I'm way behind in my reading of other blogs. Richard's had quite a few good posts the past week that I've not had time to respond to. Unfortunately for most the discussion has passed. A lot of his posts are about the defense from free will to the problem of evil (See here and here) He also had a good discussion about the New Atheism issue that I'd blogged about last week. The most interesting post was one he linked to in his comments on the New Atheism debate. It was from last fall on the issue of religious experience and epistemology. I wanted to make a brief comment on this.
Roughly Richard's argument is that if I have a practically private experience it ought be transferrable. So if I, as a UFO doubter, see a UFO then if that justifies my belief in space ships it ought justify others I speak to about it. This seems false to me though since the experience of seeing something simply is more trustworthy than being told about it by someone else. I'm surprised Richard would make this claim. The reason for this, as I see it, is primarily due to the amount of information that I have to justify my first person experiences verses someone I tell. So I can know, for instance if I'd had other experience that may make my account untrustworthy. I know if I've been blacking out, seeing things that aren't there, have mental illness and so forth. All facts that a person I talk to doesn't.
While I think first person privilege to mental states is often overstated by philosophers, clearly there is something to it. I typically (although not always) have more information about how my thinking is functioning. Put an other way, even though I have problems with Plantinga reliabilism, it does seem the case that I can tell if my perceptions are functioning reliably that a person I talk to typically doesn't have access to.
Now I say typically since there are places where this is reversed. Where I think I'm thinking in a reliable fashion whereas third persons can quickly see I'm not. However I don't think this need pose an epistemological problem. It all depends upon how you take sufficient justification for knowledge. If you require justification to be such that there is no way new information could falsify my belief then of course this is a problem. If you require the looser sense of knowledge where we can be wrong, then I may be justified with one set of evidence but new evidence changes the justification.
Put an other way, I think the way one considers this case is tied up in terms of how one views justification. Some writers on epistemology end up demanding justification strength that is so strong that it entails certainty or close to it. Unfortunately in that case I find the debate strangely uninteresting since almost no belief we hold rises to that level.
This really gets at the heart of one of the problems with over-reliance on science. Science only deals with phenomena that can be repeated, and is only concerned with the properties of individual phenomena to the extent that those properties can be repeated in subsequent phenomena. The sorts of evidentialism/verificationism that Richard is accepting, even if distinguishable from bare scientism, still require the same metaphysics. That is, it requires a metaphysics in which the only things that can be true are those that are repeatable. This may be problematic when faced with individual experiences of things in the outside world (e.g., UFOs), but it's particularly problematic when we're talking about things like the experience of the divine. Science, and verificationism that relies on the same metaphysics as science, can't do anything with statements concerning these types of experiences, and therefore either rejects them out of hand or attempts to explain them from a third-person perspective in terms of underlying (usually biological, but always physical) causes that are in fact repeatable. Of course, there are problems with the sort of reductionism/naturalism that this move requires, but even if there weren't, the question arises as to whether there are evidentiary/scientific reasons for accepting the metaphysical position in the first place. If not, then the position itself becomes inconsistent.
I've been hearing from various sources that there are various logical/philosophical problems with metaphysical naturalism. Being a neophyte, I have no idea what those problems are. The best that I find are questions about the origins of the universe and of life which are not terribly convincing as stated in the resources I found. Would someone provide some pointers?
Jonathan, the Quentin Smith article that Macht (of Prosthesis) links in his comment at Richard's blog is a good place to start:
http://www.philoonline.org/library/smith_4_2.htm
It doesn't so much argue that naturalism is wrong as that most people who are naturalists don't realize why it might be right or wrong. And as Macht notes, Smith is an atheist.
Oh, and you can adhere to some versions of naturalism without the metaphysical assumption above (the one about everything being repeatable). I'm sure Clark could discuss this more knowledgeably, but since Hegel (maybe since Kant's third critique), there have been several attempts at a philosophy of the particular or the individual, and some of those attempts have been within a naturalist metaphysics.
My own problem is that naturalism and physicalism are quite hard to define in a coherent fashion. This is a historic problem. I have no problem with methadological naturalism. I even argue that Mormon theology entails some kind of naturalistic metaphysics. But what is naturalism?
