Too late, too late! Alas!
Spiritus,
I started to read that article. Pretty nice. If I were to finish reading I don’t think I’d ever manage to get to post a thing here.
I can certainly see how, for example, Rand actually toggled between both forms of a priori knowledge, all the while refuting the idea of anything a priori. I have never claimed to be able to justify everything Rand. Good thing, too, since I can refute some it myself. 
I am sure you would have no problem with “Introduction to Objectivist Epistemology.”
However, enough of that. Onto more elaboration, at least of how I see things.
I am actually a huge fan of “cogito ergo sum,” and have since based my entire world-view on it. In a way.
Ever since I read his justifications for that infamous (in latin, anyway) statement I have seen it as self-evident(a danger in itself!). Everything could be false, a trick, an illusion. Except, of course, the fact that there was some thing I call “Me” that was going throgh the process of doubting. Hence, cogito ergo sum. Fantastic.
If I were to be called anything it might very well be a rationalist. I’m not sure of all the things that entails, so you can be the judge.
It seems obvious to me that a priori knowledge and epistemic systems which exist sort of in the same way universals exist. That is, there are a set of things that can only be founded by generalizing specifics. Nothing tells us it is valid to do so. I would hazard to say that we are “hard-wired” to have a set in our epistemology bag at birth, in this case. Nonetheless, abstraction is possible by the standard “reflection” process. It isn’t guaranteed to happen, of course. I dare say I know a few people who find reflection into matters a bit distaseful.
Anyway, once a process of reflection occurs, we find that we don’t know what makes our initial espist. set valid, or indeed if it is valid at all. Here is where I put your initial axiom in. This is the beginning of the first fabricated epist. system.
We postulate a thing based on no evidence at all and see how that effects views/observations/morals we have already aquired just by living in a society(implicit contract? ). No validity is involved in this initial assumption; checking the changes it implies is in line with whatever validity-checking set was already “har-wired” into us, or perhaps, now, imposed on us through the society we live in (the latter, especially, falls in with the negative utopian books we discussed previously).
No doubt our assumption caused a schism in our current set. To have done otherwise would be either impossible or totally improbable(not sure which yet). From here we find that a person could stop the questioning process and go fundamental. Otherwise, we see that the schism might be implying a problem in our current set.
Yadda yadda yadda.
So we come to strong set(epistemic set) by this process, postulating (and perhaps keeping previous axiomatic ideas) new ideas and seeing “what happens.” Can I trust my own memories? etc.
I strongly disagree with the implication that the set is initially empty or indeterminate. I would rather say it is not explicit. The act of reflection and of forming explicit a priori concepts tends to probe this initial set and force it to become explicit, but it is clearly already known what it is, if you see what I mean. In fact, I would say, questioning knowledge is not a part of this initial set (which is why I say that initial rejection of reflection leads to fundies of any set) but more something that happens as a side-effect of desiring efficiency. Perhaps I should stop here for some critiques, showing that I don’t know that I think what I think haha.
My cat’s breath smells like cat food.
Prove it.
We should definitely clear a couple of things up first:
Do you want to talk classical epistemology or real world compromise? When you start talking about hard-wired respocnses, etc. you are no longer speaking epistemology, you are speaking theories of consciousness. (Which might, indeed, be an epistemology).
Do you want to discuss Descartes? i.e. Was he just too silly to see his tautology, or was he actually was trying to express, in slightly different terms, an epistemological “first element”?
Do you want to be rigorous? When you say things like, “It seems obvious to me that a priori knowledge and epistemic systems which exist sort of in the same way universals exist.” you are making multiple levels of epistemological assumption, none of which are justified in a strictly classical sense. You might be able to make a case for them as empyrical a priori assertions, but if we are going to talk epistemology, empyricism is groundless (until an epistemological set is developed from which empyricism can be derived).
** Not a question – a caution (or a corollary to the above)**. If we are really going to talk meaningfully about the way epistemological sets can be “developed”, you need to make a concerted effort to let go of preconception and “common sense”. Both of those are epistemologies, and of course we cannot rely on them until we have a means of validating their membership in our set. Statements like, “once a process of reflection occurs, we find that we don’t know what makes our initial espist. set valid, or indeed if it is valid at all. Here is where I put your initial axiom in.” reflect that mistake. You cannot “find” anything about knowledge, nor “know” anything about your reflection, until that first element is asserted. Remember, “how do you decide whether a priori knowledge belongs in the set of epistemologies?”
