Arif Ahmed, morality and empiricism

The latest episode of the Unbelievable? radio show is out this weekend. This time it features a discussion between me and Arif Ahmed, an atheist from the University of Cambridge.

We were discussing the moral argument for theism. The discussion was certainly interesting enough, although it was divided up so as to fit into the show’s schedule. It wasn’t structured like a formal debate, so we didn’t get to respond in depth to each comment that we might have liked to. Both of us remarked after we had recorded the show that we could have gone on for a couple more hours unpacking the material that we had briefly touched on. Well, we couldn’t do that, but I have a loaded blog and I’m not afraid to use it.

Arif denies that there are any moral facts at all. I know of a few arguments that some philosophers use for that conclusion; several arguments from queerness (especially the epistemic and the ontological arguments from queerness), the argument from motivation or the argument from moral diversity would be among these arguments. To be honest I wasn’t really expecting the reason that Arif gave for not believing that there are any moral facts: The fact that there is no empirical evidence of their existence.

As you might detect from the recording, I was a little taken aback by this. Now, there are several things to be said about this, but (again, time constraints and all that) I said only one thing. I noted that the demand for empirical evidence before we believe something is far too restrictive. It would even rule out purely deductive logical arguments. We couldn’t even accept that the axioms of logic were correct!

Arif’s reply was that actually, we know that logical is grounded empirically, as argued by philosopher William Quine. So even if we only believe things that are empirically demonstrable, we could still accept logical arguments as authoritative. There was a major response to be made at that point, but the show was moved on to other questions (and what’s more, the response is fairly technical in nature). But this was actually a fairly significant point, so I wanted to pick it up here.

The claim that Quine has shown that even logic itself is grounded in empirical evidence is about as contentious a claim as one can make about logic. What (I strongly suspect) Arif was referring to is the fact that Quine denies that there is any distinction between analytic and synthetic truths. (See my blog in the nuts and bolts series,  synthetic vs analytic truth, for an earlier blog entry where I explain the difference between the two.)

The first thing to note is that it is somewhat misleading to just declare that Quine has successfully shown that there’s no distinction between analytic and synthetic truths, or that logical truths are really empirical truths after all. The fact of the matter is that many – from what I can tell most – philosophers do not share Quine’s view at all.

A chief concern for Quine is how claiming that logical truths are “analytic” is really any different from saying that they are “obvious” in the sense that a whole host of empirical facts is obvious (e.g. “the universe is enormous” or “the earth has existed for millennia”). But is it really an empirical fact that 2 + 2 = 4? Or that “all bachelors are unmarried men?” Surely in order to be empirical facts they need to be open questions. They must be the kind of thing that we need to investigate before we know to be true, and they are claims that we could (in principle at least) be wrong about. But could we? I just don’t see how. We might (however unlikely we think it actually is) discover some evidence that reveals that the universe is not that old after all. Even if that’s never going to happen, we can at least imagine what that would be like. But can we really imagine discovering that some bachelors have five wives? Or that some triangles – just a few of them, mind you – have four corners instead of three? Try imagining those scenarios some time and let me know how it works out for you!

A more complex concern of Quine (one that I will not venture into here) is over whether or not the idea of one word being a synonym of another is adequately defined. The best I can suggest here for readers who are interested is to make a judicious use of Google. What will emerge before long is that it can’t simply be noted that Quine raised these concerns and then assumed that the matter is at rest. Quine may have raised them, but they have attracted plenty of criticism too!

But setting Quine aside, it seems rather obvious that if every claim stands or falls based on whether or not there is empirical evidence for the truth of that claim, then at least one claim should be rejected from the outset. Here it is:

Every claim stands or falls on whether or not there is empirical evidence for the truth of that claim.

If consistent empiricism is the way to go, it doesn’t just rule out morality, it rules out empiricism!

Facebooktwittergoogle_plusredditpinterestlinkedinmailFacebooktwittergoogle_plusredditpinterestlinkedinmail

38 thoughts on “Arif Ahmed, morality and empiricism

  1. I am wondering: if the main evidence for moral facts is usually taken to be that it is through moral facts that we best explain the specifically moral sphere of our experience, such as feelings of being obligated and of guilt and contrition, then does a view that only sees as legitimate evidence that is empirical (and sees our moral experience as not empirical) likewise commit itself to denying that our other experiences can be evidence for other things? And if the experiences of ours that we identify through introspection are not empirical (and so are not evidence at all), then what of consciousness? Isn’t consciousness something we only know about in virtue of our own experience of being conscious? After all, we could all be conscious-less “zombies,” as the philosophers say, and from an external point of view continue act just as we presently do. But if we can’t conclude to moral facts by virtue of our moral experience, since such experience is inadmissible as evidence by virtue of not being empirical, then neither can we conclude to consciousness by virtue of our conscious experience, since it is just as unempirical. But if consciousness (which is not to say an immaterial mind, this is Glenn’s blog after all!) is a necessary condition for having beliefs — for what but a conscious being could have beliefs? (E.g. what would it mean for an electron, or a stick, to “believe” anything?) — then any position, view, theory, what have you, that does not affirm the existence of beliefs (given that it cannot affirm a necessary condition for belief, in this case consciousness) is self-defeating, cannot be believed in, etc.