Often it is a denial of the irreducible reality of some first person reality. (Which need not entail Cartesian dualism - just consider property dualism) But that seems an odd thing to hinge things on. Often it is defined in opposition to supernaturalism. But supernaturalism is even more problematic to define than naturalism. I've heard a few who argue it in terms of everything being knowable scientifically. And that kind of naturalism I'm actually fairly sympathetic to. The problem there is that one has the vague notion of what science is. Further clearly the science in that definition is more than the science we have now. Also just because science could know something doesn't entail that it does. (i.e. it doesn't make scientific knowing the only way to know; just that if something is knowable it must also be scientifically knowable)
There are other senses of physicalism. Personally while I don't have trouble with loose use of physicalism (and I do it myself) one has to be careful.
I'd second Quenton Smith on the problem. I think everyone has a vague idea of what physicalism entails and therefore don't really give it much thought. Not that this is necessarily bad. I think many basic concepts are hard to give definition to. It's just that folks attribute definitions to physicalism quite frequently which are completely unworkable. It's even worse when they do this without realizing they are doing this.
Chris, I don't accept any form of verificationism. I think it's perfectly meaningful to posit the existence of an undetectable being, for instance. The hypothesis could even be true. But that doesn't mean we have any reason to believe it.
In particular, I don't object to private experiences on the grounds that they "can't be repeated", or anything like that. (Replicability doesn't interest me in the slightest.)
My argument was instead as follows:
(1) If there's any such thing as essentially private justification, I obtain it through my experiences.
(2) My experience E informs me of precisely the following objective fact F: "Richard just had an experience E."
(3) Either fact F is evidence that C is true, or it is not. (This doesn't change depending on who's looking at it.)
(4) In principle, fact F is knowable by others besides myself (e.g. through my testimony).
Therefore:
(5) There's no such thing as essentially private justification.
Now, Clark has helpfully pointed out that *in practice* the experience is likely far richer than can be captured in our descriptions. Plus others may lack the background knowledge to assess my reliability. So this all makes a difference (relevant to premise 4).
Still, as a matter of principle, it's not as if the first-personal fact that *it was my experience* makes any intrinsic difference. In other words, the evidence provided by experiences is not essentially "private" or non-transferable (even if in practice we are never really able to transfer all of the information at our disposal).
I've assumed here that justification is propositional. It isn't the subjective experience itself that directly justifies anything. Rather, what's epistemically relevant is the *information* provided by our experience. (And information, of course, is tranferable.)
That's my assumption. Again, it's got nothing whatsoever to do with "repeatable" data. I don't know where you got the idea that I'm a "verificationist", because that isn't supported by anything I've written.
I agree (with numerous reservations) that first had experience is generally more reliable than second hand experience. However, I would argue that second hand private experiences are regularly transferred (at least symbolically since the original experience is ephemeral and direct memory cannot be transferred as can an abstract of the experience). We continually access information about our environment through this process. Of course, what allows us to draw conclusions as to the ultimate validity of any new information must hinge on direct experience. When I receive an abstract of a remote private experience, I must directly experience the transfer of information through some sensory medium. In a vacuum, however, this information is of no more consequence to me than a single pixel of color on a computer screen. It is only in context that the information acquires meaning. In addition to making some kind of sense of the information (which requires at least a tentative integration with my current mental and psychic map), I must also draw conclusions as to its reliability. For this I must make an evaluation of source and the trustworthiness of source while comparing the relative strength of those evaluations to the reliability of conflicting beliefs which I have previously accepted. Of course, the evaluation of source and trustworthiness are grounded in my previous direct experience.
What I believe is more germane to this discussion is anticipated in the statement that this matter is "tied up in terms of how one views justification." IMHO justification is neither perfect nor absolute. It is not perfect in that it can never be complete since original assumptions can always be doubted. It is not absolute in the sense that evidence sufficient for justification in one set of circumstance may be insufficient in another set. My reason for this second contention relies upon the first. If justification can never be complete (that is, at least some residual doubt must remain since we cannot confirm our axioms nor the ultimate reliability of our perceptions) then to talk meaningfully about justification we must define a level of certitude that we will accept. In my way of thinking, this brings us to thinking in terms of sufficient justification. Admittedly this is not as psychologically satisfying as the illusion of perfect knowledge but such an illusion is difficult to justify philosophically.
