Mobile Science

In Part 1 of this extended argument (quite a while ago, I know) I looked at the claims made about logic in the Transcendental Argument for God (TAG) and why they don’t hold up to scrutiny.  If you haven’t read that post, don’t bother reading this one yet.  Click the link, have a quiet sit and think, and then come on back – I’ll wait.

The next major theme I’d like to deal with is the TAG claim that only the christian god provides a rational basis for the use of induction.  The claim rests on the supposed Problem of Induction and ‘solves’ this supposed problem by a naked assertion that the only source of uniformity in nature is the uniform nature of god.  Of course, no proof is actually offered for the uniform nature of god except a reliance on ‘scriptural authority’, but you can’t have everything…

TAG proponents argue that inductive reasoning assumes the uniformity of nature – that is, we assume that how the world worked yesterday is pretty much how it will work today.  They assert that this assumption is invalid – unless we also assume that there is something (god) that can account for/provide for the uniformity of nature.  They claim that the uniformity of nature is, and can only be, accounted for by God’s uniform nature.

You will have noticed that there is nothing here but naked assertions.

Before picking apart the many fallacies in this view, let’s first take a look at what the problem of induction actually is.

An inductive argument is one where the conclusion is said to be supported by the premises, but not necessitated by the premises (as would be the case in a deductive argument).  There is an example of each in Part 1 of this argument, but I’ll give another:

Premise: the sun has risen every day of my life to this point.

Conclusion: the sun will rise tomorrow.

The problem here is that the premise is a collection of previous experiences, none of which necessarily has any bearing on the future.  Perhaps tomorrow will be the first time that the sun does not rise.  You cannot know for sure based only on past experience – there is no absolute logical basis for a ‘therefore’ statement between the premise and the conclusion (of course, the lack of a real therefore is actually the difference between induction and deduction – a justified therefore shouldn’t actually be expected since inductive arguments are probabilistic.  More on that later).  David Hume pointed this out in the mid-18^th century in his book Inquiry Concerning Human Understanding.

Notice that the assumption that nature is uniform does not actually fix this problem.  The statement that ‘nature is uniform’ does not presuppose that the sun will rise every day – what if part of the nature of suns is such that once every thousand years they don’t rise?  For that matter, what of the much more supportable (and, incidentally, true…) notion that suns ‘die’?  Will it rise then?  To say that the uniformity of nature would justify induction would only be true if we knew everything there was to know about how nature operates.

The uniformity of nature is a necessary but insufficient condition for the use of induction.  In other words, even if the uniformity of nature were a proven fact, it still wouldn’t, by itself, provide justification for the use of induction – a key point that TAG proponents constantly miss.  This is unfortunate because it blows their entire argument out of the water.  Stating (by fiat) that the nature of god is such that it provides for the uniformity of nature does not actually fix the problem that it was intended to solve.

We can also ask, “On what grounds does the TAG proponent ‘know’ that the nature of god is uniform?”  The answer is rather funny since it turns out to be an inductive argument itself.  A christian ‘knows’ that god upholds the uniformity of nature because god said so and promised to continue.  Let’s leave aside the obvious unsupported assertion (that god said so) – on what grounds do christians trust god’s statements and promises?  Because god has never lied in the past.  Leaving aside yet another unsupported faith statement, we can now see that TAG proponents are relying on past events to support their claim to knowledge of future ones – an inductive argument.  Using inductive arguments to justify the validity of inductive arguments?  Then you’ve just failed Philosophy 101.

Assuming the uniformity of god (UoG) to avoid the need for a naked assertion of the uniformity of nature (UoN) just pushes the problem back a step – the uniformity of god is still a naked assertion.  The assumption of the UoN and the assumption of the UoG both do the same amount of work when it comes to questions of certainty of knowledge (otherwise known as epistemology – how we know what we know).  If we are faced with a choice between assuming the UoN to provide the necessary but not sufficient conditions for the use of induction, and using the assumption of the UoG to provide the same – which is the more reasonable choice?  Occam’s ever-sharp razor makes the choice here – which choice is the more parsimonious?  Which choice contains the least number of extra assumptions?

I don’t really have to answer that, do I?

The TAG proponent asserts that atheism has a problem, states that only the christian notion of god can overcome that problem, and then quietly fails to recognise that they have precisely the same problem that they were sure was such a death blow to atheism.

Well done, guys.  Well done.

After I’ve finished this series of three posts (the next one will look at the TAG take on morality) I’d like to come back to induction and look at the question of how much of a problem it really is.  The assumption that induction presents an insoluble problem seems to rely somewhat on a strawman version of what type of conclusions it’s supposed to be able to provide, and completely ignores the role of probability – but more on that later.  Before I finish up this post I’d like to just point out a few things.

First, many christian TAG proponents assume that Hume (along with Bertrand Russel) was the last word on the problem of induction – he wasn’t.  There have been many at-least-partial solutions proposed to it over the years since, and many justifications of the use of induction.  Hume himself proposed a reasonable justification for the use of induction (if not a technical logical justification) in the same book in which he raised the problem.  Hume proposed that induction was a habit born of necessity on the purely pragmatic grounds that we’d starve without assuming the uniformity of nature.

Or, as I like to put it: Science – it works, bitches!

Second, it must be noted that the assumption of the uniformity of nature is not an irrational assumption.  Not everything that cannot be strictly logically justified is irrational.  There is nothing about the assumption that is inherently contradictory or absurd – though, given the way TAG proponents speak about a universe without a foundational god, you might have a hard time gleaning that from their ‘arguments’.  Just because we can’t say for certain that the sun will rise tomorrow does not justify you saying that it won’t – all we can say is that we don’t know for certain.  But, again, there is nothing necessarily inherently irrational about things that are not justified or provable.  Just try proving logically that your name is really your name – can’t be done, but that doesn’t mean you shouldn’t continue to use your name.

Third, some TAG proponents seem to use a similar tactic to creationists – pick holes in an opponent’s position and declare yourself the default winner.  By pointing out the technical problem of the lack of justification for induction as used by atheists (and everyone else), and glossing over the many shortcomings and holes in their own attempt to use the notion of god as a justification, they attempt to declare themselves the only alternative position by fiat without ever actually illuminating precisely what that position is, examining it critically or seeking other possibilities.  Not that I actually expect them to behave with intellectual integrity, but it would be nice to be surprised once in a while…