  2. Hi Glen,

    I really enjoyed the blog; I appreciate that you strive for a clear writing style, one which makes it easy for both your critics and your allies to locate which premises you are challenging and which you are proposing. I also appreciate the reminder to your readers that you have now had more time to argue your case than Arif, as Arif doesn’t have the recourse to his own personal blog. Furthermore, I think you scored some very important points against my anti-realist comrade, in that Arif’s reliance on Quine does make the ground on which he walks a little shaky. But also, naturally, I have some disagreements.

    First: I didn’t hear the original radio dialogue, but the way you phrase Arif’s reason for not believing that there are any moral facts is “that there is no empirical evidence of their existence.” But then you seem to equate this with a demand that we have “empirical evidence before we believe something.” But the first is about what we should believe EXISTS, while the second is about ALL beliefs*. As the truth of the axioms of logic are not (obviously) claims about what exists, then Arif is not (if your paraphrase of him is correct) committed to denying the truth of the axioms of logic.

    Second: Even if we did demand that ALL of our beliefs be supported by empirical evidence, I get the impression that you are using a very narrow interpretation of ‘empirical evidence.’ But what of a scientific-type view of evidence, that passing empirical tests is the best type of evidential support a theory can get? One of your old professors takes precisely this line when discussing the axioms of deductive logic: he says that the reason we should believe the axioms of deductive logic is because they have survived empirical tests (unlike the axioms of inductive logic). This seems to be a coherent view of empirical support, and one that is not prima facie too restrictive.

    Third: You don’t really give an argument for your supposedly crushing blow to ‘consistent empiricism.’ You state (or rather imply) that there cannot be empirical evidence for the claim that “Every claim stands or falls on whether or not there is empirical evidence for the truth of that claim” – but this is not analytically true, I presume; and yet you give no argument for its truth.

    But overall it was an interesting read.

    Cheers,

    Matt.
    * Supposing that by ‘something’ you mean ‘anything,’ which I think is required for your argument.

  3. Matt, you say:

    the way you phrase Arif’s reason for not believing that there are any moral facts is “that there is no empirical evidence of their existence.” But then you seem to equate this with a demand that we have “empirical evidence before we believe something.” But the first is about what we should believe EXISTS, while the second is about ALL beliefs*. As the truth of the axioms of logic are not (obviously) claims about what exists, then Arif is not (if your paraphrase of him is correct) committed to denying the truth of the axioms of logic.

    A claim about the existence of facts isn’t really a claim that there is something out there called a fact existing somewhere. It’s just a clumsily worded way (perhaps clumsy on my part, the audio will be available soon) of saying that certain claims really are facts (i.e. they are really true), and Arif doesn’t think that there’s empirical evidence for this claim so he doesn’t accept it.

    Second, I don’t know how “narrow” you think my understanding of empirical evidence is. But let me just say that in order for empiricism to be a tolerable view, I think the concept of empirical would need to become so broad as to make the label “empirical” worthless. For what it’s worth, I don’t think we have any requirement to engage in experimentation before believing the axioms of logic. There’s no need to test whether or not A = A. I think anyone who hesitates in giving assent to the claim “A = A” until some empirical testing has been done has clearly failed to understand what is being claimed.

    And thirdly, no it’s not analytically true that there’s no empirical evidence for the claim that “Every claim stands or falls on whether or not there is empirical evidence for the truth of that claim.” And yet, there’s still no empirical evidence for it. As such, empiricism commits us to the rejection of the principle.

  4. 2 + 2 = 4 may be confirmed by some empirically, but the definitions of the numbers and the operators are hardly empirical. But mathematics itself is not empirical. 2 clear reasons for this.

    One is that empiricism implies that mathematics is a posteriori, but we can know for certain that 1452796485 + 58942365785 = 60395162270 even though no one has probably done that particular sum before, and certainly no one has counted out that many beads in 2 piles the counted them out together.

    Secondly, it would imply that current mathematical truths are in principle disprovable if we had more evidence, but we can know for certain that some mathematical truths are true and no future evidence can disprove this. (Important note, this does not imply that we can know every mathematical truth, or that that there may be some true statements we cannot prove).

  5. Bethyada, I actually don’t see how we can confirm empirically that 2 + 2 = 4 in a meaningful way, given that “4” is just the token that we attach to the product of 2 + 2.

