So, before we can get into what it means for things to be ways of knowing and whether there are any other ways of knowing than science, we need to define what it means to be a way of knowing. And that, of course, requires us to decide, at least loosely, what it means to know.
As might be expected, I start and pretty much end with the philosophical definition, which is “justified true belief”. However, this loose terminology can introduce some confusions since the meanings of pretty much all of those words are not always clear or consistent, so let me expand it to the full definition:
S knows that p iff:
S believes that p.
p is true.
S is justified in believing that p.
Where p designates a proposition and S is a subject.
What it means to believe something isn’t entirely simple, but it should be simple to give a quick-and-dirty definition that will work for this. For me, what it means to believe is to simply think or hold that a proposition is true. Whether this is simply a psychological state or has more epistemic import is left to the side, for now. However, some people distinguish between religious belief and normal belief. I don’t. I consider all religious beliefs that are worthy of the name belief to be simply cases where a proposition is held to be true. Anything above and beyond or separate from that is no longer belief, but is something else that may either spawn the belief or be the result of it. But for me belief is simply thinking or holding that a proposition is true.
True, however, can get a bit more complicated philosophically, and has been discussed over many, many years as to what precisely it details. However, for me what it will be for a proposition to be true, loosely, will be that the proposition is actually the case in a domain; it reflects the way things are in a domain. Specifying “in a domain” is for me quite important because some propositions are true in some domains and not in others. For example, take the statement “Santa Claus flys around the world bringing presents to good boys and girls on Christmas Eve”. As a proposition about the real world, it’s false; Santa Claus does not exist. However, as a proposition about the concept of Santa Claus, it’s absolutely true. So used in the latter domain — the conceptual domain or perhaps the fictional domain — the proposition is true, but not in the former domain. Some would argue that this ambiguity simply means that I haven’t defined my propositions properly; if I had, I’d see that they were separate propositions and so there would be no problem and no need to specify a domain. But it seems that even in that case I would have to add the domain to the proposition to make the distinction, which then ends up with pretty much the same result as specifying that truth is domain-dependent. And so for a loose definition of truth, it doesn’t seem that bad.
Okay, now the big one: justification. This is what has driven epistemology for thousands of years. It’s an even bigger question than truth, philosophically, although some might argue that. What does it mean to justify a belief to the level required for knowledge? How justified is that? And what does it mean?
I adopt the “reliabilism” definition, which was in vogue when I was taking epistemology courses, although I also think it’s a pretty good one. It replaces “S is justified in believing that p” with “S’s belief that p was produced by a reliable truth forming faculty”. I also append to that “under the conditions where that faculty is reliable”, since a reliable truth-forming faculty may have conditions where it won’t produce truths reliably. For example, sense data about your backyard when the backyard is lighted reliably produces truths, but not in the dark. So, then, justification is done by a faculty that has proven itself generally reliable. How has it done that? Well, that’s not all that important for my definition of knowing, but it may become important for ways of knowing later on.
So, then, what does it mean to know? You have a belief that is produced by a faculty that you can generally rely on to produce true beliefs and the belief you have is in fact actually true. But note that when we talk about knowing, the “true” part isn’t all that important. How do we know that a particular belief is true or not? Well, beyond the “justification” part, we don’t. This may seem problematic, and certainly bothered Ophelia Benson. Replying to Hutchinson’s original post, she takes him to task over justification and truth:
He (Ian Hutchinson) gives the gist in the first para.
One of the most visible conflicts in current culture is between “scientism” and religion. Because religious knowledge differs from scientific knowledge, scientism claims (or at least assumes) that it must therefore be inferior. However, there are many other important beliefs, secular as well as religious, which are justified and rational, but not scientific, and therefore marginalized by scientism. And if that is so, then scientism is a ghastly intellectual mistake.
Notice that he carefully leaves out the “true” in “justified true beliefs” – the standard philosophical definition of knowledge.
But when we’re dealing with ways of knowing, do we need anything more than “justified”? I argued once with a professor that including “true” in the definition of knowledge caused issues because we needed a way to determine if the belief was actually true, and he replied that that only applied to knowing that you know, not about whether you actually knew. So if it is true, then you know it, even if you don’t know that you know it. I advanced the argument to argue that including “p is true” in the definition ran us right back into Cartesian certainty, which we had rejected as being a requirement for knowledge, but with this distinction that objection goes away.
In Benson’s case, the same counter argument applies. While we may not know that we know, the question is if it is reasonable for us to believe that we know. But that, of course, comes down to nothing more than justification. If my belief is justified, by the definition of knowledge it becomes reasonable to believe that I know it. At this point, and in this debate, there’s no need for anything more. Thus we can see that the whole debate here turns on justification; with justification the beliefs are reasonable, and so if Hutchinson has a method that justifies then he can make a claim to knowledge that is at least reasonable.
So it seems clear that being a way of knowing will, in fact, rely critically on justification. Belief is exceptionally common in our everyday lives, and truth as truth is difficult for even ways of knowing to guarantee that they get (and would itself require a justification). So justification is the key, and ways of knowing, then, will provide justification for our beliefs by at least being a reliable way of producing beliefs that are true. Note, however, that there is an ambiguity on “justified” when we apply it to beliefs. Often, people will claim that if a belief is not justified, it means that it shouldn’t be believed; it would automatically become an unreasonable belief. However, doing this would mean that beliefs are held to the standards of knowledge for their “reasonableness”, and thus would become knowledge itself. Thus, you would end up only believing that which you knew to be true. Putting aside the problem just raised about how we only reasonably believe that we know something (generally), this seems to leave us with two unpalatable options. First, we can lower the standards of knowledge to include everything that we think it reasonable to believe. Alternatively, we can keep the standards of justification high and refuse to believe some very common everyday beliefs, like believing someone when they tell us what time it is. While some may take either of these options, it far easier in my opinion to simply understand that it may be reasonable to believe something that you do not know. What this does, then, is introduce cases where person X may reasonably believe p while person Y reasonably believes ~p, because neither of them have sufficient justification to go beyond reasonable belief to knowledge. There’s obviously far more to say about knowledge itself but I think I’ve belaboured this point enough for now.
