| Posted 10/23/11 at 10:16 PM||Reply with quote #1 |
|It seems to me, that agnosticism with regards to a proposition, is belief in the possibility of the truth of the proposition.|
If this is the case, then agnosticism with respect to the truth of premise 3 of Plantinga's Ontological Argument, is to believe that premise 3 is possibly true.
Premise 3: It is possible that there is a being that has maximal greatness.
To be agnostic w.r.t premise 3, is to believe that it is possible that premise 3 is true.
To say that premise 3 is possible, is to say that it is possible that it is possible that there is a being that has maximal greatness.
To say that a proposition is possibly possible, is to say that it is possible.
Therefore, to say that premise 3 is possible, is to say that premise 3 is true.
Therefore, to be agnostic w.r.t premise 3, is to affirm premise 3.
Analagous to Flew's 'Presumption of Atheism', the default position of the person considering the existence of God, is absense of belief until valid argument is produced. Might the burden of proof in regards the truth of premise 3, be upon the one affirming that premise 3 is not true? Neither to affirm nor to deny a proposition's truth, is agnostic, but to affirm it's possibility, seems to be. In the case of propositions regarding possibility, such as in premise 3, it is inherently agnostic and to affirm it's truth, is to be an admission that one is agnostic as regards the certainty of there existing being possessing maximal greatness.
| Posted 10/23/11 at 11:10 PM||Reply with quote #2 |
Well, there are at least two different senses of "possibility." There's subjunctive possibility, which (among other things) involves possibilities insofar as they do not entail a logical contradiction, and there's epistemic possibility, which involves possibilities as far we know.
For example, you might say that it is possible that the trillionth digit of Pi is 8. However, this is only true under the epistemic interpretation of the word "possibility," because the trillionth digit of Pi, whichever single digit it may be, in terms of subjunctive modality, is necessarily the trillionth digit of Pi. So while the atheist may be charitable with regard to the epistemic possibility of God's existence, s/he can still plead ignorance with respect to God's subjunctive possibility.
The way I see it, the problem for the atheist is not that s/he has to shoulder the burden of proof with respect to the third premise; rather, the atheist's problem is that the evidence in favor of the premise is stronger than that of the contrary (myself, I don't believe in "default positions"). But then, whatever supports the possibility premise will probably, on its own, prove the existence of God, rendering the argument superfluous.
I actually prefer Anselm's version of the argument, for this very reason.
| Posted 10/24/11 at 12:29 AM||Reply with quote #3 |
|thank you chuckq1982,|
Although I do not believe that ontological arguments can provide us with the truth-value of propositions, in virtue of fact rather than by semantic rule, theyI find them very interesting to contemplate. I have since read 'On Denoting', as you had earlier advised me to do -- is there anything on the distinction between subjunctive and epistemic possibility, that you would recommend?
| Posted 10/24/11 at 01:32 AM||Reply with quote #4 |
|Alright, so I read the wikipedia article on subjunctive possibility.|
It seems that epistemic possibility, is subjunctive possibility, as far as I know.
Which is to say, that to affirm that something is epistemically possible, is to believe that according to all the knowledge I am aware I have, it is subjunctively possible.
To take into consideration your example of the trillionth digit of pi, it is epistemically possible that it is 8, which is ti say that it is subjunctively possible, as far as I can see. However, if it were actually, 5, then it would not actually be subjunctively possible, all things considered -- it would only be subjunctively possible as far as I can see.
To say that something is subjunctively possible as far as I can see, speaks less of the truth of the proposition,, and more of what is believed and understood by the person. However, this seems not to be a strong argument against my original position, precisely because everything we believe to be true, is believed to be true as far as we know. In what way can we ever say that a state-of-affairs is subjunctively possible? The only way we can, is to say that it is subjunctively possible, as far as we know, which is to say that it is only epistemically possible. Otherwise, we would have to be omniscient. If wholism is true, in regards to affirming the truth of propositions, then pragmatically and necesssarily, we can only ever speak of epistemic possibility, if we could affirm anything at all.
