Tuesday, November 27, 2012

Presumptions and the first law of general ignorance

As I sit here typing on my computer, it’s easy to get overwhelmed by the wealth of knowledge and benefits science has brought us.  In fact, it’s so easy that we have to a great extent forgotten the appalling depths of our ignorance.

How can I say that?  Very simple: the quantity of unknowns in the universe is by definition unknowable.  This gives us Layne’s First Law of General Ignorance: We don’t know how much we don’t know.  If we know, then, that the extent of our ignorance is unknowable, we know that at least one thing is unknowable.  But we don’t know if there’s anything else that Man cannot comprehend or will not be able to comprehend at some future date.  Therefore, the corollary to the First Law: We don’t know how much we can’t know.

Got a headache yet?

Most thought systems have to start with at least some assumptions that neither need nor admit of proof.  For instance, you can’t get anywhere in plane geometry if you don’t accept that “a line is the shortest distance between two points”, or in algebra if a2 = b2 + c2 is merely an opinion.  Likewise, reason has a fundamental assumption that “a thing cannot both be and not-be at the same time and in the same manner”.

Why can’t we take atheism as self-evidential?  This seems to be the answer the New Atheist prefers, given that philosophers since Socrates have known that it’s impossible to prove that something does not exist.  In law, we take it as a necessary presumption of justice that a person accused of a crime is innocent until proven guilty.  Are there not situations in which we can safely presume that a theory is false until proven true?

Theoretically, at least, such a presumption is an appeal to ignorance (argumentum ad ignorantiam), which is a material fallacy in informal logic, and which takes two forms:

There is no evidence for p.
Therefore, p is false.
There is no evidence against p.
Therefore, p is true.

But some logicians maintain that there are cases in which presumption of one answer in absence of proof of its opposite is a reasonable practice that doesn’t reach a diagnosis of logical fallacy.  Here it would pay to see if atheism meets the test for such an exemption.

The presumption of innocence is one such case, where to prevent the incarceration or execution of the innocent the prosecution must meet the standard of proving guilt beyond reasonable doubt.  In civil law, however, where no such overriding moral concern obtains, judges and juries are free to decide based on the preponderance of the evidence.  In similar fashion, no mortal danger makes it morally necessary to assume atheism without proof; if anything, the least potentially perilous of all courses is to assume Christianity right and live it authentically, which was the point of Pascal’s Wager.

Likewise, we often lay a burden of proof on a person making a new or improbable claim.  Certainly, though, the claim that at least one god exists can hardly be called new.  And improbability is only an appropriate ground when probability obtains.  If we recall the problem of contingency, the existence of a Necessary Being upon which the conditional existence of all beings depend is by definition a zero-sum equation: He is or He is not; He cannot “probably” exist or not exist, save as a figure of speech.  We can say we doubt the existence of God; but then it’s only reasonable to lay a burden of proof on the theist if it’s reasonable to doubt God’s existence, which is not an accepted, objective fact.

Another proposed case is extensive investigation has been undertaken.  This qualification atheism fails miserably.  As I pointed out in the previous post, how can you test for the presence or absence of artifacts left by an immaterial Being when you have no substantive theory of how the immaterial would manifest itself in the material world?  The basic theoretical legwork hasn’t begun in earnest yet, let alone been successful.

In essence, the proposed exemption would obtain if we’d performed, say, 47 tests of a specific hypothesis and successfully predicted the outcome each time.  However, if the outcome is always “no”, then how do we know the test is appropriate?  How do we know we’re not using a metal detector to look for unicorns in a sock drawer?  We would only know if the outcome were “yes” on a being with similar properties … angels, perhaps.  Otherwise, your test may as well be an ammeter without a needle, the measurements in megahertz rather than amperes.

Despite the centuries-old problem of induction, scientists have used and continue to use inductive reasoning (with remarkable success) because 1) to wait for a true and perfect “closed system of knowledge” to declare a hypothesis proven would be an intolerable and impractical burden to meet, not mention that it would require a God-like omniscience that would make scientific investigation moot; and 2) such failures as occur after X successful iterations of a test create opportunities to refine theories. 

But moreover, inductive reasoning works for the physical scientist because he studies objects that have no real freedom of action; such randomness as they display comes from their interactions with other objects or substances that follow the dictates of cause-and-effect.  She can predict the intersection of a comet’s path with the orbit of Jupiter, for instance, because the comet can’t decide it needs a change of scene and go off to make a busman’s holiday of orbiting Betelgeuse for a while.  Humans, by contrast, are less bound by cause-and-effect relationships (if we were, we couldn’t be said to “reason” in any meaningful sense); we don’t know whether the immaterial plane — if it exists — is bound by cause-and-effect at all.[*]

Which brings me to the final possible exemption: 

We sometimes have meta-knowledge — that is, knowledge about knowledge — which can justify inferring a conclusion based upon a lack of evidence. For instance, schedules — such as those for buses, trains, and airplanes — list times and locations of arrivals and departures. Such schedules usually do not attempt to list the times and locations when vehicles do not arrive or depart, since this would be highly inefficient. Instead, there is an implicit, understood assumption that such a schedule is complete, that all available vehicle departures and arrivals have been listed. Thus, we can reason using the following sort of enthymeme: 

There is no departure/arrival listed in the schedule for location L at time T.
Suppressed Premiss: All departures and arrivals are listed in the schedule.
Therefore, there is no departure/arrival for location L at time T.

This kind of completeness of information assumption is often called the “closed world assumption”. When it is reasonable to accept this assumption — as with plane or bus schedules — it is not a fallacy of appeal to ignorance to reason this way.[†]

Do we have such meta-knowledge that justifies a “closed world assumption”?  Rightfully speaking, we don’t, because we don’t have a truly objective standard of “enough”.  More to the point, atheism itself hasn’t satisfied the condition of being a verified hypothesis, and is not therefore a valid basis of inductive reasoning.  It’s certainly not a scientifically necessary presumption, as nothing we’ve learned or created through the scientific method will be imperiled or falsified if God does indeed exist.  Nor has anyone succeeded in refurbishing the closed-loop, origin-less universe (or rather multiverse) that needs no First Cause or Necessary Being to explain its existence.

Understand, at this point I’m not arguing for a presumption that God exists; neither theism nor atheism is strictly necessary for the scientific method to work.  What is needed is intellectual honesty.  Presumption is a triumph atheism has yet to earn.


[*] In Hindu and Buddhist cosmology, karma is the cause-and-effect relationship between natural action and supernatural consequence.