More on the falsifiability of intelligent design


In response to “Behe’s bait and switch: on the falsifiability of intelligent design,” EnlightenmentLiberal comments:

I’ll take the stronger position that you can falsify intelligent design outright, and that is has been falsified, along with the rest of supernatural claims.

The following paper includes the simple and obvious argument that every supernatural explanation in the past has never born out, or it’s been falsified, and investigations into “natural” causes and mechanisms has worked out pretty well, and therefore, the only reasonable conclusion is that there are no supernatural forces.
> How not to attack Intelligent Design Creationism: Philosophical misconceptions about Methodological Naturalism
> (final draft – to appear in Foundations of Science)
> Maarten Boudry, Stefaan Blancke, Johan Braeckman
https://sites.google.com/site/maartenboudry/teksten-1/methodological-naturalism

In the following blog post, PZ Myers makes much the same argument through metaphor.
https://freethoughtblogs.com/pharyngula/2013/03/03/thor/

Sean Carroll makes a particular and stronger form of this argument, which is that with the discovery of the Higgs Boson, the standard model of physics is complete, and the fundamental physics of everyday life is entirely known, and there is no room in known physics for supernatural forces due to the math and rules of quantum field theory and the standard model. The only way to insert supernatural forces is to invoke special pleading, contrary to all accumulated evidence. Sean Carroll makes this argument in several places IIRC, including this public talk / lecture:
> Particles, Fields and The Future of Physics – A Lecture by Sean Carroll
https://youtu.be/gEKSpZPByD0?t=2868

As is often the case in these discussions, much depends on how we define the terms, in particular ‘falsify’ and ‘intelligent design’. I think EnlightenmentLiberal and I are on more or less the same page when it comes to defining intelligent design. Since I was responding to a specific argument by Michael Behe, I did my best to define it in a way that is consistent with his thinking:

By “intelligent design” I mean to imply design beyond the laws of nature. That is, taking the laws of nature as given, are there other reasons for concluding that life and its component systems have been intentionally arranged?

We know that Dr. Behe accepts the evidence for both common descent and natural selection. He’s said as much on many occasions, for example in The Edge of Evolution:

Common descent is true. [p. 72]

…there’s also great evidence that random mutation paired with natural selection can modify life in important ways. [p. 3]

So Dr. Behe’s view is that evolution (meaning just change over time in this case) results from a combination of natural and supernatural (“beyond the laws of nature”) factors. Not everyone in his camp agrees about common descent, for example Casey Luskin, Jonathan M., Stephen C. Meyer, Jonathon Wells, and John G. West spend much of their time arguing against it (and for this reason I think it’s fair to call them creationists). Yet my sense from reading what other cdesign proponentsists have to say is that most of them believe in a combination of natural and supernatural factors.

Perhaps the minimal statement of intelligent design is “supernatural factors have, on one or more occasions, played a role in biological evolution.” There are certainly stronger ways to formulate it, for example “supernatural factors play a large role in biological evolution” or “supernatural factors play the dominant role in biological evolution.” But the first definition is minimal in the sense that if it doesn’t fit that description, it’s not intelligent design. Furthermore (and this is relevant to Dr. Behe’s argument that evolution is not falsifiable), if that statement is true, if supernatural factors have ever played a role in evolution (or anything else), then the materialist worldview is false.

Now what about falsification? I think EnlightenmentLiberal and I mean different things by this. I agree that “the only reasonable conclusion is that there are no supernatural forces.” I agree that “The only way to insert supernatural forces is to invoke special pleading, contrary to all accumulated evidence.” But I mean something more formal by falsification, something more along the lines of Karl Popper: could any conceivable experiment ever, in principle, demonstrate that intelligent design is false?

As a sidenote, I’m aware that there are problems with Popper’s view. Falsification isn’t the be-all and end-all of science, and Popper didn’t definitively resolve the demarcation problem. But my post was responding to a specific claim by Dr. Behe: intelligent design is falsifiable, “…open to direct experimental rebuttal.”

Suppose my claim is this: in some human ancestor, some time between the origin of life and the Pliocene, an omniscient creator intervened to cause a mutation (my claim doesn’t specify which one). Through its omniscience, the creator foresaw that this particular mutation at this particular time in this particular individual would have a butterfly effect leading, through subsequent natural processes, to the evolution of humans.

Is it reasonable to think this? No. It’s an extraordinary claim, and I’ve provided no evidence, only my assertion that it happened. Is it special pleading? You bet, but here I am pleading it. Can you falsify it? Not a chance. I haven’t specified which mutation, so you’d have to show that every mutation in every human ancestor back to the origin of life occurred naturally. Some of those mutations have subsequently been lost, so you’d need a time machine. Let’s say you have one (this is “in principle”, remember), and that you have identified every mutation in every human ancestor. How will you distinguish random mutations from divinely caused ones?

