I have not gotten far into this article, but I would like to flag that this paragraph is a pretty big misrepresentation of theistic philosophers.
“Another example is the constant insistence by theistic philosophers that dress up the idea (in some capacity) that we need an external agent (God) to “give” us morality, when there is abundant literature of how moral instincts in humans are evolved, observable in babies, and yes, in non-believers and non-religious societies.”
There are two broad categories of moral arguments put forth by apologists which could match what you are describing. (I’m presuming you definitely aren’t referring to the much more heavily written upon moral argument found in Kant)
The first is Craig’s moral argument. In every rendition of his argument, down to his 5 minute videos, Craig explicitly distinguishes between a need for God in morality and a need for Belief in God for morality. His argument is about moral ontology and says nothing of moral epistemology. Thus, your criticism cannot apply to this argument.
The second is the moral knowledge argument, which is at least about moral epistemology. Nonetheless, this argument also asserts that all humans have moral knowledge. It also affirms that they gained this knowledge through the accepted evolutionary processes. What they deny is only that a naturalist can say that this process would select for the objectively correct moral beliefs. This argument was first put forth by naturalist anti-realists, such as Sharon Street in this enormously influential paper (for a philosophy paper). Your claim that “but we did evolve true moral beliefs” would be question-begging in reply to this argument.
I can happily point you towards resources that respond to the actual moral arguments made by theists if you so choose. 😁
I’ll concede that my language was sloppy here (I added this little section at the last second) and I probably should have said something along the lines of “popular Christian Apologists” instead of theistic philosophers (we can probably agree there’s a substantive difference between the two!).
Having said that, I believe my characterization is probably still true of Craig. When I say that theistic philosophers say we need God to “give us” morality, I’ll admit that sounds epistemological, but what I’m getting at here is that the concept of objective moral facts coming from (in some capacity) a subjective agent of some sort (which I would argue God is). “Giving to” may sound like an epistemological description, but my actual meaning is that the theist position is that moral facts (and thus morality) cannot exist without God. My presumption is that it’s a little silly because it just seems absurd, if not internally inconsistent (as I’m implying but not communicating clearly), then empirically wrong (the baby example). Lurking here is a meta-ethical disagreement (I’m a moral non-realist). Craig and other theists present as if moral facts exist, come from God, etc and thus humans who do not accept them are not moral in the same way. He may not explicitly say this, but when he’s pressed on it his arguments and tone revert to Turek levels of pearl clutching that lead one to believe he thinks atheists can’t be moral or that we wouldn’t be truly moral or maximally moral without Christianity; the same goes with objective meaning (see: his debate with Shelly Kagan), but that’s another story.
As for the second aspect, I don’t think it really matters as it’s completely different from what I’m arguing. I don’t think moral knowledge exists! My point is simply that, to the degree that anything resembling “morality” does exists, it’s informed by a psychology that evolved to coexist in a group setting. We can describe the attributes that inform this psychology (the Jonathan haidt moral foundations thing) and like all of our sensory perception and instincts, they don’t exist to establish absolute knowledge, but as reactions to our environment, steering us in the direction of an equilibrium of cooperation. That’s not the same as knowledge! And that’s my frustration with moral philosophers in general, is that outside of virtue ethicists, there’s really no mainstream reckoning with these findings in psychology or refinement with these concepts. So no, my argument is not the question begging “but we did evolve true moral beliefs,” it’s “this is not how people compute and conceptualize morality and this trail of thought leads us in the opposite direction of the truth.”
So, I fully concede that I didn’t communicate this paragraph optimally, but I still stand by the spirit of the criticism. Out of laziness and giving you due credit, I won’t change it and if anyone has any further comments in the similar vein, I’ll refer to them this discussion :)
You're right that fine-tuning is a scientific debate, and evaluating it properly requires much deeper understanding of physics than philosophers normally have. I'm only beginning to understand the physics underpinning the fine-tuning argument, but it's obvious that Adelstein knows less about it than I do. For example, in his fine-tuning articles, Adelstein barely mentions the work of Fred C. Adams, who has submitted the fine-tuning argument to detailed scientific scrutiny. Until he read my work, Adelstein seems also to have been unaware of the most powerful critique of fine-tuning, which is that we can't know what would happen if we tweaked the physical constants of our universe. I discuss all this in my article:
Yes! I read your article and my first thought was "whew, thankfully I can just cite this/reference and not come up with the argument on my own, or whatever I inevitably write on design will be much shorter" lol I highly recommend it!
He also said in a comment to me (though I think it was an accidental comment maybe meant for you, ironically, but I could be wrong on that) which says he emailed Adams and said that Adams agreed with his portrayal of the physics and call the agreement trivial. Which tells me BB didn't understand the nature of the disagreement/physics more than anything.
I'm working on an FTA post, but I think I'm going to avoid talking about BB, or at the very most, he'll be a side character. Even if you disagree with him/the FTA, I think he is doing innovative things with it, if you look how it's in discourse with other thinkers. That doesn't mean I find it convincing or agree! But it is interesting and worth considerations. Unfortunately, I think he does adopt views of the science that aren't necessarily true, and conceptions of philosophy of science one need not adopt as an atheist (but on the latter point, I think that's common of many FTA proponents)
Adams unambiguously rejects the fine-tuning argument. See the link below for quotes. What he's saying is that fine-tuning arguments are superficial because fine-tuners don't consider all of the ways in which the universe might support life. I give his analysis of carbon production as an example in my essay. He does not think that we have the final word on all of the ways in which the universe might be life-supporting. He's saying, based on what physicists know now, there are more ways that the universe can support life than the fine-tuners acknowledge.
I liked the beginning of this essay when you made more general points about apologists, but I found the part specifically addressing Adelstein on the nature of probabilistic reasoning to be bad. I generally think objections to the fine tuning argument on the basis that we don't know the probabilities involved to be the worst ones. The fine-tuning argument doesn't need you to know the exact probability of fine-tuning, just that the probability given naturalism is extremely low, and proponents give arguments as to why the probability is low. You could object to this by arguing that the probability isn't actually as low as they think, but the objection that it's impossible to assign any probability at all, even a rough approximation, is silly and would rule out basically all ordinary reasoning. Some of the objections you made assume that in order to have a probability of fine-tuning at all, you need to have some physical process that could set the constants in certain ways and had a certain probability of setting them in each way (for example, the quote complaining that there's no roulette wheel makes this assumption). But this is absolutely false - if it was true, we couldn't assign any probability, even a rough estimate, to any hypothesis about physics being true, since after all, whatever the most fundamental laws of physics are, there's no process that causes them to be what they are and could have made them behave differently (otherwise, the laws that govern that process would be the true fundamental laws) - they just are. So this objection would imply that we can never judge one hypothesis about the nature of reality or fundamental laws as more plausible than others. Every time we do an experiment in physics, we are really just obtaining evidence to update our probabilities that the laws of physics work a certain way. So if we had pure, Knightian uncertainty, we wouldn't even be able to do physics.
The point about possibility is conflating between different types of possibility. On the one hand, there's physical possibility, which says that something could have been a certain way without changing the laws of physics. That's obviously not relevant to the fine tuning argument, since it's about the probability of having different laws of physics. And it's not relevant to probability assignments over various hypotheses either: Unless we're absolutely certain of the true laws of physics, we can obviously assign nonzero probability to scenarios with differing laws of physics, and, as explained in the previous paragraph, we have to do that in order to even do physics. The FTA is based on the low prior probability of fine-tuning, and there are no a priori reasons to rule out different values.
The form of possibility that actually matters to probability assignments over possible laws of physics is metaphysical possibility, which just means "any way the world could be". If something is metaphysically impossible, so that there aren't even any hypothetical scenarios in which it's true (e.g., a logical contradiction), then sure, you can assign probability zero to it to a first approximation (you actually shouldn't do it even then, unless you're 100% certain it's metaphysically impossible). But different values of the constants are definitely metaphysically possible - the only people who dispute this are necessitarians, but necessitarianism also completely prevents you from making the "metaphysically impossible -> 0 probability" inference, so even their objections to this point aren't relevant. And metaphysical possibility is firmly within the realm of philosophy, so the charge of philosophism doesn't apply here as it does to the Kalam argument or Thomistic metaphysics (both of which it definitely does apply to).
A lot of objections to the specific points he's making seem to misunderstand his points. His thought experiment about birth control is not a model of how the world works - it's a hypothetical scenario used to illustrate a point about probability. Objecting that the assumptions made in the thought experiment aren't true in real life misses the point. Also, it's obviously true even in real life that someone's existence is evidence that their parents have had sex without birth control. I'm not sure why you deny this is the case (evidence in favor of =/= 100% proof, so the fact that some people are conceived despite their parents using birth control or without their parents having sex is not an objection to this point).
Similarly, your objections to his use of the principle of indifference are strawmen. No one thinks that we should assign the same prior probability to, "It's Monday," and, "It’s not Monday." All formulations of the principle of indifference include some qualifier stating that there are no relevant differences between the options or that they are all "the same" in some way, which is not true in the case of Monday vs. not Monday. Maybe you think such qualifiers are too vague, fair enough, but you yourself used the principle of indifference when you claimed that actually the probability of it being Monday should be 14%. You understand that there's no reason to consider any day more likely than any other a priori, and so you assign the same probability to each. Similarly, in the election example, you shouldn't apply the principle of indifference because you have information about the candidate's positions, polling data, etc., that might make you think one candidate is more likely to win than the other. The principle of indifference does not deny that you should update your probability based on this evidence - that would be crazy. It would only apply in the election example if you knew nothing at all about the candidates, in which case assigning them an equal probability of winning sounds pretty reasonable. The principle of indifference is really just a statement of the fact that you should not arbitrarily consider one possibility more likely than other unless there is some reason to think it is. Also, note that the quote from Adelstein is actually making a claim much weaker than the principle of indifference. All that's required for his quote to be true is that the probability of the options don't differ wildly from each other, nothing that they're all exactly equal.
