Wednesday, 18 September 2013

How to 'falsify' Islam

Disclaimer: This post may seem to endorse re-interpretation of scripture, but this does not mean it endorses non-orthodox theology or interpretation. I keep the nature of "acceptable" re-interpretation open for discussion by theologians and academics. So re-interpretation here should be understood only as a loose principle and not relating to any particular (especially modernistic) school of thought.

How effective are so-called 'contradictions' at disproving Islam?

A more or less common trend especially prevalent among the internet non-theistic intelligentsia is this: refuting a religion is easy, just one or two supposed internal/external contradictions in the literal/apparent reading of the scripture spells its death sentence.

Not to mention that even given the truth of this claim, many of the attempts to establish contradiction claims are wrong headed (see this series of articles for example, where the author takes apart a series of claims about so-called internal contradictions in the Qur'an), the claim itself, I think, is overly simplistic. If an apparent/literal reading of scripture appears contrary to fact, can't the believer adopt a phenomenological/metaphorical interpretation instead?

The critic at this point may say this is ad hoc, and the only reason to attempt such "alternative interpretations" is for face-saving purposes.

Now I don't think this is true at all. To understand my point of view, consider what it takes to disprove a scientific theory. The following excerpt is taken from the introduction section of "The Soul Hypothesis: Investigations into the Existence of the Soul" (available online) by Stewart Goetz and Mark Baker:

How does one normally detect that the leading idea of a theory is (probably) false in scientific and philosophical research? Not just by showing that one preliminary version is false, but rather by showing that every version that has been seriously proposed is false. It can happen that an idea is proposed, then refuted, then rescued by adding auxiliary hypotheses, then refuted again, then rescued by adding still more auxiliary assumptions on top of the first set, only to be refuted once more. Once a hypothesis has been discredited through an extended process of this sort, it could be reasonable to give up on it, even if a few zealots still cling to it.
This paragraph makes a crucial point. When we find that one version of a theory (relevant to our discussion, one interpretation of scripture) is wrong, this shouldn't warrant disbelief in that theory (or scripture). Rather, auxiliary hypotheses should be introduced to try and "save the phenomenon". Why is this warranted and not ad hoc? Because the theory in question has a lot of other explanatory virtues, which would be lost were we to abandon it. Only when the evidence against the theory outweighs the evidence against it, or when the explanatory virtues of another theory becomes more than those of the previous one, do we consider abandoning it. Consider the shift from the geocentric to heliocentric model of the solar system: we only abandoned geocentricism when the calculations required to maintain it became too hairy and complex. Heliocentricity, on the other hand, provided a relatively simple and workable hypothesis. In other words, the explanatory virtues of the latter hypothesis became more than the former. This provided ground enough for geocentricism's rejection.

When this principle is applied to the case of Islam, the "quick refutation" theory of the anti-Islamist seems to evaporate. A (genuine) contradiction in the literal reading of the interpretation gives us a choice between the following two options:

a) Declare the Islam hypothesis to be false
b) Re-interpret Islamic scriptures to fit the data.

The reason (b) is not ad hoc, I think, is because Islam as a hypothesis has certain explanatory virtues no other worldview-hypotheses can explain, e.g. the conjunction of the evidences for existence of consciousness and God with the evidences specific to Islam, Islamic religious experiences, the multiplicity of world religions (to an extent), and so on.

Here's a different way to look at it. Scripture can be seen as analogous to nature, and exegesis can be seen as analogous to science. When a scientific theory gets it wrong, it's no reason to say nature is somehow incoherent (if that even makes sense), rather we read nature wrong to begin with. Similarly, if scripture seems to be in genuine conflict with established facts, it's perfectly plausible to assume we read the scripture wrong to begin with, rather than saying the scripture is wrong.

The take-away point of the discussion thus far is this: it's insensible to be hasty in our judgment in religions. "Inconsistencies" in religion should always be treated in the broader context of the religion's other explanatory qualities, and a decision should be made with this broader context in mind. As such, there is no quick-and-simple falsification test for Islam as the critics would have it.

What's the falsification test for Islam?