  6. I don’t think maths is empirical. I just think if you define numbers (say up to 10, you can get a pile of 2 beads and another of 2 and join them together and count the result you find there is 4 (as previously defined). Although this is provable analytically, we often “prove” it “empirically” (by way of demonstration) to children when teaching them.

  7. On the self-evident but non-empirical nature of the rules of logic (and morality), CS Lewis claims they are calling them “intuitive”. You can argue about the premises, but if someone disagrees with, say, the law of non-contradiction you cannot demonstrate to them otherwise.

    If you don’t mind me saying so, every time I see your gravatar, the name Che Guevara comes to mind. 🙂

  8. Hi Glen,

    Thanks for the response.

    1) I disagree. I think it is a perfectly coherent view to say that there are rules written out there somewhere in the sky which say things like “Don’t kill puppies” and that these constitute moral facts. When I read a sentence like “There is no evidence that moral facts exist,” I presume that it refers to this perfectly coherent view. It will be interesting to hear the original dialogue, at least.

    2a) “Let me just say that in order for empiricism to be a tolerable view, I think the concept of empirical would need to become so broad as to make the label “empirical” worthless.”

    Interesting. Worthless in what sense? In the sense that you personally wouldn’t find it interesting, or worthless in the sense that it could do no work? The concept of empirical evidence that I suggest in my previous post would allow belief in the axioms of logic; but not in, say, invisible elephants that circle the moon. That is, every time we find an A, we see that ‘A=A’ holds, so we can believe ‘A=A;’ but as there have been no tests successfully conducted (that I know of) indicating invisible elephants around the moon, we ought not to believe in such elephants. So it’s not like this conception of empirical evidence can do no work.

    2b) The foundations for the axioms of logic is an interesting topic, no doubt. But one thing I’m not a fan of is ‘self-evidence.’ I’m happy to say that if we find an A that isn’t an A, we should stop believing A=A! You might think that it’s crazy to even entertain such an idea, but people said the same thing about Euclid’s axioms. But look, there’s lots to say on this subject (trying to argue for the foundations of logic without first presuming the truth of those axioms is always going to be tricky!), but in any case you said a demand for empirical support for all of our beliefs would rule out the axioms of logic; and that’s simply not true. The axioms of deductive logic, at least, pass tests; whether we need those tests or not.

    3) You say: “And thirdly, no it’s not analytically true that there’s no empirical evidence for the claim that “Every claim stands or falls on whether or not there is empirical evidence for the truth of that claim.” And yet, there’s still no empirical evidence for it.” But, a) then it’s simply an empirical claim for which you give no reference, and b) we would then not be rejecting it from the outset, would we? We’d be rejecting it after at least some empirical investigation (unless we reject every non-analytic truth ‘from the outset,’ but then your criticism hardly seems devastating).

    Thanks Glen,

    Matthew.

  9. Matt, you can have a thought experiement about rules being written out there as things that exist as written rules – but neither Arif nor I had such things in mind.

    2a – worthless in the sense that there would be nothing to distinguish emppirical evidence from things that we don’t normally think of as epirical. it would be called empiricism for the sake of the name only.

    2b – “You might think that it’s crazy to even entertain such an idea” – Yes I do.

    3 – ‘We’d be rejecting it after at least some empirical investigation’

    If the claim is that we need empirical evidence before we believe something, then we shouldn’t start believing the empiricist claim itself unless and until we do find some evidence for it.

  10. 1) No, I’m just going on how you paraphrased Arif.

    2a) So it’s a semantic thing? It could still be a worthwhile thing, you just wouldn’t want to call it empirical?

    2b) Doesn’t matter; not the point. What we say about the foundations of logic is interesting and difficult and frought with problems, but: “you said a demand for empirical support for all of our beliefs would rule out the axioms of logic; and that’s simply not true. The axioms of deductive logic, at least, pass tests; whether we need those tests or not.”

    3) Then that’s true of all non-analytic truths. Is that really what you meant by “[this] claim should be rejected from the outset”?

    Thanks again; interesting debate.

    Matthew.

  11. 1) Well with all respect, your intepretation was pretty strange. When most people speak of there being moral facts I’m almost certain that they’re not talking about objects. I didn’t interpret Arif that way, I’m sure he didn’t interpret me that way – I think you’re all alone!

    2a) Yes it’s semantic. After all, your objection was semantic. You were objecting that (you suspected) I had a very narrow semantic range for the word “empirical.” I think that if we widen the semantic range up far enough to make empiricism believable, we end up making pretty much everything empirical, so it’s no longer a helpful distinguishing term. But yes of course it’s semantic.