So, if justification is to be evaluated based upon its sufficiency, the question arises, "Sufficient for what?" I would suggest that it must be sufficient for what all knowledge is used for. It must be sufficient to act upon. Such action can be physical or mental but it must be based upon at least a tentative acceptance of validity sufficient to precipitate choice (passive or active). It is in this context that justification becomes relative since the gravity of the contemplated action determines the level of sufficiency.
If you will forgive a trite example: Suppose I am at a track with pari-mutuel betting and I receive a hot tip from a friend on the next race. To affect my betting behavior I will need to assess the quality of the information, the trustworthiness and reliability of my friend, the consistency with other information I have previously accepted, etc. After making those evaluations I can assess whether the information is justified. However, the evidentiary threshold required to place a $20 bet is far different from the threshold to bet my mortgage. Moreover, those respective thresholds will be different still for another bettor, both because his direct experiences affecting his assessment of the evidence will be different but also because the consequences of a $20 bet and a bet of his homestead will have a different value than for me.
Though the differentiation between direct experience and secondhand experience is a useful distinction, ultimately the sufficiency of justification for knowledge or beliefs is the degree of certitude required for a particular act contemplated by a particular actor. That sufficiency of justification is relative to the act and actor at a particular moment. Therefore the reliability of secondhand private experiences will always be a shifting target.
However, this principle of sufficient justification has relevance beyond the reliability of secondhand private experiences. It implies that the justification for any information must be sufficient within the world view of a particular actor. That makes each of us the arbiter of the information which is presented to us (though that, perhaps, is a topic for another thread).
PS-I am new to this forum. If my comments are out of character to the types of issues that are discussed, let me know and I will spend more time observing before chiming in again.
Richard, I'd argue that the problem is with (2). That is you've left out an essential but unstated part.
(2) My experience E against a background B informs me of precisely the following objective fact F: "Richard just had an experience E."
It's that background that allows me to interpret my experience E as entailing F that is key. Without that background then one can't communicate E as entailing F to others. So the problem isn't so much the richness of E, although that is true in part. Rather the problem is how an experience becomes meaningful. (And of course then we start quibbling about what constitutes an experience)
Just to make that more explicitly stated, being in (or having) experience E doesn't entail we can derive a proposition P describing E in an unmediated fashion. We're always stuck in the hermeneutic circle. (Yeah, this is the Peircean in me coming out)
I'd throw in the stronger claim that if E is the set of true propositions about E, then P, the set of propositions I interpret about E will have fewer members than E. Further there will be propositions in P that aren't in E. Put in simpler English, I won't be able to capture everything about the experience nor will I be able to only interpret it correctly.
Richard, I'm sorry if I've misinterpreted you. Still, I think that lurking in your argument are assumptions about the repeatability of experience. That is, presumably an entirely unique experience, or perhaps even an experience that an individual believes to be entirely unique (that is, they can't find anything to compare it to) will be difficult if not impossible to communicate. That would make convincing others difficult/impossible. The only way for an experience to be communicable is for it to have something that's describable in a language developed to communicated shared aspects of experience. So measuring the validity of an experience based on a person's ability (even in the ideal) to communicate it requires assumption about the repeatability of the components of that experience, if not the structure of it. It at least requires that the components and/or structure of that experience be analogous to something that is repeatable (i.e., that more than one person can have experienced).
Since certain types or aspects of religious experience are, by their very nature, completely private and non-repeatable, that creates problems when talking about religion and evidence/convinceability. For example, Jesus presumably had experiences that we will never have (e.g., his access to God the father in his adult life), and which he could not adequately communicate to people. Presumably Christians are convinced that he had these experiences for reasons entirely unrelated to Jesus' ability to communicate them (e.g., prior belief in his divinity and therefore his connection to God the father). The same goes for the experiences of prophets and mystics. They often don't communicate about the experience itself, but what that experience says about the empirical world (the world that science can study), e.g. in the form of predictions. People don't buy those predictions because they understand/are convinced of the nature of the experience, then. And I'm not sure such a method of conviction can be the right one to use for evaluating such experiences. To use such a method is to rule their possibility out, in essence.