Regarding Boudry et al., I read this differently. Their argument is that a lack of falsifiability is not an intrinsic problem with all supernatural explanations:

We conclude that the argument from ‘intrinsic unfalsifiability’ misconstrues typical immunization strategies and ad hoc manoeuvres of creationists as general and intrinsic problems with supernatural explanations. [emphasis in the original]

I agree. Some supernatural claims are falsifiable. But Boudry et al. don’t even argue that intelligent design as it is actually formulated is falsifiable:

If a hypothesis is designed to be immune to falsification, scientists are justified in dismissing it, and in that sense we completely agree with Scott and Pigliucci. Critics have also pointed out that proponents of supernatural claims, notably IDC theorists, often make use of evasive manoeuvres that render their theories immune to empirical falsification. For example, in response to the argument from imperfection and bad design, Michael Behe has simply replied that we cannot gather any scientific information about the character and intentions of the Designer, that His reasons are unfathomable and that any speculation about them is pure metaphysics.

So Boudry et al. are arguing that even though intelligent design could be formulated in such a way as to be falsifiable:

Thus, if only they chose to do so, IDC proponents could easily equip an alleged supernatural Designer with specific attributes and intentions in such a way that the design hypothesis would yield unexpected predictions and is not “compatible with any and all observations of the natural world,”

the way it generally is formulated is “…immune to empirical falsification.”

As for PZ Myers “Thor” example, I take this too as supporting my position. Just like intelligent design, his encounter with a hungry god of thunder is unfalsifiable. It’s certainly an extraordinary claim, and there’s no actual evidence that it happened, but I can’t prove that it didn’t. If I’ve badly misinterpreted the metaphor, maybe Dr. Myers can weigh in to set me straight.

So I think EnlightenmentLiberal and I are arguing different things. He is saying (and again, feel free to correct me) that supernatural explanations in general have such a miserable track record (“never born out”) that it’s reasonable to dismiss them, and I agree. Unless and until we’re presented with evidence more compelling than “it’s too complex to have evolved naturally,” the natural mechanisms that we know work remain the preferred explanation. But, at least the way I think about falsifiability, Dr. Behe’s arguments and intelligent design arguments in general are not falsifiable. They’re designed not to be. No pun intended.

Comments

  1. consciousness razor says

    Let’s say you have one (this is “in principle”, remember), and that you have identified every mutation in every human ancestor. How will you distinguish random mutations from divinely caused ones?

    When you use terms like “random,” it suggests to me that you don’t know certain things about the mutation. But in principle, you could know them and do experiments with them.

    Think of it from something like Sean Carroll’s point of view. Presumably, biological mutation is not a fundamental aspect of the physical world but something which does rely on more fundamental mechanisms, so we have to get to the bottom of that to explain how those occur.

    If there were some specific “divinely caused” mutation, you could observe that the physical laws we see applying ubiquitously in the whole universe were violated in that specific instance. Try to imagine concretely what sort of situation we’re talking about here. For example, some molecule or whatever was jostled this way or that, causing the mutation (or whatever it is that physically happens when there is a mutation of this sort). Or maybe that molecule just appeared out of the void and found itself in the right place, because some supernatural thing put it there in such a way that it will do its mutating work. Either physical laws are consistent with this event so that it’s a natural “random mutation” as you call it, or they are not so that it’s a “divinely caused mutation.”

    There could conceivably be some supernatural energy source, which urged the molecule in a direction it “shouldn’t” go according to physics. If so, we can (in principle) figure out enough about the natural world to know natural things didn’t do it, that something else other than the natural world had precisely that effect. If that’s what you found, then you should be satisfied that some kind of supernatural thing is necessary. So you look at the molecule and its environment, to empirically check whether or not physical laws (doing their thing without violation) suffice to describe the phenomena in question.

    We don’t have the means to do that for every event in the history of the world, especially not for obscure events in the distant past, but that is in principle what you would be doing. If the procedure of empirically checking for such things isn’t going to be enough to give us a definitive answer, for practical or technological reasons (given what we’re able to do currently), if that’s all the problem comes down to when you analyze it as closely as you can, then it isn’t something that counts as unfalsifiable in principle.

  2. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    Hmmm… CR, it seems like now you’re talking about whether it’s unfalsifiable-in-principle by time-traveling aliens with unfathomable technologies of investigation or unfalsifiable-in-principle by humans that actually exist, in this universe, on this earth, at this time.