I'm not really sure what you're trying to say in the section on agnosticism. Agnosticism isn't some third possibility aside from theism and atheism, so there is no "probability of agnosticism." Agnosticism is a description of someone's state of belief or knowledge, not of how the external world actually is. By definition, either theism or atheism is true, and agnosticism is just the position that we don't know which, or a personal suspension of judgement. And I don't think Bentham's Bulldog has ever said that any theory is better than agnosticism. He just thinks the evidence for God us good enough to motivate belief.
okay, so we’re clearly talking past each other so this response is not going to be as long. I’m not responding to everything because I don’t have time, so I’m going to reduce my criticism as succinctly as possible: The fact that you say “many physicists endorse the fine tuning argument” at the end of your first response highlights the problem and how BB and others mislead people and it leads to talking past each other. Because the physicists use the phrase fine tuning to mean different things from philosophers. I’m not denying that, the OG victor stenger didn’t either. The whole issue is whether or not there is empirical basis (i.e external validity to current models) to make high-confidence statements about what the world would be like if one variable changed, or if it was completely different. *We can’t model life arising in our current universe with our current models.* Some of this is computational (we don’t have the computing capacity to model billions of years). Some of this is empirical (our current models will say the universe at time x should not have complexity such as a galaxy, but using different methods, like redshifts and telescopes, we find this to be false constantly - there’s a great discussion about this in both the documentary I cited and a follow up by the same creator)
The skeptic (and as far as I know, physicist position) is skepticism toward the *causes* of fine-tuning (which is the subject of debate). No one is disputing that life wouldn’t arise on earth if the constants were different, but we can’t make an informed guess to probabilistically model how things were different *because we know so little.* It’s not a permanently unknowable truth, of course, but it’s still very hard at the moment. I used this example in a note yesterday, but it’s like me asking you who the king of a country was in a certain year, while providing only one data point and you have no other background information. Sure, you can come up with a model, but it wouldn’t be rational to trust it just because there are intuitively compelling reasons to believe it. Because the world is usually counterintuitive, and there are so many lurking defeaters to the proposition. That’s why most models are built first by observation. For instance, the multiverse (and other hypotheses) are informed not just by the realities of fine tuning, but other math that is beyond my comprehension. The FTA is an affirmative proposition (that life could not arise given the conditions of the universe and therefore there is a creator). We don’t know either clause is true!
The thing that does piss me off about BB (and when I say this I’m not mad at him as a person, but something he does), is that he oversells his case on issues like this by either not understanding or not properly communicating probability correctly, to say nothing of the physics, and so we have lots of people parroting his arguments. It leads to falsehood.
I don’t want to speak for the guy, but I saw a few months ago Bryan Frances (who has a master’s in physics, was thinking about getting a PhD in physics but instead went the philosophy route) spar on here with BB and basically “this is not what you say it is.” The same goes with physics-trained philosophers like James Fodor, in his own conversations with BB. The same goes for actual physicists who disagree with Luke Barnes; the most common response I see to him is “you didn’t understand what I said.”
You may be correct about all these things on probability and physics, but I think the medium of our discourse is not conducive to communicating it. I’m not following it! It may be your writing, the fact that my dad is in town and I’m distracted, or that it’s hard to say these things succinctly. Regardless, I can’t put much credence in it because, to use bayesian language (which I hate because my hot take is that people don’t think as bayesianly as though), I have such a low prior of it being true because of how I learned the scientific process (approach a subject without priors for long as possible) and because of the preceding paragraph. As you’re defending BB, I can’t really tell how much of it is you trying to steel man his arguments or is just your own opinion, informed by your prior knowledge. I suspect it’s a combination of both, but I think you give way too much credence to BB’s knowledge on the subject.
I highly recommend watching his livestreams with James Fodor (who iirc has an MA in physics) as Fodor’s content does a good job of pointing out the science-based flaws in FTA, but also in direct conversation with BB is good at point out how he jumps to conclusions pretty quick. I also recommend listening to the comments of Niayesh Afshordi on Paul Halper’s documentary on FTA (and the follow up video discussing responses to it). These people are smarter than me, know physics and probability better than me, but the reason why I trust them is because scientific modeling is based on two key concepts (among others!) : external and internal validity. They’re able to communicate why the main premises of FTA depend on hypotheses that lack both forms of validity, and so we should give low credence to them working. To me, that’s the main reason we accept agnosticism. The reality of fine tuning (as physicists understand it) is real, but there’s really no good explanation for it either way (yet!).
For your first paragraph, I think you’re wrong on a lot of point. The idea that we don’t know the probability comes from *the physicists themselves.* In the language of science, it’s unclear what the external validity of the modeling shows, and that we can’t generalize it to other universes, because they seem to be falsified in our own. On the point of saying that it’s impossible to set probabilities, I’m going to say, yes that’s true, and it’s not absurd. We just don’t know enough, at this point. Maybe impossible isn’t the correct word, we could feasibly get to a point where we have enough data and working models, but we’re not at that point. What’s more, I don’t think “ordinary reasoning” is something one uses in physics. Your complaint about the roulette wheel is not valid IMO, the argument is not that we can’t make any probabilistic assumption in physics, it’s that we lack any understanding about the specific concepts of life, universal laws, and how they happen, etc. We have theories that may work out, but we know very little. The same can’t be said about experiments within physics (like, I dunno, gravitational waves). I don’t think it’s necessarily inaccurate to say that a lot of (but not all) data collection/experimentation in physics is updating our probabilities, finding outliers, and trying to find new models to explain the outliers that don’t conform to your current models. I’m not sure if I’m understanding you correctly here - I’d appreciate elaboration on this paragraph.
Secondly, as I said in the follow up to this post, there’s no falsification criterion for the FTA, so it’s hard to say that naturalism is improbable relative to theism given that they don’t really say what the probability of life would falsify their theory. I know it’s cringe to talk about these things in scientific terms but the FTA tries to resemble a scientific theory.
For your second paragraph, the whole point is that we don’t really know if the laws of physics are changeable. They could be brute facts (and some polling seems to suggest a plurality of scientists believe that’s the case. I have 2 brothers. What if I had five? It’s a silly what if, that raises interesting questions, but it doesn’t elucidate possibility. More data needed. I’m not going to argue with you on the probability theory (because I’m bad at probability theory!). However, the whole force of the FTA seems to be a spotlight effect triggered by not having falsifiable criterion for the theistic hypothesis. Namely, yes, it could be true that things are different, and we should metaphysically take that seriously, but unless we have scientifically-informed constraints about what is or is not possible, we can’t narrow it down. So, when I say it’s a spotlight effect, I mean we’re spotlighting how we don’t know something or something seems weird. It is weird! But they’re trying to impose the constraint that it can’t be different or it can only be different in a specific way. We don’t know that! Creating that probability architecture (or calculus? I dunno) while not yourself having a good theory that fits the evidence and also creates testable theories seems very off to me.
As for the third paragraph, my point in criticizing those specific examples (even if I didn’t say so explicitly) was that I thought they were really half assed in how he wrote them. I saw him on a stream do a better example of the 500 shooter thought experiment, and just how clumsily expressed it here made it open to easy criticism. As for the actual point about evidence, I think I made an edit about this, not sure if it saved (as I was on my phone and substack sucks in browser for long post edits). I think I did get that part slightly wrong, but not so wrong. BB seems to think that any evidence that fits within a model is good evidence for a model, when a piece of evidence can fit a model, while still not being compelling. I use the example of having shoes on. We know the killer had shoes on, and our prime suspect also had shoes on. Vioala! Good evidence. Even though all suspects were wearing shoes.
As for the fourth paragraph, I was just relaying the principle of indifference as a general idea (as expressed in the documentary). It is a really fallacy people commit, and understanding that helped me express my next point. Not going to retract that (I do sincerely think you’re flat out wrong - the baseline probability of it being any day of the week is closer to 14%, not 50%). That whole section is about how models can’t predict correct outcomes while still being kinda garbage, and we don’t elevate them *just* because they were correct
Finally, (and again, I hate to come off as mean - I’m not! I promise!), you’re wrong about agnosticism! There are different kinds of agnosticism, one of which is just suspension of belief. You can be a suspension of belief agnostic on lots of things. I think regardless of one’s beliefs about god (atheist or not) we should be agnostics about the cosmological design because there’s just so much here we don’t know. Our position on the state of the world - as it relates to cosmic origins - should be agnosticism until we get more data. I think BB and FTA proponents oversell the certainty and the science here. I have seen him argue with actual physicists and at least one physics-philosophy PhD who tells him he’s wrong. To put it more simply, a non-agnostic position would say that, within current knowledge either: we have good reason to think it was designed and therefore convert to Xenu OR we know for sure it wasn’t designed and therefore burn down the churches. An agnostic simply says “wait and see.” I’ll admit I’m rigging the scales a bit by facetiously attaching actionable steps to the hypothesis, but that’s the stakes we’re talking about here, in the same way, say, one takes bets on predictions. I don’t think anyone in good faith can say that we have enough information here to conclude the FTA is true. The fact that most physicists don’t agree with someone like Luke Barnes tells me that there’s more going on here, more data to be collected.
thanks for the comment or discourse! I hope I wasn’t mean here (haven’t eaten in hours). Next time, write this is a post that way it can be signaled to me that I need to take more time to respond (switching tabs in the substack iPad app is terrible so I did a lot of this from memory)
Before beginning my response, I just want to say that in no part of your response did you come off as mean, so no need to feel guilty about that! Also, I should warn you based on what you said at the end of your last response that my response here is very long.
> The idea that we don’t know the probability comes from *the physicists themselves.*
The physicists you cite are saying that there's no physical process that you can use to assign a probability to different values of the physical constants. That's fine for them to say, but it has nothing to do with the fine tuning argument because that's not how the fine tuning argument comes to the conclusion that it would be very unlikely a priori for the universe to be finely tuned unless someone intentionally fine-tuned it. That conclusion is based on the fact that, supposedly (I say supposedly because I'm skeptical of this part of the FTA), the constants all had to be within an extremely small range, relative to their actual values, in order for any conscious observers at all to exist. The fact that they had to take on such specific values is what makes the prior probability low, not the fact that some process by which the values were set was unlikely to choose those particular values. The fact that we don't know what process caused them to take on these values, or if there even was such a process, doesn't affect this assessment, because even if there was a process that made the constants very likely to take on values within a small range, it could be any range, not just the specific range of life-permitting values.
