Laws, damned laws, and statistics

One key part of the argument John Wesley brings for there being particular providence (see previous post), as against only general providence, is that the latter necessarily consists of the sum of the former:

You say, “You allow a general providence, but deny a particular one.” And what is a general, of whatever kind it be, that includes no particulars? Is not every general necessarily made up of its several particulars?

As an Arminian, I suppose he would have tended to exempt human choice from providence, and placed it under the kind of Molinist foreknowledge adopted by Arminius, and I would say that, if so, this is an inconsistency. After all, how many of the answered prayers he cites as evidence for special providence depend directly on free human decisions? Paul would not have survived to reach Rome (fulfilling many prophecies) had the Roman commander in Jerusalem not believed, and freely acted on, the report of an assassination plot against him.*

Indeed even the “singular events” such as “the rise and fall of empires”, which are granted by the “general providentialists” Wesley opposes, are principally dependant on human decisions (for example, a single tactical decision of Admiral Jellicoe at the Battle of Jutland, according to historians, may have determined the outcome of the First World War and subsequent Eurpean history). Does providence not include that?

But as Wesley rightly perceives, the denial of special providence in his day, as in ours, stems from the assumptions of then-contemporary science, and Wesley says a fair bit about its application to the natural world – our main interest here – before he gets on to the matter of humanity. And I find that his argument is somewhat prescient, and instructive, of the way that science developed after his time, when James Clerk Maxwell applied statistics to scientific laws.

It is still regularly claimed that God governs the world by natural laws alone – or that exceptions to them are rare, possibly even limited to the resurrection (for some Evangelicals – and not even to that for more liberal Christians). This conclusion then leads to seeing God as sitting aloof to the details of evolution, maybe intending its overall direction but not specifically the existence of the bullfinch or the Anopheles mosquito. And as for governing the welfare of the individual finch or mosquito, that is as far beneath his dignity as Wesley considers the poet Pope to believe.

Wesley argues that the whole is necessarily the sum of the parts,and that general providence cannot exist without particular providence. Statistical laws both illustrate this, and demonstrate its applicability to the natural world. Let me start with a fairly clear example from human life. If I Google “average UK income”, I’m offered a Daily Mirror headline that says “UK average salary is £26,500 but figures reveal huge pay gap”. Now, let’s disregard the loose journalistic use of terms like “average”, but this is a case where one might well find that relatively few people actually earn that “average” figure – there are very many poor people with less, and a few fat cats with such huge incomes that it skews the results. Those differences are, clearly, far more important in God’s scheme of things than the average.

One may argue about whether God only permits those disparities, or governs them (but before you answer, consider how you’d answer your struggling church member who rejoices, “God answered my prayer and I got the promotion!”), but clearly it would be foolish to conceive that God simply makes a general decree that the average salary in the UK will be £26,500, and then lets the market sort the details. Instead, the economic metric is an abstraction of all the complexities of millions of human decisions, political trends, physical realities and so on. At the same time, the general level of observation obviously has significance for both God and man, because there are countries where the average income is vastly lower – and those countries are exporting individual people in vast numbers, affecting the rise and fall of empires, partly because of that statistical fact.

The same is no less true in the world of nature. I always turn to Boyle’s Law as an intuitively obvious example of a statistical law. The destiny of any individual molecule of gas in a compression experiment is completely unknowable, and very complex. But, as we all learned at school, the collective behaviour of the gas in terms of pressure and volume is, within limits, simple and mathematical. This, we conclude, is “one of God’s scientific laws”. So do we really believe that God’s priority in the creation of molecules (when in gases) was the average of their behaviour? Did he not, rather, decide how individual molecules would behave? Or rather, did he not in one intellectus ordain both their individual properties, with the specific outcomes that would result, and their average behaviour, that would determine the way that the weather works, stars form and so on?

A commenter on BioLogos recently reasoned that if God made laws, it was rational for him to abide by them himself. I think that meant not “breaking” natural laws by miracles and so on, though taken at face value it would mean God himself has to exert greater pressure in inverse proportion to his volume. Taken either way it would also imply that any rules that you write into a computer program, you’re rationally and morally bound to follow yourself, which is absurd. But the primary fault of the argument is the old (and still widely ignored) truth that God has not revealed any hint that he has made the creation obey rigid laws: “law” is just one, johnny-come-lately way of abstracting the regularities observed in the world to human comprehension.

I’ve mentioned before (with reference to “chance“) how there is a distinctive statistical pattern of letter distribution which differs from language to language. It’s easy thereby to distinguish between languages, or even the same language at different historical periods. Given large enough samples, one might even distinguish between the styles of individual writers in the same language (as has been done, with far too small a sample size to be significant, to deny the authorship of some Pauline epistles).

Across the entire literature of a language, that distribution would be statistically predictable enough to constitute a “law”. Yet not only would it be entirely false to suggest that anybody, at any time, had prescribed that “law”, but it would be looking at entirely the wrong evidence for human intention, which lies in a pretty dilute way in the evolution of language, but in its true form in the meaning of the literature, of each individual document that has been written in the language. The “law” is a complete epiphenomenon, though indirectly a marker of purposive activity.

This last point has a bearing on the “weak” anthropic principle, of which Wikipedia says:

The anthropic principle states that [cosmic fine tuning] is a necessity, because if life were impossible, no living entity would be there to observe it, and thus would not be known. That is, it must be possible to observe some universe, and hence, the laws and constants of any such universe must accommodate that possibility.

And that sounds perfectly compatible with an Epicurean recourse to chance until you replace the word “laws” with “language”, a move which is not only justificable from the paragraphs above, but from the original conception of “scientific law” as “God’s command” – a conception that has never been satisfactorily replaced in secular science. It is really like saying that the Universe only seems intelligible to Englishmen because it speaks English rather than Gobbledegook. Which is as much an explanation as saying it only seems uniquely designed because of its unique design.

So, if nature is one of the two books written by God, it is not inappropriate to view scientific laws as merely a statistical analysis of the language spoken by the Logos, or at least the language he uses when he is dealing with this Creation. the “laws” may be interesting or useful in a limited way, but in terms of understanding divine government, one would gain far more by hearing what he says in that language – and that constitutes special providence – the “particle” in Wesley’s sermon.

* Molinism, in effect, places special providence over human choices at the point of creation, by having God create only those people who will make choices consistent with his will, and the circumstances that will accord with them. Molinism has a number of other problems too, including this: God knows by infallible “middle knowledge” what any specific person he might create would freely do in any potential circumstance. Simultaneously he also knows by infallible actual knowledge which people and circumstances he will actually create, so what is the point of all the counterfactual possibilities?