    2b) Well I think it is the point. The point is that it’s very clear to all of us (I hope) that the axioms of logic are a clear case where we don’t need to do any empirical testing, and any theory that says otherwise is manifestly wrong. What’s more, it’s not obvious that the axioms of logic are even subject to testing in a way that didn’t presuppose thee axioms of logic. You’d need an example before I could grant that.

    3) There are lots (and lots and lots) of synthetic truths that we start believing because of evidence or reasons of some sort that is encountered (although not always empirical evidence of course). Probably all of them actually. It’s not like we wander wround thinking that there might be a tree, and then eventually we see it, thus confirming the belief, right?

    But the empiricist claim is clearly not like that. Nobody believed it because of some empirical evidence or reason that they encountered. And further, it’s anything but clear that any empirical evidence exists for the claim. So it’s a self defeating principle.

  12. For what it’s worth, I’m happy to embrace what I’d call “modest empiricism,” which is the rather mundane claim that empirical evidence for a claim is a sufficient reason to believe that claim. Empiricism proper, of course, makes empirical evidence a necessary condition.

  13. Hi again,

    1) You phrased it as an existence claim; so I took it as an existence claim. If that’s strange, then so be it. (I also don’t think I’m all alone, but let’s not get sidetracked).

    2a) I think I’ve already responded to this. To quote myself:

    “The concept of empirical evidence that I suggest in my previous post would allow belief in the axioms of logic; but not in, say, invisible elephants that circle the moon. That is, every time we find an A, we see that ‘A=A’ holds, so we can believe ‘A=A;’ but as there have been no tests successfully conducted (that I know of) indicating invisible elephants around the moon, we ought not to believe in such elephants. So it’s not like this conception of empirical evidence can do no work.”

    You’ve not made it clear in what sense you disagree with this. Your posts seem to imply that you disagree, but you’ve not said where you think I’m going wrong.

    2b.a) You say: “The point is that it’s very clear to all of us (I hope) that the axioms of logic are a clear case where we don’t need to do any empirical testing, and any theory that says otherwise is manifestly wrong.” That’s a bold hypothesis: care to justify it in any way? “No no, Glen, everything you say is manifestly wrong!! Mwah hah hah hah hah!” I thought you had a little more to say than this.

    2b.b) Anyway, I still don’t see how that’s relevant. You said “a demand for empirical support for all of our beliefs would rule out the axioms of logic,” and that’s what I took issue with. Whether we NEED testing or not is another thing — but your point wasn’t that empirical support was irrelevant; it was that we couldn’t have it.

    As for what tests there are: get a bunch of As. See if they are As. You find an A that isn’t an A, ‘A=A’ would have failed the test. You find that all As are As, you’ve got empirical support for ‘A=A.’

    3) You say: “Nobody believed [the consistent empiricist cliam] because of some empirical evidence or reason that they encountered.” That’s not an argument: that’s an assertion. Further, you said in your article that we should reject the empiricist claim from the outset. If all that you mean is that it’s a non-analytic (or ‘synthetic’!) claim, then it seems a rather weak objection. If you mean something else by ‘from the outset,’ then I ask you again: what do you mean?

    Sorry to fill this post with lots of quotes; I just feel like maybe we’re veering off track every now and again, and I’m hoping that using the original phrasing of an argument/objection will help to keep focus.

    Thanks,

    Matthew.

  14. Matthew, every claim is an assertion. Pointing out that a proposition isn’t an argument is not, in my view, important. With that in mind:

    2a) I don’t really think you explained how I was construing “empirical” evidence too narrowly at all (or maybe I’m just not getting it).

    2b) a) I have always, and will continue to, take for granted that there are things that are manifest to us. I maintain that to understand A = A is to realise that it is true. I note that you didn’t actually deny this, you just called it a bold claim. I daresay that you yourself probably believe this.

    b) But in any event, you appear to be saying that via induction we can empirically prove that A = A, and that is how we can say that laws of logic are empirically grounded. I beg to differ. A = A is not at all a statement about observation. It’s a staement about identity. There’s a sense in which empirical testing can’t tell us if two things have numerical identity. That’s a metaphysical question. But what the logical axiom assures us of is that whatever A is, it’s numerically identical with A.

    3) Yes it’s an assertion when I say that “Nobody believed [the consistent empiricist claim] because of some empirical evidence or reason that they encountered.” Since I think it’s a true assertion, I don’t see the problem.

    I don’t see what is obscure in my previous point 3. My point is that when it comes to synthetic truths, we do not postulate them and then test them. Instead we come to hold them because of exposure to the right sort of stimuli. But we never encountered the empiricist principle that way. Somebody (or some people) formulated it, and they did not do so (so say I) because of any empirical evidence they had encountered. What’s more I don’t think that there exists empirical evidence for the principle, so it is self defeating. I see no need to “ask again” for ground that (so say I) I have covered already.

