it — and they do that for good functional reasons!

      … if I am someone mostly interested in harvesting food from the sea, then classifying a whale as a “fish” is more useful to me than classifying it as a “mammal”, and so I should then conclude, with perfect validity, that whales are really fish.

      You could indeed do that, and so long as everyone was clear on the different meanings it would be fine.

      Also, if the differences between O16 and O18 are critical to my work, it should be perfectly valid for me to say that O16 and O18 are not, in fact, both still “oxygen”, despite the cases where they DO function the same.

      Yes, sure! In the same way, people might adopt different names for H-1, H-2 and H-3. Maybe “hydrogen”, “deuterium” and “tritium” for instance.

      At this point, there is no such thing as a natural kind, and all assignments of concepts are PERSONALLY subjective, and based entirely on what is important to me.

      Nature would still show patterned-ness, so there would still be natural kinds. Whether your language aligned with the natural kinds would then be up to you.

      This would definitely get confusing … and clearly ISN’T what we do.

      I would suggest that it is indeed what we largely do. Except that the whole point of language is to communicate, and thus it is useful to abide by common conventions, rather than to continually have to explain your own personal usages.

      The conceptual properties are the properties that we use to identify whether something IS a car at all, and the accidental properties are the properties that a car might happen to have but that can be different — or potentially even missing — while the instance remains a car.

      This seems to me a very functional concept. Let me guess, the “conceptual” properties are those that affect the basic function of the car, whereas “accidental” ones, such as the colour of the paintwork do not?

      So, in the context of the play, you would have to say that the 8 year old OUGHT to call the fool’s gold gold, and consider fool’s gold and gold to be the same thing. But no one actually ever does that; even if the teacher says “Get your gold” everyone knows that what she means is to get the fool’s gold that is representing the gold.

      This all seems fine to me. The only reason that it seems a bit weird is that we would all know about the other contexts in which fool’s gold is not gold. But, if you took a naive 8-yr-old who did not previously know about fool’s gold, you would not have to explain the difference.

      In short, it is absurd because it is not how we, in fact, view those things.

      The only reason that we do not take that view is that we’re inevitably aware of a much wider range of contexts, and so we adopt language suitable for that wider array of contexts.

    33. verbosestoic

      Which, again, privileges one particular account of what water “is” over the others.

      Since the choices are to have one “privileged” account of water — ie a “real” meaning — or else to have it be completely open, I’m comfortable with that. YOU, on the other hand, seem to constantly flip-flop between the two.

      Question: would you say that people in the 15th century didn’t know what water was? If the answer is yes, would you then say that scientists in the year 1900 didn’t know what water was, because, despite knowing that it was H2O, they didn’t know what oxygen was made of? How about scientists in the year 1950, who knew that water was H2O and knew what hydrogen and oxygen were in terms of protons and neutrons, but didn’t know what protons and neutrons were made of?

      Given the ACCOUNT I’m using here, I’d say that in the first case they didn’t, because water is critically H2O, so until they knew that they didn’t know what water was. In the other cases, however, I’d say that they DID know what water was, but didn’t know what oxygen and protons and the like really were. This is so natural a phrasing that you fell right into it in your comment.

      To compare it to my other examples, the last two questions are like saying that you know what the concept of a “car” is, but because you don’t know what steel is you don’t know what a car is (because a car is made of steel). That doesn’t seem relevant to the concept of “car”, nor does your other examples seem relevant to the concept of “water”.

      You’ve accepted that 5-yr-olds and adults mean different things by it, and yet 5-yr-olds and adults manage to talk to each other.

      You’re conflating reference and meaning/concepts again. 5 year olds can REFERENCE water in a way that adults can understand, but they don’t really know what it means for something to be water — which holds true for a lot of things.

      Yes. Is that a problem, provided the meanings overlap sufficiently to be useful?

      Well, yes, because it implies, for example, that we NEVER mean the same thing when we use the word, which means that the person travelling to Twin Earth really DOESN’T mean the same thing as anyone there does … which is pretty much the conclusion you REJECTED, isn’t it? Taking this, saying that “Water means X” is an utterly meaningless, useless, and misleading statement … but all of our language AND conceptual analysis AND science is based on that NOT being the case. If you’re violating our usage of terms that badly, it would behoove you, I think, to go back and question whether your answer is right or not.