If we want to be able to affirm the truth of any proposition, it seems that we have to make do with states-of-affairs, as far as we know, rather than states-of-affairs as they objectively are.
I can not help seeing that in order for us to affirm the subjunctive possibility of something, rather than the epistemic possibility of it, require's a God's-eye view of reality. As long as we have to make do with our partial knowledge of things, everything we affirm can only be affirmed to be true, as far as we know. For us fallible humans, if we do not want to be complete philosophical skeptics, then we must live as if the coherent content of our minds corresponds to reality. The default position, is that we can know, or affirm things, on the foundation of the existing content of our minds.
| Posted 10/24/11 at 10:02 AM||Reply with quote #5 |
|As I understand it, subjunctive possibility requires a context (which can either be explicit or implicit). In particular, we need a conceptual model of how the world works, and then we can talk about what the model permits and what it disallows. Given our understanding of a certain system, "it might have been" that one thing or another happened. So for instance given my understanding of how human beings behave, it might have been that my this thread was never made. Maybe if I had had a better understanding of your psychology I would have known that this thread was inevitable. But I wasn't, and hence, from my point of view---in the context of my conceptual model of human behavior---it might have been otherwise.|
But subjunctive possibility results from a different division of modality than does epistemic modality. That is to say, an entity can be subjuctively possible but not epistemically possible, and vice versa. Neither supervenes on the other.
chuckg1982 really took the words out of my (proverbial) mouth with his example about the digits of pi. Let me ask you---and don't look the answer up on Google, just use the knowledge you have right now---is the following proposition true or false:
(D) Possibly, the 76th digit of pi is 5.
Given the broad logical interpretation of possibility (which is the interpretation Plantinga has in mind), the 76th digit of pi is only possibly the digit that it "really is." It could not possibly be any other digit in the broad logical sense. But without looking it up on Google, what are you going to say about (D)? Is it true or false?
In fact, unless you have an encyclopedic knowledge of the digits of pi, you don't know what is the actual 76th digit of pi, and so you're not in a position to rule on the truth or falsity of (D). In particular, you do not know whether or not (D) is true. But according to your OP, this requires us to affirm
(E) Possibly (D).
But the "possibly possibly" operator chain collapses to a single "possibly," and (E) therefore implies (D) on your view. So the agnostic to (D) has accidentally fallen into a trap whereby he needs to affirm (D) to be consistent.
Clearly something is wrong here, though. Plantinga has constructed an argument that seems almost magical---all we need to do is affirm the possibility of God's existence, and, viola !, we obtain the actuality of God's existence! But Plantinga's magic, like all magic, is founded in misdirection. One of the tricks that he hides from us is that he's using a very rigid sense of possibility which cannot be interchanged with our everyday usage of the term. (He has another trick up his sleeve for this argument, but that's beyond the scope of this thread.) We're not really getting something for nothing with Plantinga's argument. He just tries to make it look that way.
Hopefully that helps. Also thanks to chuck for the spot-on comments.
EDIT: Oh, and if you're curious, (D) is false. The 76th digit of pi is actually 6.
| Posted 10/25/11 at 12:59 AM||Reply with quote #6 |
Thank you hatsoff,
I believe I understand the difference between the two, nevertheless, I am skeptical of the claim that we can ever affirm something to be subjunctively possible, a posteriori.
What is the justification for our believing that there is the subjunctive possibility that 6 is the 76th digit of pi? It is conceivable that however we discovered the fact, we may be incorrect, but we are warranted in affirming the truth of the proposition within the constraints of our present knowledge. What I am arguing, is that there is a distinction between a proposition having the subjunctive possibility of being true, and our being warranted or justified in believing, that the proposition has the subjunctive possibility of being true.