Anyway, back to ways of knowing. Note that the initial appeal to reliablism talked about faculties. Is a way of knowing, then, just a faculty? I don’t think so. For something to count as a way of knowing, in my opinion it must be a method of knowing. So it must have a methodology that it follows, and that defined methodology must in part be why we think that way of knowing is reliable and that it therefore justifies beliefs and that it therefore produces knowledge. Now, by this I do not mean that we cannot have faculties that produce knowledge that are themselves not ways of knowing. I think, in fact, that sense experience is a faculty that generally does produce knowledge as it reliably produces justified true beliefs. But there is no methodology to reading sense experiences; we just get them and translate them into true beliefs. Sense perception, then, may well be the only faculty that reliably produces true beliefs that is not also itself a way of knowing, but far more work would have to be done to demonstrate that (testimony is potentially an example as well). But suffice it to say that for me, to be a way of knowing means that you have to have a set, identifiable, defined methodology. And to count as a way of knowing you would have to have a unique methodology; if you use the methodology of another way of knowing, one of you might be a way of knowing, but if you are not the same way of knowing then one field simply uses the other as its way of knowing.
Thus, we can see that in general religion will not meet this standard because we don’t really have a unique defined methodology for it. Religion uses history and literary analysis and philosophy and theology and testimony and social beliefs, but it doesn’t seem to have anything methodologically specific to it. However, I can already predict that most people are already asking “Well, what about faith? Isn’t that a methodology that is unique to religion?”. Well, that depends on what is meant by faith. I’m actually pretty sure that no one has defined what this methodology of faith would be, and so it wouldn’t meet my standards for being a way of knowing, at least not yet. I’m open to being convinced, but I’d need to see the argument first.
But a key part of this argument would be getting around what I think faith is. I’ve already talked about what I think faith is before, and taking that definition it is clear that no such methodology is possible. For me, faith is nothing more than believing something to a degree of certainty higher than is strictly warranted by the evidence. Thus, for faith to give knowledge by my definition would mean that you are holding your belief as being something you know — ie you believe you know p — while strictly speaking by the justification you have available you don’t know that p. Now, I don’t argue that this is necessarily a bad thing, but we can clearly see here that this is a psychological knowledge, not an epistemic knowledge. You feel like you know, but epistemologically the justification is just not there. For ways of knowing, I think it quite reasonable to hold them to the standards of epistemic knowing and not merely to the standards of psychological knowing. So while faith may not be bad and may not be limited to only religion, it in and of itself is not a way of knowing, and without that it seems there is no unique epistemic methodology that could make religion a way of knowing. But someone could, of course, argue that point.
However, we can find clear examples of ways of knowing by my definition. Science, of course, reliably produces true beliefs and has its own defined method, the scientific method. Philosophy, I argue, also reliably produces true beliefs and has a defined method, even if that method is more general and less precisely defined than science’s. Mathematics as well reliably produces true beliefs and has a different method than philosophy or science. Everyday reasoning is again reliable and produces true beliefs but doesn’t do what science or philosophy or mathematics does. While we may need to do more work to define what these methods are (see my attempt to clarify the philosophical method on this blog), that they have defined and distinct methods is not that controversial, and so that they are separate ways of knowing at least seems like a reasonable hypothesis.
Now, some may argue that out of all of those, only science produces true beliefs. Let me address some potential objections here:
1) They don’t reliably produce true beliefs because they make errors (this would be one that could be aimed most productively at everyday reasoning). Simply making mistakes doesn’t eliminate something as a way of knowing. And good thing, too, because science makes enough errors (that it eventually corrects) that it would be eliminated right from the start. The key would be how many it produces as compared to what it gets right and in the case of things like science, everyday reasoning, philosophy and maybe mathematics how it goes about catching its own errors. If there is a method for correcting false beliefs that is workable and will catch most of them, then it is still a way of knowing. Science and everyday reasoning both share at least a semblance of a testing model, while philosophy and mathematics work on an analysis model. But it’s clear that all of these not only have methods for correcting errors, but that they do correct errors. So making errors does not disqualify something from being a way of knowing as long as part of the methodology includes error detection and error correction.
2) They haven’t produced a lot of true beliefs; thus, their content is low. This is an objection that’s probably most profitably aimed at philosophy. However, that doesn’t disqualify something from being a way of knowing either. This is not a contest where we add up the number of true beliefs generated and say that if you don’t generate a certain number, you aren’t really a way of knowing. Ways of knowing, like truth, apply to domains, and some domains may produce knowledge at different rates. Moreover, more conservative or skeptical ways of knowing will produce knowledge slower than those that are more open. The amount of knowledge, then, is not a criteria for determining if something is a way of knowing or not.
3) They don’t produce true beliefs because they don’t produce truths about the actual/natural/real world. This objection is one that could be raised against both philosophy and mathmatics. Again, ways of knowing have domains, and produce truths about those domains. They can only be judged by what they produce in their domains of study. If those domains do not interest you, then you do not have to care about those ways of knowing … but you cannot deny that they are indeed ways of knowing.
This, I think, covers most of what I’d want to say about ways of knowing as a preface to the underlying argument about scientism. Later
Tags: Scientism 101