We've all heard of NOMA, Stephen Jay Gould idea that religion and science are not really incompatible, because according to him, science is about 'what' and 'how' regarding the natural world, while religion is about values, morals and purpose.

This is problematic on multiple levels. I'll outline four of them, before explaining why I think how we approach and treat religion is of great importance.

 1) While this may be a nice idea in theory, it is completely devoid of any relevance to how theists actually behave and employ their faith in real life. Religion, of all stripes, inherently trespass into the purview of science, even if it is not the explicit intent of the believer, the belief or claim itself.

 Claims about reality-- that is, claims regarding fact, objective truth, and history--necessarily rest on an understanding of reality and the specific object of the individual claim. Claims about reality are therefore inherently knowledge claims, and as such, demand evidence. Anyone making such claims imply--by virtue of making the claim--that they have good reason, some justified foundation, for the claim. If they do not have this evidence, then they are either just speculating without justification, or they are merely expressing a desire, and either way have no business making the claim about reality.

Religion is inherently about the nature of reality, and is built on claims and beliefs regarding fact, truth, and history, and as such, everything I outlined above applies to religion. Religious beliefs and claims, being about reality as they are, necessarily imply evidence, lest the beliefs and claims are without justification. Science and religion therefore inherently overlap here; they both make claims about reality, which is supposed to be the purview of scientific inquiry, and because of this, science is justified in responding to the claims and demanding the necessary evidence.

 2) What is perhaps the biggest flaw of NOMA is its central claim that morality and purpose are to be dealt with by religion, not science. One can initially claim that even if it is the case that morality and purpose are not within the purview of science, that does not mean that religion has any special authority on such issues, indeed, while religion can only speculate and/or declare by fiat, as to what is moral and what our purpose is, if morality and purpose have objectively true answers, surely it would be scientific inquiry that would deliver these answers, not religious declaration? If, however, such questions do not have objectively true answers; if morality is purely subjective, and we have no ultimate metaphysical purpose, then religion is of not more an authority than any other domain of inquiry. Either way, assigning morality and purpose to religion is erroneous and only provides religion with a credibility and authority which it does not deserve.

 In any case, that being said, I would agree with Gould that questions of morality do in fact belong in a specific domain of inquiry, but it isn't religion. A very good case can be made for why a scientific morality (a morality derived from scientific values and methods) is not only superior to a non-scientific morality, but is in fact the only legitimate, justified, moral system. This is a position argued by the historian and philosopher Richard Carrier in his book "Sense and Goodness without God." Roughly, it goes as follows:

 There is necessarily one thing that people want more than anything else, so, first you empirically determine what it is that people want more than anything else. Then you empirically determine which behaviours and lifestyles are more likely to statistically achieve that desire. For example, if people desire happiness more than anything else (as most would agree this is the case), then we can empirically determine which behaviours and lifestyles are statistically more likely result in happiness. These behaviour and lifestyles would by definition be moral in so far that they are the means of achieving what people desire most.

 Obviously I cannot do his position justice in a few lines, so here are two talks he gives on the topic:

 On natural morality:

http://www.youtube.com/watch?v=Dce8mE0q4zA

(Part 1 of 6)

 Why science is better than religion as always has been:

http://www.youtube.com/watch?v=PNMUz1iediU

(Part 1 of 14, although Carrier only speak in parts 1-6, after which it's PZ Myers)

 3) The philosopher Daniel Dennett as rightly noted that even if it is true that science and religion occupy different domains, where science does what it does, and religions does what it does, and neither overlap, it does not however exclude science from studying what religion does, and there would be no reason for why it shouldn't. It would not mean that science was actually doing what religion does. So even if we ignore the above two points and agree with Gould's distinction between science and religion, because science gets things right so often, why not use it to study religion as a natural phenomena, and to assess the objective claims it makes?

 4) If we strictly adhere to Gould's boundaries, we find that most of religion is in fact self-excluded (!) from Gould's domain of religion, indeed religion is inherently built on factual empirical claims, and yet according to Gould, matters of fact belong in the domain of scientific inquiry. I will give various examples:

 The virgin birth is inherently a scientific claim; it is a claim about biology and physics. This equally applies to the resurrection of Jesus, a central tenet of the Christian religion. Both of these are claims about matters of fact; they are historical claims about what factually happened, and as such belong in Gould's "magisterium" of science. The ideal of an afterlife also trespasses into science in that it contains the scientific claim that the self, our individual identity, is more than the brain and can therefore survive the decomposition of our brain. Other claims such as miracles, the nature of certain books, the historicity of apparent religious events, and even the supposed existence and activity of a creator within the natural world are all necessarily claims to fact; they are all claims which carry scientific import. Even if it is the case that we cannot or currently cannot scientifically test these religious claims, it is still the case that they are claims about matters of fact (which, by definition, have an answer); they are objective claims about the universe, and as such, they fall under the purview of scientific inquiry.

 5) Finally, on the wider issue on religion in society, which I think relates this topic. It must be said that the root of the problem (the problem being the prevalence of irrationality in society) of is of course not religion, but rather faith-based thinking (e.g. any belief not based on evidence) and dogma, which can present itself in various forms, including conspiracy theories, alternative medicine, denialism, cults, secular political ideology, and of course religion. As skeptics and critical thinkers surely we want to foster critical thinking, reason and scientific skepticism to the degree that it severely and negatively impacts dogmatic and faith-based thinking, however we will never achieve this if we maintain the current social double standard of holding religion as somehow distinct from all the other forms of nonsense, pseudoscience, and non-critical thinking that we do criticise. It is inconsistent to endeavour a society which respects critical thought, while still allowing, or even respecting, the most persistent form of non-critical thought to persist. This doesn't mean that we are to be critical without restraint, and/or critical at inappropriate times, we must of course show composure, however we must at least acknowledge that overall religion is detrimental to the goal of a society which appreciates scientific thought. To pretend that there is no conflict between the core values and methods of science and religion, when there clearly is, and to pretend that religion does not inherently trespass into the domain of science, when it clearly does, I feel, gives religion a standing which it does not deserve, which in turn makes it harder to criticise when it needs to be, and therefore makes it more prevalent, and above all else undermines the goal of fostering critical thinking.

 I also recognize that the a stronger critical approach may make the theist more defensive and therefore less open to discussion and reviewing their beliefs. This is clearly true, but I don't think this necessarily makes that approach the wrong way to go, since the goal should never be for the person to immediately renounce their beliefs, or even to necessarily to get them to renounce them at all, rather, it is to plant a seed, an idea, which they will someday examine on their own, and I don't think their initial defensive stance will affect that, indeed surely it is natural to initially be defensive when someone is in some way critical of our views, however later we can examine the issue with a clear head.

 Many have said that instead of emphasising why religion is wrong, we should instead emphasise why science is right, why it works, and why skepticism it the best approach, something which I can agree with, especially with regards to those who are 'on the fence'. Unfortunately however many religious people are already of the opinion that their beliefs are correct, and that science is incorrect/or that it supports their belief, so simply discussing why science is so good may not always be effective. Instead, I would contend that we should discuss both why science is right/accurate, and why religion is wrong/inaccurate. By first demonstrating a flaw or inaccuracy in their belief, then demonstrating the benefits of science, may in fact have a greater affect than had we just focused on the science.

 The harder vs. softer approach to religion is not mutually exclusive, however it is often portrayed as being so. How we criticise and deal with religion should depend on the individual person and on the context, but we should deal with it, in some way, in all cases.