    Even the phrase “in principle” typically includes certain background assumptions and/or data: it has to, we’re applying those principles to one or more real life things. By saying that ID creationism is unfalsifiable – whether in principle or in practice – we’re taking it for granted that we know what ID creationism is. That itself entails that we have an understanding of what humanity is and where its borders lie.

    Saying that we could identify violations of the “laws of nature” in a created universe (how are they the “laws of nature” if “nature” itself was designed? – but that’s outside the scope of this, I suppose) either in principle or in practice entails that we, in fact, know what the laws of nature are. There are two ways to go here: either presume that we don’t know the laws of nature completely/thoroughly/correctly, in which case falsification is not possible in principle, since we literally don’t know what standard we’re testing events against or we do know the laws of nature completely, thoroughly, and correctly (at least as they apply to the masses and energies being discussed as part of evolutionary events).

    In that case, our knowledge of the laws of nature comes into play: given the statistical physics of the quantum realm, how will you observe the behavior of individual atom-to-atom bonds within specific molecules without changing that behavior? In principle it is not possible to observe this behavior without changing this behavior. Therefor, in principle it is not possible to falsify ID creationism, because even if there is undiscovered physics that allows us to go back in time to do this research, when we investigate phenomena at masses & energies at which mutations take place, the answer to the question “Has this molecule been affected by intelligent intervention affecting its behavior” will always be “Yes” for any molecule we observe.

    Since it is in principle not possible for any answer except one to be returned, no information can be gained from such investigations. Since we can gain no information from such investigations, we cannot (rigorously, in the manner discussed in the OP) falsify ID creationism. This is true in practice, but it is also true in principle, for the subset of cases where our knowledge of the natural laws which apply to these masses & energies is complete and correct. Since we have already considered the cases where our knowledge isn’t complete and correct for these circumstances, and since in those cases it is also impossible in principle to falsify ID creationism, we can, in fact, say that ID creationism in the broadest sense, the sense discussed in the OP requiring only one intelligent intervention, is unfalsifiable in principle.

    I know that you agree that you have to make some pretty generous assumptions to get falsifiability “in principle”. I know that we’re generally on the same side of this. But I think you make too much of the phrase “in principle” in your comment. Your comment’s conclusion would entail that we change the laws of the universe for the purpose of our investigation of IDC in order to make falsifiability possible “in principle”. I would suggest that if something requires a change in the laws of the universe to achieve, that is exactly the kind of achievement which is properly labeled impossible in principle.

  3. Matthew Herron says

    Conciousness razor, that’s a fair point, and after all I’m the one who said this is about “in principle” falsifiability. I do think, though, that we’re stretching even the broadest meaning of “in principle.” When we have to stipulate a time machine and a complete accounting of every subatomic event in the history of the universe to even conceive of a way to test a hypothesis, that hypothesis is either poorly designed or well designed to be immune to testing. I still maintain that intelligent design isn’t testable in any meaningful sense, because it was carefully constructed not to be.
    I wasn’t going to go into the quantum stuff because, truth be told, I don’t understand it in more then the shallowest sense, but as I was typing this answer, Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden went where I dared not.

  4. consciousness razor says

    Crip Dyke:

    Even the phrase “in principle” typically includes certain background assumptions and/or data: it has to, we’re applying those principles to one or more real life things.

    No, you’re not applying anything, because this isn’t a reference to methodology. The idea is that there is a physical fact which implies the truth or falsity of the proposition (however it’s formulated). “Could there be a fact which would imply true or false?” is asking about that, not “does anybody have a way to know what the truth is, by applying some kind of principle?” If a proposition is falsifiable, there will be something in the world which is different, depending on whether it is true or false. If not, then not. You don’t go about applying principles to anything, to figure out the answer — that’s not what the “in principle” language means (that’s suggesting a priori reasoning, which is pretty much the opposite of the standard being invoked here). Principles that you pulled out of your ass don’t do that — the world does. Something in the world would pinpoint that difference if it’s falsifiable, or nothing would if it’s unfalsifiable. The question of whether we can know about it is just a separate, logically distinct question (which is also interesting and worth asking, of course).

    Saying that we could identify violations of the “laws of nature” in a created universe (how are they the “laws of nature” if “nature” itself was designed? – but that’s outside the scope of this, I suppose) either in principle or in practice entails that we, in fact, know what the laws of nature are.

    As I described recently on Ranum’s blog about causality, physical laws are our best system for describing the world. They are just descriptive facts like any other facts, which are true, real, mind-independent, etc., features of the world. The only difference, if it matters to you, is that laws are most informative about the most stuff (whereas other facts may not be very useful), given our human systems of representation like languages, math, logic, and so forth.

