Believe me Apos, I’m fairly sure that Eris has put forward some of my difficulties with your argument. It’s just that I haven’t even been able to establish certain preliminaries with you. Or so it seems - I’m still not sure.
Furthermore, engaging in an argument about definitions usually precludes substantive debate. Personally, I usually prefer to accept the opposing side’s definitions provisionally, just so I can get on with addressing the underlying issues. Here, I haven’t been able to.
Preliminaries
Just as I can’t define atheism merely by looking it up in the dictionary (which will typically give a definition of strong atheism), so (I contend) it is invalid for Apos to mandate that “Agnosticism/gnosticism deals with whether or not the person can claim KNOWLEDGE about the validity of the god claim.” As far as I can tell, that is not the way empirical agnostics typically define themselves.
Of course there is the substantive issue about whether “Withholding judgement regarding the existence of God”, is a simple dodge. But I think I’ve defined this position reasonably, to wit: ~Believe(X) and ~(Believe(~X). At any rate, it is not answered by definitional discussions.
I see nothing wrong with Apos proposing a definitional structure and proceeding to a substantive argument. What I have a problem with is when he insists that agnosticism is thus and so, especially when this definition flies in the face of common usage. (Sort of. There is certainly some foundation for Apos’ framework.)
The Point: Definitions are tools; it’s better to address the underlying contentions than to argue about the “true” definitions or to redefine words so that the meaning of the opposition’s points are lost.
Nonpreliminaries, which I haven’t gotten to yet
There is the issue of whether Apos’ framework is reasonable. I have a few difficulties:
a) Counter-framework: Belief could be construed to be a stance towards an alleged fact: this stance is not necessarily binary.
b) Continuity difficulty: Although you can always restate a belief within a binary framework, this tends to over-simplify beliefs that are continuous. (So when belief varies by intensity, as in gambling or probabilistic contexts, you throw a great deal out at best and generate absurdities at worst.)
c) Qualitative difficulty: Although Professor Hosper takes the dictionary’s route and equates atheism with strong atheism, at least he acknowledges a variety of stances that can be held theologically. Now Apos’s 4 box grid does allow a certain variance. It would be better though, if (IMHO) the stated beliefs of the various parties formed the starting point of the analysis.
d) If you are a cognitivist or a Decision-theoretic cognitivist, belief does not arise independently of perceived knowledge. So the Apos’ key distinction between belief and knowledge is not unambiguously sharp.
A Confession
I believe I was wrong when I stated that the Infidel link proposed a different definition of atheism than the about.com link. To review, ““Weak atheism” is simple skepticism; disbelief in the existence of God.” The author then goes on to say that, "Disbelief in a proposition means that one does not believe it to be true. Not believing that something is true is not equivalent to believing that it is false; " Oops. That’s the position of Apos and About.com. So my proposed definition of soft atheism as one based on probabilities lacks foundation. I retract it, pending evidence of its use.