People love to explain things. Ask a gambler about luck, or google "conspiracy theories" and you'll be buried in explanations -- and some may be true! So, how do we tell a good explanation from a bad one? One way, maybe even the only way, is to test them.
An explanation is testable when it can be disproven. The only way to disprove an explanation is to use empirical evidence, or to show the explanation to be logically incoherent. The scientific method disregards explanations that can't be tested like this.
Why would explanations that are not empirically falisfiable, or which are illogical, be bad?
This is a good question. The short answer is that such explanations are bad because there's no way to tell them apart from their contradictions. Another way of putting it is that very few things can exist within the bounds of scientific rigor, while literally everything can exist outside those bounds.
For example, One can assert that the Christian God revealed Himself as the one true God, or that the totality of nirvana revealed itself as specifically not-God. Since neither assertion can be disproven, there's no reason to recognize one over the other.
This is an example of a specific kind of argument made outside the bounds of scientific rigor: the citing of personal, inviolable experience, which cannot be tested by others. This form of argument is known as a naked assertion, which is when an argument is made by asserting the explanation without empirical evidence, or with incoherent logic. Faith-based explanations fall into this category. It should not be surprising to note that literally anything can be nakedly asserted, true things as well as false things. Contrast this type of argument to a claim that falling objects fall at the same rate regardless of their weight. This can be tested by dropping two differently weighted bowling balls from the same height.
Anything can exist outside of empirical evidence and logic, including contradictions. If we only treat falsifiable explanations as good ones, then we'll always be able to tell whether an explanations is better than another. We just test them until one gets disproven, and then we discard it or modify it.
It is only important if one recognizes that there is a reality. If there is a reality, it might be useful to distinguish between real and unreal. Faith-based explanations fall outside the bounds of scientific rigor because they cannot be disproven. However, science is sometimes accused of being as faith-based as religion. This is known as universal skepticism. This argument takes the form of asserting that the acceptance of reason -- which requires a recognition that there is a coherent reality which can be known and understood -- is an act of faith. Faith in reason, however, is a contradiction in terms because reason as used in the scientific method is founded upon the principle of falsifiability. Reason is the act of recognizing empirical evidence and processing it logically. On the other hand, faith as used in the universal skepticism argument is founded on the principle of belief due to feelings and emotion, not evidence. Faith in this case is the acceptance of explanations just because of preference or prejudice, without empirical evidence or a demonstration of logical coherency. Arguing that reason requires faith is an iconic example of doublethink. It is the same as arguing that an objective reality doesn't exist -- typically the reality being offered instead is a subjective one, subjective to God's will. The denial of an objective reality external to oneself, while unreasonable, is nevertheless a powerful rhetorical tool.
This view is related to solipsism, which holds that you can know nothing outside of your own existence. Without recourse to the scientific method and a reliance on explanations that can be disproven, the solipsistic view is quite correct. If it is accepted that any explanations is possible, then it's not possible to know what is true or not.
Without the following observations it is not possible to make an argument for or against anything:
When one steps into an elevator, one does not have faith that one will not plummet to a grisly death. Instead, one has a reasonable understanding of not only how elevators do work, but how they have worked in the past. It requires no faith to ride an elevator free of the fear of imminent death. In the same way, it requires no faith to recognize reality as empirically demonstrated.
One of the great differences between reason and religious faith is the necessity of doubt and humility. Science must doubt everything, including its own theories and observations, and proceed with the utmost humility.
We reason that there is an external world that exists independently of our minds and that we can understand it. Reason is a means to gain knowledge of reality by observing reality. Faith is not a means to gain knowledge, it is an assertion that belief can be accepted without observing reality, or even in spite of reality. It is a rejection of an objective reality. Faith cannot support reason, because faith is the end of reason. It is the thought which stops thought.
Science questions what we know all in the hope that it will result in a better understanding. If this better understanding is substantially different than what we previously believed, so much the better.
If one accepts that an objective reality exists, then it follows that it is possible to demonstrate an argument as "more true" or "less true." How can one do this? By making observations, which is what empirical evidence is, and by trying to understand what the 'coherency' of reality is, which is what logical thinking is. So, by appealing to logic and empirical evidence one can discern between good and bad arguments. This is known as critical thinking.
But why use critical thinking for everything? Couldn't some things be beyond empirical evidence and logic?
As noted above, everything is beyond empirical evidence and logic. Or rather, anything at all can be demonstrated to exist if one ignores critical thinking. Citing something as beyond empirical evidence or logic is identical to declaring that sometimes two and two make five, or that reality is contingent on something else (God, typically).
If one believes that arguments can be demonstrated to be "more true" by appealing to illogical arguments lacking empirical evidence, then one cannot discern between good and bad arguments except by arbitrary rejection of one side's personal experience. This thinking can be dangerous.
If one accepts that good arguments can be made without logic or empirical evidence, then it follows that one may accept such an argument. Thus it can be concluded that one may believe literally anything.
Or as Thomas Jefferson puts it:
Man once surrendering his reason, has no remaining guard against absurdities the most monstrous, and like a ship without rudder, is the sport of every wind.
Let me explain. If the empirical evidence and logic do not support the argument, all the professor of the argument must do is say, "this is one of those arguments that exists outside the bounds of scientific rigor." If one accepts that such arguments do exist, then one can only reject the professor's argument through naked assertion, which is itself illogical and indistinguishable from the professor's own naked assertion.
To claim one has a good argument that violates critical thinking is the same as claiming all absurdities are equally good arguments.
An individual who believes good arguments can be illogical could believe that suicide bombing will result in 72 virgins in heaven.
An individual who insists on having arguments that are logically coherent and empirically verifiable could never believe such a thing.
The suicide bombing argument is bad and illogical. But not all illogical arguments are bad. Many illogical arguments may in fact be true.
Likewise, logical arguments can be bad. Usually this is because one of the observations is incomplete or wrong. Of course, 'wrong' only has meaning if we accept there is an objective reality external to ourselves, otherwise we could claim the 'wrongness' is just a divine exception. A miracle to be taken on faith. Suffice it to say, as long as one recognizes the constraints of empirical evidence and logic, it will be possible to distinguish a better argument from a worse one.
If one rejects the necessity of empirically verifiable evidence and logic, it will never be possible to distinguish a good argument from a bad one without appealing to an arbitrary decision no less valid than one's opponent's. In fact, no discussion will be meaningful, since it has been accepted that reality is not objective, but rather contingent on some higher power's will.
Some reference links:
"Believe in the Scientific Method?" is the companion to another piece of mine, "Believe in God?"
Perhaps you are unconvinced that accepting one thing on faith necessarily results in a rejection of objective reality. Or maybe you just want to see that thought expanded upon. Check out my essay on Rhetoric: The Power of Persuasion for a demonstration. Be warned: what you learn from reading this article will make you and your arguments more powerful. Use that power for good, please!
My definition of the faith of the scientific method was inspired by this very long (and well-written) essay: Why Atheism