Also, I should note that physicists use the low prior probability of constants taking on a particular value as an argument all the time. Many of the big open problems in physics (e.g., the hierarchy problem, the vacuum catastrophe, the strong CP problem) are just observations that a certain constant has a weird value, and the demand that there must be some explanation for the constant having a value with such a low prior probability of occurring. Cosmic inflation, a theory believed to be true by most physicists, is motivated in part by the flatness problem, which asks why the density of the early universe just happened to be so close to the critical density. And many theories are rejected on the basis that they can only be made to line up with observations with extreme fine-tuning of their parameters. These arguments don't include a model that explicitly gives probabilities but are based on intuitive reasoning. In fact, they often use the exact same reasoning as the FTA. In many cases, the reason why the value is surprising is because, based on currently known physics, it requires an extremely precise cancellation that would only be possible if some value is finely tuned to factor of 1 in 10^(some large number), which physicists consider extremely unlikely to happen by chance even if they don't know what process sets that value, or if it's just a brute fact or not. In fact, physicists use this kind of argument so often (even using the same terminology of "fine-tuning") that some apologists have misinterpreted physicists talking about these problems and thought they were actually talking about fine-tuning for life (This is one of the reasons I'm skeptical of the claims about how finely tuned the universe really is for life. I know that at least one of apologists' favorite claims, the idea that the cosmological constant is finely-tuned, is false and based on a misinterpretation of things physicists said about the vacuum catastrophe).
So if you want to appeal to the authority of physicists to determine whether this type of reasoning is valid, most physicists say it is. In fact, many physicists endorse the fine tuning argument itself (though usually they use it to conclude that the multiverse exists rather than God).
> On the point of saying that it’s impossible to set probabilities, I’m going to say, yes that’s true, and it’s not absurd. We just don’t know enough, at this point.
We don’t know enough about what? We’re talking about prior probabilities here. If we know very little, then we should set our probability distribution over possible values to be very wide. That actually improves the case for assigning an extremely small probability to the constants taking on values in a very small range. The only way that you ever could say that the probability of a constant falling into some range where the lower bound differs from the upper bound by an extremely small factor isn’t low is if you already knew a lot about the constant and had strong reasons for thinking that the constant’s value was very close to that range.
> we could feasibly get to a point where we have enough data and working models, but we’re not at that point.
What data and working models do you think we need? It seems like you’re just assuming that we need a “roulette wheel” like the quote said in order to assign a probability, but we don’t. If I tell you that I have a weighted 6-sided die and roll it, your credence that I will roll a 1 should be 1/6 even though you don’t know how the die is weighted. Having a model of the die would cause you to change your credence, but not having it doesn’t mean you can’t assign a probability at all.
> What’s more, I don’t think “ordinary reasoning” is something one uses in physics.
This is beside the point, although I also think it’s false - physics uses some reasoning tools that aren’t used in everyday life, but it’s not like there’s no continuity whatsoever between reasoning in physics and ordinary reasoning. Unless you think that the reasoning we use in everyday life is completely invalid, then an argument that would rule out that reasoning must have something wrong with it.
> Your complaint about the roulette wheel is not valid IMO, the argument is not that we can’t make any probabilistic assumption in physics, it’s that we lack any understanding about the specific concepts of life, universal laws, and how they happen, etc.
The argument you quoted about a roulette wheel is claiming that we have to know what process caused the physical constants to take on particular values (if there even is such a process) in order to say anything about their prior probability distribution. But this is not true. What probability would you assign to the proposition that the 1,001 through 1,010th digits of the fine structure constant are your phone number? I think you should assign a probability of about 1/10^10 of this, and you can do this without knowing anything about how the fine structure constant came to have the particular value that it does, or if there even is a reason for its value. I think that this is obvious and should be uncontroversial. Now sure, if you collect further data, you might find out something that changes your probability assignment. But that doesn’t mean you can’t assign a probability before you have that data: Probability is supposed to update in light of new evidence, and this doesn’t mean that the previous probability was wrong - it represented a conditionalization on different evidence, i.e., a different state of knowledge and uncertainty.
To be clear, if what you’re trying to argue with that last point I quoted is that we don’t know enough about life and how the constants of nature came about to justify assigning a probability as low as proponents of the FTA say, then I actually agree with you. As I said previously, I’m skeptical that the range of life-permitting constants is really as small as they think, and even if it is that small, we do know that there’s something special about the particular life-permitting values - the very fact that they’re life-permitting is special, and it also implies other special things like the ability for complex structures to form - so we can’t use the principle of indifference to say that the probability that they’d fall into that range is the same as a comparably-sized range. The place where I disagree with you is in your rejection of using probabilistic reasoning at all just because we lack knowledge - domains where we lack knowledge are exactly where probabilistic reasoning is most relevant. Instead of saying, “We can’t say anything about the prior probability of the constants of the universe being finely tuned on atheism,” we should actually say, “The probability is much higher than FTA proponents think it is, and certainly much higher than naively calculating it by multiplying together the relative sizes of all the ranges would suggest.”
> Secondly, as I said in the follow up to this post, there’s no falsification criterion for the FTA, so it’s hard to say that naturalism is improbable relative to theism given that they don’t really say what the probability of life would falsify their theory.
I haven’t read the follow-up yet, but this seems like a different argument than the one I’m objecting to. I’m not saying you should accept the fine-tuning argument, just that a few particular objections you made were wrong.
> For your second paragraph, the whole point is that we don’t really know if the laws of physics are changeable.
I’m not talking about whether the laws of physics are changeable, but whether they could have been different. If it’s a brute fact that the laws of physics are what they are, and it’s physically impossible to change them, that doesn’t imply that they couldn’t have been different - in fact, it implies the opposite. They could have been any arbitrary values and just happen to have the values they have. On the other hand, if there is some process that led to the particular values of the physical constants, they could have been different if their values had been determined by a different process, or in a possible world where their values are brute facts not determined by any process at all.
> Namely, yes, it could be true that things are different, and we should metaphysically take that seriously, but unless we have scientifically-informed constraints about what is or is not possible, we can’t narrow it down.
I’m not sure exactly what you mean by this, unless you’re trying to state some sort of modal skepticism. Science tells you very little about what’s metaphysically possible because science just tells you what’s true about the actual world, not what would be true in possible worlds that don’t exist. But it’s pretty uncontroversial among scientists that the laws of physics are not metaphysically necessary. No one thinks the world had to behave exactly the way it does - you can easily come up with scenarios where the laws of physics are wildly different, and physicists come up with such scenarios all the time.
> But they’re trying to impose the constraint that it can’t be different or it can only be different in a specific way.
The FTA is definitely not trying to impose the constraint that physics can’t be different. Arguably, it is trying to impose the constraint that the physical constants can only be different in a specific way, since it basically assumes that if naturalism is true, the values of the constants must be brute facts, or at least determined in a way that has no correlation with life permitting values. I agree that this is a flaw in the argument, but like I said before, this just means we should say that the probability of life permitting values given naturalism is higher than the FTA says, not that a probability doesn’t exist at all outside of a specific model of how the constants are determined.
> BB seems to think that any evidence that fits within a model is good evidence for a model, when a piece of evidence can fit a model, while still not being compelling. I use the example of having shoes on. We know the killer had shoes on, and our prime suspect also had shoes on. Vioala! Good evidence. Even though all suspects were wearing shoes.
I think this is a misunderstanding of his reasoning. It’s true that, if some evidence is predicted by a theory, then it makes the theory more likely, unless it’s predicted by all possible theories. This is a theorem of Bayesian reasoning, but it’s also how scientific experiments work. Of course, as your shoe example shows, it’s not necessarily good evidence - it may increase the probability by a negligible amount. That’s why proponents of the FTA have to argue that the probability of fine-tuning on atheism is low - otherwise, it doesn’t increase the probability of theism by much. So BB doesn’t believe that *any* observation that fits with a theory is good evidence for that theory - he just thinks the probability of fine-tuning is low on atheism and high on theism, which would mean that fine-tuning is good evidence for theism.
> I do sincerely think you’re flat out wrong - the baseline probability of it being any day of the week is closer to 14%, not 50%
I think you misread that paragraph because I didn’t say the baseline probability of it being any day of the week was 50% - I said the opposite. The principle of indifference gives you a probability of 1/7.
Regarding the points on agnosticism, it’s true that there are multiple kinds of agnosticism, but none of them are alternatives to theism or atheism considered as propositions. Theism and atheism exhaust the logical space, so any version of agnosticism that said that neither theism nor atheism is true would be logically impossible. Agnosticism, when considered as a proposition, rather than just a state of belief or knowledge, is a proposition either about epistemology or the state of the evidence (e.g., “It’s impossible to know whether God exists or not,” “There’s not enough evidence to say whether God exists or not,” “Claims that God exists are unjustified, but so are claims that he doesn’t.”). So agnosticism is not an alternative to “God exists,” or, “God doesn’t exist” - rather, it is an alternative to positions like, “We should believe that God exists/doesn’t exist,” or, “I know God exists/doesn’t exist,” or, “There’s good evidence that God exists/doesn’t exist.” Maybe I just misinterpreted what you were saying about agnosticism, but this is why I objected to you talking about P(agnosticism). That doesn’t need to be considered when judging P(theism) because agnosticism isn’t some alternative hypothesis to theism - rather, it’s a statement about how good the evidence is or what we can know.
> Our position on the state of the world - as it relates to cosmic origins - should be agnosticism until we get more data.
I agree with this. I think any individual hypothesis has a probability <50%. Again, I may have misunderstood what you said because you used the phrase “more probable than agnosticism,” which made it sound to me like you were considering agnosticism as an alternative hypothesis to cosmological models when it’s really just a statement about how likely any cosmological models are. I do still think your criticism of BB is unfair, though - he doesn’t claim to have certainty, and he certainly doesn’t blindly assume that you have to be certain one way or the other or that the most likely cosmological model must have probability >50%. He just thinks the evidence for God is strong and therefore places a high probability on his existence.