    If we don’t know what the facts are, then of course we obviously can’t falsify or verify anything by using those facts which we don’t know, to determine what is logically entailed by them or what is logically consistent with them. But laws are definitionally true facts we have observed to reliably hold under all circumstances, which is how they’re useful for telling us the most about all of the things there are…. So I don’t understand where this worry comes from, that we somehow don’t know anything like that.

    Have you not been paying attention to physics for the last century or two or three? Is there a good reason why we should distrust it? It’s some of the most reliable knowledge we’ve ever gotten about anything, so if we’re supposed to be distrusting it, then I wonder how deep this radical skepticism goes. I wonder how we even got started with the question of an intelligent designer…. I mean, if we don’t know shit about shit, then let’s actually take that position for a ride: why would we be postulating even more shit, which is not only inconsistent with the first lot of shit but also has the feature that even less is known about it? How could that get us out of this trouble?

    There are two ways to go here: either presume that we don’t know the laws of nature completely/thoroughly/correctly, in which case falsification is not possible in principle, since we literally don’t know what standard we’re testing events against or we do know the laws of nature completely, thoroughly, and correctly (at least as they apply to the masses and energies being discussed as part of evolutionary events).

    No, this simply isn’t the idea which is at issue, as I tried to sketch out in the first paragraph above. Not knowing stuff (or being incapable of knowing it) doesn’t entail that there is nothing to know. If I don’t know, for example, which atom is at the moon’s center of gravity right now (or if any atom is in that location), and nobody will ever have any way of determining that, this doesn’t imply that it isn’t in fact at that location right now. I don’t know that, yet claims about it are falsifiable.

    In that case, our knowledge of the laws of nature comes into play: given the statistical physics of the quantum realm, how will you observe the behavior of individual atom-to-atom bonds within specific molecules without changing that behavior? In principle it is not possible to observe this behavior without changing this behavior.

    Why do you think the measurement problem is a problem? Positivism and instrumentalism are hopelessly misguided. Give up the ghost. You don’t need it. I’m not going to delve into this, but QM does not seem to help us clarify anything here.

    Matthew Heron:

    When we have to stipulate a time machine

    That wouldn’t be necessary, if you had a complete accounting of everything. Maybe time machines are physically impossible in our universe — doesn’t really matter. None of it literally disappears over time; it’s just becoming much more difficult to detect.

    and a complete accounting of every subatomic event in the history of the universe to even conceive of a way to test a hypothesis, that hypothesis is either poorly designed or well designed to be immune to testing. I still maintain that intelligent design isn’t testable in any meaningful sense, because it was carefully constructed not to be.

    You’re certainly right that creationism (of all types) is deliberately shielded from criticism by its proponents. But I do think your belief that creationism is false is coherent, because according to your best and most charitable understanding of it, creationism is falsifiable and the evidence you do have (and all the evidence you’re ever likely to get in the future) does strongly suggest that it is false.

    I think it’s pretty straightforward. You can pose what you take creationism to mean (e.g., an intelligent designer intervening in the world in some way) and deliberately decide that you’re not interested in shielding that from criticism. After reviewing all of the evidence available, you can decide there’s no justifiable reason to believe it’s true. So it’s false, which implies it was not in a “meaningful sense” unfalsifiable.

    Of course, those ID jokers over there are playing a different game, but their game isn’t going anywhere or doing anything constructive, so we don’t need to agree that they have any serious point to make with it. Their strategy doesn’t need to be our strategy, and it shouldn’t be because it’s dishonest and useless. Why not stick with an approach that does try very hard to be honest and useful?

  5. EnlightenmentLiberal says

    I’ll try to keep this short (and probably fail).

    Now what about falsification? I think EnlightenmentLiberal and I mean different things by this. I agree that “the only reasonable conclusion is that there are no supernatural forces.” I agree that “The only way to insert supernatural forces is to invoke special pleading, contrary to all accumulated evidence.” But I mean something more formal by falsification, something more along the lines of Karl Popper: could any conceivable experiment ever, in principle, demonstrate that intelligent design is false?

    As a sidenote, I’m aware that there are problems with Popper’s view. Falsification isn’t the be-all and end-all of science, and Popper didn’t definitively resolve the demarcation problem. But my post was responding to a specific claim by Dr. Behe: intelligent design is falsifiable, “…open to direct experimental rebuttal.”