Are you familiar with Russell? Not for epistemology directly, but his set theories/notations are useful in clarifying some things – things like, "I strongly disagree with the implication that the set is initially empty or indeterminate. I would rather say it is not explicit. Now – this confusion might be my fault. My typo a few posts ago shoud have read “membership in the set is indeterminable” not “indeterminate”. If you see a distinction between indeterminable and “not explicit” please make it explicit.
However (this is where Russell helps) – this is true for the “SET OF ALL EPISTEMOLOGIES”. But of course, that set itself (since it represents a statement of knowledge/truth/validity) could under a classical model be considered an epistemoology itself. This would lead to some interesting self-referential paradoxes. A solution can be found in Russell’s Theory of Types. Our “SET OF ALL EPISTEMOLOGIES” becomes a Class (meta-epistemology), and paradoxes of class membership disappear by definition. (BTW–Russell’s set definitions, since they avoid elements of axiomatic self-referentialism, also avoid Godel statements. Just thought you’d like to know.) [sub]damn I was going to keep this short. Really.[/sub]
Anyway – the saem is not true for any particular set of internally-consistent epistemologies (such as those used by a particular person can use to decide “what he knows”. That set is rigorously empty until populated.
[sub]Okay – that’s probably more than enough for now. We seem to have chased everyon else away already. Except Matt, brave connoiser of feline respiration that he is.[/sub]
Mine too. Do you think they’re the same cat?
I have to take a course in epistomology some time in the near future. I used to think I knew that it would make my brain explode, now I know I didn’t know that. 
Damn you.
However, I didn’t wish to discuss any theory in particular, just what I think. Draw on whatever theories you like. Also, I know less about Russel(+ whitehead?) than I do about Godel. :::pauses for laughter::: {{how is THAT possible, spiritus cries}}
Anyway, I find it very commendable to try to “start at the beginning,” but also totally impossible, so I guess we will discuss real-world compromise. It seems little more than a mental work-out to try and remove all the turtles from epistemology.
[subject]Indeed, really, what is the difference between a non-empty indeterminate epistemology “class” (more than a set of sets? Or by def’n a set which contains itself?) and the hard-wired implicit epistemology that I propose? I think we should start here in what is the difference, and where would your initial set come from?(or is that outside the realm, by definition, of epistemology)[/subject]
Yes, let’s be rigourous. I’m stopping here, at the first turtle, hopefully. Also, since you mention Russel, ever since I heard of modern set threory I’ve been awfully interested in seeing if I can pick it up. I find its relevance to the topic is completely apropos. But I really don’t know much about it other than what you pick up in later, and earlier, algebra classes (which is next to nothing, as I’m sure you know)
Descartes, too silly to see his own tautology? I think that the tautology was the point. When I use it, it is my point, anyway. Though I’ve never actually referred to is as such…actually, I’ve never heard it referred to as such. But anyway…
Or, instead of having it out in a board, feel free to email me (unless others are joining the fray). Makes for better reading, IMO. You killed me in other arguments, so we need to start really, really simple here, lol
Well, here’s the thing about picking a compromise (call it Alf) as the starting point: it can lead us to interesting extrapolations o fwhat particular set of valid epistemologies we can develop from Alf, but we are not really discussing “things we can know directly”. It is similar to picking a set of axioms and seeing what geometry they define:interesting, perhaps even useful, but carrying no implications for the “true” situation.
Absolutely!! Epistemology is nothing but mental workout. (or mental masturbation, depending on whether you enjoy it not at all, just enough, or waaaaayyy too much: Goldilocks and the three philosophers.)
subject:
Legion. I think your unfamiliarity with the theory of types makes it harder for you to see (Originally Russell’s work solo[sup]1[/sup], BTW, though he and Whitehead refined it together into a basis for number theory and arithmetic in Principia Mathematica [sub]with a nice foundation from Peano[/sub].) A class is very much different from a set of sets, though it is very much simiar, too[sub]gets clearer every second, eh?[/sub]. By definition a class cannot contain itself. A class and the members of the class are of different types (think strongly typed computer languages, for an analog of why the identity fails.)