    If you still find yourself deeply troubled by the inclusion of “from the outset” then feel free to excise those three words, and settle for this: The empiricist principle is self defeating because it is not demonstrated via empirical evidence.

  15. Hi Glen, quick question: do you sleep? I had to knock off to bed last night; props to you for seemingly never having to shut your eyes. Anyway, to it:

    “Matthew, every claim is an assertion.”

    Fair point, but they can be more than merely assertions; you could, for instance, provide a good reference for an empirical claim.

    2a) Well it seems you’re not taking into account the criterion of empirical evidence that I have proposed to you (that ‘passing empirical tests,’ thing). It doesn’t seem worthlessly broad, as it rules out beliefs in certain things (invisible moon-orbiting elephants) and not others (axioms of logic).

    2b.a) You say: “I have always, and will continue to, take for granted that there are things that are manifest to us. I maintain that to understand A = A is to realise that it is true.” Yeah, go tell it to Euclid. Look, this is all interesting, but it’s not relevant. You said we can’t get empirical support for the axioms of logic, and that’s not a claim that rests on whether we can have OTHER reasons to believe the axioms of logic as well.

    2b.b) On what do we differ here? I agree that ‘A=A’ is a statement about identity; I’m just saying that this statement about identity can be empirically supported. It’s like ‘Eric Black is George Orwell’ is a statement about identity, and one that can be empirically supported (I mean, it’s not something that ‘manifests’ itself to us, is it?) Are you taking me to be claiming that ‘A=A’ is a statement about observation, or just that ‘A=A’ could be supported by observation?

    3) So let me get this right: all you were claiming in your blog entry, when you said “If consistent empiricism is the way to go, it doesn’t just rule out morality, it rules out empiricism!” is not strictly that it rules out empiricism, but that it ruled out empiricism when the empiricist claim was formulated? Well, aside from again making an empirical claim without any reference whatsoever (good or bad), your rhetoric seems a little bit of an overkill. I know you’ve now added the ‘fact’ that there is no evidence for the empiricist claim, but that’s not in your article. As I’m sure you’ve gathered, I do have a problem with this new ‘fact,’ but that’s neither here nor there with reference to your blog post. Your article is about rejecting the empiricist claim ‘from the outset.’ You tried to score a point against the consistent empiricist claim ‘from the outset,’ and I can’t see how claiming (or even PROVING) that the claim was formulated without reference to evidence is damning or even an objection of any kind.

    Thanks again, and make sure you get some sleep!

    Matt.

    Is that what you’re claiming you meant in your blog? Keeping in mind you didn’t say

    So all you mean by ‘we should reject the empiricist claim from the outset’ is that it was formulated without reference to evidence?

  16. 2a) Matt, your example of elephants doesn’t really explain how broadly I ought to construe “empirical,” so I don’t really see how it offers any corrective to the supposedly narrow definition that I’m using.

    2b a) So… do we agree that to understand “A = A” is to realise that it’s true?

    b) I think I offered reasons for saying that we can’t empirically test A = A. I’m sorry if you don’t think I did a good job.

    3) No, I have said and I continue to say that it rules out empiricism, since there is no empirical evidence for the empiricist principle.

    I think my inclusion of “from the outset” was just fine and I offered a reason for saying so, but I could see that you were having a problem with that. For that reason I offered a simpler version which I also think is fine: If we should live by the empiricist principle, then we should reject it, for it lacks empirical evidence. It was not formed on the basis of empirical evidence (this was my earlier point), and no empirical evidence has sine been offered for it (this is the version that should be easier for you to accept). Either way, it’s sunk.

  17. 2a) I’m not saying how you ought to construe empirical; I’m saying how it is sometimes used (the passing empirical tests thing). You say that this is so broad as to be worthless, but I don’t understand in what sense. It seems to be a “helpful distinguishing term,” in that it rules out beliefs in certain things (invisible moon-orbiting elephants) and not others (axioms of logic).

    2b.a) No we do not; it’s just besides the point here.

    2b.b) Such as? That ‘A=A’ is a claim about identity? But I agree with that. I don’t see why claims about identity can’t be empirically supported (eg, Black and Orwell).

    3) Yes, you continue to say that the empiricist principle rules out empiricism, but in your article you said the reason for this is that “It was not formed on the basis of empirical evidence.” But why should this rule out consistent empiricism? If we have evidence now (which you don’t believe, but it’s not something ruled out by the fact that we didn’t have evidence when we formulated it), then the empiricist principle doesn’t rule out consistent empiricism, right? [I don’t find your new point easier to accept, either, but that’s not my point here. I’m objecting to points made in this blog entry only.]

