The Sound of Logic …

As the regulars on FTB are taking Dawkins to task over his discussions of logic and emotion, there’s a lot of talk about logic on those sites. P.Z. Myers is the latest to get into the fray, claiming that pure logic can or does lead to things like the firing of flechette bombs at Palestinian children:

We can stand aloof from the events and carry out thought exercises, and we can carefully weigh the pros and cons of war—this side did this horrible thing, that side did that horrible thing, this side has this worthy cause, that side has that worthy cause—and we can attempt to calculate who is slightly better and who is slightly worse, although even there it’s striking how often different people seem to come up with completely different sums, as if maybe, somehow human lives resist being reduced to simple numbers. Let us reason together, you say; if only we could get everyone to look at the situation logically, if only everyone would be a dispassionate observer like me, if only everyone would sit back and coldly analyze all possible actions to arrive at an optimal conclusion that maximizes idealized outcomes…

…and then we arrive at this moment where all the brilliant science and technology of our civilization culminates in this beautifully intricate weapon, designed, machined and assembled by highly educated teams of engineers and executives and politicians, aimed at a small child. One human being, persuaded by the moral calculus of their side that this action is a logical necessity, pushes a button and turns another innocent human being into shredded meat.

We don’t need any more logic. What we need now is more appreciation for the value of life.

As you read through the comments, the justification for claiming that pure logic leads to this is one that I find disturbingly common: comments that they can make a logically valid argument that has whatever horrible or insane proposition as its conclusion and claim that therefore “pure logic” validates thinking that the conclusion is true. And while pure logic classes do spend a lot of time pointing out the importance of logical validity and how logical validity does not depend on the truth of the premises or even the conclusion, if you actually learn what logic is you’ll understand that that isn’t where logic stops.

So, then, what does it mean to say that a logical argument is valid? Simply this: if the premises are all true, then the conclusion cannot be false (ie must be true). Now, what we want from a logical argument are to know what propositions are true and what propositions are false. So, given a logically valid argument, can I say anything about the actual truth of the conclusion, just by knowing that the argument is logically valid? No. All I know is that the conclusion is true if the premises are true (it’s not even an if and only if, because it is possible in a valid argument for at least one of the premises to be false and the conclusion to still be true, based on another argument), but without knowing if the premises are all true I can’t say if the conclusion is itself true or false.

Thus, the criteria of a logical argument that everyone keeps forgetting when they talk about “pure logic”: the soundness of the argument. An argument is sound, roughly, if the argument is valid and the premises are all true. If you have an argument that is valid and sound — the conclusion is true if the premises are all true and the premises are all true — then you know that the conclusion is true. If it isn’t, then you don’t, at least not from that argument.

If I had a pure and fully logical argument that said that the moral thing to do was to use flechette bombs, meaning that the argument’s conclusion was “The moral thing to do was to use flechette bombs” and the argument followed from premises such that if all of the premises were true the conclusion had to be true and the premises were indeed known to be true, and if I was the sort of person who would indeed choose to act morally, I dearly hope that no pictures of dead children would sway me from using the “pure logic” argument and in fact using them. I can’t think of a valid and sound argument for that, because I don’t think there is one, once you include the unstated moral premises into the debate. And thank God for that. But contrary to Myers’ assertions, we don’t need less logic, but more logic. We need to remind people that simply making a valid argument logically doesn’t mean that you are reasonable in thinking the conclusion true, which is one of the first things formal logic classes teach you about logical validity. And we need to remind people that they have to include all of their hidden premises, especially when dealing with morality. It’s the failure of people to use pure logic that causes problems like Myers talks about, not the fact that people use it too much instead of their own emotional reactions.

If you can make a valid and sound argument for a conclusion, then no amount of emotion, concern, or care ought to convince anyone that that conclusion is actually false. If that happens, you are being irrational, and dangerously slow. But you cannot forget that soundness part; validity is not enough.

Reason, Emotion, Experience and the Right Answer

So, after things had settled down a very small bit between Richard Dawkins and the Gnu Atheist/FTB/Atheism+ group, it all started again over Dawkins using an example where he argued that date rape was not as serious as stranger rape in an attempt to provide an example of how saying that X > Y (ie more serious, worse, etc) doesn’t mean that that person thinks that Y is a good thing. When he did that, people jumped on him for claiming that date rape was not as serious as stranger rape — with some justification — and Dawkins replied with, essentially, a sigh and a “That wasn’t my point” type of response.

