d

A Very Short Argument Against Empiricism

In empiricism, skepticism on 15/07/2011 at 5:42 am
  1. The ambition of empiricism is to reduce all relevant matters of fact to sensory qualities.
  2. Sensory qualities, even if assumed that they can be known immediately, do not justify matters of fact outside sensory qualities.
  3. The ambition of empiricism to reduce all matters of fact outside sensory qualities to sensory qualities is impossible.


I vividly dream I am on a beach when I am not in fact on a beach. There is no ocean, no smell of the sea, no sound of gulls flying overhead. The beer I’ve just drunk is, unfortunately, not actually a beer. My sensory qualities say nothing about matters of fact outside sensory qualities.

The hypothetical scenario of the vivid dream is designed solely to demonstrate that the existence of any and all matters of fact outside sensory qualities is not deductively entailed by sensory qualities. What we have is a variation on the problem of underdetermination: the available data (sensory qualities) do not permit us to make a decision between two rival theories.

In response to my positing of a vivid dream, a defender of empiricism might argue that I naturally dream about things I’ve otherwise experienced. In response, I will happily admit that there have been some times in the past when I have experienced sensory qualities of lounging on beaches. What of it? It’s far too easy to change the hypothetical scenario in any way I choose to deal with any assumption the defender makes. No matter the objection, a small nudge to the scenario means I cannot tell apart a dream from waking life, meaning sensory qualities do not justify matters of fact outside sensory qualities.

Of course, I know I live in an objective reality — but does empiricism justify such a basic intuition? Hume and Berkeley thought not, and that any attempt to pull empiricism’s irons out of the fire was doomed to failure: If sensory qualities are not necessarily caused by  matters of fact, but caused by imaginative creations of the mind, then how can I determine the causal relationship in any case? By appealing to sensory qualities? Such an attempt would, of course, beg the question.

After we came out of the church, we stood talking for some time together of Bishop Berkeley’s ingenious sophistry to prove the nonexistence of matter, and that every thing in the universe is merely ideal. I observed, that though we are satisfied his doctrine is not true, it is impossible to refute it. I never shall forget the alacrity with which Johnson answered, striking his foot with mighty force against a large stone, till he rebounded from it — ‘I refute it thus.’ (James Boswell: A Life of Samuel Johnson)

I don’t deny the existence of reality: I’m a realist. I only argue for the limitations of empiricism, not for anti-realism. Science was never about reducing matters of fact to sensory qualities at all.
//
  1. […] A Very Short Argument Against Empiricism (thephilosophyofscience.wordpress.com) […]

  2. […] A Very Short Argument Against Empiricism (thephilosophyofscience.wordpress.com) […]

  3. I’m not sure that empiricism is about reducing matters of fact of outside qualities to sensory qualities. That’s because, under Hume, matters of fact are sensory qualities! And whether it’s a dream or not, there’s no practical difference between either choice, and you have no reason a posteriori reason to pick one position over another.

    And you determine a causal relation by observing X and Y to be found together, and this, through constant experience, becomes that causal relation. No need to invoke some mysterious thing as the cause, because you’re opening up something in which there’s a logical infinity of things that could be that cause, and all contradictory from one another.

    And intuition doesn’t equal knowledge, and no defense for it when it has no practical difference.

    • allzermalmer,

      And you determine a causal relation by observing X and Y to be found together, and this, through constant experience, becomes that causal relation.

      So emeralds are grue?

      • No idea what “grue” or “green” has to do with it. And I’m aware of the “grue” problem of induction, which is based on Goodman’s nominalism. “Natural” predicates and all that.

        The main problem is your second premise, which says “Sensory qualities, even if assumed that they can be known immediately, do not justify matters of fact outside sensory qualities.”

        How did you arrive at the conclusion of some matters of fact (perhaps an abuse of the word?) outside of sensory qualities? Through induction or deduction? A priori or A posteriori?

        • allzermalmer,

          I am referring to Quine’s observation that sense-experience does not justify any preference for some properties that are in line with sense-experience over others — from sense-experience, we cannot determine if emeralds are green or grue. These matters of fact, the obvious (at least to some) properties of any emerald, swan, and so on, are not justified through sensory qualities.

          As someone that does not accept the existence of any inductive procedure as justifying any inference, I do not think it is inductive, nor do I think any deductive entailment is justified, for any preference for some properties over others rests on a priori warrant that itself encounters the Münchhausen Trilemma.

          • I’m aware of Agrippa’s Trilemma, and all the skeptical tropes that go with it like the 10 tropes and the problem of criterion. But I don’t see how any of that would support anything to think there’s matter of fact outside of sensory qualities. I think even Sextus Empricus even mentioned that we have nothing to support it.

            And at least we can determine some things through experience, like one thing follows another. What you like to call it doesn’t change that you know-how to do something, whether you know-that it’s grue or green. Those make no difference in the know-how, actually living life and using them.

            • allzermalmer,

              It’s pretty clear that there are matters of fact outside sensory qualities: whatever matters of fact there are, everyone wants to know if there are or are not rocks that are kicked, hands that are there, and whatnot. If empiricism cannot deliver, then so much for empiricism. It can’t do the heavy lifting. By its own lights, empiricism fails to satisfy its very ambition. Hume and Berkeley saw this, and consequently made very conservative claims about causation, the existence of properties, and so on. They were as minimal (and consistent) as you could get once accepting the empiricist ambition, which is to say, to acknowledge that empiricism did not deliver.

              And at least we can determine some things through experience, like one thing follows another. What you like to call it doesn’t change that you know-how to do something, whether you know-that it’s grue or green. Those make no difference in the know-how, actually living life and using them.

              No … that doesn’t follow from experience. If I know that emeralds are grue and not green, I will act differently than if I know that emeralds are green. For instance, if emeralds were grue at t, if I knew they were grue, I’d put a hefty bet on them turning true at t … but emeralds are obviously green, so I won’t make that bet … but how can I determine from experience that emeralds are green and not grue? I cannot.

              • It’s not at all clear that there are things outside of sensory qualities, because you can’t argue from the sensory qualities to something outside of the sensory qualities (two different things), and you can’t rely on experience to say that there is (that’s a contradiction). So what are you left with? Nothing. That’s not reason, or even close to it. And if empiricism makes you have to recognize this, then so much for the belief. It has to go by the wayside when there’s no evidence for it. We can ignore the evidence, if wanted.

                And I’m not sure what you mean when you say that people want to know there is or not rocks that are kicked? Did you seem something that you call a rock, or should it be roue(?),? If so, then you have your answer. There is a rock. Don’t see the problem, it’s right there before you. Why ignore what’s before you for what’s not before you?

                You still haven’t shown the relevance of green and grue. Those are just words, not experiences. And this seems like irrelevant to what you brought up. You’re just bringing up a problem of organizing the experiences, but not the experiences itself. And there’s no practical difference between grue and green, right now. Because grue relies on a definition based on a future time, which is not being currently occupied. So it makes no difference until that time comes to pass, and then you can make a change. But this is all irrelevant to your second premise, because nothing you’ve pointed out seems to be related to matters of facts outside of the senses, unless you’re just talking about conceptual organization.

                What you seem to be pointing out is that our concepts are the problem, not the experiences. But if the concepts become a problem, then that means that the extra-sensory are the problem. But nominalism basically showed that words mean what you want them to mean, just make sure you get others to follow the rules!

                But I think you could help me out if you could show what grue and green have to do with your argument.

                • allzermalmer,

                  It’s not at all clear that there are things outside of sensory qualities, because you can’t argue from the sensory qualities to something outside of the sensory qualities (two different things), and you can’t rely on experience to say that there is (that’s a contradiction).

                  One doesn’t need to rely on sensory qualities — if the theory of evolution were empirically adequate, for instance, then there exists matters of fact outside sensory qualities.

                  And I’m not sure what you mean when you say that people want to know there is or not rocks that are kicked?

                  While there is no pragmatic difference between the realist and the idealist when it comes to the existence of rocks (they’ll act the same way in all cases), there’s an epistemic difference. Empiricists have wanted, at least traditionally, and to this present day, to know facts not just about their phenomenal states, but about the physical world. The logical positivists/empiricists spent years on protocol sentences/observation sentences/basic sentences/what have you.

                  You still haven’t shown the relevance of green and grue. Those are just words, not experiences.

                  The words describe two possible properties of emeralds. I don’t see how the fact that they are not experiences matters in the least.

                  What you seem to be pointing out is that our concepts are the problem, not the experiences.

                  Far from it — I point out that experience does not justify anything outside experience. Not causality, not properties, not relations between properties, not anything about any presently unobserved properties/unobservable properties.

                  But I think you could help me out if you could show what grue and green have to do with your argument.

                  Betting. There’s no way for theory-preference to occur with an infinite number of hypothetical properties that are equally corroborated by the empirical evidence unless one chooses to follow some a priori rule that guides theory-preference, but that rule is not grounded, and so on. We covered this earlier.

                  • “One doesn’t need to rely on sensory qualities — if the theory of evolution were empirically adequate, for instance, then there exists matters of fact outside sensory qualities.”

                    Actually, if one doesn’t *need* to rely on sensory qualities, then science wouldn’t be an empirical matter. This means there’d be no need for testability or falsifiability, but science is empirical and relies on those things. Being empirical means that theories need to, eventually, come back to the human senses. And being empirically adequate means that it makes true predictions of what will be present to the human senses. And, because the theory of evolution is empirically adequate, that means it’s made true predictions of what would be present to the human senses. But again, the human senses don’t show anything beyond themselves, and thus theory of evolution doesn’t show any matter of facts outside of sensory qualities by being empirically adequate.

                    And the theory of evolution doesn’t state anything about thing outside of sensory qualities. It just states how species have changed over time. All of this change is, in principle, able to be observed with sensory qualities. At best, you can say that the theory of evolution has selected those species which believe in something beyond the sensory quailties, but not that there is such a thing.

                    “Empiricists have wanted, at least traditionally, and to this present day, to know facts not just about their phenomenal states, but about the physical world.”

                    I can definitely grant that *some* empiricist have wanted that, or people that proclaim that they’re empiricist. But they don’t look like it to me if they’re constantly going beyond the senses. No experience for it, and the logical arguments for it tend to be fallacious.

                    “The words describe two possible properties of emeralds. I don’t see how the fact that they are not experiences matters in the least.”

                    Well, I think, one could plausibly say that if they have two contradictory properties, and suppose to belong to the same thing, then “emeralds” are just contradictory and don’t exist. Emeralds can’t both be grue and green at the same time, so these properties don’t seem to be anything actually existing. But this is the conventional nature of language. Even though I do support the arguments of grue because it helps present the nature of language, IMO.

                    “Far from it — I point out that experience does not justify anything outside experience. Not causality, not properties, not relations between properties, not anything about any presently unobserved properties/unobservable properties.”

                    I definitely agree with you that experience doesn’t justify anything outside of experience, especially if someone proclaims a causal connection between something in experience being caused by something outside of experience. But it looks like the properties that we do learn of are in experience, not outside of it.

                    “Betting. There’s no way for theory-preference to occur with an infinite number of hypothetical properties that are equally corroborated by the empirical evidence unless one chooses to follow some a priori rule that guides theory-preference, but that rule is not grounded, and so on. We covered this earlier.”

                    I agree with this, and I think this is basically related to the problem of affirming the consequent within science, and keeping in mind that we create logically possible worlds and see which fit the world of experience through being testable/falsifiable. I just don’t, personally, see how this is suppose to support premise two. I definitely grant that we don’t have an a priori ground for choosing amongs them that are truth-preserving.

                    Sidenote, I recently found your blog yesterday, and I’ve found some good stuff in it. Keep it up.

  4. allzermalmer,

    Empirical adequacy, though, is different from theories that appear to be empirically adequate to date. The theory of evolution may in fact be empirically inadequate. So there would be matters of fact outside mere empirical adequacy — for instance, that a scientific theory were empirically adequate (but we cannot tell from experience if they are or are not empirically adequate), that the unobservables predicted by the theory are true (atoms existing, for instance), causal relations, and so on.

    Emeralds can’t both be grue and green at the same time, so these properties don’t seem to be anything actually existing.

    I don’t think you understand the problem. It’s not that emeralds are both grue and green, but that there is no way to determine if emeralds are grue or green from sense-experience. But emeralds are obviously green. Therefore, empiricism cannot justify theory-preference.

    But it looks like the properties that we do learn of are in experience, not outside of it.

    You keep saying this — how would one learn from the empirical evidence which properties to prefer?

    I recently found your blog yesterday, and I’ve found some good stuff in it. Keep it up.

    Thanks.

    • “So there would be matters of fact outside mere empirical adequacy — for instance, that a scientific theory were empirically adequate (but we cannot tell from experience if they are or are not empirically adequate), that the unobservables predicted by the theory are true (atoms existing, for instance), causal relations, and so on.”

      I’m not sure how you got that distinction. There’s no appears to be empirically adequate. Theory makes a prediction, you’ve found that the theories prediction is correct. There’s no appears that it made a correct prediction. It did. The rest of what is in the theory doesn’t matter to the empirical adequacy.

      “It’s not that emeralds are both grue and green, but that there is no way to determine if emeralds are grue or green from sense-experience. But emeralds are obviously green.Therefore, empiricism cannot justify theory-preference.”