"How can one's existence be evidence for something that didn't occur?"
Easily, evidence for is not the same as proof. So a can be evidence for b even if b didn't actually happen, so long as learning a) makes it more rational to believe b) happened than it was before.
As a concrete example. Suppose I actually didn't commit a fatal srabbing, but by coincidence, I was seen fleeing the area of the murder five minutes after it happened carrying a knife that matched the murder weapon. The fact that I was seen fleeing the crime scene, armed with a weapon matching the victims wounds is obviously evidence I committed the crime even though in fact I am innocent. So it is evidence for a false claim. Why is it evidence for that false claim? Well, because if someone learnt that I was fleeing the scene with a big knife, they are rational in becoming more confident that I was the murderer than they were before they got that information. That's why it would be admissible evidence at a trial, rather than barred by a judge as being irrelevant.
Yes, I'll admit when I wrote this I wasn't framing it the best way. It is evidence, certainly. I was wrong!
However, different contextual considerations can determine if it's good evidence or not. Good evidence being evidence you take seriously and doesn't have a defeater that undermines a causal explanation. IMO, that's the more important problem/distinction when evaluating evidence, not the *mere existence* of purported evidence being itself evidence, if that makes sense.
BB and design argument advocates seem to argue that "because most people are born via sexual union without contraception, you exist, therefore it is good evidence of your parents doing that, and therefore they did." The problem is that it's extrapolating a probabilistic model that is not certain as if it is. And the evidence is not actually a law of the universe, but a probabilistic deduction that was arrived at without investing relevant evidence that could have been a defeater.
Put another way, Bentham in the past and apologists have argued that the number of arguments for God's existence is itself evidence for God. That's a little silly because some of the arguments are bad, and just because you categorize bad arguments under the same category of "argument" doesn't mean they are the same thing.
We're basically incentivizing both gish galloping (throwing out a bunch of nonsensical points and tallying them up as a score) and not actually look at the facts of a specific case. To use another metaphor, it's like looking at 5 murder suspects and honing in on one, a man, over the other four, all women, because men are overrepresented in murder statistics.
As I say in the article, it's often a safe bet to be informed by statistics and probability, but I don't think it's satisfactory. I think it's too broad of a definition of evidence to basically include all data points. The more data points you have, the easier it is to over fit!
I don't really follow what your saying here, but that might just be because I don't think I've ever read all of BB's original piece that your responding to.
Returning to my past notes and this piece (tho not BB's as my coffee isn't finished yet), the issue I always have is trying to figure out if we're on the same page that probability is just a description of the most likely to be accurate guess you could make given the constraints of some information and of not having other information. When a voter brings her pen to the ballot, there isn't any probability involved in what candidate she will vote for; there is no "chance" element that messes with coins flipping.
BB is aware that probability is not describing, literally, the mechanics of what would happen if you reset the universe a bunch of times, but rather is a tool for dealing with his lack of knowledge of what did happen. But it would only work as an argument *that something was true* if you thought it meant something more than that, or at least, that's how it appears to me. And when you say things like "the argument is majorly supported," that sounds to me like that's what you think -- that the fact that you would feel more comfortable putting money on your bet actually has some effect on whether the bet was a good one. But I don't want to treat him as if he's saying that, because that's... crazy.
Which idk, to me this just brings us back to Pascal's wager. My natural tendency is toward a sort of apatheism, so unless I hear a good argument otherwise I don't really see why I have a need to gamble here. I truly have no idea if my default belief ("As far as I can figure it kinda just seems like stuff has always existed bc that's what stuff does") has good odds of being correct, what matters (insofar as this matters at all) is that it either is or it isn't, and that I personally am unlikely to ever be able to know which it is.
Follow up: OR to be even more narrow, it’s like an example I saw Joe Schmid make a few years ago (I watched it recently). If model 1 has a jar of 80% orange marbles and 20% brown marbles, and model 2 the inverse, if we draw a marble blindly and it’s orange, it supports model 1. One retort to this is that this isn’t good evidence. To which he glibly cocks his eyebrow, smirks, and says “I’m not sure what that means?” From secondhand observations, I think Schmid and BB agree on this. There’s no criterion for bad evidence, which is actually a little silly. It’s like saying I believe Liverpool will win the premier league because Lebron James is a minority owner. How are the two connected? This is where theistic and agnostic philosophical types really disappoint me. You can create a model for anything, but the model they’re putting forth kinda sucks because it’s not really predictive for the future; it’s just retrospectively fitting the data.
Yeah I think this is a good way of putting it. It feels like there’s a lack of preceding conversation of what this is all *for.* I think the likelihood vs. evidence distinction gets at the difference in thinking - it’s like, Monty Hall brain vs. “the car has 100% odds of being behind the door it is behind” normie brain. And obviously one *is* better math/stats! But that’s a *tool* to get us less blind toward truth, not itself the goal.
To riff in a totally different direction, this is also my issue w/ takes on shrimp welfare - there isn’t an underlying discussion of what morality is or what it’s for. I had this interaction w/ him the other day - if you press hard enough on “but why should I care about shrimp pain” there’s nothing past “it’s prejudiced not to care about shrimp and immoral not to care about pain,” no argument that we shouldn’t be prejudiced and immoral.
Yeah, I don’t care about shrimp. I think this is an instance of “defining words in a certain way and then taking it to absurd conclusions.” Like, I said this somewhere else, but people get into ethics to 1) live a better life 2) help people act better toward one another and 3) help other people live a better life. The shrimp welfare (and its contemporary “maybe human extinction wouldn’t be so bad”) perspective is very much a reductio ad absurdism - and obviously so! How do you go from “let’s help people” to “maybe it would be ‘good’ to genocide all sentient beings who have the capability to suffer.” Not only are normal people going to think you’re weird at best and a public safety threat at worse, it’s like going into Christianity to convert people to atheism or to going into architecture to be an arsonist! It’s prima facie wrong.
i think some of this comes back to some intermediate or higher level of reasoning. Like, what is evidence and what is probability? when we evaluate evidence, is it a probability machine in our head going up and down? In some ways, this is true, but when I try to go all metacognitive on my evaluation of the evidence, it’s not like a dial that’s going up and down. It feels like it when it comes to Bentham and Bayesian theists.
Like the idea that a piece of evidence (circumstantial evidence?) is good evidence just because you expected it to be the case, given the conclusion. Let’s say I’m cutting vegetables with a knife and I cut my hand and have to get stitches. But while I’m gone, someone steals that knife to commit the murder. If they looked for DNA and found my DNA on that knife, that would be evidence that I committed the murder. It’s what you would expect if I committed the murder. Obviously, there are defeaters for that theory, given access to other information (my medical records, hospital CCTV, etc), but in this sort of situation (i.e proving God) we don’t have access to such information. So, I would be the murderer (or likely the murderer) in that situation.
And this is where I think BB’s formulation is wrong. Because we’re either evaluating theories super mathematically/scientifically in which we can make specific measurable predictions or we’re doing it narratively. By narratively, I mean we evaluate evidence and give no credence, high credence, or low credence. We evaluate that credence based on our informed understanding of reality (which we can’t do with the beginning of the universe!). So, if my case was presented to the jury, the fact that DNA was on the knife would prima facie be a high credence evidence, but with the other context and defeating evidence, it becomes 0. Not .1%. 0. People (or at least I) do not attach numbers to these probability, it’s either 2 thumbs up (definite), 2 thumbs down (definitely not) or 1 thumb up/down (who knows!).
But with Bentham’s model, it seems likely he’s just using a dial and going credence up/credence down. Maybe this is Bayesian, but the evaluation of the inputs is a lot more credulous than me! You mention Pascal’s wager and I think that’s a good illustration. On its face it’s silly if you’re a skeptic or an atheist (I don’t believe in Hell even if God is real) so the calculation is silly to me. But using BB’s model I feel like you have to take it seriously and not really evaluate inputed evidence in the same way (and thus infinity skews the outcome and thus you’d have to believe). This isn’t philosophy, it’s naive math, in my view
Not having read BB's original post, I didn't really get the point of the discussion about conception and birth control. Maybe you could ad a sentence or two saying what the claim is, for those coming to this cold... But I might have missed something as I am in bed with a virus.
What I saw of BB's logic in his "argument from moral knowledge" didn't really impress me. I think theists who have strong convictions inevitably make logic subservient to those convictions, and their arguments can look a little ridiculous to those outside the framework.
We have no idea what it takes for a universe exist and, until we do, discussions of probability entirely miss the point. As you say, if multiple universes are possible, the anthropic principle dictates that a very small fraction of possible universes will be observed. No matter how small that fraction is, it would be a mistake to appeal to its value as evidence for divine intervention.
Being even less of a mathematician or physicist, I can never tell if I’m missing something, if BB is missing something, or Bayes is missing something, but I always walk away with the impression that we do not share an understanding of what statistical chance is.
My honest assessment is that 1) BB has a better logic/math brain than I do 2) However, that does not mean he is correct 3) Many of these arguments have implicit premises/evaluations that amount to personality attributes and 4) As I said in the piece today, if you understand that Bayes is about what is *probable given possibility* and that it involves big data sets, you realize that it’s not a good way of evaluating God’s existence.
I *do* however think he oversells his case and often opines about things he doesn’t know about. I once saw him arguing with a philosophy/physics PhD on here about the design argument, and the PhD was like “seriously, it doesn’t mean this” lol
Howdy,
I have not gotten far into this article, but I would like to flag that this paragraph is a pretty big misrepresentation of theistic philosophers.
“Another example is the constant insistence by theistic philosophers that dress up the idea (in some capacity) that we need an external agent (God) to “give” us morality, when there is abundant literature of how moral instincts in humans are evolved, observable in babies, and yes, in non-believers and non-religious societies.”