In order for me to know that there is the subjunctive possibility of 6 being the 76th digit of pi, I would have to know that it is the case. Until then I can not reasonably affirm or deny the proposition. Herein lies the problem -- in order for me to know that the 76th digit of pi is 6, I must have justified, true belief that it is the case. In order for me to know that it is in fact subjunctively possible, I must know that my justified, true belief , is in fact true. It is obvious that we find ourselves against an infinite regress. We may have justified true belief, but we can not be sure that our belief is in fact true. Likewise, we can believe that which is subjunctively possible, but we can not know that it is in fact subjunctively possible, ie. we can not know that we know, ad infinitum, but we can be justified in believing that what we believe, corresponds to reality.
Epistemic possibility, is what we believe to be a subjunctive possibility, as far as we know.
Subjunctive possibility, is any identity or predication given certain constraints, eg. Logic, the laws of nature, metaphysical necessity.
Standing in front of my house, and seeing it not on fire, I would not be justified in affirming the subjunctive possibility of it being on fire, but as I walk down the street, although the actual subjunctive possibility does not change, I am more justified in believing that there is the subjunctive possibility of it being on fire. If I were out of town, and I received a phone call from a fireman saying that my house is on fire, I would be justified in believing it to be the case. However, if he were to tell me so, as I stood in front of my house, and I observed that it was not on fire, I would not be justified in believing him.
In regards to the possibility of there existing a being possessing maximal greatness, what we want to know if it is subjunctively possible, within the constraints of logic. When we want to know if something is subjunctively possible within science, ie. nomologically, we make a decision upon our limited understanding of what the world is like, even if we understand that current laws and models may be replaced which could affect the truth of our affirmation. But this is nothing more than epistemic possibility. All claims regarding subjunctive possibility may be contradicted with further knowledge of the way things are – all claims as to the possibilities of things being the case, are in respect to epistemic possibility. The flaw, as I see it, between the distinction between epistemic and subjunctive possibility, is that all knowledge claims are claims as far, and only as far as we know – and as far as I know, there is the subjunctive possibility that there is a being which possess maximal greatness.If I could think further than my thoughts, and go beyond the horizon of the Universe, which is my mind, so as to determine whether what I believe, corresponds to what is true, then I might be able to know if something is objectively, subjunctively possible. In the meantime, I will be here waiting for your reply.
Well, there are at least two different senses of "possibility." There's subjunctive possibility, which (among other things) involves possibilities insofar as they do not entail a logical contradiction, and there's epistemic possibility, which involves possibilities as far we know.
If premise 3 does not entail a logical contradiction, it is a subjunctive possibility.
Premise 3 does not entail a logical contradiction, as far as I know.
Therefore, premise 3 is a subjunctive possibility, as far as I know.
All beliefs, including beliefs as to a subjunctive possibility, are beliefs as far as one knows.
It is reasonable to affirm beliefs, which are believed to be true as far as one knows.
I have the belief that it is a subjunctive possibility, as far as I know, that a being possesses maximal greatness.
Therefore, it is reasonable for me to believe, that there is the subjunctive possibility, that there is a being which possesses maximal greatness.
To my mind it seems that, to be agnostic with respect to the truth of a proposition, is to affirm the subjunctive possibility that it is logically coherent. If we did not affirm even the logical possibility, would we still be agnostic w.r.t the proposition?
| Posted 10/25/11 at 08:21 AM||Reply with quote #7 |
It seems like you're taking the position that if we cannot find a logical contradiction in some notion then we should assume there is no such contradiction, i.e. that it is broadly logically possible. However this is clearly an untenable policy. For an illustration, consider the alternative hypothesis that
(H) It is (broadly logically) possible that there is not a being that has maximal greatness.
If you can find a logical contradiction in (H) then you have no need of the ontological argument! But if we assume that (H) is broadly logically possible just because we cannot find a logical contradiction, then given our definitions we obtain the unwelcome (from a theistic point of view) conclusion that God does not exist. Employing the same policy on Plantinga's original argument, we conclude that God does exist---and now we can find a contradiction!