Of course, all of this gets into the problem of other minds as well. Perhaps people can be convinced about the experience of a prophet or mystic by hearing it described, but only because through the prophet/mystic's communication of the experience, people mistakenly compare that experience to disanalogous experiences that they've had. In such a case, they could be right or wrong about the validity of the experience, but for reasons that aren't really related to the validity of the experience.
I just wanted to say that I hope at least some part of my last comment made some sense. When I read through it, even I was confused.
The question, Chris, is whether a totally unique experience is even intelligible. I'd argue that to be understood the experience (or at least elements of it) must be repeated. Now this repetition might itself be private. But repetition is always necessary so as to subsume the experience under some sign.
The problem I see with Richard's position (possibly) is confusing repetition with public repetition which aren't the same. For instance the way I experience blue might be completely unique but since it is repeated coherently I can communicate it because elements of it are shared and repeated.
Clark, I see your point. I'm not really sure what I think about the question. I'll have to give it some more thought. For now, I think it might be important to make a distinction you find in phenomenology a lot (e.g., I was just reading The Essence of Truth, where it's discussed), between explanation and understanding. Understanding involves personal relevance, or life relevance, and an experience that can't be repeated can be relevant (say, related in meaningful ways to other experiences) in a way that is perfectly comprehensible, but it can't be explained. That is, we can't determine its causes, but we can understand what it means to us. Anyway, I'm not sure that's what I actually think, but I'll definitely meditate on it a bit.
I think the understanding vs. communicable point is important. And often misunderstood. Understanding for Heidegger is very much about a kind of projection, not just in terms of concepts or signs but also practices. Understanding in a real sense is about competence or capability. I can communicate about running, for instance, but it is only in terms of my competence in running that I understand running. An obvious point we all notice, but one often forgotten. Put an other way, there's much more than meets the eye to language. But we all experience this in one sense or an other.
The stereotypical example of this is traumatic events like military service. You can communicate what happened, but a certain understanding is missing if you don't have competence in the practices that make the events intelligible. I was trying to touch on that in my adding to Richard's (2) the notion of a Background. (I was being vague and not trying to inject too much phenomenology or pragmatism into the discussion unnecessarily)
You are also right that this notion of understanding entails that singular events that aren't repeated can thus be meaningful in a certain way. This is actually a big point in Heidegger's philosophy of science. (I'd written a post a while back contrasting it with Kuhn) For Heidegger a single experiment can radically change how one thinks because we have a shift in the understanding of Being. (Ultimately all understanding is understanding of Being) Being and Time §69 really has a good discussion of all this.
Anyway, ultimately what one has to note is meaning is more than propositions. We might be able to communicate propositions but meaning might well be considered (ala Grice if one wants to avoid Continental thinking) as involving practices and changes.
Whether this entails a difficulty for epistemology is something I'll leave alone for now. I do think it entails that epistemological justification can't merely be a matter of propositions as I think Richard takes for granted. But my ultimately point is that even if we ignore that problem the conclusions Richard raised don't follow due to the "depth" of experience in terms of explication as propositions and the surrounding propositions that allow one to draw inferences.
Clark, thanks. That's what I was trying to get at, but you put it much better.
I should add that Heidegger makes a distinction, relative to understanding, between authenticity and inauthenticity. But that's obviously getting too complicated for this post. It's probably sufficient to say that understanding relates to the totality of the context (practices, references, etc.) that enable us to understand something or an event.
The obvious problem then of sharing understanding is that we never share the same context.
Ah, that was a helpful exchange! Framing the issue in terms of 'repeatability' still strikes me as a little strange, though I think I have a better idea what you mean by it now.
But I'm unsure: do you [both] think there can be ineffable content (i.e. my experience represents a state of affairs X as being the case, but X cannot be put into words), or that qualities of the experience itself (besides what's represented) have deep epistemic significance? Or is this not a helpful distinction here?