  18. 2a) I’m sorry but the point of your objection is now lost on me. You started out by saying that I have too narrow a concept of the empirical. I have said that if we make the idea so broad that it includes just anything, then calling it empiricism is empty. Now you’re talking about the fact that there’s en empirical way to check for elephants. But I still want to know what’s wrong with the way that I have construed/defined the empirical.

    In short, I agree that we can apply empirical testing to see if there are elephants orbiting the moon. I never doubted this. But what’s wrong with my concept of the empirical?

    2b a)Well if we don’t agree on this, then that’s a very important disagreement.

    An important distinguishing feature of analytical truths is that they are known a priori – prior to any investigation whatsoever. If we can know that A = A without any investigation, then that matters. It means that we are justified in holding some beliefs sans any empirical testing. But if that’s the case, then it’s not true that we should only believe claims for which we have empirical evidence. And if that’s false, then we shouldn’t be empiricists. So it’s not beside the point and can’t be dismissed so quickly.

    b) Again, I believe I’ve already supported my claim that we can’t test for A = A. I don’t mean to be irritating, but on a blog where I often encounter people who disagree with me, I have a general policy of not revisitng arguments that I think I’ve offered already without an adequate rebuttal being offered first.

    3) I have stated two things that I think are true: Firstly, the empiricist principle is not formed on the basis of exposure to empirical evidence. Secondly, we don’t have empirical evidence for it now. But since the empiricist principle requires just this sort of evidence before we believe anything, it does rule itself out. I think this is among the most obviously true claims I made in this blog post. But to break it down:

    1) If empiricism, then all claims require empirical evidence
    2) If a claim requires empirical evidence, then no empirical evidence = no basis for the claim
    2) Empiricism is a claim
    3) There was not and is not any empirical evidence for empiricism
    4) Therefore if empiricism then there is no basis for empiricism

    (Add to this the linguistic convention of “no basis” meaning the same thing as “it is ruled out.”)

    PS: Regarding the question of whether or not I sleep, I work during the week, so most bloggers and commenters (many of whom do not work during the week) have an advantage over me. I am not able to post or comment very much during the week, so I make up for it on weekends.

  19. 2a) Well of course if we make the idea of empirical evidence so broad as to include anything, it would be empty. But I never called for that. There’s nothing even necessarily wrong with your conception of empirical evidence; I’m merely saying that Arif may have been using another common conception of empirical evidence, one that would (if my other claims are right) allow belief in the axioms of logic.

    2b.a) It may be important, but it’s besides the point here. Your point wasn’t that empirical support for the axioms of logic is irrelevant; it was that we couldn’t have it. There may be other reasons for not being a consistent empiricist (that, since we ought to believe the axioms of logic without reference to empirical support, the consistent empiricist principle is wrong); but, as I say, I’m commenting on this blog post entry only.

    2b.b) You’re welcome to not respond to whatever you wish — as you say, it is your blog, and I will not hold it against you. However, it remains the case that your argument for the claim that “We cannot get empirical support for ‘A=A'” is that ‘A=A’ is an identity claim; but you’ve failed to say why we can’t have empirical support for identity claims (such as in the Black/Orwell case).

    3) Yes, you have stated two things; but, even according to you, only the first was what you stated in the blog entry (“[The consistent empiricist principle] was not formed on the basis of empirical evidence (this was my earlier point)”). And I don’t see how the first is in itself a criticism. The first doesn’t rule out us having evidence for the consistent empiricist claim NOW; so the first doesn’t rule out consistent empiricism.

    PS: You’re doing well; I don’t think I could have the stamina you have. Very impressed!

  20. Well Matt, speaking of stamina, I hate to disappoint, but I think we can both see that we’d be able to reply to one another until the great heat death. I’ll leave your most recent comment as the last say.

    PS – I’m right. 😉

  21. I completely understand; if I had as many posts to respond to as you, I’d’ve given up a lot longer ago. But cheers, it’s been interesting, and I’ll continue to read the blog.

    PS: Well that is manifestly true!

  22. Where’s the broadcast audio? The “Unbelievable” web site only lists an episode with Alister McGrath and Caspar Melville as the latest episode.

  23. Jared,
    I actually think it will be next weekend (October 16th). This weekend they will have John Polkinghorne and Hugh Ross, as well as some atheist physicist, discussing Hawking’s new book.

  24. (Oh, and BTW, I heard Justin Brierley say in an interview on Apologetics315 that he is going to have a show coming up with Alvin Plantinga–that should be a good one!)

  25. Empiricism does have evidence for it, namely the success of science. Science, an empirical methodology, has continually discovered facts about the world and every other way has continually failed. This has been the case since science was invented, which is almost conclusive empirical evidence that the empiricist principle is true.

  26. Hi Benjamin, and thanks for your comment on this ancient post!

    Here’s the thing: As it turns out I agree that the empirical method is great, and the success of science is indeed because of our use of the empirical method. Long may it continue!