Otherwise I don't know how you think we can conclude that premise 3 does not involve some hidden logical contradiction.
| Posted 10/25/11 at 04:38 PM||Reply with quote #8 |
There doesn't seem to be any way to prove the possibility premise except by begging the question. See below:or
(1) If it is not possible that God exists, then the cosmological argument is unsound.
(2) The cosmological argument is sound.
(3) Therefore, it is possible that God exists.
(1) If it is not possible that God exists, then everyone who believes in God is confused.
(2) At least one person who believes in God isn't confused.
(3) Therefore, it is possible that God exists.
With the versions of the argument posited by people like Hartshorne, Plantinga and Malcolm, most if not all attempts to advance the argument to an opponent will likely end at an impasse--a dialectical stalemate, if you will. The theist is within his or her epistemic rights to affirm the possibility of God; the atheist is in the same position to deny it. That's why, for me, the best versions of the argument dispense with the possibility premise altogether.
| Posted 10/26/11 at 02:56 AM||Reply with quote #9 |
I initially thought you had defeated my argument, nevertheless, I am not quite sure anymore.
Proposition H may be broadly logically possible, but like P3, we are only justified in believing it to be true, given the content of our existent knowledge.
Take the statement S: There is a swan, that is not white.
Before the discovery of the black swan, it was believed, as I am sure you are aware, that all swans are white. From this historical context, hypothetical ‘John’ may have affirmed:
A) It is broadly logically possible, that S is true.
B) It is broadly logically possible, that S is false.
From within the constraints of all true propositions, either A or B is correct, but John has epistemic warrant in affirming B, ie. that all swans are white, unless a swan of different colour is presented to him.
(H) entails the positive claim, that there is a true proposition that contradicts the existence of a maximally great being. P3 entails the absence of one. The absence of such a proposition in our beliefs is compatible with both P3 and H, ie, we don’t have to know it for it to exist. However, I am warranted in affirming that the next new proposition I believe to be true, will be compatible with the existence of a maximally great being.
John’s experience was that the meaning of the term ‘swan’, was indicative of an object, attached to it the predicate of ‘whiteness’. It may be broadly logically possible that there is not a being that has maximal greatness, just as it is true that not all swans are white, but measured against my existing belief-set, P3 has always found to be compatible, and in my conceptual experience, every proposition to be believed as true, was indicative of something attached to it the predicate of, ‘compatible with P3’. The more propositions I believe, which upon discernment are compatible with P3, the greater my belief that P3 is true. What about H, can the same be said for it? H is also compatible with all the propositions I believe to be true, in that it is always possible that we may come to believe a proposition, which happens to contradict P3, ie. We may discover the black swan. But I have no reason to believe that there is such a black swan.
The agnostic with respect to P3, would have, upon comparing P3 with his’ own believed propositions, would not be an agnostic if he had found a contradiction. His own experience seems to indicate there is nothing incoherent in the idea of a maximally great being. Ofcourse, his own experience would seem to indicate that there is nothing incoherent in the idea of a maximally great being, not existing. Because we are dealing with the idea of a maximally great being, we are not left at an impasse. To say that atheism is possibly true, is to say that it is possible that there exists a true proposition which contradicts the idea of the existence of a maximally great being. The fact that atheism does not contradict any of the believed propositions, is no evidence for the existence of a true proposition which contradicts the existence of such a being. All that P3 requires is the absence of such a proposition. The existence of a proposition, contradicting the idea of there existing a being which has maximal greatness, is out of the ordinary, for the agnostic.
He withholds belief in P3, because there is conceptual space for P3 to be false, but this is as rational as withholding believing that the sun will rise tomorrow, because there is conceptual space to think the proposition, “the sun will rise tomorrow”, is false. The sun not rising tomorrow, is out of the ordinary, and one is not warranted in believing that it is true. Of course it is possible, and the proposition is compatible with every other proposition held to be true – if it were not, I would not believe it to be possible. I however, am justified in believing that the sun will rise tomorrow, because that is the ordinary course of events. The ordinary course of events in my mind is that the existence of a maximally great being, is coherent, and I am warranted in believing that there is no true proposition which will contradict this. I place this elusive, defeater for the possibility of the maximally great being, in the same category as the Loch Ness monster, and Bigfoot. Until I am provided with good reasons, I will believe that all plesiosaurs are dead, all bigfoot photographs are doctored, and all true propositions are compatible with the maximally great being existing.I can not say with certainty that (H) is false, just as I can not say with certainty that Bigfoot does not exist. However, because (H) means that there exists a proposition contradicting the maximally great being's existence, and P3 means that there is no such proposition, I am in a similar position to Antony Flew's 'weak-atheist'.