      And certainly, given that, your claim that you had a simple and obvious answer seems false. There is much philosophy in that answer, and much questionable philosophy at that.

      Sure! And science and other fields often adopt a technical meaning of a term that is different from how it used outside the field.

      Technically terms are (ahem) technically DIFFERENT WORDS, that don’t mean the same thing as the non-technical usages. So you wouldn’t be talking about the same word/concept at all, strictly. This is how the fallacy of “equivocation” gets started.

      Yes, sure! In the same way, people might adopt different names for H-1, H-2 and H-3. Maybe “hydrogen”, “deuterium” and “tritium” for instance.

      Are they all still hydrogen, or not? If the former, then this is just parent concept->sub concept linking, as I talked about in my discussion of concepts. If not, then they are all expressing different concepts, and so are not the same thing at all. Take your pick.

      Nature would still show patterned-ness, so there would still be natural kinds. Whether your language aligned with the natural kinds would then be up to you.

      But we could easily argue that — even scientifically — the right concepts are those that are empirically validated — ie align with the real patterns in nature — and so the natural kinds are just the RIGHT meaning for those phenomena, at which point saying “Water is H20” is describing what it REALLY means. So if these patterns exist and are “real” in that sense, then why wouldn’t we say that what the thing is really IS that pattern? Even if functionally it might be more useful to not do that, it’s quite easy to say that the RIGHT answer is the one that aligns with the natural kind. Otherwise, you seem to justify responses saying “God is love” is that makes sense to a theist.

      I would suggest that it is indeed what we largely do.

      Nope. Science insists on using terms that align with natural kinds, and as we’ve seen people insist on treating fool’s gold as not being gold even when they could be used interchangeably. In fact, there are SO many cases in this sub-thread where you’ve had to deny what people would explicitly claim that I see absolutely no evidence that people do this at all. So, to make this claim, you had better provide some evidence, methinks.

      Let me guess, the “conceptual” properties are those that affect the basic function of the car, whereas “accidental” ones, such as the colour of the paintwork do not?

      Why are you “guessing” here? I went through an example in detail, and pointed that the concept “Red Car” HAS the paint colour as a conceptual property. It’s hard to explain anything to you if you aren’t going to read it.

      What I’m talking about here is neither functional nor compositional. The bare bones approach really is nothing more than asking “If an instance of this concept didn’t have this property, would it still be an instance of that concept?” Things get more complicated than that, but that’s the starting point. Some of those properties will be functional. Some will be compositional. At the end of the day, it’s all about, really, what it really MEANS to be an instance of that concept.

      But, if you took a naive 8-yr-old who did not previously know about fool’s gold, you would not have to explain the difference.

      But we would, because treating the two of them as being the same is considered to WRONG; fool’s gold is not gold. You can assert that it’s just that we know of more contexts, but at the end of the say we say that fool’s gold is not gold, end of story. That you so contradict our every day usage doesn’t bode well for your theory unless you can provide a really good reason for adopting it, and again it is clear that it is not a simple and obvious answer to the Twin Earth problem.

    34. Coel Post author

      Since the choices are to have one “privileged” account of water — ie a “real” meaning — or else to have it be completely open, …

      Why can’t we have “water” as referring to a package of concepts, where that package is (to some extent) dependent on context?

      I’d say that in the first case they didn’t [know what water was]

      Isn’t that a rather weird thing to say about people who would spend their lives drinking water, washing in it, cooking with it, watching it fall from the sky, et cetera? If you gave them a jug of water, a rock, a piece of wood, and a ball of wool, they’d have been quite capable to telling you which one was water.

      the last two questions are like saying that you know what the concept of a “car” is, but because you don’t know what steel is you don’t know what a car is (because a car is made of steel).

      But you’re the one insisting that what it is made out of the privileged account. Would a car made out of aluminium or carbon fibre also be a “car”? Or is “what it is made out of” privileged with regard to water but not with regard to cars?

      Well, yes, because it implies, for example, that we NEVER mean the same thing when we use the word, …

      But why is that a problem, so long as the concepts overlap sufficiently? You’re right, no two people would mean 100% exactly the same by it, but so what? So long as the two people’s concepts overlapped sufficiently to disambiguate, then the concept works.

      which means that the person travelling to Twin Earth really DOESN’T mean the same thing as anyone there does … which is pretty much the conclusion you REJECTED, isn’t it?