And then he made a post about it, which Ophelia Benson replied to, which asks this question that Dawkins invites people to consider and that Benson is considering:

Are there kingdoms of emotion where logic is taboo, dare not show its face, zones where reason is too intimidated to speak?

After eliminating interpersonal cases, Benson goes on to bring it down to discourse:

Discourse by definition rests on at least minimal reason and logic. But does that mean emotion must be banished?

Being Stoic-leaning, I’m inclined to answer that with a simple “Yes”. But even I realize that that’s far too trite an answer. And yet, the answer depends a lot on what is meant by saying that emotion must be “banished”. A reasoned and logical decision shouldn’t be settled by emotion or an appeal to emotion, but that doesn’t mean that the facts of people’s emotional states are always going to be irrelevant. For morality, I think it should be, but others disagree. It is reasonable, however, to think that when you want to predict or influence the behaviour of people who are not Stoics that you will need to consider their emotions. So the key, I think, is this: the facts of emotions — what emotional states people are in, what emotional states they will be, and so on — will be relevant to logical and rational discourse as facts, as states of the world. But emotions are not arguments, nor should they attempt to stand in for them. So, in a rational discourse, an emotion is on the same level as, say, a colour or a solid object or a conceptual truth: things that are true and are assessed for their relevance to the argument. But an emotional state does not make an argument true in and of itself; they only affect the truth of an argument if joined to it by solid logic and reasoning.

Benson goes on to talk about how she thinks emotion can impact discourse:

But more to the point, it isn’t just random daft meaningless “emotion” that make people wary of discussions of, say, abortion. It’s emotion about things like consequences and experience and the difference between being someone vulnerable to the harm under discussion and being someone who is not vulnerable to it.
So we could have another discussion about the morality of trying to discuss moral issues that have huge impacts on one kind of people but no impact on you. Does that make a difference? Should it make a difference? Is it possible that, for instance, a very rich person who has always been very rich and has no personal experience at all of what it’s like to be poor – that such a person would have a shallow understanding of the consequences of, say, a wage cut for bottom-tier workers in a company? Should very rich people be the only people deciding what wages get paid? Is that a question about reason and logic, or emotion, or both?

They might, indeed, have a shallow view of the experiences of the very poor, and thus might be missing some of the relevant facts … particularly, those facts of the experiences of the very poor. And assuming that those experiences are relevant, then they might be missing important facets of the situation that will make for a bad logical argument. However, the flip side is this: if those facts are relevant to the argument at hand, someone ought to be able to convey them in a way that doesn’t require that very rich person to have had those experiences. The key is this: if you are making an argument that is objectively true, then it has to depend on facts that are objective as well, meaning that anyone can get access to them. If you are relying on subjective facts, then only those who have access to those facts can see the truth of the argument, and so it has now become a subjective argument. And a logical and rational argument is objective, at least for any argument that you want other people to accept using logic and reason.

So if you end up arguing that people have to have experienced what you’ve experienced to see the truth of your argument, at that point you have to consider that you have either made a bad argument, or an irrational/illogical one. (Note, sometimes you want irrational/illogical arguments, or at least subjective ones. But the case listed here is not one of those; we should be able to logically assess whether the wage cut is the right thing to do by means that are accessible to everyone.)

For an example: suppose you get a group of prosperous comfortable well-fed men having a rational logical discussion of rape. Is it excessively emotional to point out that a group like that would be simply talking over the heads of the people most vulnerable to rape? I don’t think it is. I don’t think it’s excessively emotional to point out that there’s something blood-chilling about seeing people who are safe talk calmly about the risks or tragedies faced by people who aren’t like them.

Would those people be talking over the heads of the people most vulnerable to rape … or simply possibly telling them what they don’t want to hear. And I personally think that people should be able to, in general, talk calmly about issues like this without it being considered “blood-chilling” or in any way wrong. It is indeed the rational ideal that such arguments should be made calmly and, more importantly, without bias. Those who are most impacted by the choices are clearly biased, and it’s perfectly natural that they would have a bias. But if people who have a vested interest in the outcome of a discussion are the only or main ones who can argue over it, they risk introducing bias into the arguments, even unconsciously. And a rational, logical and dare I say scientific argument wants to remove bias as much as is humanly possible. Again, it comes back to what I said above: if the argument isn’t one that’s accessible and demonstrable to everyone, then it’s at best subjective and at worst illogical/irrational.