| Posted 10/26/11 at 11:19 AM||Reply with quote #10 |
It seems to me that in your last post you have presented an entirely new argument. But hopefully you do agree that if the (3)-agnostic affirms possibly-(3) then this does not collapse into simply (3) due to the different flavors of possibility involved.
As for your other argument, I'm afraid I have to take issue with some of your assumptions. You say that it is as rational to affirm (3) as it is to affirm the sun rising. But our belief that the sun shall rise again is based on inductive inferences from our experience. Are you suggesting the same of (3)?
Now, I certainly agree that just because we can conceive (3) to be false does not prevent us from affirming that (3) is true. We don't need absolute certainty in anything. However, I do claim that we must have reason to think (3) is true before we will be justified in believing (3). The challenge for the OA apologist, then, is to find a reason to affirm (3).
Such a reason does not consist of the principle
(P) If we cannot find a logical contradiction in S, then we are justified in believing (or tentatively assuming) S to be broadly logically possible.
And to justify my denial of (P) I pointed to the fact that it leads us to a contradiction when we alternatively consider (H) and (3). In fact we could spin various examples all day, but I thought (H) and (3) fit nicely into the context of our discussion. The point, of course, is that (P) is false.
Now, it looks like you think that even though (P) is false, something like (P) is true. Maybe you have this in mind:
(Q) If we cannot find a logical contradiction in S and S does not require us to affirm a logical contradiction in any other statement, then we are justified in believing (or tentatively assuming) S to be broadly logically possible.
So even though we can't find a logical contradiction in (H), it requires us to affirm a contradiction in (3), and so (Q) is not applicable. But (3) does not require us to affirm a contradiction in anything, so (Q) is applicable to it. But here is where I must again disagree: For (3) indeed requires us to affirm exactly the same kind (i.e. broad logical) of contradiction in (H) as (H) requires us to affirm in (3). They are mutually exclusive in terms of broad logical possibility.
Maybe I've misunderstood you, but in any case I just can't find anything special about (3) when compared to (H). They both seem to be on the same footing.
| Posted 10/28/11 at 01:50 AM||Reply with quote #11 |
You seem to understand me quite well, except that you disagee.
I hope that by the end of this post, you will see that the qualification I have made to (Q), is not arbitrary.
We are justified in not believing the existence of p, unless we have reasons to believe in p, but if the nonexistence of p implies the existence of q, where q is a universe more likely given that p does not exist, and we have no reasons to believe in q, then we are not justified in believing in the nonexistence of p.
In our context:
We are justified in not being theists, unless we have reasons to believe theism, but if atheism implies the existence of a universe more likely given that atheism is true, and we have no reasons to believe that what we know about the universe is more likely given that atheism is true, then we are not justified in believing that there is no God.
This seems to be the position of the agnostic. He sees no evidence in the universe which confirms either the existence or the non-existence of God. When he considers the Ontological Argument, since he is an agnostic, he must see no contradiction in the properties of a maximally great being – so in the spirit of agnosticism, he neither affirms nor denies the existence of such a contradiction. However, the fact that the idea of a maximally great being seems consistent with every held belief, seems to be evidence that the idea of a maximally great being is coherent.
You told me that the 76th digit of pi is broadly logically possibly 6. I have never calculated pi to the 76th digit, but I don’t see anything incoherent in 6 being the 76th digit of pi. I don’t have the means to confirm it to be true, and I don’t have the means to confirm that confirmation, ad infinitum. Nevertheless, I am warranted in presuming it to be logically possible, until I have reasons to believe otherwise.