There are two broad categories of moral arguments put forth by apologists which could match what you are describing. (I’m presuming you definitely aren’t referring to the much more heavily written upon moral argument found in Kant)
The first is Craig’s moral argument. In every rendition of his argument, down to his 5 minute videos, Craig explicitly distinguishes between a need for God in morality and a need for Belief in God for morality. His argument is about moral ontology and says nothing of moral epistemology. Thus, your criticism cannot apply to this argument.
The second is the moral knowledge argument, which is at least about moral epistemology. Nonetheless, this argument also asserts that all humans have moral knowledge. It also affirms that they gained this knowledge through the accepted evolutionary processes. What they deny is only that a naturalist can say that this process would select for the objectively correct moral beliefs. This argument was first put forth by naturalist anti-realists, such as Sharon Street in this enormously influential paper (for a philosophy paper). Your claim that “but we did evolve true moral beliefs” would be question-begging in reply to this argument.
I can happily point you towards resources that respond to the actual moral arguments made by theists if you so choose. 😁
I’ll concede that my language was sloppy here (I added this little section at the last second) and I probably should have said something along the lines of “popular Christian Apologists” instead of theistic philosophers (we can probably agree there’s a substantive difference between the two!).
Having said that, I believe my characterization is probably still true of Craig. When I say that theistic philosophers say we need God to “give us” morality, I’ll admit that sounds epistemological, but what I’m getting at here is that the concept of objective moral facts coming from (in some capacity) a subjective agent of some sort (which I would argue God is). “Giving to” may sound like an epistemological description, but my actual meaning is that the theist position is that moral facts (and thus morality) cannot exist without God. My presumption is that it’s a little silly because it just seems absurd, if not internally inconsistent (as I’m implying but not communicating clearly), then empirically wrong (the baby example). Lurking here is a meta-ethical disagreement (I’m a moral non-realist). Craig and other theists present as if moral facts exist, come from God, etc and thus humans who do not accept them are not moral in the same way. He may not explicitly say this, but when he’s pressed on it his arguments and tone revert to Turek levels of pearl clutching that lead one to believe he thinks atheists can’t be moral or that we wouldn’t be truly moral or maximally moral without Christianity; the same goes with objective meaning (see: his debate with Shelly Kagan), but that’s another story.
As for the second aspect, I don’t think it really matters as it’s completely different from what I’m arguing. I don’t think moral knowledge exists! My point is simply that, to the degree that anything resembling “morality” does exists, it’s informed by a psychology that evolved to coexist in a group setting. We can describe the attributes that inform this psychology (the Jonathan haidt moral foundations thing) and like all of our sensory perception and instincts, they don’t exist to establish absolute knowledge, but as reactions to our environment, steering us in the direction of an equilibrium of cooperation. That’s not the same as knowledge! And that’s my frustration with moral philosophers in general, is that outside of virtue ethicists, there’s really no mainstream reckoning with these findings in psychology or refinement with these concepts. So no, my argument is not the question begging “but we did evolve true moral beliefs,” it’s “this is not how people compute and conceptualize morality and this trail of thought leads us in the opposite direction of the truth.”
So, I fully concede that I didn’t communicate this paragraph optimally, but I still stand by the spirit of the criticism. Out of laziness and giving you due credit, I won’t change it and if anyone has any further comments in the similar vein, I’ll refer to them this discussion :)
You're right that fine-tuning is a scientific debate, and evaluating it properly requires much deeper understanding of physics than philosophers normally have. I'm only beginning to understand the physics underpinning the fine-tuning argument, but it's obvious that Adelstein knows less about it than I do. For example, in his fine-tuning articles, Adelstein barely mentions the work of Fred C. Adams, who has submitted the fine-tuning argument to detailed scientific scrutiny. Until he read my work, Adelstein seems also to have been unaware of the most powerful critique of fine-tuning, which is that we can't know what would happen if we tweaked the physical constants of our universe. I discuss all this in my article:
https://open.substack.com/pub/eclecticinquiries/p/the-fine-tuning-argument-cant-get?r=4952v2&utm_campaign=post&utm_medium=web&showWelcomeOnShare=false
Yes! I read your article and my first thought was "whew, thankfully I can just cite this/reference and not come up with the argument on my own, or whatever I inevitably write on design will be much shorter" lol I highly recommend it!
He also said in a comment to me (though I think it was an accidental comment maybe meant for you, ironically, but I could be wrong on that) which says he emailed Adams and said that Adams agreed with his portrayal of the physics and call the agreement trivial. Which tells me BB didn't understand the nature of the disagreement/physics more than anything.
I'm working on an FTA post, but I think I'm going to avoid talking about BB, or at the very most, he'll be a side character. Even if you disagree with him/the FTA, I think he is doing innovative things with it, if you look how it's in discourse with other thinkers. That doesn't mean I find it convincing or agree! But it is interesting and worth considerations. Unfortunately, I think he does adopt views of the science that aren't necessarily true, and conceptions of philosophy of science one need not adopt as an atheist (but on the latter point, I think that's common of many FTA proponents)
If you can find that comment where Adelstein talks about his email to Adams, I would like to read it.
Adams is a prominent critic of the fine-tuning argument so it would be odd if he said what Adelstein claims he did.
Adams agrees with fine tuning, but disputes the degree of fine tuning. You can read his dissertation to verify for yourself.
Adams unambiguously rejects the fine-tuning argument. See the link below for quotes. What he's saying is that fine-tuning arguments are superficial because fine-tuners don't consider all of the ways in which the universe might support life. I give his analysis of carbon production as an example in my essay. He does not think that we have the final word on all of the ways in which the universe might be life-supporting. He's saying, based on what physicists know now, there are more ways that the universe can support life than the fine-tuners acknowledge.
https://open.substack.com/pub/eclecticinquiries/p/the-fine-tuning-argument-cant-get?r=4952v2&utm_campaign=post&utm_medium=web&showWelcomeOnShare=false
I liked the beginning of this essay when you made more general points about apologists, but I found the part specifically addressing Adelstein on the nature of probabilistic reasoning to be bad. I generally think objections to the fine tuning argument on the basis that we don't know the probabilities involved to be the worst ones. The fine-tuning argument doesn't need you to know the exact probability of fine-tuning, just that the probability given naturalism is extremely low, and proponents give arguments as to why the probability is low. You could object to this by arguing that the probability isn't actually as low as they think, but the objection that it's impossible to assign any probability at all, even a rough approximation, is silly and would rule out basically all ordinary reasoning. Some of the objections you made assume that in order to have a probability of fine-tuning at all, you need to have some physical process that could set the constants in certain ways and had a certain probability of setting them in each way (for example, the quote complaining that there's no roulette wheel makes this assumption). But this is absolutely false - if it was true, we couldn't assign any probability, even a rough estimate, to any hypothesis about physics being true, since after all, whatever the most fundamental laws of physics are, there's no process that causes them to be what they are and could have made them behave differently (otherwise, the laws that govern that process would be the true fundamental laws) - they just are. So this objection would imply that we can never judge one hypothesis about the nature of reality or fundamental laws as more plausible than others. Every time we do an experiment in physics, we are really just obtaining evidence to update our probabilities that the laws of physics work a certain way. So if we had pure, Knightian uncertainty, we wouldn't even be able to do physics.
The point about possibility is conflating between different types of possibility. On the one hand, there's physical possibility, which says that something could have been a certain way without changing the laws of physics. That's obviously not relevant to the fine tuning argument, since it's about the probability of having different laws of physics. And it's not relevant to probability assignments over various hypotheses either: Unless we're absolutely certain of the true laws of physics, we can obviously assign nonzero probability to scenarios with differing laws of physics, and, as explained in the previous paragraph, we have to do that in order to even do physics. The FTA is based on the low prior probability of fine-tuning, and there are no a priori reasons to rule out different values.
The form of possibility that actually matters to probability assignments over possible laws of physics is metaphysical possibility, which just means "any way the world could be". If something is metaphysically impossible, so that there aren't even any hypothetical scenarios in which it's true (e.g., a logical contradiction), then sure, you can assign probability zero to it to a first approximation (you actually shouldn't do it even then, unless you're 100% certain it's metaphysically impossible). But different values of the constants are definitely metaphysically possible - the only people who dispute this are necessitarians, but necessitarianism also completely prevents you from making the "metaphysically impossible -> 0 probability" inference, so even their objections to this point aren't relevant. And metaphysical possibility is firmly within the realm of philosophy, so the charge of philosophism doesn't apply here as it does to the Kalam argument or Thomistic metaphysics (both of which it definitely does apply to).
A lot of objections to the specific points he's making seem to misunderstand his points. His thought experiment about birth control is not a model of how the world works - it's a hypothetical scenario used to illustrate a point about probability. Objecting that the assumptions made in the thought experiment aren't true in real life misses the point. Also, it's obviously true even in real life that someone's existence is evidence that their parents have had sex without birth control. I'm not sure why you deny this is the case (evidence in favor of =/= 100% proof, so the fact that some people are conceived despite their parents using birth control or without their parents having sex is not an objection to this point).
Similarly, your objections to his use of the principle of indifference are strawmen. No one thinks that we should assign the same prior probability to, "It's Monday," and, "It’s not Monday." All formulations of the principle of indifference include some qualifier stating that there are no relevant differences between the options or that they are all "the same" in some way, which is not true in the case of Monday vs. not Monday. Maybe you think such qualifiers are too vague, fair enough, but you yourself used the principle of indifference when you claimed that actually the probability of it being Monday should be 14%. You understand that there's no reason to consider any day more likely than any other a priori, and so you assign the same probability to each. Similarly, in the election example, you shouldn't apply the principle of indifference because you have information about the candidate's positions, polling data, etc., that might make you think one candidate is more likely to win than the other. The principle of indifference does not deny that you should update your probability based on this evidence - that would be crazy. It would only apply in the election example if you knew nothing at all about the candidates, in which case assigning them an equal probability of winning sounds pretty reasonable. The principle of indifference is really just a statement of the fact that you should not arbitrarily consider one possibility more likely than other unless there is some reason to think it is. Also, note that the quote from Adelstein is actually making a claim much weaker than the principle of indifference. All that's required for his quote to be true is that the probability of the options don't differ wildly from each other, nothing that they're all exactly equal.
