This is actually something that I’ve put a great deal of thought into in recent months, it’s not completely coherent at this point, but I’ll give it a shot.
In short, I am a theist, but I don’t see the morality passed down by God as being anything more than sort of a set of basic guidelines that, generally, will lead us toward the goal. Of course, different people have different ideas of what that ultimate goal is but, given that goal, and the level of understanding/enlightenment of man at a particular time, that one could ultimately derive a set of morals remarkably similar.
I think this idea is best illustrated with an analogy. The game of chess is, given a sufficiently powerful computer, ultimately computable. As such, given a sufficiently powerful computer one could, at any given state, determine the ideal move toward a particular goal (usually winning). As humans, we’re not able to maintain such a massive statespace in our heads, and even the greatest masters can only look so many moves ahead. As such, we cannot rely on a definite result for a given state and, instead, have to look ahead at expected outcomes and generate a set of guidelines that we can generalize to some set of states and then know what are likely a good set of moves as a result. As a player becomes more skilled and experience, he learns a larger set of rules that have a finer resolution over the set of states.
To my knowledge, as best as we can tell, the universe is quantized, and thus is also computable. As such, in this analogy, the nature of God would be equivalent to such a powerful computer that is able to determine the ideal move in a given state toward a particular goal. Thus, the morals passed down by God are much like the set of guidelines in the game of chess, where God is providing a set of morals at a level roughly equal to how progressed humanity is. And the interesting thing is, much like how the guidelines for chess were slowly built up over time, even in the absence of God, given the same goal, humanity is able to test out various guidelines at our level of enlightenment and determine how effectively they further that goal.
The problem is, each approach has some drawbacks. From a top down approach, the guidelines are only approximations based on our ability to understand, so even if they are divinely provided, they can lead to sub-optimal decisions. The bigger problem, though, is that they have to be re-evaluated constantly, which is something that many do not do. As an example, if we take the Bible as such a revelation, we have to realize that those guidelines are based upon the culture and progress of those people, thousands of years ago. As such, that they show issues today is not necessarily a sign that they were bad then, but simply that we have changed, and that with our higher degree of understanding and development, we can uncover how they apply today or even how they are being revealed to us today, and unlikely to be in such a manner.
With a bottom up approach, you have the exact opposite problem. Things seem to be furthering a goal, but we could go quite far down a particular decision branch before we realize that the whole thing is a dead end. This can easily result in a lot of spiralling immorality (at least as it appears to us now) that could be very difficult to recover from. Worse, unlike the other approach, we can never really be that sure how far back we screwed up.
Or to summarize with another analogy. Imagine morality is some arbitarily large dimensional curve that we’re trying to model with an n-polynomial curve such that n is our level of progress and enlightenment. The top-down approach would be if someone took the exact curve as it exists, could take an arbitrarily large set of samples, and did an n-polynomial fit to those data. When we see that polynomial, we know it’s the best that is achievable for that degree, and it’s probably pretty close over most of the curve, and it’s better than anything we could achieve for that degree, but it’s still not exactly on. As we’re able to understand higher and higher degrees and are better able to measure the distances between the model and the actual, the error may become more apparent we compared to other methods for fitting the data we have. With the bottom-up approach, we have a methodology for fitting the nth degree polynomial, but we’re only able to get so many samples and we really have no idea how representative various samples are, only how much error they produce with the fitted polynomial. As such, when we get a larger set of data or an ability to fit a higher degree, we can become more aware of a better model.
As such, I really see the best approach, to mean the fastest convergence, as a sort of hybrid of these two, where we can have definite points of reference for good approximations, and we use that context and our growth since as a basis for working up to where we are. And this, I think, is where a lot of theists go astray, an unwillingness to deviate from a divinely provided set of rules, and where many atheists fail too, sometimes venturing wildly down the wrong path OR backing up much too far, either due to oddities in the available data and a lack of a definitive point of reference.
Now, of course, this all begs the question of what that “goal” is, and while I believe it is somewhere around freedom and growth for reasons that are probably out of the scope of this discussion, I’m unsure exactly how much it matters because, in my own thought experiments, there’s very little difference in the resultant set of morals when applying a multitude of different “goals”. I think this is due, though, to the relative infancy of humanity such that pretty much any general movement in the direction of growth will further these multitudes to some degree.
So, in short, even as a theist who believes that God can, and likely has, provided us with a set of divinely inspired moral rules, to say that it “necessitates” him for justification isn’t really a fair assessment. You see, we WILL be trying to refine these rules ourselves regardless of whether or not he exists and whether or not he provides a set of rules. Even if our goals don’t exactly match, as it seems ours over time has generally be in a direction of growth, we’re headed in the same direction. The fundamental difference, as I see it, is just that God is able to provide us that model that is roughly at our degree of understanding which, if we make good use of it, can help us converge, and thus grow, faster than we could without it.