This seems to be the case with all statements, ie. We assume that a statement in all of it’s elements, is meaningful, unless we can find it to be self-referentially incoherent. The statements: “God does not exist” and “God exists”, are to be taken as meaningful unless it is proven otherwise. The situation is different in the case of these two statements:
1)“A necessarily existent being, exists”, and
2)“A necessarily existent being, does not exist”.
In the second case, we can only affirm it to be coherent, if we presuppose that an element of itself, “A necessarily existent being”, is not coherent. This violates the principle I am following, in presuming the intelligibility of a statement in all its elements, unless we have reason to believe otherwise. The more critical the analysis of the statement in which it is not found to be incoherent, the more reasonable it is for one to believe that it is coherent.
(Q) as I understand it:
If we cannot find a logical contradiction in S and S does not require us to affirm a logical contradiction in p, where p is the referrent of S, then we are justified in believing (or tentatively assuming) S to be broadly logically possible.
Language presupposes the meaningfulness of statements and the referrents of those statements. Statements which entail the impossibility, or the self-referential incoherency of things, such as (H), make an unjustified, arbitrary assumption. (3) assumes the cogency of a maximally great being, without asserting the impossibility of another entity. Ofcourse, all propositions entail the opposite truth-value of their negations, and in respect to contingent entities, we are able to affirm the broadly logical possibilities of both their existence, and non-existence, but in regards to necessary entities, just as we assume that a contingent thing is coherent unless otherwise demonstrated, we also assume the coherencey of those things which, if possible, must exist.
As regards to statements. we assume that a proposition is meaningful unless we have reason to believe otherwise. We do this in the case of both necessary propositions, and contingent propositions. In the case of contingent propositions, we can also affirm the coherency of their negations. But with necessary propositions, we can not do this. If we are to assume the meaningfulness of a proposition unless it is shown to be incoherent, then we are to assume a necessary proposition, to be necessarily true unless shown to be otherwise.
The 76th digit of pi, if 6, is necessarily 6. If it is not 6, then it is necessarily not 6. But there is an important distinction between:
The 76th digit of pi is 6, and
The 76th digit of pi is not 6.
The first assumes that the property of being the 76th digit of pi, and the property of being 6, are compatible. The second assumes that those properties are incompatible. The reason why we don't presume that it is in fact 6, is because it could just as likely be 5 or 7. We don't assume any incompatibility between any of the digits between 0-9, and being the 76th digit of pi, but (H) does presume such an incoherency, which (3) does not. The dichotomous nature between the broadly logical possibility of there being a maximally great being, and the broadly logical possibility of there not being a maximally great being allows us to, in the rejection of the latter, to accept, albeit tentatively, the former.
| Posted 10/28/11 at 10:21 AM||Reply with quote #12 |
First of all, I need to quibble a bit about the digits of pi. (This may or may not be relevant to our main line of discussion, but I think it deserves mentioning.) The "compatibility" you mention between the 76th digit of pi and each of the digits 0,1,2, etc., is purely epistemic. We would not be justified in assuming any of these to be logically compatible, except in the very narrow sense that we can logically examine our limited epistemic position. So you are not warranted in assuming it to be logically possible that any particular one of these digits is in fact the 76th digit of pi. You are only warranted in affirming the epistemic possibility.
But like I said, I'm not sure if that will bear on our main discussion. Maybe it will, I don't know. Anyway, moving on....
So you make a qualification to (Q). Let's call it (Q') to distinguish it from the unqualified (Q). It is:
(Q') If we cannot find a logical contradiction in S and S does not require us to affirm a logical contradiction in p, where p is the referent of S, then we are justified in believing (or tentatively assuming) S to be broadly logically possible.
In the context of our discussion, at first glance it looks like (3) satisfies the conditions outlined in (Q'), whereas (H) does not. But consider the statement
(H') The statement (H) is true.