      No, most of the bag of concepts the Earthling means by “water” would overlap very well with most of the concepts that the Twin-Earthling means by “water” — so, to a very large extent, they both mean the same thing by it.

      Any problem arises only given that insistence that “water” can only mean one very particular thing, and that is your account, not mine.

      Are they all still hydrogen, or not?

      That depends on the context! You are very essentialist in your questions!

      But we could easily argue that — even scientifically — the right concepts are those that are empirically validated — ie align with the real patterns in nature — and so the natural kinds are just the RIGHT meaning for those phenomena, …

      Agreed.

      … at which point saying “Water is H20” is describing what it REALLY means.

      No, because all the other concepts relating to water (drinkable, wet, falls as rain, fish live in it) also “align with the real patterns in nature” and so are just as much “what water means”.

      So if these patterns exist and are “real” in that sense, then why wouldn’t we say that what the thing is really IS that pattern?

      We do, but all those other concepts about water are just as natural, just as much real features of nature.

      The bare bones approach really is nothing more than asking “If an instance of this concept didn’t have this property, would it still be an instance of that concept?”

      Fine, but I’m suggesting that the very definitions of those concepts are functional.

      You can assert that it’s just that we know of more contexts, but at the end of the say we say that fool’s gold is not gold, end of story.

      Yep, and we say that because of all the other contexts.

      That you so contradict our every day usage …

      That’s simply because everyday usage takes account of the range of contexts encountered in daily life.

    35. verbosestoic

      Why can’t we have “water” as referring to a package of concepts, where that package is (to some extent) dependent on context?

      Why WOULD we?

      If you gave them a jug of water, a rock, a piece of wood, and a ball of wool, they’d have been quite capable to telling you which one was water.

      Sure, they’d have known how to REFER to it, but not what it really was. Someone could learn enough about oxygen to be able to, say, identify it and say that it’s the think we breathe in and yet not know that it and carbon are actually different gasses, or that it’s in water, or what atomic number it has, and so on. Do we really understand what oxygen is if we don’t understand it’s atomic number or, even worse, think that it has a different atomic number than it actually has?

      But you’re the one insisting that what it is made out of the privileged account. Would a car made out of aluminium or carbon fibre also be a “car”? Or is “what it is made out of” privileged with regard to water but not with regard to cars?

      What a car is made out of it not a conceptual property of a car. That a car is made of steel has no relation to what a car is, as shown by the fact that it is clear to us that a car made from aluminum or even ice is still a car. But for water, we tend to think that what it means to be water is that it’s H20. It’s certainly not its functional properties since almost all of those are a) shared by other things and b) if someone for some reason couldn’t use water for that function even to THEM it’d still be water.

      Again, we have to retreat to the question: if an instance doesn’t have property X, is it still an instance of that concept? That a car is made of steel isn’t, that water of made of Hydrogen and Oxygen is, but the properties of hydrogen and oxygen are properties of THEM, not of water.

      But why is that a problem, so long as the concepts overlap sufficiently?

      That we all actually have different concepts of the same thing is a problem, even if they sufficiently overlap. Communication relies on transferring concepts from one person to another. If a different concept ends up on the other side, we’re stuck with saying exactly what I said: we don’t mean the same thing when we try to express a concept. Thus, there is NO meaning for a word or concept. That’s … rather eccentric, to say the least, and there seems to be no reason to actually take that line. You have given no reason to think that there isn’t a “right” concept that we learn or ought to learn, and taking that tack leads to massive issues that you have failed to address. So I return to the question above: why in the world would we ever want to accept your view?

      No, most of the bag of concepts the Earthling means by “water” would overlap very well with most of the concepts that the Twin-Earthling means by “water” — so, to a very large extent, they both mean the same thing by it.

      No, they don’t, because what they mean when they use the word is THEIR concept of water, which is not ours, and arguably no one shares the same concept. Thus, they never mean the same thing. They can both refer to water objects in the world, but always mean different things when they talk about water. That’s the inescapable conclusion that you’re trying to dodge by insisting that, well, it’s close enough. It’s not.

      That depends on the context! You are very essentialist in your questions!

      I repeat my explanation of the consequences of each answer:

      If the former, then this is just parent concept->sub concept linking, as I talked about in my discussion of concepts. If not, then they are all expressing different concepts, and so are not the same thing at all. Take your pick.