I'm not really sure what you're trying to say in the section on agnosticism. Agnosticism isn't some third possibility aside from theism and atheism, so there is no "probability of agnosticism." Agnosticism is a description of someone's state of belief or knowledge, not of how the external world actually is. By definition, either theism or atheism is true, and agnosticism is just the position that we don't know which, or a personal suspension of judgement. And I don't think Bentham's Bulldog has ever said that any theory is better than agnosticism. He just thinks the evidence for God us good enough to motivate belief.
okay, so we’re clearly talking past each other so this response is not going to be as long. I’m not responding to everything because I don’t have time, so I’m going to reduce my criticism as succinctly as possible: The fact that you say “many physicists endorse the fine tuning argument” at the end of your first response highlights the problem and how BB and others mislead people and it leads to talking past each other. Because the physicists use the phrase fine tuning to mean different things from philosophers. I’m not denying that, the OG victor stenger didn’t either. The whole issue is whether or not there is empirical basis (i.e external validity to current models) to make high-confidence statements about what the world would be like if one variable changed, or if it was completely different. *We can’t model life arising in our current universe with our current models.* Some of this is computational (we don’t have the computing capacity to model billions of years). Some of this is empirical (our current models will say the universe at time x should not have complexity such as a galaxy, but using different methods, like redshifts and telescopes, we find this to be false constantly - there’s a great discussion about this in both the documentary I cited and a follow up by the same creator)
The skeptic (and as far as I know, physicist position) is skepticism toward the *causes* of fine-tuning (which is the subject of debate). No one is disputing that life wouldn’t arise on earth if the constants were different, but we can’t make an informed guess to probabilistically model how things were different *because we know so little.* It’s not a permanently unknowable truth, of course, but it’s still very hard at the moment. I used this example in a note yesterday, but it’s like me asking you who the king of a country was in a certain year, while providing only one data point and you have no other background information. Sure, you can come up with a model, but it wouldn’t be rational to trust it just because there are intuitively compelling reasons to believe it. Because the world is usually counterintuitive, and there are so many lurking defeaters to the proposition. That’s why most models are built first by observation. For instance, the multiverse (and other hypotheses) are informed not just by the realities of fine tuning, but other math that is beyond my comprehension. The FTA is an affirmative proposition (that life could not arise given the conditions of the universe and therefore there is a creator). We don’t know either clause is true!
The thing that does piss me off about BB (and when I say this I’m not mad at him as a person, but something he does), is that he oversells his case on issues like this by either not understanding or not properly communicating probability correctly, to say nothing of the physics, and so we have lots of people parroting his arguments. It leads to falsehood.
I don’t want to speak for the guy, but I saw a few months ago Bryan Frances (who has a master’s in physics, was thinking about getting a PhD in physics but instead went the philosophy route) spar on here with BB and basically “this is not what you say it is.” The same goes with physics-trained philosophers like James Fodor, in his own conversations with BB. The same goes for actual physicists who disagree with Luke Barnes; the most common response I see to him is “you didn’t understand what I said.”
You may be correct about all these things on probability and physics, but I think the medium of our discourse is not conducive to communicating it. I’m not following it! It may be your writing, the fact that my dad is in town and I’m distracted, or that it’s hard to say these things succinctly. Regardless, I can’t put much credence in it because, to use bayesian language (which I hate because my hot take is that people don’t think as bayesianly as though), I have such a low prior of it being true because of how I learned the scientific process (approach a subject without priors for long as possible) and because of the preceding paragraph. As you’re defending BB, I can’t really tell how much of it is you trying to steel man his arguments or is just your own opinion, informed by your prior knowledge. I suspect it’s a combination of both, but I think you give way too much credence to BB’s knowledge on the subject.
I highly recommend watching his livestreams with James Fodor (who iirc has an MA in physics) as Fodor’s content does a good job of pointing out the science-based flaws in FTA, but also in direct conversation with BB is good at point out how he jumps to conclusions pretty quick. I also recommend listening to the comments of Niayesh Afshordi on Paul Halper’s documentary on FTA (and the follow up video discussing responses to it). These people are smarter than me, know physics and probability better than me, but the reason why I trust them is because scientific modeling is based on two key concepts (among others!) : external and internal validity. They’re able to communicate why the main premises of FTA depend on hypotheses that lack both forms of validity, and so we should give low credence to them working. To me, that’s the main reason we accept agnosticism. The reality of fine tuning (as physicists understand it) is real, but there’s really no good explanation for it either way (yet!).
For your first paragraph, I think you’re wrong on a lot of point. The idea that we don’t know the probability comes from *the physicists themselves.* In the language of science, it’s unclear what the external validity of the modeling shows, and that we can’t generalize it to other universes, because they seem to be falsified in our own. On the point of saying that it’s impossible to set probabilities, I’m going to say, yes that’s true, and it’s not absurd. We just don’t know enough, at this point. Maybe impossible isn’t the correct word, we could feasibly get to a point where we have enough data and working models, but we’re not at that point. What’s more, I don’t think “ordinary reasoning” is something one uses in physics. Your complaint about the roulette wheel is not valid IMO, the argument is not that we can’t make any probabilistic assumption in physics, it’s that we lack any understanding about the specific concepts of life, universal laws, and how they happen, etc. We have theories that may work out, but we know very little. The same can’t be said about experiments within physics (like, I dunno, gravitational waves). I don’t think it’s necessarily inaccurate to say that a lot of (but not all) data collection/experimentation in physics is updating our probabilities, finding outliers, and trying to find new models to explain the outliers that don’t conform to your current models. I’m not sure if I’m understanding you correctly here - I’d appreciate elaboration on this paragraph.
Secondly, as I said in the follow up to this post, there’s no falsification criterion for the FTA, so it’s hard to say that naturalism is improbable relative to theism given that they don’t really say what the probability of life would falsify their theory. I know it’s cringe to talk about these things in scientific terms but the FTA tries to resemble a scientific theory.
For your second paragraph, the whole point is that we don’t really know if the laws of physics are changeable. They could be brute facts (and some polling seems to suggest a plurality of scientists believe that’s the case. I have 2 brothers. What if I had five? It’s a silly what if, that raises interesting questions, but it doesn’t elucidate possibility. More data needed. I’m not going to argue with you on the probability theory (because I’m bad at probability theory!). However, the whole force of the FTA seems to be a spotlight effect triggered by not having falsifiable criterion for the theistic hypothesis. Namely, yes, it could be true that things are different, and we should metaphysically take that seriously, but unless we have scientifically-informed constraints about what is or is not possible, we can’t narrow it down. So, when I say it’s a spotlight effect, I mean we’re spotlighting how we don’t know something or something seems weird. It is weird! But they’re trying to impose the constraint that it can’t be different or it can only be different in a specific way. We don’t know that! Creating that probability architecture (or calculus? I dunno) while not yourself having a good theory that fits the evidence and also creates testable theories seems very off to me.
As for the third paragraph, my point in criticizing those specific examples (even if I didn’t say so explicitly) was that I thought they were really half assed in how he wrote them. I saw him on a stream do a better example of the 500 shooter thought experiment, and just how clumsily expressed it here made it open to easy criticism. As for the actual point about evidence, I think I made an edit about this, not sure if it saved (as I was on my phone and substack sucks in browser for long post edits). I think I did get that part slightly wrong, but not so wrong. BB seems to think that any evidence that fits within a model is good evidence for a model, when a piece of evidence can fit a model, while still not being compelling. I use the example of having shoes on. We know the killer had shoes on, and our prime suspect also had shoes on. Vioala! Good evidence. Even though all suspects were wearing shoes.
As for the fourth paragraph, I was just relaying the principle of indifference as a general idea (as expressed in the documentary). It is a really fallacy people commit, and understanding that helped me express my next point. Not going to retract that (I do sincerely think you’re flat out wrong - the baseline probability of it being any day of the week is closer to 14%, not 50%). That whole section is about how models can’t predict correct outcomes while still being kinda garbage, and we don’t elevate them *just* because they were correct
Finally, (and again, I hate to come off as mean - I’m not! I promise!), you’re wrong about agnosticism! There are different kinds of agnosticism, one of which is just suspension of belief. You can be a suspension of belief agnostic on lots of things. I think regardless of one’s beliefs about god (atheist or not) we should be agnostics about the cosmological design because there’s just so much here we don’t know. Our position on the state of the world - as it relates to cosmic origins - should be agnosticism until we get more data. I think BB and FTA proponents oversell the certainty and the science here. I have seen him argue with actual physicists and at least one physics-philosophy PhD who tells him he’s wrong. To put it more simply, a non-agnostic position would say that, within current knowledge either: we have good reason to think it was designed and therefore convert to Xenu OR we know for sure it wasn’t designed and therefore burn down the churches. An agnostic simply says “wait and see.” I’ll admit I’m rigging the scales a bit by facetiously attaching actionable steps to the hypothesis, but that’s the stakes we’re talking about here, in the same way, say, one takes bets on predictions. I don’t think anyone in good faith can say that we have enough information here to conclude the FTA is true. The fact that most physicists don’t agree with someone like Luke Barnes tells me that there’s more going on here, more data to be collected.
thanks for the comment or discourse! I hope I wasn’t mean here (haven’t eaten in hours). Next time, write this is a post that way it can be signaled to me that I need to take more time to respond (switching tabs in the substack iPad app is terrible so I did a lot of this from memory)
Before beginning my response, I just want to say that in no part of your response did you come off as mean, so no need to feel guilty about that! Also, I should warn you based on what you said at the end of your last response that my response here is very long.
> The idea that we don’t know the probability comes from *the physicists themselves.*
The physicists you cite are saying that there's no physical process that you can use to assign a probability to different values of the physical constants. That's fine for them to say, but it has nothing to do with the fine tuning argument because that's not how the fine tuning argument comes to the conclusion that it would be very unlikely a priori for the universe to be finely tuned unless someone intentionally fine-tuned it. That conclusion is based on the fact that, supposedly (I say supposedly because I'm skeptical of this part of the FTA), the constants all had to be within an extremely small range, relative to their actual values, in order for any conscious observers at all to exist. The fact that they had to take on such specific values is what makes the prior probability low, not the fact that some process by which the values were set was unlikely to choose those particular values. The fact that we don't know what process caused them to take on these values, or if there even was such a process, doesn't affect this assessment, because even if there was a process that made the constants very likely to take on values within a small range, it could be any range, not just the specific range of life-permitting values.
