The agnostic wants more evidence and is therefore unsure; the atheist insists there is ample evidence and is therefore certain. What follows is a dialogue between an Agnostic (AG) and an Atheist (Ath), who eventually come to a meeting of minds...
but not without struggle.
AG: Agnosticism is more reasonable and more logical than atheism. Your absolutism is simplistic and the certainty of atheism is impossible.
Ath: You're right that I cannot be certain that there is no God in the same way that I can be certain that two plus two equals four. But lack of mathematical certainty does not preclude me from drawing conclusions with a high degree of confidence.
AG: So you are agreeing that you can't be certain, in which case it turns out that you are conceding agnosticism!
Ath: The reverse. I can insist it is provisionally true that there is no God, while admitting the possibility that I'm wrong, just as I can insist on being certain that neither Zeus nor leprechauns exist, notwithstanding the possibility that they might. I am not certain in the hard mathematical sense, but in the softer scientific sense.
AG: Well you're just trying to have your cake and eat it too by playing with the word "certain" - claiming you don't believe in God but admitting that you're not absolutely positive.
Ath: Acknowledging the possibility of being wrong neither makes me agnostic nor undermines my atheism. We're quibbling about what it means to be certain, but this confusion is why I argue that our difference is merely semantic, and when this confusion is eliminated, agnosticism is not compelling.
AG: The term agnosticism was coined by biologist Thomas Huxley in 1869 to refer to the insoluble nature of metaphysical matters, specifically spiritual ones. Huxley was adamant that you can't claim certainty about something unless there is sufficient evidence to justify it. So he rejected atheism on the same grounds that he rejected theism. He would not be on your side.
Ath: I'd like to think he could be convinced on the basis that there is sufficient evidence because our understanding has come a long way in the past 140 years since Huxley was writing.
AG: What evidence?
Ath: I think there are three strong lines of evidence - not proofs in the mathematical sense, but support in the scientific sense. The first relates to the concept of God.
AG: I assume you're going to argue that suffering and evil render a benevolent god impossible, but a theist would simply counter-argue that suffering is the cost of freedom and that, as God taught Job, humans are not endowed with sufficient wisdom to understand His reasons: God offers his presence, not answers to our questions.
Ath: That's my point exactly - as mysterious as God may claim to be, it's unintelligible that he would limit knowledge of his presence to a very select group. The evidence against a caring god is not so much the problem of evil or suffering, but the fact that billions of people don't believe in God. If belief in God is the sine qua non of human meaning and purpose, not to mention the gateway to a rewarding afterlife, why would a benevolent god deny this privilege to billions of people? In fact, why did God ignore all the Homo Sapiens who lived for tens of thousands of years prior to His revelation to Abraham?
AG: You've merely demonstrated the intellectual difficulty, not the impossibility, of the Abrahamic version of God, but there are other ways of thinking about god - as a causal, organizing spirit, or as a life force, or as...
Ath: "God" connotes specific traits that are shared among monotheists, and is problematic enough without aggravating the semantic chaos by suggesting that "God" can mean whatever imaginative thing that anyone stipulates.
AG: Fair enough. But humans have believed in some version of a god for a long time, so there is clearly something at the root of this belief.
Ath: Of course there's something, but it's not a supernatural something. The second line of support for atheism is basic human psychology: animals such as us, with complex brains that seek to understand in order to feel in control of our environment, will always yearn for ways to make sense of and cope with the existential anxiety of life's day-to-day stresses, not to mention the dark cloud of impending death. On top of that, religion was presumably useful to us, as we began living in larger communities - religion served group cohesion and rule-making. There are many outdated superstitions that we pass on for no other reason than cultural habit.
AG: Ok, but an antiquated need for believing does not disprove God.
Ath: Not on its own, but it's the second of three insights, the third being the evolution of religion over millennia, including the relatively recent invention of monotheism. Homo Sapiens have been around for about 200,000 years, with fully developed consciousness for about 50,000 years and elements of spiritual-based rituals since then. But monotheism didn't come until much later. It was likely the Babylonian destruction of Jerusalem in 586 BCE that gave monotheism a boost, as a way for the exiled Jews to re-invigorate their religion. During this period, much of the Hebrew bible was compiled, singling out "YHWH" (Yahweh), who was likely named after the patron god "YHW" of the Shasu people. But Israelites initially worshipped multiple gods and monotheism competed with polytheism for a long time.
AG: And of course the resurrection of Jesus complicated the issue of monotheism.
Ath: It took nearly three centuries after Jesus' death, when the Nicene Creed was formulated, to establish agreement on the status of Jesus' divinity as equivalent to God's. Scholastic consensus is that the historical Yeshua of Nazareth was a Jewish, itinerant, apocalyptic preacher - one of many at the time who rallied against corruption by the high priests of the Temple and, in the spirit of Second Isaiah, preached God's imminent creation of a perfect human society on earth. The first Christians were Jewish Christians and the two religions only began a gradual separation after the destruction of the Second Temple by the Romans in 70 CE. Most Romans considered the Christians to be atheists since they didn't worship the pagan Roman gods and Christianity may have disappeared altogether if it weren't for Constantine's legalization of and conversion to it, likely in order to unify the sprawling Roman Empire.
The origin of Christianity is parallel to the origins of Buddhism and Islam: neither Siddhartha Gautama nor Muhammad ibn Abd Allah aspired to start new religions. In the latter case, Muhammad wanted to bring Judeo-Christian monotheism to the Arab people. He grafted his revelations onto popular rituals of the time, including the Arab pilgrimage to worship at the Kaaba in Mecca, as well as the Jewish ritual of fasting. And parallel to the evolution of Christianity, many Islamic tenets were formulated after Muhammad's death.
AG: As intriguing as the history of the Abrahamic religions is, I'm not clear why you think it proves God's non-existence.
Ath: Most theists don't appreciate how contingent and path-dependent the development of their religious tenets has been. By understanding how monotheism arose and morphed over the past 2500 years, it becomes clear how God is a cultural invention. Together, the three main arguments suggest an extremely low probability of God's existence.
AG: "Extremely low" is not the same as "zero."
Ath: And we're back to my original point - that any difference between our views is superficial because the distinction between "extremely low" and "zero" is not meaningful. It's a distinction without a difference. What separates a learned agnostic from an atheist is nothing more than an irrelevant dispute over what it means to be "certain." I can make the argument for the impossibility of a specific god, such as the Abrahamic God, on the basis that it is incoherent, self-contradictory and unintelligible. But I can also make the argument for the improbability of any god, on the basis of the evolution of religious belief, explained by our natural, psychological need to make sense of our predicament. Whether God is impossible or improbable, the agnostic credo "we can't be certain" is a weak rejoinder that relies on a far too restrictive definition of certainty.
AG: Even if I acquiesce that the evidence favours atheism, I feel a visceral reluctance to embrace the certainty of atheism, even with your more flexible definition of certainty. Your arguments are persuasive, but I'm more comfortable concluding with the softer version: "I highly doubt there's a god."
Ath: I highly doubt it too. And that, my friend, is why, semantic quibbles aside, we ultimately share the same view.