That brings us to another question: what would constitute a defeater for Islam? In other words, what would falsify Islam?

It does seem like an explanatory virtue for Islam to be falsifiable- as Karl Popper said, the key virtue of a hypothesis is that it makes risky and novel predictions successfully. The Qur'an itself opens itself up to the possibility of falsification numerous times- by saying it can never be surpassed (in terms of literary superiority, among other things), the Qur'an could not have been produced by the Prophet (saws) and answering relevant objections, claiming that there is no "inconsistencies" in it, and so on. So how could a falsification test for Islam be developed?

To get the proverbial ball rolling, consider the following verse in Surah Nisa:

Do they not consider the Qur'an (with care)? Had it been from other than Allah, they would surely have found therein much discrepancy. [Qur'an 4:82]

The popular interpretation of the verse is that it is referring to external contradictions i.e. the Qur'anic information clashing with established facts. To my mind, this interpretation of the verse doesn't present a very impressive falsification test. First, the Qur'an, unlike some parts of the Bible, rarely narrates events in a systematic, chronological manner. The key feature characterizing almost all Qur'anic discourse is allusion- it alludes to scientific facts and past narratives without fully expounding them. So there is precious little "information" that is to be gleaned from the Qur'an to see if it matches with reality. Second, the Qur'anic literary style (it's openness of language, among other things) often allows a multiplicity of interpretations. Some verses of the Qur'an depicting descriptions of natural phenomena may be understood to be literal as well as phenomenological, and in their case the falsification in terms of possibility of contradicting external facts doesn't seem at all interesting. So I think the interpretation of 4:82 is less restrictive than popularly made out to be in order for the challenge to be meaningful (it may include literary inconsistencies, for example, as Laurence Brown suggested in his God'ed). While I don't think the lack of external contradictions in and of itself provides a good enough falsifiability test, maybe a degree of falsifiability can be gathered from Qur'an's agreement with established facts about reality as well. Cases in point here are the putative arguments from Qur'anic prescience about science, history, or future events in general.

But that's not the main (or most productive) route to falsify Islam. I think the main falsification test for Islam, or any religion for that matter, is whether it has evidence or not. The fact that a religion doesn't have any evidence is enough to suggest, to my mind, that it fails the falsifiability test. Consider the aether or the geocentricism hypotheses for example: none of them were really disproved. I like the point this link makes:

Did these experiments "disprove the ether"? I wouldn't put it that way. None of these, or any other experiment could disprove the general idea of an all-pervading ether. The point is that there is simply no experimental evidence for the ether, and no need for it in any of our physical laws or theories.

This seems about right to me, the evidenceless-ness of a hypothesis seems to imply superfluousness, and superfluousness implies lack of justification*. Hence, Islam is falsified if and only if there is no evidence for it, and it passes the falsification test provided there is evidence for it. More precisely, Islam passes the falsification test if and only if the evidence for it outweighs the evidence against it. In other words still, Islam passes the falsification test if and only if it is more probable to believe in Islam than not.

If you agree with this definition of Islam's falsifiability, it seems to merit more discussion than we may have thought. After all, whether a given hypothesis has any evidence or not, or what counts as evidence and what does not- are deep and discussion-worthy questions. Included in this discussion would be things like the Qur'an's literary inimitability, the Prophet's sincerity, the Qur'anic prescience, explanatory superiority of the Islamic hypothesis over other ones, alleged arguments against Islam, and so on. These can be read in terms of what Popper called "novel predictions", in ways like the following: "if Islam is true, we should expect to find evidences X, Y and Z for it".

I think this analysis of Islam's falsifiability is more productive and avoids unnecessary confusion.

-----------------------------

*Notice I said "irrationality" and not "falsehood". The falsification test we are talking about is a test of a hypothesis' justification, not it's falsehood. For all we know, aether may as well exist and dinosaur bones may as well have been buried by demons. Whether these beliefs are true isn't the interesting question. What is more interesting is what contemporary philosopher Alvin Plantinga calls the de jure question: is it justified to hold such beliefs? Likewise, an interesting falsification principle would try to prove or disprove the justification of a belief, not its truth per se.

1 comment: