This is the basis of all superstition. Our crops were doing poorly, but then I burned a chicken and now they're doing well. Let's keep burning chickens.
You have to have an explanation of what counts as a valid empirical relation. And that depends on a whole lot of things worked out by philosophers and scientists.
Including, for example, that you can safely ignore correlations that can't have a basis in physical laws. Or that you can write certain symbols on a blackboard and compute the probability that your correlation is worth spending more time on.
I think we have different definitions of empiricism. Empiricism is the belief that sensory experience is necessary for knowledge, not that it is sufficient. It stands in contrast with the belief that knowledge can be justified by pure logic without reference to any sensory experience.
Your claim was that empiricism justifies itself, I was explaining why that's not the case.
Your definition is overly broad IMO. Empiricism is usually taken to be the belief that the sole source and justification of knowledge is ultimately sensory data. See e.g. Wikipedia or Britannica.
It stands in contrast with rationalism sure. But more importantly in the context of your comment and the parent, empiricism also stands in contrast to the only sane POV (IMO), which is that knowledge is combination of empirical and rational sources, as pointed out for example by Kant.
"empiricism is an epistemological view which holds that true knowledge or justification comes only or primarily from sensory experience and empirical evidence."
Note the phrase "or primarily".
> knowledge is combination of empirical and rational sources
The "or primarily" hedge is all I need to refute you (or at least to show that we are not actually in disagreement). However, rationality can itself be justified by empiricism if the Church-Turing thesis is correct, which I believe it is.
I suspect that you're using empiricism somewhat more broadly than philosophers do, and that you may include things like science as a whole. Science is more traditionally IMO considered to be a recursive process with rational (i.e. theory) and empirical (i.e. sense data) steps. In science, sense data provides a constraint on theory and theory (together with psychology) provide a constraint on the raw sense data (in the same way, for example, an AI provides structure to the raw input vectors).
If that's something along the lines of what you believe, then sure I'd say we're on the same page and we're not actually in disagreement.
But I wouldn't describe that belief system as saying empricism is self-justifying. Empiricism in practice goes off the rails pretty quickly with things like subjective idealism and the belief that there are no external entities, only sense data etc.
The thing we have now, where theory and experience mutually constrain each other seems obviously correct to me both in the academic sense (e.g. I think that Kant was right that the mind provides structure to raw sensory data). I also think it's true in the broader sense (e.g. that the applied sciences tend to move more slowly on their own and advance much more rapidly when combined with theory).
> Science is more traditionally IMO considered to be a recursive process with rational (i.e. theory) and empirical (i.e. sense data) steps.
That's true, but like so many traditions, it's wrong. Science can be justified entirely in terms of empiricism.
To be clear, it is not at all obvious that this is possible (it relies on the Church-Turing thesis), and so people can be forgiven for making this mistake. But it's still a mistake, and it's grounded in ignorance.
> Empiricism in practice goes off the rails pretty quickly with things like subjective idealism and the belief that there are no external entities, only sense data etc.
Unless your sensory experiences are radically different from mine, then you have to concede that those experiences behave, to a very close approximation, as if external entities exist, and a good explanation for that is that external entities do in fact exist.
Ironically, that explanation turns out to be wrong, but to see that you need to dive into quantum mechanics. And the justification for quantum mechanics grounds out in the sensory experience of perceiving the results of experiments. There is no escape from empiricism in the world we live in.
I do think we must have pretty different sensory experiences then, because my awareness of sensory data is entirely of processed data and nothing at all like the raw output of my retina (which I personally don't have accesss to although we can access it in animals if we cut their heads open). My ears have a high pitched ring constantly, and although it's indistinguishable from a sensory perspective from hearing a high pitched noise out in the world, my brain has somehow never attributed it to an outside object. Similarly with occular migraines.
I think a reasonable test would be whether you could build a useful embedded device out of only a sensor and no logic to interpret the output of that sensor.
The thing that lets us even talk meaningfully about computable functions is that they have a logical description. We wouldn't be able to say anything meaningful about the entire class of such functions if we were restricted to those we have experience of. It's not even clear what it would mean to universally quantify over a class of logical structures in a purely empirical world. We don't even need to get as fancy as computable functions. We couldn't even meaningfully talk about the class of natural numbers.
And the thing that lets us translate little dots on some quantum mechanical detector is that we have a mathematical theory that predicts that dots will look one way if such and such is the case. And they will look another way if such and such is the case. If sense data are all you need, then you have to give an account of why a trained experimental physicist can see different things in those dots than a toddler does.
> my awareness of sensory data is entirely of processed data and nothing at all like the raw output of my retina
Yes, obviously.
> I think a reasonable test would be whether you could build a useful embedded device out of only a sensor and no logic to interpret the output of that sensor.
A more useful test would be if you could build a useful embedded device with no sensors.
> The thing that lets us even talk meaningfully about computable functions is that they have a logical description.
But that logical description is a description of a physical process. The whole notion of computability is inherently physical. Even the lambda calculus is a description of a process of symbol-manipulation, and symbols are physical things.
> If sense data are all you need
That's a straw man. Empiricism does not claim that sense data is all you need. It is the claim that sense data are the primary source of knowledge. All knowledge -- even mathematical knowledge -- starts with sense data, is ultimately grounded in sense data, but is obviously not just raw sense data.
Nobody is disputing that sense data are necessary. The standard rationalist position is, to quote Leibniz since it's easily accessible on Wikipedia
> The senses, although they are necessary for all our actual knowledge, are not sufficient to give us the whole of it, since the senses never give anything but instances, that is to say particular or individual truths. Now all the instances which confirm a general truth, however numerous they may be, are not sufficient to establish the universal necessity of this same truth, for it does not follow that what happened before will happen in the same way again.
So sense data are necessary but not sufficient, especially for logic. If you believe that empiricism is self-justifying then it does seem like you have to start with the manifold of sense data and somehow build up logic from it. Otherwise you're going to end up #include-ing logic somewhere that may not be obvious, and you've landed to the position you were originally arguing against that you need a reason-based framework to justify empiricism.
A lot of smart people have tried to work out a version of empiricism that attempts to build of logic from sense data. For example, Russell's work and early Wittgenstein. But it always ends up getting pretty crazy.
The easiest way, IMO, to see why it won't work to build up human knowledge from sense data is to realize that the human knowledge system is a distributed system. Empiricists are working with the wrong unit of analysis. The words we have for thought are things like "logic" from logos or "word" and "rational" from ratio or "account" or "reckoning". They're both words that are fundamentally about multiple parties (in CS we can think of them as nodes) doing computation together. Some of the oldest ideas we have about reasoning are dialogs, and that basic idea persists to this day with things like game theoretic semantics. There seems to be something fundamentally "distributed systems" about thought and knowledge.
The individual nodes and their sensors aren't the whole story or even the bulk of the story. That's even more true when you consider that how we interpret sense data depends on billions of years of evolution. Yes we operate on sense data, but we do so using a logical structure that heavily constrains what we see, hear, taste and so forth.
Sense data plays a role, as Leibniz says, just as network cards play an important role in the Google data center. But I don't think anyone would really argue that the value of any large knowledge base is ultimately grounded in network packets.
> So sense data are necessary but not sufficient, especially for logic.
But they are sufficient.
> it does not follow that what happened before will happen in the same way again.
Yes, that's true. It does not follow that sense data are insufficient. All that follows is that induction is not a valid mode of reasoning (which is true -- it isn't).
> The easiest way, IMO, to see why it won't work to build up human knowledge from sense data is to realize that the human knowledge system is a distributed system.
Nonsense. Distributed systems can do logic. Being distributed is completely irrelevant.
> Mathematical knowledge can be justified by pure logic without reference to any sensory experience.
No, I don't think it can. I challenge you to give me an example of mathematical knowledge that you can justify without reference to any sensory experience, keeping in mind that reading and hearing people talk are both sensory experiences.
It doesn't even have to be math. I'll bet you can't even define the distinction between "true" and "false" without reference to sensory experience.
If you try burning a chicken repeatedly when crops are doing poorly, and it works every time, that hypothesis is supported by the evidence. There's obviously no chain of direct causality, but that is immaterial.
This reminds me of scientists who completely shit on astrology for lacking any predictive power. At the very least it has some predictive power because people believe in it and subtly conform to its predictions. Beyond that, there are obviously cycles in the universe and we know biological cycles synchronize with natural ones (melatonin is any easy example, but animals synchronize to year level cycles as well). Are the stars causing the correlations, and are all the correlations astrology talks about present? Clearly not, but there is a mountain of poorly controlled empirical observations that hint to more being there than we've laid out with regard to human character that science is content to shit on because it can't do anything else without looking bad.
You have to have an explanation of what counts as a valid empirical relation. And that depends on a whole lot of things worked out by philosophers and scientists.
Including, for example, that you can safely ignore correlations that can't have a basis in physical laws. Or that you can write certain symbols on a blackboard and compute the probability that your correlation is worth spending more time on.