    However, this is not really a non-circular way of defending the claim that the empirical method works. We can only assess the success of science by using the empirical; method, and so our doing so presupposes that the empirical method works.

    So this boils down to: The empirical method works because the empirical method tells us that it works.

    This would not be well received if we swap the terms and say “The Bible is true because the Bible tells us that it is true.”

    What we do in the case of empirical research is we employ a form of “reliabilism.” We say that provided our senses (i.e. the instruments via which we engage in empirical research) are reliable, then they provide us with true information, and we proceed as though they are reliable unless we see good reasons for thinking otherwise.

  27. Well, I didn’t expect a reply! I listened to the debate you did with Arif Ahmed of whom I’m a big fan, and I thought you were a worthy opponent. I came to this page to hear your after-thoughts and commented on the post, and I did this before I knew how long ago it was all originally posted, so thanks a-lot for the reply, I must however disagree with it.

    “Every claim stands or falls on whether or not there is empirical evidence for the truth of that claim.”

    This was the definition of the empiricist principle in the post, and if that principle is to be justified then by its own mandate it needs empirical justification, this is the problem you identified. All that needs to be done for empiricism to work is to eliminate the paradox of always asking for empirical evidence, but not having any to justify the question. Unlike the bible the empiricist principle doesn’t make any claims outside itself, so it need not have any justification outside itself. So one can’t say that it’s presuppositionalist or circular in any way to provide empirical evidence for the principle, it’s just logical sleight of hand, tails you win, heads I lose.

    I think you got a bit lost when you started talking about methods, my statement was specifically aimed at the empiricist principle, all it needs is empirical justification, I think you agree that the successes of science, and the failure of other alleged routes to knowledge, provides empirical evidence that empiricism is the only way to go.

    I’m not aware of any epistemic theory that completely denies the use of empirical evidence or thinks that evidence in general is irrelevant, so even if what I just said was complete rubbish, it still wouldn’t be true that you couldn’t prove the principle to be correct without presupposing its truth. Remember the principle isn’t simply that empirical evidence works as a means of obtaining truth, this is undeniable, to learn about the world we need to observe it, it’s that simple. The religious, and as far as I’m aware everybody else agrees, just turn on a light if you doubtful, the crucial part is that it’s the only way of obtaining truth.

    One could think that empirical evidence is a completely unreliable way of getting to know the truth, and that intuition is the best, it’s just that this hypothesis if tested would prove to be false, and has been. Anyone could easily prove the empiricist principle false by simply showing another way of knowing produces real knowledge, but despite religions repeatedly trying, this has never happened, and for one good reason, subjective experience can’t be made applicable objectively.

    But again you must remember we’re only asking, and only need to ask, wether the empiricist principle has empirical support, and I think the answer is yes.

    I don’t think reliabilism works because it just reduces to empiricism. One needs to ask oneself the question, “How to I determine whether my senses and instruments are reliable?” And I think the answer must be, “By empirical evidence.” Science couldn’t be done by merely presupposing that the instruments that were used work, because the answers they gave couldn’t be trusted.

    Don’t feel obligated to reply, this is a very old post, however, I’m looking forward to learning exactly how, and to what crushingly large extent I am completely and utterly wrong, so I do hope you oblige. Oh, and it would be cool if you and Arif could do another debate, but I won’t get my hopes up.

  28. “And I think the answer must be, “By empirical evidence.” ”

    The reason that reliabilism was developed was because of the realisation that this is viciously circular. If you don’t agree that it is, I suppose I am appealing to (what I think is) the obvious.

    “So one can’t say that it’s presuppositionalist or circular in any way to provide empirical evidence for the principle, it’s just logical sleight of hand, tails you win, heads I lose.”

    This strikes me as clearly not true. “I think you agree that the successes of science, and the failure of other alleged routes to knowledge, provides empirical evidence that empiricism is the only way to go.”

    This is indeed precisely the sort of circularity that raises the concern in the first place. “science” here is an empirical discipline. And there’s really no way around it: You are, in fact, saying that outcomes that our empirical observations of the outcomes of science justify our reliance on the insistence of the use of empirical observation. But why think that the right way to assess the outcomes of science is by the empirical method? Now don’t get me wrong – that is a crucial way of assessing such outcomes, but it doesn’t justify the principle in question. It merely employs it.

    If you’re looking at the same claims of yours that I’m looking at, and you still don’t agree with my assessment, oh well, we’ll have to disagree.

  29. I should probably have added: Even if the problem of assessing empirical findings using the empirical method weren’t an issue – pointing to the success of science (namely, an empirical method of acquiring knowledge) and being able to assess it from outside would still only show us that there are cases where empirical methods (namely, sense perception) can justify some beliefs and produce knowledge. But this is a world away from the empiricist principle, which claims even more than this, and says that the empirical methods is the only way to acquire knowledge.