      Remember, if you take your extreme view then in some contexts lemonade and water are the same concept, which is somewhat absurd since even in those contexts we know that they express different concepts that can have the same functional use in some contexts. That’s a much simpler theory as well than the one you’re talking about.

      Fine, but I’m suggesting that the very definitions of those concepts are functional.

      You need to look back at my “Red Car” example to show why you can’t suggest that without ignoring my entire argument.

    36. Coel Post author

      Why WOULD we?

      Because it makes more sense than privileging one particular aspect of water and declaring that that is what water “really is”.

      But for water, we tend to think that what it means to be water is that it’s H20.

      I agree that that is one important aspect of what water “is”, but I don’t see what is gained from declaring that aspect to be privileged such that it is what water “really is”.

      That a car is made of steel isn’t, that water of made of Hydrogen and Oxygen is, …

      No, because water made of XYZ is also “water”!

      That we all actually have different concepts of the same thing is a problem, even if they sufficiently overlap. Communication relies on transferring concepts from one person to another.

      Sure, but we encounter and cope with this problem all the time. For example, what two different people mean by the concepts “justice”, “poverty”, “safe”, “large”, et cetera are all different. Yet we cope; all that we need is for the concepts to overlap sufficiently that we can use them for communication.

      … we’re stuck with saying exactly what I said: we don’t mean the same thing when we try to express a concept. Thus, there is NO meaning for a word or concept.

      That would mean there is no meaning for the word “justice” because different people mean different things by it.

      You have given no reason to think that there isn’t a “right” concept that we learn or ought to learn, …

      Words and concepts are human creations, created to be useful. The idea of a “right” concept, as oppose to a useful one, is rather bizarre.

      … and taking that tack leads to massive issues that you have failed to address.

      I’m under the impression that I have addressed them; it’s just that you dissent from my replies!

      No, they don’t, because what they mean when they use the word is THEIR concept of water, which is not ours, and arguably no one shares the same concept. Thus, they never mean the same thing.

      You could say exactly the same thing about “justice”, “poverty” and all sorts of other words that don’t have precise definitions.

      Remember, if you take your extreme view then in some contexts lemonade and water are the same concept, which is somewhat absurd …

      Yes, but I don’t regard it as absurd. If someone sees something just catching fire, and shouts: “quick, pour some water on it!”, then if someone pours lemonade then that meets the request. It is sufficiently watery to be “water” in that context. Presumably you’d agree if it were sea water? Yet sea water is just water with added salts whereas lemonade is just water with added stuff.

  8. Ron Murphy

    verbosestoic,

    When theists wonder why atheists don’t take the notion of God seriously, they sometimes ask questions that imply the possibilities are a) God (their God, usually), or b) a long list of naturalistic ideas, like variations on multiverses, our lone and only universe, eternal universes or finite universes, …

    Not only do they tend not to consider twin gods, good gods and bad gods, committees of gods, an ancestry of gods begat gods, with all these supernatural gods creating and tinkering with universes of naturlaistic entities, they also don’t consider the hypernatural and hypergods that create supernatural and supergods, to any level one might imagine. If any of those alternatives are true we seem to have been lumbered with a particularly crappy god.

    But in that vein, is it worth doing meta-metaphysics, about the really real reality? How would you know it’s worth it, other than the mental satisfaction of persuading onesself that there is a real reality beyond the mundane reality science shows, or a really real reality? Isn’t metaphysics just imaginative speculation with no means of knowing if one is being clever or dumb?

    Don’t get me wrong, I’m all for metaphysics, and the use of speculative imagination generally. It is not only an unbounded outlet for curiosity that requires no special skills, it might also occasionally lead to some interesting insight that in turn leads to some useful insight in science. But outside those specific uses of satisfying curiosity and potentially useful insight, is there any reason to take it more seriously, as if it’s actually doing anything, actually showing anything?

    “… but how can we know if it REALLY is? In short, how do we justify that what that tells us reflects reality at all?”

    Indeed. Metaphysics does not answer that any better than science does – in fact it is a lot worse. Metaohysics can’t even tell us if that’s a remotely meaningful notion, ‘how things really [really, really, …] are’.

    Reply
    1. verbosestoic

      Not only do they tend not to consider twin gods, good gods and bad gods, committees of gods, an ancestry of gods begat gods, with all these supernatural gods creating and tinkering with universes of naturlaistic entities, they also don’t consider the hypernatural and hypergods that create supernatural and supergods, to any level one might imagine. If any of those alternatives are true we seem to have been lumbered with a particularly crappy god.

