Thursday, January 18, 2018

Reasoning About Causation

     In the course of composing my recent essay about Occam’s Razor, I got to thinking about important principles of causation, how they apply to human reasoning, and how seldom they’re addressed for general consumption. Mind you, this isn’t about whether causation is “real,” an ontological requirement for reasoning at all, but about how to think usefully about it.

     Hey, if you’re a regular patron of this joint, you know how I can get after a poor night’s sleep...or a good one.

     Many years ago I was powerfully struck by an insight I’d encountered in a relatively obscure science fiction novel:

     You can never do only one thing – Marc Stiegler

     This is an especially important insight in economic thought, which I resolved to explore in detail. It applies to many settings other than economics and political economy with equal force.

     If we reword Stiegler’s insight in the most general possible terms, here’s what comes out:

Every cause has more than one effect.

     This is the most compact possible wording of what’s commonly called the Law of Unintended Consequences.

     While it can be demonstrated in innumerable settings, that law cannot be proved. Likewise, an important corollary to that law:

At least one consequence of any action will be undesirable.

     ...cannot be proved, though it too has been demonstrated endless times. It is that corollary that mandates honest, judicious analysis and evaluation of both the benefits and the costs of every political proposal, both a priori and a posteriori. Politicians practice evading such evaluations, for a simple reason: whatever the costs might prove to be, the politician doesn’t intend to pay them himself.

     In his remarkable little book Economics in One Lesson, the late Henry Hazlitt gives an economic interpretation of the law:

     The art of economics consists in looking not merely at the immediate but also at the longer effects of any act or policy; it consists in tracing the consequences of that policy not merely for one group but for all groups.

     ...on the third page of the first section. The rest of the book consists of demonstrations of its soundness.

     But unintended consequences – their inevitability and the inevitable undesirability of at least one of them – are only half the story.

     The second critical principle of causative reasoning is harder to grasp – enough harder, in fact, that some supposedly intelligent persons have striven lifelong to deny it:

Every effect has more than one cause.

     Once again, this thesis cannot be proved, but the confirmations of it are beyond enumeration.

     If we look at a special case of this principle – recipes for various edible dishes – it becomes empirically clear. There’s no such thing as a dish with only one ingredient. There’s always at least one more: the labor of the preparer required to transform that ingredient into the differently named and recognizably distinct dish to be served. For example, “toast” is not merely bread, but bread that has been subjected to a deliberate process that absorbs time and energy.

     Part of the cost of producing anything lies in determining what ingredients must go into it. Part of the reasoning that results in deciding whether or not to make it lies in comparing the total cost of the ingredients, plus the labor required, to the expected benefit plus the costs to be incurred from the unintended consequences. Every toaster must eventually be cleaned.

     John Locke’s theory of property rights uses the principles above in a direct and illuminating fashion:

     He that is nourished by the acorns he picked up under an oak, or the apples he gathered from the trees in the wood, has certainly appropriated them to himself. Nobody can deny but the nourishment is his. I ask, then, when did they begin to be his? when he digested? or when he ate? or when he boiled? or when he brought them home? or when he picked them up? And it is plain, if the first gathering made them not his, nothing else could. That labour put a distinction between them and common. That added something to them more than Nature, the common mother of all, had done, and so they became his private right. [from Locke’s Two Treatises of Government]

     The unnamed subject of Locke’s example above creates food by applying his labor to available acorns. Two causes; one effect. He thus assuages his hunger, but he does more: he creates waste matter: the discarded shells of the acorns. He also creates waste heat and some degree of personal fatigue. One undertaking: several effects.

     And that’s the simplest example I could produce!

     A third principle of reasoning is as useful as the ones above, though it’s somewhat lighter in tone. I think of it as the Anti-Nitpicking Principle:

In determining whether to embark on an undertaking, some costs, whether they arise from causes or effects, may legitimately be neglected.

     We all have thresholds for neglect: how large a cost must be, or how large an undesirable consequence must be, to be significant in one’s decision making. I encountered a case this morning that will serve as an example.

     After morning Mass, I stopped at a nearby supermarket for a few items. Due to a recent edict of the legislature, a shopper who requests bags at checkout must pay $0.05 each for them – and I had forgotten to bring bags from home. I needed two bags; thus, my little shopping trip cost $0.10 more than the cost of the goods I purchased. The extra cost was insufficient to deter me from purchasing what I wanted.

     When I got home, the C.S.O., who has no client to visit today, upbraided me for paying for bags “when we have plenty already.” I was aware of that undesired consequence when I made my purchase, and decided to neglect it; I mastered the art of ignoring petty criticisms from accountants – male and female – long ago. A simple “yes, dear” sufficed to deal with it.

     Of course, to proceed in this fashion requires that one have accurate foresight about undesired consequences. It also requires an accurate estimate of one’s ability to deal with costs, whether deliberately embraced or arising from undesired consequences. Most politicians lack both characteristics, though it might be otherwise were they required to spend their own money for the boondoggles they advocate, and to remediate the undesired consequences personally. I propose it as an appropriate subject for experimentation. Extensive experimentation.


furball said...

Absolutely agree with your last paragraph!

Somewhere along the road to here, politicians started getting off completely scott-free from suffering consequences of their poor choices in law-making.

Of course, having a civil, informed discussion on - for example - the actual costs and effects of Obamacare would be beyond not only the attention spans of most viewers, but beyond the intellectual capabilities of most in the traditional media.

furball said...

Oops! I forgot that my username here doesn't include my real name and Francis convinced me long ago to have the guts to stand behind my postings. (Pretty wise fellow, that Fran.)

Tim Turner

Mark said...

The political elites don't just get the causes wrong, the completely and utterly fail to provide any objective measure of what the expected results of their interference should be.

I'm no engineer, but even I know that inventing a "solution" is only a small part of process improvement. If I make some change to the way I heat my home in order to save money and energy, I should also state what my expected savings will be. Then, monitor my energy use and cost. If I don't realize the expected savings (or even see an increase), then I reverse what I did, and come up with a different solution. Only a fool or a politician (but I repeat myself) would fail to assess the success/failure of a newly-implemented plan.

Because congress-critters insulate themselves from any effects, they don't even bother to state what they expect to see, nor provide any way of measuring success/failure. Or, their egos are such that they are so sure of themselves that they can't even entertain the possibility their plan may fail. As such, they don't have to worry about reversing a failed plan -- they simply pass a bill, declare success, and go on to conduct the next bit of legislative carnage.

And, as that wise guy, (er... man of great wisdom), Fran has encouraged us to do, I fully stand behind my thoughts....