    You are right to mention this. I have some severe problems with Popper’s take on philosophy of science. I endorse a Bayesian approach to epistemology and science, such as outlined by Richard Carrier in his book “Proving History”. The short of it is this: More or less nothing is known for certain. Falsification in the sense of absolute certainty that the hypothesis is false – that is impossible to achieve. There is no set of experiments that should absolutely convince you beyond all doubt of the absolute truth (or absolute falseness) of some claim. You should always have some doubt. For example, I consider it to be extremely unlikely, in an epistemological sense, that monkeys will fly out of my ass in 5 minutes. Yet, I also consider it to be possible, in an epistemological sense – of course I also consider the chances to be very, very small. Similarly, maybe the sun will not rise tomorrow – the word “maybe” is again in an epistemological sense.

    Necessary clarification: Now, it may be that the laws of physics, as they are, “require” that the sun will rise tomorrow. It may be that it’s physically impossible for the sun to not rise tomorrow. However, in the context of epistemology, aka “that which I claim to know and believe”, I am not absolutely certain that the sun will rise tomorrow, and I do not have perfectly reliable knowledge about the natural laws that govern our reality.

    But I mean something more formal by falsification, something more along the lines of Karl Popper: could any conceivable experiment ever, in principle, demonstrate that intelligent design is false?

    In the simple and obvious sense, no one experiment should ever convince you that intelligent design is false. However, as I argued before, and as consciousness razor notes above me, we have such a massive wealth of evidence in favor of materialism that it is perverse to not accept the obvious truth of materialism e.g. philosophical naturalism, and the consequence that the supernatural does not exist, and the consequence that (religious) intelligent design is clearly false. (Similarly, we can conclude that all religious supernatural claims are obviously false.)

    IMHO, the point that the Boudry et al paper makes is that philosophical naturalism is not a prereq of doing science. Rather, philosophical naturalism is a consequence of science. We know that philosophical naturalism is true because of the scientific evidence, and that is why we adopt provisional methodological naturalism when we do further science. Boudry et al might not come out and say this, verbatim, but this is the clear takeaway from the paper, IMHO. And it’s also clearly correct.

    Suppose my claim is this: in some human ancestor, some time between the origin of life and the Pliocene, an omniscient creator intervened to cause a mutation (my claim doesn’t specify which one). Through its omniscience, the creator foresaw that this particular mutation at this particular time in this particular individual would have a butterfly effect leading, through subsequent natural processes, to the evolution of humans. Is it reasonable to think this? No. It’s an extraordinary claim, and I’ve provided no evidence, only my assertion that it happened. Is it special pleading? You bet, but here I am pleading it. Can you falsify it? Not a chance. I haven’t specified which mutation, so you’d have to show that every mutation in every human ancestor back to the origin of life occurred naturally. Some of those mutations have subsequently been lost, so you’d need a time machine. Let’s say you have one (this is “in principle”, remember), and that you have identified every mutation in every human ancestor. How will you distinguish random mutations from divinely caused ones?

    Yes, I can falsify it, in the normal sense of “showing that it is false beyond all reasonable doubt”. I won’t falsify it by showing the actual origin of every mutation, one by one. I’ll falsify it by falsifying the existence of your purported for-the-sake-of-argument “omniscient creator”. We know that there is no such creator. This is what the PZ Myers story shows. This is also what the Sean Carroll argument shows. We know this. We know that there is no supernatural stuff, as surely as we know that the sun will rise tomorrow (approx).

    But Boudry et al. don’t even argue that intelligent design as it is actually formulated is falsifiable:

    I take a stronger stance than Boudry et al. Where “intelligent design” means “design by a supernatural designer”, it is clearly falsified, because the existence of all supernatural designers is clearly falsified.

    As for PZ Myers “Thor” example, I take this too as supporting my position. Just like intelligent design, his encounter with a hungry god of thunder is unfalsifiable. It’s certainly an extraordinary claim, and there’s no actual evidence that it happened, but I can’t prove that it didn’t. If I’ve badly misinterpreted the metaphor, maybe Dr. Myers can weigh in to set me straight.

    Yea. Politely, I think you missed the clear point of the story. The point of the story is that even Thor knows that there are no gods, and he explains how he knows that there are no gods through his candle in the dark room metaphor. He doesn’t know everything, metaphorically there are still dark and hidden places in the room, but Thor does know something, and he sees enough of the room to conclude that there is no great enemy there, e.g. Thor knows that there is no god. Maybe there’s a mouse, e.g. maybe there’s some new natural law that he doesn’t know, or maybe some other sort of mundane or unusual creature or thing out there. Thor doesn’t know everything, but Thor does know enough to conclude that there are no gods.

    The story was done (presumably?) because it was humorous, that even a god knows that there are no gods (due to the ambiguity in the meaning of the word “god”). Are there sufficiently advanced aliens ala Arthur C Carke’s third law? Maybe. Are there supernatural creatures like the Christian god? No.