Note again that this doesn’t mean that their concerns are irrelevant; they should be demonstrable facts. But the fact that people who are in those situations have those concerns says nothing, in and of itself, about the truth of the argument. It seems to me that hinting that there’s something wrong or something missing if a bunch of uninvolved people had a dispassionate discussion of an issue is advocating for a subjective argument: you can’t see the truth of the argument unless you’re involved. And that’s wrong. It begs for an answer of “Demonstrate these facts objectively and then we’ll incorporate them into the argument, and see where they take us.”

Also note two things:

1) Dawkins’ actual complaint was about shutting down the argument or consideration of it at all due to emotional reasons, which is about an even stronger stance than I’m talking about here.

2) My take here is a bit stronger than Benson’s phrasing would insist on; she could say that she just wants the situations considered, and not to trump discussion.

In summary, emotion is not useful as part of the method of rational discourse, and only introduces bias and gets in the way. However, facts about emotions may be necessary to produce the right argument, and so should be limited to that role in rational discourse, not banished entirely. Rational discourse has to depend on things that can be demonstrated to everyone, and emotions and personal experiences can’t. Thus, rational discourse should follow the “Just the facts” model, where sometimes the facts include personal experiences and emotional states.

God in the Age of Science, Part 1

So, I did receive “God in the Age of Science?” on July 16th, and immediately read four chapters. And then got busy and didn’t read any more since then. But I’d like to comment briefly on my very first impressions of the book and on the first couple of chapters or so. I have a more detailed commentary planned for chapters 3 and 4 … whenever I get around to it [grin].

First, on tone, the tone isn’t particularly aggressive, which is a plus, in my opinion, for a book like this. A few stylistic notes: he doesn’t seem, at least not yet, to have Prinz’s obsession with presenting all of the counter-arguments in as detailed and precise a manner as possible, although he does indeed address counter-arguments, which is nice. The second is that he seems to be fond of Derek Parfit’s style — which, to be fair, is fairly popular in philosophy in general — of stating what he sees as the obvious conclusion to an argument as if it was, well, obvious, except Philipse does it through rhetorical questions while Parfit uses out-and-out statements. Unfortunately, both of them have a tendency to do it without giving any further explanation and in cases where the conclusion is not or at least may not yet be clear, which always chafes me a bit. Philipse’s approach is a bit less annoying, leaving you wanting more rather than automatically wanting to challenge him, but still it would be better if these things were presented as the results of full arguments rather than as asides.

But that’s not all that important. The big goal in the first section is to demonstrate that you can’t have or rely on revealed knowledge or, perhaps, revelation in general to justify a belief in God, but instead even for revelation will have to rely on either empirical evidence or reasoning to justify your case, or else you’ll be irrational. This last part is actually pretty controversial, because it risk conflating rational belief and being justified in claiming to know something, and a lot of the arguments particularly against reformed revelation (in Chapters 3 and 4) do rely on that. But remember that theism is a belief in the existence of one or more theistic gods, not necessarily a knowledge claim. As someone who doesn’t make the knowledge claim, it’s going to be easy to say “Can we believe that rationally, though?” as a response to most of this. More on that when I look at chapters 3 and 4 specifically.

The biggest flaw in the first two chapters, though, is probably in his discussions of contradictions in the Bible and how they can’t be resolved through revelation, which for him seems to be “Reading the Bible and thinking really hard about it, which may include noting the contradictions and resolving them through the text”. While few people will deny that there are some at least difficult things to resolve in the Bible, for this point to work they have to be virtually impossible to resolve, and his examples aren’t that hard to resolve. For example, he cites a contradiction between Jesus and Paul over whether works or following the Laws are what is required to get to heaven, and notes that this one is unresolvable. Except it’s very easy to resolve for most Christians: if there’s a contradiction between something that Paul said and something that Jesus himself said, you go with Jesus, and Paul either got it wrong or should be taken another way. Philipse could have found a similar contradiction between the purported words of Jesus himself, but I suspect that those would be easier to reconcile on interpretation. Now, for Philipse’s point disowning Paul’s revelations might support not trusting revelation since there are times when it gets things wrong, and you don’t really know when it’s getting things wrong or not, but that isn’t the tack he’s taking in those chapters, and thus it’s about the contradictions being unresolvable … but he leaves himself open to the counter of “Says who?”.