Also, I should note that physicists use the low prior probability of constants taking on a particular value as an argument all the time. Many of the big open problems in physics (e.g., the hierarchy problem, the vacuum catastrophe, the strong CP problem) are just observations that a certain constant has a weird value, and the demand that there must be some explanation for the constant having a value with such a low prior probability of occurring. Cosmic inflation, a theory believed to be true by most physicists, is motivated in part by the flatness problem, which asks why the density of the early universe just happened to be so close to the critical density. And many theories are rejected on the basis that they can only be made to line up with observations with extreme fine-tuning of their parameters. These arguments don't include a model that explicitly gives probabilities but are based on intuitive reasoning. In fact, they often use the exact same reasoning as the FTA. In many cases, the reason why the value is surprising is because, based on currently known physics, it requires an extremely precise cancellation that would only be possible if some value is finely tuned to factor of 1 in 10^(some large number), which physicists consider extremely unlikely to happen by chance even if they don't know what process sets that value, or if it's just a brute fact or not. In fact, physicists use this kind of argument so often (even using the same terminology of "fine-tuning") that some apologists have misinterpreted physicists talking about these problems and thought they were actually talking about fine-tuning for life (This is one of the reasons I'm skeptical of the claims about how finely tuned the universe really is for life. I know that at least one of apologists' favorite claims, the idea that the cosmological constant is finely-tuned, is false and based on a misinterpretation of things physicists said about the vacuum catastrophe).
So if you want to appeal to the authority of physicists to determine whether this type of reasoning is valid, most physicists say it is. In fact, many physicists endorse the fine tuning argument itself (though usually they use it to conclude that the multiverse exists rather than God).
> On the point of saying that it’s impossible to set probabilities, I’m going to say, yes that’s true, and it’s not absurd. We just don’t know enough, at this point.
We don’t know enough about what? We’re talking about prior probabilities here. If we know very little, then we should set our probability distribution over possible values to be very wide. That actually improves the case for assigning an extremely small probability to the constants taking on values in a very small range. The only way that you ever could say that the probability of a constant falling into some range where the lower bound differs from the upper bound by an extremely small factor isn’t low is if you already knew a lot about the constant and had strong reasons for thinking that the constant’s value was very close to that range.
> we could feasibly get to a point where we have enough data and working models, but we’re not at that point.
What data and working models do you think we need? It seems like you’re just assuming that we need a “roulette wheel” like the quote said in order to assign a probability, but we don’t. If I tell you that I have a weighted 6-sided die and roll it, your credence that I will roll a 1 should be 1/6 even though you don’t know how the die is weighted. Having a model of the die would cause you to change your credence, but not having it doesn’t mean you can’t assign a probability at all.
> What’s more, I don’t think “ordinary reasoning” is something one uses in physics.
This is beside the point, although I also think it’s false - physics uses some reasoning tools that aren’t used in everyday life, but it’s not like there’s no continuity whatsoever between reasoning in physics and ordinary reasoning. Unless you think that the reasoning we use in everyday life is completely invalid, then an argument that would rule out that reasoning must have something wrong with it.
> Your complaint about the roulette wheel is not valid IMO, the argument is not that we can’t make any probabilistic assumption in physics, it’s that we lack any understanding about the specific concepts of life, universal laws, and how they happen, etc.
The argument you quoted about a roulette wheel is claiming that we have to know what process caused the physical constants to take on particular values (if there even is such a process) in order to say anything about their prior probability distribution. But this is not true. What probability would you assign to the proposition that the 1,001 through 1,010th digits of the fine structure constant are your phone number? I think you should assign a probability of about 1/10^10 of this, and you can do this without knowing anything about how the fine structure constant came to have the particular value that it does, or if there even is a reason for its value. I think that this is obvious and should be uncontroversial. Now sure, if you collect further data, you might find out something that changes your probability assignment. But that doesn’t mean you can’t assign a probability before you have that data: Probability is supposed to update in light of new evidence, and this doesn’t mean that the previous probability was wrong - it represented a conditionalization on different evidence, i.e., a different state of knowledge and uncertainty.
To be clear, if what you’re trying to argue with that last point I quoted is that we don’t know enough about life and how the constants of nature came about to justify assigning a probability as low as proponents of the FTA say, then I actually agree with you. As I said previously, I’m skeptical that the range of life-permitting constants is really as small as they think, and even if it is that small, we do know that there’s something special about the particular life-permitting values - the very fact that they’re life-permitting is special, and it also implies other special things like the ability for complex structures to form - so we can’t use the principle of indifference to say that the probability that they’d fall into that range is the same as a comparably-sized range. The place where I disagree with you is in your rejection of using probabilistic reasoning at all just because we lack knowledge - domains where we lack knowledge are exactly where probabilistic reasoning is most relevant. Instead of saying, “We can’t say anything about the prior probability of the constants of the universe being finely tuned on atheism,” we should actually say, “The probability is much higher than FTA proponents think it is, and certainly much higher than naively calculating it by multiplying together the relative sizes of all the ranges would suggest.”
> Secondly, as I said in the follow up to this post, there’s no falsification criterion for the FTA, so it’s hard to say that naturalism is improbable relative to theism given that they don’t really say what the probability of life would falsify their theory.
I haven’t read the follow-up yet, but this seems like a different argument than the one I’m objecting to. I’m not saying you should accept the fine-tuning argument, just that a few particular objections you made were wrong.
> For your second paragraph, the whole point is that we don’t really know if the laws of physics are changeable.
I’m not talking about whether the laws of physics are changeable, but whether they could have been different. If it’s a brute fact that the laws of physics are what they are, and it’s physically impossible to change them, that doesn’t imply that they couldn’t have been different - in fact, it implies the opposite. They could have been any arbitrary values and just happen to have the values they have. On the other hand, if there is some process that led to the particular values of the physical constants, they could have been different if their values had been determined by a different process, or in a possible world where their values are brute facts not determined by any process at all.
> Namely, yes, it could be true that things are different, and we should metaphysically take that seriously, but unless we have scientifically-informed constraints about what is or is not possible, we can’t narrow it down.
I’m not sure exactly what you mean by this, unless you’re trying to state some sort of modal skepticism. Science tells you very little about what’s metaphysically possible because science just tells you what’s true about the actual world, not what would be true in possible worlds that don’t exist. But it’s pretty uncontroversial among scientists that the laws of physics are not metaphysically necessary. No one thinks the world had to behave exactly the way it does - you can easily come up with scenarios where the laws of physics are wildly different, and physicists come up with such scenarios all the time.
> But they’re trying to impose the constraint that it can’t be different or it can only be different in a specific way.
The FTA is definitely not trying to impose the constraint that physics can’t be different. Arguably, it is trying to impose the constraint that the physical constants can only be different in a specific way, since it basically assumes that if naturalism is true, the values of the constants must be brute facts, or at least determined in a way that has no correlation with life permitting values. I agree that this is a flaw in the argument, but like I said before, this just means we should say that the probability of life permitting values given naturalism is higher than the FTA says, not that a probability doesn’t exist at all outside of a specific model of how the constants are determined.
> BB seems to think that any evidence that fits within a model is good evidence for a model, when a piece of evidence can fit a model, while still not being compelling. I use the example of having shoes on. We know the killer had shoes on, and our prime suspect also had shoes on. Vioala! Good evidence. Even though all suspects were wearing shoes.
I think this is a misunderstanding of his reasoning. It’s true that, if some evidence is predicted by a theory, then it makes the theory more likely, unless it’s predicted by all possible theories. This is a theorem of Bayesian reasoning, but it’s also how scientific experiments work. Of course, as your shoe example shows, it’s not necessarily good evidence - it may increase the probability by a negligible amount. That’s why proponents of the FTA have to argue that the probability of fine-tuning on atheism is low - otherwise, it doesn’t increase the probability of theism by much. So BB doesn’t believe that *any* observation that fits with a theory is good evidence for that theory - he just thinks the probability of fine-tuning is low on atheism and high on theism, which would mean that fine-tuning is good evidence for theism.
> I do sincerely think you’re flat out wrong - the baseline probability of it being any day of the week is closer to 14%, not 50%
I think you misread that paragraph because I didn’t say the baseline probability of it being any day of the week was 50% - I said the opposite. The principle of indifference gives you a probability of 1/7.
Regarding the points on agnosticism, it’s true that there are multiple kinds of agnosticism, but none of them are alternatives to theism or atheism considered as propositions. Theism and atheism exhaust the logical space, so any version of agnosticism that said that neither theism nor atheism is true would be logically impossible. Agnosticism, when considered as a proposition, rather than just a state of belief or knowledge, is a proposition either about epistemology or the state of the evidence (e.g., “It’s impossible to know whether God exists or not,” “There’s not enough evidence to say whether God exists or not,” “Claims that God exists are unjustified, but so are claims that he doesn’t.”). So agnosticism is not an alternative to “God exists,” or, “God doesn’t exist” - rather, it is an alternative to positions like, “We should believe that God exists/doesn’t exist,” or, “I know God exists/doesn’t exist,” or, “There’s good evidence that God exists/doesn’t exist.” Maybe I just misinterpreted what you were saying about agnosticism, but this is why I objected to you talking about P(agnosticism). That doesn’t need to be considered when judging P(theism) because agnosticism isn’t some alternative hypothesis to theism - rather, it’s a statement about how good the evidence is or what we can know.
> Our position on the state of the world - as it relates to cosmic origins - should be agnosticism until we get more data.
I agree with this. I think any individual hypothesis has a probability <50%. Again, I may have misunderstood what you said because you used the phrase “more probable than agnosticism,” which made it sound to me like you were considering agnosticism as an alternative hypothesis to cosmological models when it’s really just a statement about how likely any cosmological models are. I do still think your criticism of BB is unfair, though - he doesn’t claim to have certainty, and he certainly doesn’t blindly assume that you have to be certain one way or the other or that the most likely cosmological model must have probability >50%. He just thinks the evidence for God is strong and therefore places a high probability on his existence.