    So I think EnlightenmentLiberal and I are arguing different things. He is saying (and again, feel free to correct me) that supernatural explanations in general have such a miserable track record (“never born out”) that it’s reasonable to dismiss them, and I agree.

    The exact meaning is very important here. I do not simply “reject” them in the sense “I do not accept that they are true”. I reject them in the sense that “I am convinced beyond all reasonable doubt, e.g. to a very high degree of confidence, that all supernatural claims are false”.

    PS: I’ve skimmed the comments. Let me read them in full to see if there’s anything else I need to address in another post.

  6. EnlightenmentLiberal says

    I also must post the following two videos, which IMHO are required watching for anyone who wants to take part in this conversation.

    > Nobel Physicist Richard Feynman explains how it’s impossible to explain how magnets work:
    https://www.youtube.com/watch?v=MO0r930Sn_8

    > God, Science and the Problem with Nature – Scott Clifton (Theoretical Bullshit) – Skepticon 7
    https://www.youtube.com/watch?v=LQMLFpQEOI8

    Here, Scott Clifton makes a very compelling argument that the words “supernatural” and “natural” serve little to no useful purpose, and they cause a great deal of confusion. Often, the word “supernatural” is used simply as an excuse to avoid proper critical thinking and scientific thinking on a topic. In particular, for example, if there is a god, that is a fact about the natural world, and with the right framing and background, the word “supernatural” is incoherent.

    IMHO, at best, the word “supernatural” simply means “anything other than materialism”, and in practice, most people who make claims that “supernatural things exist” often use the word “supernatural” to implicitly make the wrong-headed argument that “science and critical thinking doesn’t work on examining this fact about our reality, because I say so, e.g. special pleading”.

  7. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    @Consciousness Razor:

    The idea is that there is a physical fact which implies the truth or falsity of the proposition (however it’s formulated). “Could there be a fact which would imply true or false?” is asking about that, not “does anybody have a way to know what the truth is, by applying some kind of principle?” If a proposition is falsifiable, there will be something in the world which is different, depending on whether it is true or false. If not, then not.

    :Sigh:

    Sorry, this is just wrong. Were not asking if a proposition can be false. In that case, you would be correct. We’re asking if a proposition an be falsified. In order to be falsified, something has to be shown to be false.

    When you say that this is “not” about

    does anybody have a way to know what the truth is, by applying some kind of principle?

    then you’re saying that propositions can be “falsified” without anyone ever knowing whether or not those propositions are false. That would be such a drastic change in the meaning of either “falsifiability” or “science” (or both!) as to make philosophy and scientific discussions inconsistently intelligible for decades to come.

  8. EnlightenmentLiberal says

    We’re asking if a proposition an be falsified. In order to be falsified, something has to be shown to be false.

    I’m pretty well convinced of the truth of the following principle. If it’s conceptually possible, epistemologically possible, to show a claim to be true, then it’s also conceptually possible, epistemologically possible, to show that the claim is false. They come together. You cannot have one without the other. Of course, it might be “harder” to show it’s true than it is to show that it is false, or vice versa, but must must be conceptually possible, else neither is conceptually possible.

    This principle is a simple consequence of the fact that all proper empirical reasoning is Bayesian. According to the principles of Bayesian reasoning, if you find a bit of evidence X that increases your estimation of the truth of some proposition P, then necessarily it follows that the absence of that evidence X entails a decrease of your estimation of the truth of that same proposition P. Again for emphasis, the increase and decrease in the estimated odds of truth are not equal – they often are very different. For example, if I find a bloody knife in the home of a murder suspect for a victim that was stabbed to death, then this greatly increases my estimated odds of guilt. On the flip side, if I do not find a bloody knife, that necessarily decreases my estimated odds of guilt, if only by a little bit.

    It is this same kind of reasoning that I have used to conclude that the supernatural does not exist. Every time I look for an example of the supernatural, I never find it. If I found any one piece of expected evidence, that would have a great impact on my estimated odds of the existence of the supernatural. Whereas, not finding one piece of evidence would have but a miniscule impact on my estimated odds. However, all of those little bits of absence of expected evidence add up. They accumulate, and we have such a mountain of that evidence that the accumulated effect on the estimated odds is huge, so huge that it is perverse to deny the obvious conclusion that the supernatural does not exist.

  9. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    @EnglightenmentLiberal:

    I’m pretty well convinced of the truth of the following principle. If it’s conceptually possible, epistemologically possible, to show a claim to be true, then it’s also conceptually possible, epistemologically possible, to show that the claim is false.

    I’m mostly with you on this – though I think I anticipate where you’re going, and I’m going to disagree for certain claims. It’s a good point to remember, I’m just disagreeing that it’s universally true rather than conditionally true depending on your epistemology. For SOME claims under the standards of SOME epistemologies, this wouldn’t work out.