I can't find any referent of (H') in which we would be required to affirm anything like a logical contradiction. You might be tempted to object that while (H') does not explicitly refer to such an object, it is closely related to one. However, the same can be said of (3). For consider:
(3') It is not necessarily the case that a maximally great being does not exist.
Notice that (3) and (3') are logically equivalent. Relationships don't get much closer than that! But even though (3) satisfies the conditions of (Q'), since (3') refers to the allegedly incoherent concept of the nonexistence of a maximally great being, it does not satisfy (Q').
Now, maybe you can refine (Q') to the point where I can no longer find counter-examples. But why do so? What is motivating the refinements, if not a desire to justify your preference of (3) over (H)? Contrary to what you were hoping, it all seems very arbitrary to me.
| Posted 10/29/11 at 05:03 AM||Reply with quote #13 |
Every necessarily true proposition, entails the incoherence of it's negation, but there is no parity between (3') and (H').
Although (3') entails the incoherence of it's negation, ie, "a maximally great being does not exist", (H'), when we understand the meanings of the words, entails not only it's negation, but also the concept of a maximally great being. There is a principle named the Presumption of Charity, in which when dealing with a set of beliefs or propositions of others, as much coherence is to be presumed with respect to those beliefs. This principle is also the only way in which we can resist falling into radical skepticism, because the first ideas and concepts a mind has, much be presumed in order for our beliefs to approach truth. In learning language, a presumption of coherence must be adhered to, in order to be able to form sentences, and to form our beliefs in the form of propositions. We have to maintain an agnosticism in regards to the coherence of the ideas and thoughts of our minds, ie. we don't know if they really are coherent, but we have to presume them to coherent, even if not all of them are. As our belief-set becomes more complex, as in science, the greater chance there is of determining whether our beliefs really are coherent or not. Our beliefs, and their coherence, approach truth as they become more complex, but this can only be done if we presume as much of our beliefs to be coherent as possible.
Imagine that (H') was the first belief you ever had. It would seem to be coherent, and in order for you to believe it, you would have to presume it to be coherent. However, in coming to an understanding of the definitions of the terms, we notice that (H) can only be coherent, if the concept of a 'maximally great being' is incoherent. A choice must be made -- either affirm (H'), and therefore presume it's negation to be incoherent, but also the idea of a maximally great being, or affirm (3'), which entails it's negation to be incoherent, but allows us to presume the idea of a 'maximally great being', to be coherent. To choose to affirm (H'), is to maximise the amount of incoherence presumed by one's beliefs. Why not presume all of the words to be incoherent, or live as if we can not make a reasonable,even if tentative, judgement on the matter? If we are to presume that we can have rational beliefs at all, it seems we must presume all of the ideas entertained in our minds to be coherent, even though we don't know which ones are, and which ones are not. If we must presume seemingly properly basic beliefs to be coherent, until it is shown that they are in fact not, then I believe this should extend to all beliefs derived from our properly basic beliefs. Euclid's axioms were presumed to be coherent, and so were the deductions made therefrom. The form of agnosticism which does not presume the coherence of the content of our minds, looks a lot like radical skepticism to me.
I might ask the agnostic, "why do you presume your basic beliefs to be coherent, but do not presume the compositions or inferences of those basic beliefs to be coherent". The agnostic, if he really did only presume that his basic beliefs were coherent, would then only believe those beliefs. The agnostic, in fact, does presume the compositions, deductions, and inferences of those basic beliefs to be coherent. When the concept of a maximally great being is brought to the attention of the agnostic, why does he not also presume it to be coherent? I understand your objection, that it presumes the incoherence of it's negation, but because the concept is now brought to the attention of the agnostic, he must make a choice, presume (3) to be incoherent, and also the 'maximally great being'; presume (H) to be incoherent, and it alone so far; or live as if one need not presume the coherence of the ideas in one's mind, until it is proven to be the case. (3) minimises the presumption of incoherence, (H) maximises it, and agnosticism is radical skepticism.