I am often perplexed by the assumptions I come up against in conversations with evangelical atheists (those who seem to feel compelled to preach atheism and win converts to it). Among the assumptions is the opinion that if a belief is held by someone who cannot justify it, then it is a false belief or a belief in something false. Thus, disproof of a belief may be accomplished by demonstrating that it is a belief held by fools. Since belief is a human trait, I would bet that a case could be made that all beliefs are held by fools, but I do not believe (whether justified or not) that the merit of a belief can be measured by who holds it to be true. Nor do I accept that anyone could function in this world without beliefs. Even the most ardently empirical among us cannot get by entirely on personal observation and reasoning. The overwhelming majority of the mental model we have of the reality in which we live comes to us second or third hand, at best. We accept most of what we take to be true, or as true as we care about, mostly on the authority from which we heard/read it, or merely from the truthiness of it, and this goes for scientists as well as theologians. Even if we have some area where we carefully examine and question every detail, for the bulk of our daily thoughts, it is mostly lore, not experience that fills the gaps. That is our belief structure, and we all have one unless we are in a vegetative state, and perhaps even then.
Where the confusion seems to lie is in the way we describe belief. Belief ranges from certainty (the feeling, not the veracity of the belief) to mere suspension of disbelief. A person may feel certain, for example, that there is a God, but be very doubtful that their understanding of the nature of God is accurate beyond the vaguest assumptions. Such a person may have experienced a positive effect from prayer, and yet be doubtful that a supernatural force intervened as a result. This is not an inconsistency of belief, but merely a state of belief that might appear contradictory to anyone who has made assumptions that all people believe one standard story or another rather than each person having a worldview of their own. How true or factual that worldview is in any objective sense will vary, but none of them is perfect any objective sense and in many cases, mutually exclusive worldviews are equally consistent with observable reality as long as the definition of observable is not closed by one system or the other. Hence, the tendency to dismiss the belief system of another simply because it is inconsistent with our own is just a manifestation of our assumption that our belief is more correct for everyone.
One of the characteristics of this willful blindness is that the definitions we use personally to understand our belief structure are either kept secret from the other participants in a discussion, or defended as the only meaningful definitions to use in the discussion. For example: if I have an idea that God would only really be God if He was a) a ‘He’, b) personally attentive to me and each and every person and c) making constant decisions about my future, and then I say “I don’t believe in God”, I have pretty well stacked the deck in that conversation and the discussion is essentially pointless. If, on the other hand, I say “I seriously doubt that God would be ‘He’ or ‘She’ in the way that we are, or that the level at which a Supreme Being would operate would be on a gazillion case-by-case bases, but I do believe that we are not at the top of the sentient entity category…” we have the start of a conversation. From there we could at least agree on what we mean when we disagree or agree on whether something is, or even could be, real. More importantly, we can then see where what we agree on or disagree on is even relevant to the conversation at hand. Whether there is a God who responds to prayer is probably not contingent on whether the Universe was created in six days, but because some believe it was so, many think that this point is the central question to the larger questions. For every person who believes that the Nature of God is such that creation stories from scripture can be and in fact are true, there are many more who find it far more likely that the creation story need not be literally true for there to be an aspect of truth to the larger narrative. Those who stick to the literal truth model tend to agree with those who claim they have no belief at all on one point: To doubt any of it is essentially not to believe. This all or none definition of belief simply defies any reasonable discussion. The Agnostic is seen by these two extremes as simply too intellectually lazy to commit to a belief. I suggest that any rigorous examination of the assumptions will show that an Agnostic approach is, in fact, more likely to withstand unbiased scrutiny than strict Atheism (not just the lack belief in god(s), but a belief in the lack of god(s)) or a religious Fundamentalism of any bent.
At this point, I must clarify a bit. I do not assert that whatever you believe to be true is true for you. You have to be using those terms in very non-standard ways to justify such a position. Whatever reality is like, there are consistencies that are independent of the observer, and some worldviews match up better than others with what is real. However, there are mutually exclusive descriptions of the same reality that fit those consistencies with roughly the same success. In some cases this would be because one system focuses on a different area then does another system, and in some cases the amount of subjectivity in play renders any objectivity largely moot. But if you believe that you can jump off a cliff and no harm will come to you, then you may have to redefine “harm” in some less physical and more spiritual way in order to maintain that belief when you get to the bottom. In other words, there is something to the idea of reality, and some belief structures can be wrong about it far more than others. Still, where two models of understanding reality focus on different aspects of reality, it is not necessarily inconsistent or hypocritical to maintain different descriptive systems for each. An idea of what “supernatural” forces, if any exist, for example, would simply not be part of any system limited to “natural” forces. Truly supernatural forces would not necessarily be observable in terms of natural law, because there is no reason to assume that natural law would be affected by a supernatural force. If a Tornado picks up a child and deposits the child unharmed elsewhere, there is a natural explanation for it and this should not be surprising or troubling to anyone who also believes that a supernatural event also occurred in that same example.