      The theologians generally DO consider them. They argue for why that isn’t relevant. Take the “Ground of All Being” argument. You can’t have an additional, deeper “god” from that because, well, the argument starts from the lowest possible level and then works its way BACK to God. So not usually an issue, and if it is the question is not that they haven’t considered those “possibilities”, but that they aren’t starting from the right starting point.

      But in that vein, is it worth doing meta-metaphysics, about the really real reality? How would you know it’s worth it, other than the mental satisfaction of persuading onesself that there is a real reality beyond the mundane reality science shows, or a really real reality? Isn’t metaphysics just imaginative speculation with no means of knowing if one is being clever or dumb?

      Metaphysics, as a field, is about getting to the lowest possible point that we can get to, so any “meta-metaphysics” would be just metaphysics. I think you’re mischaracterizing metaphysics by defining it in relation to an answer, not in relation to a question. The question is “What is really reality?”. When someone posits an answer to that, then we analyze it to see if it’s justified and if they can know that that is the case. With experiences, we KNOW that reality is not simply as we perceive it. The question is, then, what reality is really like, and how or even if we could know that.

      Indeed. Metaphysics does not answer that any better than science does – in fact it is a lot worse. Metaohysics can’t even tell us if that’s a remotely meaningful notion, ‘how things really [really, really, …] are’.

      How does SCIENCE tell us if that’s a meaningful notion? If anything can do that, it’s philosophy/metaphysics, and they’ve made far more progress on that than science has.

    2. Coel Post author

      How does SCIENCE tell us if that’s a meaningful notion? If anything can do that, it’s philosophy/metaphysics, and they’ve made far more progress on that than science has.

      Out of interest, what progress has philosophy/metaphysics made on that question?

    3. verbosestoic

      Well, note that metaphysics is something like 8th on my list of philosophical interests (after luge) so I’m not up on the latest metaphysics. If you’re really interested — and want to criticize it — you might want to pick up an introductory book on it, if you haven’t already. But, in general, there has been some progress made philosophically on these issues.

      First, the one issue that pretty much all philosophers agree has been resolved, which is that logical positivism doesn’t work; it at best ends up being circular, and at worst holds that it knows something that it cannot demonstrate using its own criteria, and so is self-defeating. Note that this is the view that you are trying to promote, which explains why philosophers are so skeptical of your position.

      Additionally, we know that our experiences do not give us direct access to reality and so cannot, in and of themselves, be used to simply validate it, and that the addition of science doesn’t help that because, as Russell points out, it only demonstrates that we don’t have that direct perception of reality.

      I’m sure that there are a number of others as well, but these are the ones most relevant for this discussion. Note that your view doesn’t have to be a strictly logical positivist position, which would leave some room for you to argue it.

    4. Coel Post author

      The problem with Logical Positivism, as it was states, is that it was not fully self-consistent. It didn’t need metaphysics to point that out, just logical reasoning (and logical reasoning is as much a part of science as of metaphysics). Anyhow, I think that a “scientism” position can improve on Logical Positivism in this respect (as I’ve argued here).

    5. verbosestoic

      You asked what progress it had made; you didn’t ask what it could do that science couldn’t. Which is also a bad question, because we can ask what science can do that philosophy/metaphysics can’t, and the answer comes up, oh, pretty much the same.

      Also note that whether or not we needed philosophy/metaphysics to get that, they definitely got there FIRST. They got TO logical positivism AND refuted it before science did, so they definitely should get the credit, and I’m sure you don’t want to end up with science saying huffily “Well, we could have done it, too! If we’d wanted to!” [grin].

    6. Coel Post author

      Also note that whether or not we needed philosophy/metaphysics to get that, they definitely got there FIRST. They got TO logical positivism AND refuted it before science did, so they definitely should get the credit, …

      It was logical thinking that pointed out the flaw in Logical Positivism. Yes, logical thinking is important, but it isn’t the special preserve of metaphysics.

    7. verbosestoic

      It was logical thinking that pointed out the flaw in Logical Positivism. Yes, logical thinking is important, but it isn’t the special preserve of metaphysics.

      If you say this, then how can you claim that empirical investigation is the special preserve of science?