Your “hard-wired implicit epistemology” is certainly a member of the class, though we have no means of determining whether it is until we place it into our Alf-set. But it is no more “priveledged”" a member than the epistemology “things which make me happy are true”.
Well, the set “comes from” the existence of its members. More precisely: the class exists as a necessary corallary of a determinable Universe. If we do not have a determnable Universe (internal or external), then epistemology is a null-proposition. Membership in the class is indeterminable, as we have said. It is not possible to fully enumerate the class of epistemologies.
rigor
It doesn’t look that way to me, I’m afraid. It seems to me that there are a number of a priori assertions that underlie your “hard-wired” epistemology, as well as the non-testable “Ur epistemology”, of course. That’s okay, though. We can start from turtle 97-3/a. It really doesn’t matter at which point we decide to say ‘go’, so long as we don’t delude ourselves that we have found a natural beginning.
descartes
I do too, but I think the question of whether Descartes understood the epistemological tautology underlying his proposition is open to debate. Certainly, there is evidence that he understood that the “I” in question was itself an undefined subject. But there is also evidence that his epistemological expansions from the tautology at some point assume a definitive subject without ever actually developing it. Similare critiques have been leveled against his proof of the existence of God (Third Meditation, IIRC). Russell called the pronoun completely unjustifiable, but I am not certain I agree with that. Then again, I think the tautology has some very interesting epistemological nuances that I doubt Descartes ever intended (and quite likely would not have agreed with). Even then, I think it is useful as an evocative metaphor, but not useful as a purported basis to rational dualism. YMMV, of course.
It is a clear tautology, though. The subject is the conclusion. If you read it literally, that is. I like it as a metaphor, or perhaps a koan.
I have long thought that western philosophy suffers from an underappreciation of koans.
stuf
If it stays “just between us” for long enough, I might take you up on that. On the other hand, the number of views keeps climbing even though the number of posts does not. Perhaps our ravings are at least providing amusement to our fellow dopers. [sub]My god, how long can they go before the veins pop?[/sub]
Oh – and I hope we are not going to argue in this thread. Arguing about epistemology is like fighting over whose kid is better looking. (Actually, you might win that one. From the last ultrasound, mine still looks like a cross betwen an Amazonian tree frog and a Whitley Schrieber alien.)
[sup]1[/sup] [sub]If you are really looking to nail down the Theory of types, Russell introduced it in hiw Principles of Mathematics. Later, he and Whitehead refined it in Principia Mathematica. What I have really been talking about is the Ramified Theory of Types developed in that later work. This combines Russell’s Theory of Types with the Theory of Order, which allows them to deal with semantic paradoxes of similar construction to the Russell paradox as well as the syntactic paradoxes for which the simple Theory of Types accounted. It didn’t seem necessary to make that distinction initially, but if you are going to seek further readings I don’t want to be a source of confusion. I highly recommend Principia Mathematica, BTW, it is probably one of the 3 most influential books on logic ever written. And Russell is much easier to read than Wittgenstein, or Descartes.[/sub]
Nah, hardly plan to argue in this thread. Just want the ideas on the table like a nice buffet.
Man, the turtles are making my head spin now:
Eh? Then the most important question is to see whether or not there is a determinable universe. This cannot be done without a validity-tester/epistemology of some sort. So here, we need to assume that the epiSet(to stick with computer analogs!) is not null, regardless of whether we define an epiSet member or “find one” through reflection (my hard-wired set). I can’t escape a turtle-stack here! If we assume that the epiSet can store an epistemology, then we will never find that the universe is unknowable, because the epiSet member will itself define a knowable universe (a member of detUniSet–determinable universes that could exist). Without assuming an intrinsic member of epiSet, we can never EVER be sure that the “real” universe is a member of detUniSet unless the member of detUniSet we found is completely renumerable/knowable. In this latter case, it seems that so long as no ideas-about-reality clash with observations-of-reality we really do know what we think we know.
But I already see a problem with my outline above. We don’t even “know” that observation is useful for anything yet. Ugh. And I’m sure I’ve still imposed a few a priori assertions in there, too. Little bastards.