  30. Thanks for your reply, and sorry mine was so late. I’ll respond to your last comment as it seems to be the most relevant, I hope you’ll see why.

    I think that your confusion comes from the particular formulation of the principle that you had in the post. If stated negatively I think you might see what I’m getting at.

    “There are no ways to obtain substantive knowledge about the world in which we live, other than by experience and observation.”

    I added the qualifications in the definition to avoid philosophical obfuscation about brains in vats and the nature of logic and mathematics, just so you know. This is the empiricist principle I have in mind when discussing these things, I could go on to extrapolate a justification principle out of that, but I don’t think I need to.

    My argument would only be circular if I presupposed the fundamental principle of empiricism (seen above) in order to prove the same principle, I see no reason why I would even try to do this. Like I said before, unless an anti-empiricist philosophy completely denies the use of empirical evidence (I’m not aware of any that do), the fundamental principle of empiricism can be empirically verified in principle and in practice, without being circular. I could use rationalism as a starting point, or reliablilsm, or some christian epistemology, in order to empirically investigate the truth of the empiricist principle. I cannot see how this could possibly be construed as circular.

    My first comment on this was a bit general and has thus produced an irrelevant tangent in your responses, I wasn’t evoking the name of science to try and prove the efficacy of experience and observation of the world, that would be needlessly convoluted. That experience and observation can provide knowledge is a self evident position that no-one denies, including yourself, so I need not waste my time there. I mearly wanted to emphasise that science has given us, (at-least in part) a true picture of reality, the fact that we are using the internet is enough to establish that for my purposes.

    The crucial claim was that all other alleged means of gaining knowledge have, (if not conclusively) been shown by science to falter. This is enough for the empiricist principle to stand empirically, and again without being circular. So despite the principle having an admittedly high burden of proof, it’s one that it’s consistently meeting and will continue to do so. If you want me to go over the specific evidence in favour of the principle I can, although I don’t think that it’s necessary for my argument to stand.

    However, there’s an important and interesting hypothetical here, it could be the case that on investigating we find solid, objective and verifiable evidence that there are ways of gaining substantive knowledge other than experience and observation, and if that happens my empiricism would be falsified. So the other common reaction against empiricism that says it’s dogmatically anti-metaphysical, also needs to be dropped on its face. It takes the step that religious philosophies always fail in doing; holding itself accountable to evidence and taking the chance that it might fail.

    I think I gave a pretty knock down case that empiricism is neither self-refuting or circular, but whatever you think, Empiricism is a philosophy that the religious should take much more seriously. I see no reason why a theist interested in evidence need be scared away by this philosophy, it simply means that they need to give solid, objective empirical evidence as justification for their belief, if I was a theist I would want to take up the challenge, and I simply encourage them to do so.

    Thanks for taking the time to engage in a conversation with me, it’s always fun to test your beliefs against those who do not share them, and I hope I gave you at-least something to mull over.

  31. Thanks for the comment Ben.

    “My confusion,” as you put it (calling it “your confusion”), if indeed it is mine, would arise from the fact that you’re undertaking two quite different projects here. I respond to one of your lines of argument, and then you responded with another.

    Your initial claim here was “Empiricism does have evidence for it, namely the success of science. Science, an empirical methodology, has continually discovered facts about the world and every other way has continually failed.”

    So just the fact that science (i.e. empirical observation) tells us truth was given as evidence of the empiricist principle itself. I pointed out the circular way in which the success of science is confirmed (namely, empirical observation is used to justify empirical observation), a circularity from which something like reliabilism can rescue us.

    When you respond to this, however, you bring back into the picture the empiricist principle, which is obviously a much, much stronger claim than that of the success of science. And if the empiricist principle is to avoid an infinite regress in terms of justification (since you’ve rejected reliabilism), it will need a non-empirical termination.

    And yet you don’t think it needs one. “The crucial claim was that all other alleged means of gaining knowledge have, (if not conclusively) been shown by science [i.e. empirically] to falter.” Well, this is stronger than the claim that science ventures where other methods have not been able to. This is the claim that science has shown that every non-empirical pursuit of truth fails. And this simply overstates the case, otherwise the very notion of a priori truths could not be seriously entertained.

    So let’s celebrate the success of empirical methods, without overstating them. Indeed our trusting empirical methods can only ever be based on the use of empirical methods in assessing the empirical method. Justifying our reliance on sense perception itself (the basis of empirical methods), our confidence in the existence of other minds or an external world, our appeal to the axioms of logic that we use in our toolkit when examining the world etc, all of this militates against the strong empiricist claim that empirical investigation is the only way of justifying beliefs.

Leave a Reply

Your email address will not be published. Required fields are marked *

 characters available