      At any rate, if we’re going to talk about the results and usefulness of the field, noting what it actually did while trying to do what it saw as its purpose, again, is something to count in its favour. If you don’t accept that, then I claim that science has no successes because everything it did could have been done by philosophy, and we wouldn’t have needed science at all. I think that’s a patently ridiculous think to say, but it seems to follow just as much as your claim.

    8. Coel Post author

      If you say this, then how can you claim that empirical investigation is the special preserve of science?

      I really don’t care that much about what labels we use for things. In this sort of discussion people often equate “empirical investigation” with “science”, but if you prefer that I use the term “empirical investigation” instead of “science” then ok.

      We are mis-communicating somewhat because by “science” I mean “empirical-evidence-based enquiry” and by “metaphysics” I mean “not empirical-evidence-based enquiry”. You, however, are not using the terms that way.

    9. verbosestoic

      We are mis-communicating somewhat because by “science” I mean “empirical-evidence-based enquiry” and by “metaphysics” I mean “not empirical-evidence-based enquiry”. You, however, are not using the terms that way.

      Despite your assertions, this is also not how people doing metaphysics use the terms. A big problem with scientism in general is taking arguments that don’t use their expanded definitions and then criticizing them as if they really WERE using those expanded definitions. No matter how broadly you define “science”, there is still a difference between doing science and doing philosophy that might matter to the arguments, and the same thing applies here.

  9. richardwein

    @verbosestoic

    1. Even if we’re in the Matrix, that wouldn’t change the fact that, within the context of our reality, the Earth really is round. After all, when we say “Earth”, we’re talking about our Earth, the Earth within the reality we know. So we do know that the Earth really is round. And putting “REALLY” in capitals doesn’t change that.

    We can use the word “really” in the course of making some particular distinction. For example, we might say that President Merkin Muffley was only president in a fiction, but President Obama is really president. That makes sense as long as we make it clear what distinction we have in mind. When you say simply “how can we know if the Earth is really round?”, you are failing to make any such distinction, and so the word “really” only serves for emphasis. (Using capitals just strengthens the emphasis.) Effectively then you are just asking “how can we be sure (really, really sure) the Earth is round?”. We can be sure the Earth is round because the evidence is overwhelming.

    We can make sense of what you’re saying if we take you to be drawing the following distinction: our reality is the ultimate one, or our reality is only a limited reality (like the Matrix) within some greater reality. If that was the question you had in mind, you should have said so. Otherwise, you can’t expect people to understand that that’s what you meant by “really”. After all, this question seems more like idle speculation than a serious subject of study, so I don’t think you can expect people to guess that this is what you meant.

    2. Metaphysics doesn’t even seem to be any use for studying this question. In the movie, Neo didn’t use “metaphysics” to discover the existence of the Matrix. He made an empirical inference, based on the evidence of his senses. But so far we have no evidence that any such thing is the case.

    So it seems we currently have no way of studying this thing that you say metaphysics studies. And, if we did, we wouldn’t use metaphysics. We would use empirical inference. You’ve also given us an example of how metaphysicists tend to confuse themselves by using language in misguided ways, because they don’t properly understand how language works. Wittgenstein addressed the philosophical misuses of language at some length in “Philosophical Investigations”.

    Reply
    1. verbosestoic

      Otherwise, you can’t expect people to understand that that’s what you meant by “really”. After all, this question seems more like idle speculation than a serious subject of study, so I don’t think you can expect people to guess that this is what you meant.

      And I don’t. I talked a lot about what I meant by that. That you tend to reply by grabbing one word/statement, elevating that to the level of that being all I’ve said, and then try to interpret that based on what you think and know and not what I do is not my problem. Case in point:

      1. Even if we’re in the Matrix, that wouldn’t change the fact that, within the context of our reality, the Earth really is round. After all, when we say “Earth”, we’re talking about our Earth, the Earth within the reality we know. So we do know that the Earth really is round. And putting “REALLY” in capitals doesn’t change that.

      Why are you still talking about the “Is the Earth really round?” question when I explicitly said that that question was not a good metaphysical question, and pointed out the two different ways we can mean it and which one relates to metaphysics? If you’re talking about the world of experience/appearances, the Earth really is round. If you get into metaphysics, then that’s an open question. If you don’t want to do metaphysics, don’t. But you can’t then pretend that all of the questions are answered because you refuse to engage the actual questions.