[side note]Descartes’ tautology: A friend of mine had a philosophy teacher who insisted that the true interpretation of cogito ergo sum was not, “I think, therefore I am,” but “I think: I am.” While some language literalists might equate the two, you might note that that is like saying the translation of “2+2=4” is “Two plus two equals four.” Know what I mean?[/side note]
[Edited by bibliophage on 09-26-2001 at 01:47 PM]
So close. . . but we cannot have a valid epistemology until the determinability of a Universe is established. This is the self-referential paradox (one of them) that we need Russell’s theory to escape.
Not quite. Remember a class is distinct in type from its members. To continue the analogy, we need to establish that epiSet exists, no matter the value of epiSet. For instance, if we could determine that our Universe is not determinable, that would be a proposition about our class of epistemologie (epiSet==NULL). That determination itself is not a member of epiSet, but represents another step up the heirarchy and is a member of the class of proposition about teh class of propositions about epistemologies. Warning, turtles crossing.
The important distinction (at this level) is that our ability to make statements about the class of epistemologies is distinct from our ability to make statements about epistemologies. They are different types, so the statements deal with different realms. Note: this does not help us actually identify an “ur proposition” in any of our nested classes that “matches reality”.
Exactly! How are those veins doing?
Russell allows us to speak meaningfully about these ideas, but he does not help us define the initial membership in a way that is logically assured of matching reality. He cannot, because it can’t be done. What the Theory of Types does allow us to do, once we choose a starting point, is to build a set of epistemologies that avoid those nasty self-refersential paradoxes like {X| X is not a valid epistemological statement of epiSet[~]} It also (for me, at least) helps by making the rigorous distinction between a class and a member of a class, which makes it easier to detect when we have allowed unwarranted a priori assertions to sneak into our constructions.
See, and having started to embrace this epistemological insanity you already find yourself able to find the flaw in your observational construction. And they said we were mad, MAD I tell you [sub]okay, so we’re mad. It doesn’t have to be a bad thing. The medications are good, and you don’t have to worry about what to wear.[/sub]
On the side
I understand the distinction your friend’s teacher was making, but I don’t think it is supported by the text. Descarte pretty explicitely “deconstructs” his epistemology down to the identity of thought and then uses thought to imply a subject. He then uses that subject to construct all else. (Well, he hedges a bit on allowing other elements to slip underived into his epistmology, but that’s another issue.)
My favorite introduction to Epistemology is called “Labyrinths of Reason” by William Poundstone
Frog, eh? And here I have pro-lifers telling me fetuses are nothing like animals. Feh.
So I can’t say a thing on the previous statement except: propositions? I can clearly see that propositions about the existence of epiSet are not epiSets themselves. I can, with the aid of the programming-class analogy, even see that an epiSet can’t contain itself. What I can’t see is the top. Ever read Godel Escher Bach? This sounds like the GOD problem. If you haven’t read it, GOD is an acronym for God Over Djinn, right? So we rub our lamp, and djinn comes out. We get a wish, and we wish for more wishes. That’s a meta-wish, and this Djinn needs to ask GOD. So he asks the meta-djinn, who has to refer to the meta-meta-djinn, etc etc.
I haven’t read the book in a while so I can’t remember how it ever ends, but in any case there was never a cieling as such. But there is clearly a conclusion. In this case, we have propositions about classes of sets, and propositions about those propositions, etc. Though there is no ceiling, there is some conclusion. Wouldn’t this, as well, be an elaborate tautology of multiple steps? If it is, can we not accept a smaller tautology like cogito ergo sum? If it isn’t a tautology, how can an initial assumption or proposition NOT define itself as a subject, be its own predicate, etc?
Nobobdy knows anything in the future. I assume I will be sitting here six seconds from now, but a bomb could hit my house and I could be vaporized.
There is no end to our ability to formulate propositions about a class. Now, don’t confuse these turtles with our epistemological turtles. Any given set of “valid” epistemologies must have an axiomatic beginning. Said beginning is not a priori in either the classical or the empyrical sense because it can never be subjected to revision or validation from any epistemology in the set derived from it. It is importaant to make that distinction because once within the framework of an epistemological set we will be blind to any “errors” resulting from our choice of epiSet[0]. The shadow of our axioms cannot be escaped; therefore, we must try to keep it as small as possible.
Regarding class heirarchy, though. The process is infinite. It is simple to demonstrate that for any class of Order k we can formulate a proposition of order k+1. In other words, I can’t see the top, either. Nor can anyone else, since there is no top.
I am not certain what you mean by conclusion. Can you explain it in some detail?
Well, we can certainly try to phrase a tautology if we choose. The structure itself does not imply one, though. Class structure is not a proposition about class structure. In fact, because the Theory of Types was formulated explicitely to disallow syntactic and semantic aradoxes of self-reference, it is not a trivial matter to create a tautology from “within”. I am not even certain it is possible. We can, of course, create a tautological statement about class structure which would not be valid within a class structure.
We can accept any tautology we choose. Honesty and rigor, however, require us to understand that by doing sso we forfeit any claim that our epistemology is “natural”, “rational”, “logical” or any of the other descriptors which have been used to imply that a given set of epistemologies was “more valid” than another.
I can’t think of any way for an “Ur proposition” to be epistemologically valid (or invalid, for that matter). I also cannot think of any way for the process of selecting an epistemological Ur statement to be anything but invalid under the epistemological set it defines.
Thus my very first post in this thread. [sub]in which the “set of epistemologies” refers to any particular epiSet not the class of all possible epiSets[/sub]
djbdjb2: I haven’t read it. Is it a historical survey/overview or does it argue for a particular epistemological set?
Then the top is where we say, “I’m done with postulating about postulations.” We reason about a system from outside the system, and that is a meta-epistemology. Of sorts, a way of resoning about reason itself, and so on and so on up the chain.
You provide a way out by stating that our initial assumption is the core of the system and cannot be checked by any means. In other words, it isn’t an a priori statement, its more of a definition.
Here’s where I sort of see a problem, though.
Earlier, you said,
Now, are you implying that knowable reality is not true? If not, what is the difference between our first axiom and the latter quote?
Certainly, one can stop making meta-propositions about classes, but that is a different question. The next turtle is still there, even if we choose not to look at it.
Of course – that is the first turtle: the class of propositions about epistemolgies. One member of that class, is teh set of all possible epistemologies we discussed earlier.
No. I am saying that we cannot know whether it is true. Not the same thing. More precisely, I am saying that under any epistemological system we derive it is not possible to verify the corellation of our truth statements to an objective reality. That relationship is always hidden by the shadow of our initial assumption(s).
None. The second quote merely discusses some of the ramifications of our using an axiomatic beginning for our epistemolical derivations.
First, what I meant by conclusion earlier:
This is the “by inspection” argument in mathematics, we reason about a system to achieve that conlcusion.
I looked for Principia Mathematica but to no avail. That Russell was a writing fiend, tho.
Anyway, we’ve so far found that we’ve moved our “First Mover” argument out of creation and plopped it into epistemology. I find this wholly dissatisfying. This, essentially, validates all sorts of philosophies so long as they do not logically refute themselves. Plato was just as right as Sarte, in the beginning.
I’m not searching for the “One True Epistemology,” of course, but it still seems that we should be able to avoid infinite regression of “reasoning about the way we reason” and avoid what could rightly be called “The God Axiom.”
Anyway out of this that you have heard of?
Actually, to modify that, could there even be anything but the duality of “beginning”/“infinite regression”? Is this a totally either-or proposition?
In that case, I can see several conclusions which can be made about the class heirarchy: for instance “heirarchical ordination is infinite”. I am not sure how useful, if at all, such conclusions are to questions of epistemology.
Not at all, unless by “First Mover” you mean something other than the idea that cause and effect implies a guiding consciousness to any observed phenomenon.
Quite the opposite. It demonstrates that no epistemology can be validated from within it’s own framework.
No. One of them may have been perfectly right while the other was perfetly wrong. The rigorous way to express the idea is closer to: We have no epistemologically valid means of determining the “truthfulness” of any epistemological set.[sub]where truthfulness is an expression of correspondence to a determinable internal or external universe.[/sub]
Many people do. Human beings, as a rule, prefer simple answers which provide surety and security, hence the plethora of “absolute truths” which have been asserted through the ages.
I see no third alternative. If God appeared before me and told me the truth of the Universe I would still need some method of deciding whether God’s word was a valid epistemological starting point. A God axiom if I’ve ever seen one.