I think there’s something very useful and very perceptive in the definitions of atheism vs. agnosticism that I’ve seen advocated by atheists. Essentially, it separates atheism/theism from agnosticism/gnosticism. The assertion is that there are two separate parameters here, and saying “I’m not an atheist — I’m an agnostic!” is missing the point.
If we were to represent this in a chart, it would look something like this:
People whose beliefs fall into the gray boxes we would call “agnostic,” according to this approach.
Still, something about this definition is so out of line with the common understanding of the word “agnostic” that I’m not fully comfortable with it. Language only exists insofar as it’s a shared set of definitions, and if everyone agrees that a word means a certain thing, a small group of us can’t just decree it to mean another thing and then act all pouty when people don’t know our new definition. (Yes, even if it is more “technically” correct.) More importantly, I see some sense in the common understanding of the word. Some agnostics really don’t consider themselves atheists or theists. They feel unable to pick sides. The jury is still out for them. Yes, there are atheist agnostics and theist agnostics, but then … there are clearly some agnostic agnostics (as well as some agnostic gnostics, people certain about the lack of an answer to the god question — call them ignostics).
If we redrew that table as a coordinate plane, these missing people would show up as the x-axis. Agnostics are the people marked here by gray and by blue.
I know, I know, it doesn’t really make sense to talk about a binary question like “are there any gods?” as a continuous spectrum between “yes” and “no.” The gnosticism-agnosticism axis makes a little bit more sense as a continuum, given that degrees of certainty are possible, but I don’t know where the transition between “certain” and “uncertain” would best be placed. Cut me some slack here — I just needed an infinitesimally thin line between atheists and theists, and I’m not trying to imply anything more than that by the diagram.
The only way I think you could prefer the top chart to the bottom one is if you consider “atheist” to mean “someone who lacks any belief in a god” rather than “someone who believes that there are no gods.” If that were the case, atheists would be the set of people who are not theists. And that’s a legitimate way to define that word, for sure. But I thought the whole point of this two-different-parameters scheme was to separate the ideas of belief and certainty, which necessitates pigeonholing people into compartments based on their beliefs. It seems to me that we can’t have it both ways.
I understand the value of telling most of the people who claim to be “agnostics” that we think they are actually atheists. Based on the discussions I’ve had along these lines, I’m confident that we’re right about that the majority of the time — so many of them think they have to have proof that no gods exist before they can legitimately say they are atheists. However, it strikes me as misleading and perhaps disrespectful to completely exclude from our terminology this category of people who actually claim lack of an answer as their position.
Jojo the hun on 
Ubi Dubium
/ October 7, 2010 at 8:45 amI like your thoughts about using a continuous spectrum to discuss belief/non-belief. I think that the coordinate plane might be a little difficult to use in everyday discussion. I prefer Dawkins’ 7-point scale as being more linear:
1 Strong theist. 100 per cent probability of God. In the words of C. G. Jung, ‘I do not believe, I know.’
2 Very high probability but short of 100 per cent. De facto theist. ‘I cannot know for certain, but I strongly believe in God and live my life on the assumption that he is there.’
3 Higher than 50 per cent but not very high. Technically agnostic but leaning towards theism. ‘I am very uncertain, but I am inclined to believe in God.’
4 Exactly 50 per cent. Completely impartial agnostic. ‘God’s existence and non-existence are exactly equiprobable.’
5 Lower than 50 per cent but not very low. Technically agnostic but leaning towards atheism. ‘I don’t know whether God exists but I’m inclined to be skeptical.’
6 Very low probability, but short of zero. De facto atheist. ‘I cannot know for certain but I think God is very improbable, and I live my life on the assumption that he is not there.’
7 Strong atheist. ‘I know there is no God, with the same conviction as Jung “knows” there is one.’
Using a scale like this, someone can easily say, “Oh, I’m around a 2.5″ or “I’m at about 6.8 right now”, etc. I find it a good aid to conversation about the subject. (Dawkins has said he considers anyone who declares themselves to be either a 1 or a 7 on this scale to be delusional.)