"How can one's existence be evidence for something that didn't occur?"
Easily, evidence for is not the same as proof. So a can be evidence for b even if b didn't actually happen, so long as learning a) makes it more rational to believe b) happened than it was before.
As a concrete example. Suppose I actually didn't commit a fatal srabbing, but by coincidence, I was seen fleeing the area of the murder five minutes after it happened carrying a knife that matched the murder weapon. The fact that I was seen fleeing the crime scene, armed with a weapon matching the victims wounds is obviously evidence I committed the crime even though in fact I am innocent. So it is evidence for a false claim. Why is it evidence for that false claim? Well, because if someone learnt that I was fleeing the scene with a big knife, they are rational in becoming more confident that I was the murderer than they were before they got that information. That's why it would be admissible evidence at a trial, rather than barred by a judge as being irrelevant.
Yes, I'll admit when I wrote this I wasn't framing it the best way. It is evidence, certainly. I was wrong!
However, different contextual considerations can determine if it's good evidence or not. Good evidence being evidence you take seriously and doesn't have a defeater that undermines a causal explanation. IMO, that's the more important problem/distinction when evaluating evidence, not the *mere existence* of purported evidence being itself evidence, if that makes sense.
BB and design argument advocates seem to argue that "because most people are born via sexual union without contraception, you exist, therefore it is good evidence of your parents doing that, and therefore they did." The problem is that it's extrapolating a probabilistic model that is not certain as if it is. And the evidence is not actually a law of the universe, but a probabilistic deduction that was arrived at without investing relevant evidence that could have been a defeater.
Put another way, Bentham in the past and apologists have argued that the number of arguments for God's existence is itself evidence for God. That's a little silly because some of the arguments are bad, and just because you categorize bad arguments under the same category of "argument" doesn't mean they are the same thing.
We're basically incentivizing both gish galloping (throwing out a bunch of nonsensical points and tallying them up as a score) and not actually look at the facts of a specific case. To use another metaphor, it's like looking at 5 murder suspects and honing in on one, a man, over the other four, all women, because men are overrepresented in murder statistics.
As I say in the article, it's often a safe bet to be informed by statistics and probability, but I don't think it's satisfactory. I think it's too broad of a definition of evidence to basically include all data points. The more data points you have, the easier it is to over fit!
I don't really follow what your saying here, but that might just be because I don't think I've ever read all of BB's original piece that your responding to.
Tbh a valid reason!
Returning to my past notes and this piece (tho not BB's as my coffee isn't finished yet), the issue I always have is trying to figure out if we're on the same page that probability is just a description of the most likely to be accurate guess you could make given the constraints of some information and of not having other information. When a voter brings her pen to the ballot, there isn't any probability involved in what candidate she will vote for; there is no "chance" element that messes with coins flipping.
BB is aware that probability is not describing, literally, the mechanics of what would happen if you reset the universe a bunch of times, but rather is a tool for dealing with his lack of knowledge of what did happen. But it would only work as an argument *that something was true* if you thought it meant something more than that, or at least, that's how it appears to me. And when you say things like "the argument is majorly supported," that sounds to me like that's what you think -- that the fact that you would feel more comfortable putting money on your bet actually has some effect on whether the bet was a good one. But I don't want to treat him as if he's saying that, because that's... crazy.
Which idk, to me this just brings us back to Pascal's wager. My natural tendency is toward a sort of apatheism, so unless I hear a good argument otherwise I don't really see why I have a need to gamble here. I truly have no idea if my default belief ("As far as I can figure it kinda just seems like stuff has always existed bc that's what stuff does") has good odds of being correct, what matters (insofar as this matters at all) is that it either is or it isn't, and that I personally am unlikely to ever be able to know which it is.
Follow up: OR to be even more narrow, it’s like an example I saw Joe Schmid make a few years ago (I watched it recently). If model 1 has a jar of 80% orange marbles and 20% brown marbles, and model 2 the inverse, if we draw a marble blindly and it’s orange, it supports model 1. One retort to this is that this isn’t good evidence. To which he glibly cocks his eyebrow, smirks, and says “I’m not sure what that means?” From secondhand observations, I think Schmid and BB agree on this. There’s no criterion for bad evidence, which is actually a little silly. It’s like saying I believe Liverpool will win the premier league because Lebron James is a minority owner. How are the two connected? This is where theistic and agnostic philosophical types really disappoint me. You can create a model for anything, but the model they’re putting forth kinda sucks because it’s not really predictive for the future; it’s just retrospectively fitting the data.
Yeah I think this is a good way of putting it. It feels like there’s a lack of preceding conversation of what this is all *for.* I think the likelihood vs. evidence distinction gets at the difference in thinking - it’s like, Monty Hall brain vs. “the car has 100% odds of being behind the door it is behind” normie brain. And obviously one *is* better math/stats! But that’s a *tool* to get us less blind toward truth, not itself the goal.
To riff in a totally different direction, this is also my issue w/ takes on shrimp welfare - there isn’t an underlying discussion of what morality is or what it’s for. I had this interaction w/ him the other day - if you press hard enough on “but why should I care about shrimp pain” there’s nothing past “it’s prejudiced not to care about shrimp and immoral not to care about pain,” no argument that we shouldn’t be prejudiced and immoral.
Yeah, I don’t care about shrimp. I think this is an instance of “defining words in a certain way and then taking it to absurd conclusions.” Like, I said this somewhere else, but people get into ethics to 1) live a better life 2) help people act better toward one another and 3) help other people live a better life. The shrimp welfare (and its contemporary “maybe human extinction wouldn’t be so bad”) perspective is very much a reductio ad absurdism - and obviously so! How do you go from “let’s help people” to “maybe it would be ‘good’ to genocide all sentient beings who have the capability to suffer.” Not only are normal people going to think you’re weird at best and a public safety threat at worse, it’s like going into Christianity to convert people to atheism or to going into architecture to be an arsonist! It’s prima facie wrong.
My favorite as a vegetarian-turned-pescatarian has been “and that’s why shrimp pain is bad but shrimp death… may be good” lol.
i think some of this comes back to some intermediate or higher level of reasoning. Like, what is evidence and what is probability? when we evaluate evidence, is it a probability machine in our head going up and down? In some ways, this is true, but when I try to go all metacognitive on my evaluation of the evidence, it’s not like a dial that’s going up and down. It feels like it when it comes to Bentham and Bayesian theists.
Like the idea that a piece of evidence (circumstantial evidence?) is good evidence just because you expected it to be the case, given the conclusion. Let’s say I’m cutting vegetables with a knife and I cut my hand and have to get stitches. But while I’m gone, someone steals that knife to commit the murder. If they looked for DNA and found my DNA on that knife, that would be evidence that I committed the murder. It’s what you would expect if I committed the murder. Obviously, there are defeaters for that theory, given access to other information (my medical records, hospital CCTV, etc), but in this sort of situation (i.e proving God) we don’t have access to such information. So, I would be the murderer (or likely the murderer) in that situation.
And this is where I think BB’s formulation is wrong. Because we’re either evaluating theories super mathematically/scientifically in which we can make specific measurable predictions or we’re doing it narratively. By narratively, I mean we evaluate evidence and give no credence, high credence, or low credence. We evaluate that credence based on our informed understanding of reality (which we can’t do with the beginning of the universe!). So, if my case was presented to the jury, the fact that DNA was on the knife would prima facie be a high credence evidence, but with the other context and defeating evidence, it becomes 0. Not .1%. 0. People (or at least I) do not attach numbers to these probability, it’s either 2 thumbs up (definite), 2 thumbs down (definitely not) or 1 thumb up/down (who knows!).
But with Bentham’s model, it seems likely he’s just using a dial and going credence up/credence down. Maybe this is Bayesian, but the evaluation of the inputs is a lot more credulous than me! You mention Pascal’s wager and I think that’s a good illustration. On its face it’s silly if you’re a skeptic or an atheist (I don’t believe in Hell even if God is real) so the calculation is silly to me. But using BB’s model I feel like you have to take it seriously and not really evaluate inputed evidence in the same way (and thus infinity skews the outcome and thus you’d have to believe). This isn’t philosophy, it’s naive math, in my view
Not having read BB's original post, I didn't really get the point of the discussion about conception and birth control. Maybe you could ad a sentence or two saying what the claim is, for those coming to this cold... But I might have missed something as I am in bed with a virus.
What I saw of BB's logic in his "argument from moral knowledge" didn't really impress me. I think theists who have strong convictions inevitably make logic subservient to those convictions, and their arguments can look a little ridiculous to those outside the framework.
We have no idea what it takes for a universe exist and, until we do, discussions of probability entirely miss the point. As you say, if multiple universes are possible, the anthropic principle dictates that a very small fraction of possible universes will be observed. No matter how small that fraction is, it would be a mistake to appeal to its value as evidence for divine intervention.
He said says anthropic principle” implies that your existence isn’t evidence that your parents had sex or didn’t use effective contraception.”
Being even less of a mathematician or physicist, I can never tell if I’m missing something, if BB is missing something, or Bayes is missing something, but I always walk away with the impression that we do not share an understanding of what statistical chance is.
My honest assessment is that 1) BB has a better logic/math brain than I do 2) However, that does not mean he is correct 3) Many of these arguments have implicit premises/evaluations that amount to personality attributes and 4) As I said in the piece today, if you understand that Bayes is about what is *probable given possibility* and that it involves big data sets, you realize that it’s not a good way of evaluating God’s existence.
I know I got sufficiently thrown by this once to badger my mathematician friend about it and he seemed to agree on the last point 😅
One thing for certain is I agree w/ your point of praise for BB, as I now have the strong urge to dig back into this.
I *do* however think he oversells his case and often opines about things he doesn’t know about. I once saw him arguing with a philosophy/physics PhD on here about the design argument, and the PhD was like “seriously, it doesn’t mean this” lol