      It’s actually not at all obvious that emeralds are green. As I think you’ve even admitted, you haven no way to determine if it is green or grue through experience, or outside of experience. So it’s not obvious that emeralds are green. They’re just what you call them. That’ the convention of language, not of anything of reality but what humans call things in reality. Reality has no language, but humans do. You seem to be focused on something that is really irrelevant.

      “You keep saying this — how would one learn from the empirical evidence which properties to prefer?”

      Simple, by the one’s that show up more often than the others or through what you’re taught. If you grow up in a society, and they teach you green instead of grue, then you have a reason to take green over grue. Society teaches you how use words and which to use. Language doesn’t mirror reality, it mirrors what humans decide.

      • allzermalmer,

        Empirical adequacy is a very technical term that refers to a very select number of scientific theories that are in line with the empirical evidence in all cases; there are an infinite number of empirically inadequate theories that happen to appear to be empirically adequate, since they are in line with the available empirical evidence, but are not in line with the empirical evidence some time in the future. It’s the basic problem of underdetermination directed at empirical adequacy.

        It’s actually not at all obvious that emeralds are green.

        It may be obvious (I, for one, take it to be obvious), but perhaps I should qualify my words: while it may be obvious, in no way does its obviousness justify its truth. I do not want to justify its truth; the scientific realist empiricists can attempt that all they want. Green emeralds, however, cohere with background knowledge while grue emeralds do not.

        Simple, by the one’s that show up more often than the others or through what you’re taught. If you grow up in a society, and they teach you green instead of grue, then you have a reason to take green over grue. Society teaches you how use words and which to use. Language doesn’t mirror reality, it mirrors what humans decide.

        As you note, it says nothing about reality. It may be interesting to you, and it may provide a non-arbitrary choice for theory-preference (very Wittgensteinian), but it doesn’t justify any choice unless you think language is privileged. But you deny such privilege.

        • “a very select number of scientific theories that are in line with the empirical evidence in all cases; there are an infinite number of empirically inadequate theories that happen to appear to be empirically adequate, since they are in line with the available empirical evidence, but are not in line with the empirical evidence some time in the future.”

          I’m aware of this, but the same is true in the opposite as well. I’ll be using 8 to mean infinity. So since science affirms the consequent, we can create a theory that that states “If A then 8”. Here 8 stands for an infinity of predictions that the theory makes. So we can say “If A then 8. 8. Therefore A (or ~A).” Now let’s hypothetically assume that we have observed all the predictions that theory makes, which was 8. Now there’s also an infinity of theories that made the same exact predictions but are ~A, ~A would be B and ~B, and ~B would be C and ~C, and ad infinitum. Thus, we have theories that are empirically adequate and make false predictions in the future, but there’s also theories that are empirically adequate and make true predictions of the future. But let’s assume that we make all the observations that are possible. We’d still have an infinity of hypothesis that are all empirically adequate, but contradict one another. So we have an infinity of theories that make false predictions and also an infinity of theories that make true predictions. Nothing is really solved.

          “Green emeralds, however, cohere with background knowledge while grue emeralds do not.”

          True, but this is the same with a hypothetical group that holds to the background knowledge of grue emeralds. There’s still no way to decide between them, unless you’re going to let the society or group decide. So it’s not at all obvious that they are green, except in a society that says it’s green instead of grue.

          • Your use of the term ’empirically adequate’ is at odds with its use by van Fraassen: empirically adequate theories cannot, by definition, be empirically inadequate. Theories that appear to be empirically adequate may be empirically inadequate, however.

            There’s still no way to decide between them, unless you’re going to let the society or group decide.

            That’s why I accept epistemic voluntarism — the decisions are not epistemically grounded, although decisions may be preferred on non-epistemic grounds.

            • “Your use of the term ‘empirically adequate’ is at odds with its use by van Fraassen: empirically adequate theories cannot, by definition, be empirically inadequate. Theories that appear to be empirically adequate may be empirically inadequate, however.”

              I don’t see how this is the case, because va Fraassen says that empirically adequate theories are theories that make correct predictions. One remains agnostic on what we can call the antecedent. For example, If X then Y. X is the antecedent, and Y is the consequent. Y would be the prediction. And a theory that makes correct predictions are empirically adequate, because their consequents have found to be the case. And yet many different theories would all be empirically equivalent, which means they all make the same exact predictions, or have the same exact consequents. They just have different antecedents. This is all perfectly consistent with van Fraasen. I’m not aware of him bringing up him accepting empirical adequacy for this point, but this is probably the strongest reason for a position like his, which is to remain agnostic over the antecedents, at least those that are “unobservable”. The theory has to account for the “actual phenomena”, which would just mean that the actual phenomena have to be consequents of the antecedent. I don’t see much of a difference.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s