Jesse
/ October 23, 2010 at 7:31 amI prefer the top chart for the reason you gave. I think your definition of atheism is too narrow because, at core, atheism just means godlessness and there is more than one way to be godless. One can be godless by believing that the whole of existence does not include any gods, sure, but you can also be godless due to just not having a god, as would be the case with the agnostic-agnostics that you describe. I think atheism should be understood in this broader sense, for three main reasons:
First, the meaning of the word since the earliest times has been rather clearly inclusive of agnostic-agnostics. The people who coined the word, for example, were the people who were building bonefires for heathens, and they would not have spared for an agnostic-agnostic. They would not have said, “Oh, you’re godless in the other sense of the word? Well that’s excellent because that saves me from having to use so much of my firewood!” The idea that the word atheism does not refer to the agnostic position is revisionary and is promoted mainly by the Christians who want to forget that their ancestors were throwing people into bonefires and by the Agnostics who recently converted away from Christianity but have yet to reconsider the issue without wearing the rose-colored glasses and by the Agnostics who have been misled by the other Agnostics. The freethought literature, on the other hand, has been faithful to the original meaning.
Second, it makes perfect sense etymologically, semantically, and logically. The word atheism is comprised of two morphemes: a- and theism. Both of these morphemes have clear meanings. Combine the morphemes into atheism and combine their meaning into “not theism” and you have everything you need to know. That’s about as straight-forward as any semantic derivation of meaning could be. In addition, this forms a logical disjunction of the form A or Not-A. This means that everyone who thinks God-talk is meaningful (everyone who is not noncognitivist) can be classified as one or the other without exception. This can all be achieved with the 100+ year old definitions of each word and morpheme. This approach is about as “hyper-rational” as you can get with a natural language.
Third, the problem you were having fitting agnosticism into the chart only arose for you because you were using the narrow definition of atheism. Broaden the definition to bring it in line with its historical usage and agnosticism fits just fine.
If you want greater nuance, you can always subdivide. To get an idea of what I mean, start with your first chart but with my definition of atheism. Then, divide atheism into “positive atheism” and “negative atheism”. The “positive atheism” box would contain the atheistic positions that you included in your original chart. The “negative atheism” box would contain all the people who do not have gods but who do not take the position of positive atheism, which means that your agnostic-agnostics would be found there. The boxes can be infinitely subdivided for as much nuance as you could stomach. This single subdivision of atheism is all the nuance I really want for most purposes–if more nuance is needed in any given conversation, then it can be inserted in the conversation rather than the chart itself.
As you can see, this approach to classification is not misleading and it does not exclude the agnostic-agnostics, which addresses both of the major concerns you had with this approach. (The agnostic-agnostics actually have more room to stretch out their legs in this system, being neither excluded nor shoved into tiny blue line.) It’s actually less misleading than your own approach because your approach does not allow for nearly as much nuance or clarity.
This is what my classification system looks like. That is a quick mock-up and could be improved, but it does give the general outline.
As for the scale proposed by Dawkins, it has two major flaws. The first is the use of probability percentages. As someone who refuses to pull percentages from his butt and who has no means to arrive at such percentages mathematically, I cannot even be located on Dawkins’ scale. The most precise thing that could be said about me is “covering the range from 2 to 6″, which would likely cause confusion. The second flaw is the capitalization of the word “god” without specifying whether that’s important. When someone uses the scale, are they rating the probability of the Christian god, the Islamic god, the Deist god, all gods, some gods, or what? If someone told they were at a 4 on his scale, what should I take from this? That they are epistemologically cautious for not ruling out naturalism or deism because they two look so much alike or should I consider them fruitcakes for thinking that talking snakes were that plausible? The capitalization creates so much ambiguity that the scale is useless for that reason alone, never mind the first flaw. Dawkins’ scale is useless when you don’t know how something thinks, because you don’t know what they are actually rating, and it is a clumsy redundancy when you do know what someone thinks.