    The problem is one of human imperfection. Imagine Russell’s teapot, floating between Mars & Jupiter. Imagine it’s big enough that it can, in theory, be seen by humans because we are at a tech level that our combined abilities 1) to travel closer to the asteroid belt, and 2) to use telescope technology to see better from distances than our naked eyes allow, and 3) to record the light coming into the telescopes used to survey the asteroid belt. However, these technologies are insufficiently developed for any person to fly out to the asteroid belt and see for themselves Bertrand’s teapot. It’s a major social effort for humans to survey the belt, and it is something that can only be accomplished with major collective effort.

    …more later.

    If we

  10. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    If we look at the epistemological import of collective effort we must realize that in any enterprise where we’re relying on others for some part of our knowledge (e.g. that the data storage system which contains -or doesn’t- the photographic evidence hasn’t been tampered with, that the focal length of our telescopes is, in fact, correct, etc.) a negative result is dependent on our faith in the work and communications of others in a way that a positive result is not.

    If we get a good pic of the teapot during the survey, we can rephotograph that area, send a new craft to the area the teapot is expected to occupy by the time the craft can get to the asteroid belt, etc.

    If we survey the belt with sufficient resolution that it would be impossible for the teapot not to show up if our survey instruments were working correctly, we have only demonstrated that
    1) the teapot does not exist
    OR
    2) our instruments were not working correctly

    Not being able to rule out #2 means that proving the claim false faces a different hurdle than proving the claim true.

    Likewise, let’s go back to the original claim that got us here: intelligent design.

    Suppose that we can travel back in time once. Maybe its a money constraint, maybe its something about the physics of time travel, maybe its just that by traveling into the past we’re going to change the future in a way that chaos theory can explain post-facto (given enough observations) and can statistically anticipate (the new future will be different in ways that are statistically constrained) but that cannot actually be predicted (it is impossible to state the exact vector variance before the variance occurs). Since any individual is unlikely to exist if we mess with the distant past, we can’t say that if we visit the past we won’t prevent ourselves from existing in a time stream that re-merges us with that modified-time-stream’s time-equivalent of our present. We get to go back and do our test (once), but we’re screwed if we try to rejoin our time stream.

    Now, IN THEORY, we could manage to choose the perfect place/moment to observe intelligent design, record our observations, and store the evidence for future persons to examine. But if we choose a place/moment to examine, we can only say that ID **did, in fact, occur** OR that ID **might have occurred, but not at this particular place/moment**.

    So, i think that your original assertion holds true for the vast majority of propositions that are made with respect to reasonable epistemologies, but I think you can admit that it is possible to have a cockamamie epistemology (think conspiracy theorists that are so out there other conspiracy theorists find them frustrating) where your assertion doesn’t hold AND (more importantly for us) it is possible to imagine cases at the edge of our ability to know that would contradict your assertion even when employing a reasonable epistemology.

    Your statement, then, is a wonderful guideline, but can’t be used to build any logical proof.

    So let’s go back to you and see where you were going with this.

  11. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    This principle

    …for those following along, this just refers to EL’s original statement that if something is possible to prove true, it is possible to prove false …

    is a simple consequence of the fact that all proper empirical reasoning is Bayesian.

    Ah, okay, so we are on the same page, sort of. Your statement that this is a consequence of Bayesian epistemology is an implicit admission that in non-Bayesian epistemologies exist, and that within them your original assertion may or may not hold. Good to have our cards on the table, and with you limiting your claim to Bayesian epistemologies, you remove one of my major objections.

    My other major objection is anticipated by you in the next bit:

    for emphasis, the increase and decrease in the estimated odds of truth are not equal – they often are very different. For example, if I find a bloody knife in the home of a murder suspect for a victim that was stabbed to death, then this greatly increases my estimated odds of guilt. On the flip side, if I do not find a bloody knife, that necessarily decreases my estimated odds of guilt, if only by a little bit.

    So we have two propositions:
    First, Cindy Lou Who is guilty of coldblooded murder through stabbing.
    Second, Cindy Lou Who is not guilty of coldblooded murder through stabbing.

    You appear to be saying (since you’re employing Bayesian epistemology) that if it is possible to collect enough evidence to be justified in believing the first, then it must be possible to collect enough evidence to be justified in believing the second.

    More generally, (in your view) it cannot be be possible for us to justify a belief in P unless it is possible to justify a belief in not-P.