If he took revelation as a method that had to reveal itself directly to the person in full form, then he’d have a point. But as soon as he allows for reflection, any philosopher should know that there are many, many ways to resolve seeming contradictions in a work (seeing how that’s done for, say, Kant, for example), and so his comment that taking a revelatory approach to the Bible leads to unresolvable contradictions is weaker than it needs to be to make his point. And if he takes that approach as requiring reason and so doing natural theology, then he seems to contradict his original discussions and, in fact, the reformed approach to revelation that he discusses in chapters 3 and 4 as if it really could save revelation. So contradictions, though a popular argument, don’t seem to support his case as well as he’d like them to. But this shouldn’t be a big issue for his overall thesis, and so probably isn’t worth worrying about.

My odd relationship with vacation …

In my job, I am “fortunate” to have a rather generous allotment of vacation. And, yes, for me that “fortunate” has to be in quotes, because for me personally it’s often more of a curse than a blessing. While here by law you have to receive vacation pay — which at one point was 2 weeks pay, but I think it’s a bit more now — and while many companies simply actually pay that out every year and make all leave unpaid leave, my company does the standard thing for bigger companies and gives you a number of days of vacation and you essentially don’t show up for a day, call it vacation, and then get paid as if you were there that day. Snazzy. Of course, that means that they don’t pay unused days out at the end of the year; it’s part of your regular compensation, essentially.

The biggest problem for me when it comes to vacation is that, in general, I don’t really do anything that I need to take vacation for. I don’t like to travel, and so never take a week or two to go somewhere warm or to take in Europe or, well, whatever it is that people do when they travel. I’m also not much into renovations and the like, so I don’t take a week or so off to fix something up that needs to be fixed up. All of my interests are generally the sort of thing that I can do in the evenings or on weekends, and so are most of my errands. My vacations, then, are me staying at home doing all of the things I’d do on the weekend, just during the week. Thus, I don’t have any reason to say “I need to take a week or two right now to get my stuff done”; all of my stuff will wait a week. Or a month. Or a few months, if need be.

Because of that, there are a number of times when it wouldn’t make sense for me to take vacation, which manage to align with all of the times that I’m busy at work. Because I never need to take a particular week off for anything, it never makes sense for me to work like crazy for a week, take a week off, and then work like crazy the week I get back. Either I should work like crazy all three weeks, or work all three weeks so that I don’t have to work like crazy at all. After all, anything that I was going to do I can do some other time, so better to reduce my stress level and just work. This also applies to taking individual days off; if I’m working time on the weekend because I’m under time pressure, it doesn’t make sense to then take a day off the next week. I would have been better off to just take the weekend off … or, at least, use the fact that I worked then to cover the day I took off.

Now, add to this that while I’m a fairly dedicated worker, I’m not quite a stupid one. If not using my vacation days meant that I’d end up with a little extra cash at the end of the year, I’d take the money and only use the vacation that I really wanted to use at times where it was reasonable for me to take it. But since that’s not the case, I’m not inclined to have a compensation package that will pay me to stay home for a number of days and then come in and work anyway without getting paid extra. So, I really do indeed want to use all of my vacation days, and the company wants me to use them (for various reasons), and so I try to use them all. Which usually ends up with my taking a lot of time off around Christmas, more than most people would. Which isn’t easy for my managers to, er, manage.

Thus, for me, vacation is actually something that I have to manage, like a resource in an RTS. Well, sure, that’s the way it is for most people, but for them it’s a resource that they run out of, while for me it’s something that I have to arrange to take without impeding my ability to do my work. That’s just one of the little ways where I can be very, very odd.

It’s in …

So, Amazon is telling me that my copy of “God in the Age of Science? A Critique of Religious Reason.” is waiting for me at the post office, so I should get it — and start reading it — tonight. I’ll try to do my “one chapter a night” thing, although I might do a bit more than that tonight since I ought to have time. I’m considering posting thoughts on each chapter as I think of it, but I hate doing that since I think that for most works it’s important to understand the whole point before breaking it down into its parts, although for a lot of books each point is independent and so can be addressed that way. So we’ll see.

But my biggest question before starting is: is this book, how can I put this, aggressive? Coyne has a strong tendency to like and recommend books and articles that utilize snark and mockery a lot, as much if not more than they utilize arguments. The initial description didn’t sound too snarky, but is this going to be a case where Coyne’s recommendation of a good book is one that doesn’t, in fact, mock the arguments but instead focuses more on addressing them? Only time will tell.

Challenge Accepted …

There has long been a line of argumentation that many atheists like to use that relates to the traditional Courtier’s Reply, which goes something like this: you keep telling us atheists that we’re ignorant of theism and can’t dismiss it until we’ve considered all of the best arguments for theism. But what about the best arguments for atheism? Can we list off a list of books and arguments that you have to read before you can be considered credible in critcizing atheism?