I don't think there can be ineffable content in logical terms but I think there can be in practical terms. By that I mean that I think you can always signify experiences and thereby talk about them. But it's kind of like Nagel's what is it like to be a bat. You can always provide a semantic description. Thus they are "effable." But there's a big difference between a semantic description that is true and understanding.
Now if you require understanding for effability then, yeah, there are ineffable experiences. However then you run into the other extreme. Almost everything is ineffable. After all, am I sure your experience of tasting salt and mine are the same? Can we express that?
As to your ultimate question, allow me to wax Peircean. I think the problem arises because we want to be able to reduce "feeling" to "representation" and I don't think they are reducible. (In either direction) So we're confusing the experience with the representation and assume that the representation can re-present (that is bring to presence) the originary experience. But that's impossible. All experiences are context dependent and contexts simply aren't replicable. You can get close and have very similar experiences. (As I'd argue is the case with salt) But that's something else entirely.
To add, I think that if one stays in the analytic field of philosophy one can get at a lot of this via Grice and Nagel. What you end up with is a view of language where meaning isn't just about representations but also manifestations. So language does things and communication isn't just about exchanging propositions but using language to generate a kind of experience (or at least memory of an experience) But that means rhetoric and not just propositions are key.
Of course speech act theory can capture quite a lot of this. But I think overall it ends up being found wanting precisely because of this problem. That is they think speech is ultimately about propositions and doing things with words is something secondary. (Some of the Searle/Grice debates are useful here, although they are arguing over a much narrower conception than I am)
I believe that what is being missed is the distinction between experiential knowledge and propositional content. It is the same debate in the philosophy of mind that is raised in Nagel's "What it's Like to be a Bat" and "There is Something About Mary." There are qualia involved in first-hand experience that are not available from a third-person perspective.
Take someone who has never smelled a rose -- or better yet a computer. Download all of the propositional content we can about the smell of a rose to computer, its chemical make-up; a complete physical description, the frequencies of light that are refracted to make it appear red and so forth. Now having conveyed all of this propositional content, does the computer know what it is to smell the rose? Of course not. The fact that the computer can successfully reproduce the relevant propositions doesn't count as the kind of knowledge that is at issue.
Religious experiences are immediate experiential knowledge. I submit that they are not "evidence" of religious knowledge from a third person or propositional perspective because the qualia of the experience cannot be conveyed. At most, a religious experience is an opportunity to say "if it happened to me, it could happen to you."
So I agree with Clark that there is an irreducible aspect of human experience -- experiential knowledge. Let me add that I cannot be sure that what you experience as the smell of the rose is the same for me either.
Don't most of the criticisms of Richard's position reduce to the narrowness of his argument rather than to its validity? If I am understanding the original argument, Richard posited that 1)a private experience had been reduced to a proposition, 2)that the elements of this experience sufficient to justify the proposition had been reliably and trustworthily communicated to another and 3)that the proposition logically led to a certain conclusion. Am I right in understanding that most of the criticisms focus on the difficulty and lack of universality (in the case of private religious experience) of satisfying #2 rather than on the impossibility? I agree with these criticisms but believe that they serve not to defeat Richard's argument so much as to drastically limit its applicability.
My biggest concern with Richard's point is that, by glossing over #2, it leaves the impression that the transfer of private religious experience is being elevated to too close a parity with experiences that tend to be more reliably transferred (e.g. those based on observations where perceptual reliability can be tested, methods of description can be standardized and instruments calibrated). The value or Richard's point is to remind us that, to the extent that #2 can be sufficiently satisfied, it is logical to draw conclusions from the private experiences of others.
Yeah, I think that's about right. The problem is the difficulty of #2. But clearly if we find someone trustworthy and the subset of propositions about E are communicated then it justifies our belief. I mean we do that all the time. If my wife says she saw a friend at the store that justifies my belief that she saw her friend at the store.
The problem is when the rubber meets the road since most accounts of religious experience aren't considered trustworthy by everyone. Understandably so.