    I have no trouble granting that any test of P is a test of not-P. I have no trouble that any evidence for or against P must necessarily be evidence against or for non-P. But remember that you have conceded in your bloody-knife example that a test might have unequal for proving P compared to proving not-P (there’s an important weakness in your example, but let’s skip it for now. The fact that it is possible to have unequal effects on justifying belief in P vs justifying belief in not-P is something upon which we agree, regardless of whether it is “proved” by anything either of us has said in this discussion).

    But if it requires 800 epistemology points to justify a belief in something, and there are quite a number of readily performable tests that will either provide 200 epistemology points in favor of P or 1 epistemology point in favor of not-P, then it is easy to see that after performing 799 such tests it is possible we be justified in believing P, but it is impossible that we will be justified in believing not-P.

    If the number of tests that we can do is greater than 3 and less than 800, then we get the situation you have said cannot occur: the possibility of “proving” (in the sense of Bayesian epistemology) P combined with the impossibility of “proving” not-P.

  12. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    It is this same kind of reasoning that I have used to conclude that the supernatural does not exist. Every time I look for an example of the supernatural, I never find it. If I found any one piece of expected evidence, that would have a great impact on my estimated odds of the existence of the supernatural. Whereas, not finding one piece of evidence would have but a miniscule impact on my estimated odds. However, all of those little bits of absence of expected evidence add up. They accumulate, and we have such a mountain of that evidence that the accumulated effect on the estimated odds is huge, so huge that it is perverse to deny the obvious conclusion that the supernatural does not exist.

    And this is where it all gets so weird.

    I actually agree with you on your conclusion. I also don’t think it’s necessary to assume (or conclude or prove) that proving not-P is necessarily possible if proving P is possible.

    All you need to support your conclusion is that
    1) the results of any test of P will, by definition, provide evidence for or against P,
    and
    2) a logical consequence of this definition is that those same results will provide evidence against or for not-P.

    In the bit I’ve quoted in this comment we simply have you saying, “We do have evidence of not-P; it would be irrational to say that we do not have such evidence; in any rational epistemology if the threshold for believing P is a mere 800 epistemology points then the threshold for believing not-P must likewise be 800 epistemology points; and we’ve done 800 or more tests of P, where negative results for P grant at least one epistemology point in favor of not-P.”

    I’m completely down with this. I think it’s unreasonable to believe in any supernatural world that has ever been described.

  13. EnlightenmentLiberal says

    To Crip Dyke
    I think we’re mostly on the same page. I don’t know if there’s any actual disagreement. Is there?

    Note, I do fully agree to this bit here:

    But if it requires 800 epistemology points to justify a belief in something, and there are quite a number of readily performable tests that will either provide 200 epistemology points in favor of P or 1 epistemology point in favor of not-P, then it is easy to see that after performing 799 such tests it is possible we be justified in believing P, but it is impossible that we will be justified in believing not-P.

  14. EnlightenmentLiberal says

    Sorry, I was too quick to post. I should also say: I think I’m in complete agreement with your analysis. My phrasing “possible to prove true / possible to prove false” was accidental shorthand. I think you did a good job fleshing out what I meant.

  15. Crip Dyke, Right Reverend Feminist FuckToy of Death & Her Handmaiden says

    I think we’re mostly on the same page. I don’t know if there’s any actual disagreement. Is there?

    I think the only disagreement anywhere is on what “falsifiable” means – does it mean that it is possible for a claim to BE false, or does it mean that it is possible for one person to demonstrate a claim to be false?

    And to the extent that the disagreement exists, I think it only exists between CR and me.

    I think you & Matthew Herron used different ideas of the burden of proof necessary to justify accepting the claim that proposition P is false and proposition not-P is true. In some cases it might be useful to use one burden of proof and in other cases it might be more useful to adopt another. To say “in theory” means – to me – adopting the highest possible burden of proof (or lowest possible, depending on how you’re looking at things), but it clearly means different things to different people. I’m okay with that.

    =================
    Long version, because I’m me:
    I think all of us – CR, me, you, Matthew Herron – have been using completely reasonable interpretations/ definitions of common terms. They simply haven’t always been the same reasonable interpretation/ definition. If, however, I adopted CR definitions, I think (I’d have to review and don’t feel like it right now) I’d accept CR’s entire argument including any conclusions. As I remember it, all the arguments here have been valid. All the logic holds internally. We’re just discussing whether or not we accept each other’s definitions and sometimes finding places where it appears that one person or another may not understand that a different definition is being used.

    It’s those times – when someone doesn’t realize a different definition is at play in another’s argument – that get really confusing and, frequently, frustrating. But whether or not that occurred in this argument, I think you and I and everyone else are, in fact, on the same page in the sense that we’re each making good arguments about a common subject. Sometimes one or more of us might be heading off on a tangent, but always (as tangents will be) obviously connected to the common ground of this thread.

Leave a Reply