Now, the theistic point isn’t usually just “You need to read all of these authors”. Most of the initial replies are people pointing out that the atheists tend to talk about particular arguments for or conceptions of God, get it completely wrong, and so really should try to understand the arguments before criticizing and, especially, before mocking the arguments. In other cases, it’s just people pushing their own preferred arguments and conceptions, because there are indeed a number of different ones. Sometimes, it’s both. But there are times when people — whom I’d tend to call “unsophisticated” — really do just toss out books and say read them. While I never approve of such things, I can approve of the underlying sentiment that makes that seem even remotely credible: in order to criticize or reject a position, you really should be well-read in not only what others say about it, but also in what those who support it say about it.

In terms of atheism, I’m doing pretty well. I’ve read Dawkins, Dennett and Harris of the Four Horsemen (I read a debate between Hitchens and someone else once which convinced me that he wasn’t worth my time), I read Kaufmann as suggested by Jerry Coyne (and wasn’t impressed, to say the least; I really should critique the religion part more directly), I’ve read Smith’s initial book, I’ve read Grayling’s take, and some others.

Now, Jerry Coyne is pushing another book:

For a good refutation of the “God off the hook” claim of Ruse, read the philosopher Herman Philipse’s God in the Age of Science? A Critique of Religious Reason. It’s the best attack on theism I know, and though it’s occasionally a hard slog, it’s well worth it. I can’t recommend it highly enough, and if a theist says he/she hasn’t read it, you can rightly say, “Well, then, you can’t bash atheism, because you haven’t dealt with Its Best Arguments.”

Well, if that’s one of the “Best Arguments” … then I shall take up the challenge and deal with it, despite the fact that it really looks like this whole challenge is one that Coyne and other atheists really don’t expect someone to accept. I’ve ordered the book and will read it when it gets in. However, from reading the description on Amazon I can already predict that it will have an uphill climb:

God in the Age of Science? is a critical examination of strategies for the philosophical defence of religious belief. The main options may be presented as the end nodes of a decision tree for religious believers. The faithful can interpret a creedal statement (e.g. “God exists”) either as a truth claim, or otherwise. If it is a truth claim, they can either be warranted to endorse it without evidence, or not. Finally, if evidence is needed, should its evidential support be assessed by the same logical criteria that we use in evaluating evidence in science, or not? Each of these options has been defended by prominent analytic philosophers of religion. In part I Herman Philipse assesses these options and argues that the most promising for believers who want to be justified in accepting their creed in our scientific age is the Bayesian cumulative case strategy developed by Richard Swinburne. Parts II and III are devoted to an in-depth analysis of this case for theism. Using a “strategy of subsidiary arguments”, Philipse concludes (1) that theism cannot be stated meaningfully; (2) that if theism were meaningful, it would have no predictive power concerning existing evidence, so that Bayesian arguments cannot get started; and (3) that if the Bayesian cumulative case strategy did work, one should conclude that atheism is more probable than theism. Philipse provides a careful, rigorous, and original critique of atheism in the world today.

So, what are my issues?

1) It starts from the Bayesian cumulative case strategy of Richard Swinburne, which I’m not familiar with.

2) That uses Bayesian analysis which I don’t care for.

3) If the end of his third argument is that atheism should be considered more probable than theism, then even there I don’t think that what we believe must be that which even we consider most probable, let alone what would be considered most probable by an abstract Bayesian analysis.

4) Knowledge certainly isn’t set by probabilities of any kind, so that wouldn’t get to a knowledge claim that atheism is true, and I don’t care much about atheism until they can claim to know that God doesn’t exist.

5) His second point about it not reaching the level where a Bayesian analysis can be done is underwhelming to me and only matters if I accept that the Bayesian route is the way to go, but since my epistemology is not Bayesian that’s going to be pretty hard to do.

6) So it will come down to his first point about not being able to state the proposition meaningfully … but since Christians can point at the Bible and most religions can point at their text that seems to be precisely as meaningful, at least to most people, as, say “Sherlock Holmes”, a word that we clearly know the meaning of, so he must mean something more advanced than that … but I don’t see why that would matter.

I hate starting a book thinking that I’ll hate it, because I find that starting with that attitude almost always ensures that you will, in fact, end up hating it. But I will read it and see if it can convince me.

So … challenge accepted.