      2. Metaphysics doesn’t even seem to be any use for studying this question. In the movie, Neo didn’t use “metaphysics” to discover the existence of the Matrix. He made an empirical inference, based on the evidence of his senses. But so far we have no evidence that any such thing is the case.

      Again, you’re elevating the “Matrix” example to being the point, and think that you can argue that what Neo did is somehow proper metaphysics just because a movie had him do that. As it turns out, what he did there was a massive failure philosophically/rationally. So, someone is trying to convince him that he’s been inside a simulator that has produced false images of what reality is like, but to such a degree that he simply couldn’t tell the difference. Then that person gives him a pill — and in the world Neo is in, it is known that taking pills can alter perceptions — and when he takes it, Neo suddenly has a new set of experiences, just as that person said. Why should Neo believe that these perceptions are now REALLY reality, as opposed to a delusion caused by the pill?

      You’ve also given us an example of how metaphysicists tend to confuse themselves by using language in misguided ways, because they don’t properly understand how language works. Wittgenstein addressed the philosophical misuses of language at some length in “Philosophical Investigations”.

      Which would be fine if we were talking about language, but we aren’t. We’re talking about concepts. Language can mislead us with concepts, but you need to show how that’s happening here, not assert it and then cite a philosopher without explaining at all how that relates to your point.

  10. Dennis Mitton

    Reblogged this on Monkey Dance and commented:
    Interesting – does science require metaphysics? Is science just another religion? Good primer on what science is and the use of weasel words by those who demand that it falls under the heading of religion.

    Reply
  11. Frederick

    It seems to me that the Templeton outfit is not just promoting a religion friendly version of science, but a specifically christian version with the simultaneous agendas that everything has to turn out to be christian, and that “free”-market capitalism as promoted by the USA right-wing think tanks is the only way forward for all of humankind.

    Reply
  12. ontologicalrealist

    Coel wrote on December 4 :-
    ” The reasons that both theoretical physicists and philosophers are interested in “concepts”, including concepts that might apply to other possible worlds,—-”
    How do you decide that which world is possible and which world is impossible?

    Reply
    1. Coel Post author

      How do you decide that which world is possible and which world is impossible?

      You’d try to do that by investigating the concepts (which is partly why theoretical physicists would indeed take an interest in the concepts themselves).

  13. ontologicalrealist

    Thanks Coel for your reply.
    Perhaps I could not communicate what I have in mind. I would try again :-

    Let us say that you, Coel, have investigated the concepts. Then on what CRITERION would you decide that that world is impossible or possible?

    Reply
    1. Coel Post author

      Then on what CRITERION would you decide that that world is impossible or possible?

      Ultimately, I’m not sure we can. We can say that a “possible” world violates basic laws of physics, and is therefore impossible, but then we have no basis for insisting that such laws must apply in all possible worlds. Or we can say that a “possible” world violates basic logic, and is therefore impossible, but again we have no basis for insisting that such laws must apply. All of our laws of physics and even logic are adopted as models of our world, so might not be universal. Still, to the extent that we can do it, it is interesting to explore our concepts by asking what alternative universes are consistent with them, or not.

  14. ontologicalrealist

    Your reply gladdened my heart! It is a sane reply and sanity seems to be rare in humans as humans are so much prey to hugely inflated self importance in the universe.

    Now I have another question. How does logic relate to possible worlds? How would you know whether a given world is logical or illogical? By logic here, I mean LNC together with all it’s entailments.

    Reply
    1. Coel Post author

      Now I have another question. How does logic relate to possible worlds? How would you know whether a given world is logical or illogical?

      I don’t think we can ever know that since we can never observe other possible worlds. All we can do is observe our world and deduce the logic that our world follows.

    1. Coel Post author

      Well, a world with four-sided triangles might be impossible. But, we can’t really trust our brains to analyse such concepts, since our brains are the product of our world and of what is possible in our world.

  15. ontologicalrealist

    That is an interesting answer! I had not thought about it before.

    Do you agree or disagree to the following and why:-

    As logic does not apply to things but only to thoughts, statements, propositions and theories etc. about things, so there can be no logical or illogical planets, galaxies or worlds etc. but only logical or illogical propositions etc. about planets,galaxies or worlds etc.

    Similarly when some people claim that quantum physics has shown that the world is Illogical, they are making a mistake as only quantum theory can be logical or illogical but not the world.

    Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s