Three Types of Explanations - Law, Chance, and Design
As William Dembski has pointed out, there are three basic types of explanations we can give for any phenomenon, E:
- Law: We can posit some nomological regularity L which allows us to predict E as a (probable) consequence of antecedent conditions.
- Chance: We can say that E had no systematic cause but was simply a coincidence or luck.
- Design: We can say that E was intentionally brought about by some agent A.
For example, suppose I pick up a rock, hold it suspended in midair, and let go. The rock falls and lands with a thud. Why? However we answer that, I'm confident that almost everyone would hazard a lawlike explanation as their best guess. Today we'd propose the law of gravity. Earlier ages would have proposed that rocks have a natural tendency to move toward the center of the universe. We probably wouldn't even consider appealing to chance or design in this case--although there's no strictly logical reason why we couldn't. There's no logical contradiction, after all, in supposing that the rock's falling was simply coincidental and that it could just as easily have gone upwards, sideways, or hung suspended in midair. Nor is there any contradiction in supposing that the rock fell because some invisible agent or spirit made it fall. But few would take such proposals seriously, and rightly so.
Contrast this with Paley's example of stumbling across a watch and noticing that it contains an intricate assembly of parts apparently working together toward a purpose, namely, measuring hours, minutues, and seconds. As Paley points out, a design type of explanation jumps right out at us, whereas law and chance type explanations strike us as woefully implausible. Again, strict logic does not force a design explanation on us. We could without contradiction suppose there to be some, perhaps very complex, natural law of watch formation or chalk the emergence of a watch up to the chance interplay of natural forces. But few would take such proposals seriously, and rightly so.
So when it comes to determining the best explanation for E, we always have a choice between three different types of answers--law, chance, or design--a choice that logic alone cannot settle for us. Other explanatory considerations must be brought in to assess plausibility, considerations like simplicity and coherence with existing background knowledge. Hence, determining whether to appeal to chance, law, or design is generally a complex matter that needs to be judged on a case-by-case basis. And there's no a priori guarantee that all rational and informed persons will arrive at the same determination.
There's another complicating issue as well. How are these three types of explanation related to each other? Considerations of parsimony encourage us to try and reduce one or more of these types of explanation to the others, and there are several proposals on how this might be done.
For example, metaphysical materialism holds that design can be reductively explained in terms of either law or chance or some combination of law and chance. There is some divergence of opinion here. On the one hand, it used to be commonly thought that everything could be reduced to law. Laplace, for example, famously claimed that given a complete understanding of the laws of nature and the state of the world at a given time, he could calculate with complete precision the state of the world at any other time. Chance, on the this view, is merely a cloak for our ignorance of the real causes of things. On the other hand, Charles Peirce at one point argued that design and law could be reduced to chance, with natural laws having evolved over countless eons from a primeval chaos.
Theistic and idealistic worldviews, on the other hand, usually try to reduce law and chance to design. Thus, why do we have the natural laws that we do? God designed the world that way. What is chance? Simply our ignorance of the real causes of things, whether law or design. Not every theist or idealist views chance as completely epistemological, however. And there are different views as to whether everything can be reduced to one designer or not. Theistic determinists like Calvin and Jonathan Edwards, hold that there is really only one designer, namely God, who has meticulously prescripted everything that comes to pass. Most theists, however, believe that God has granted a measure of free will to his creatures, such that many events cannot be fully explained without appealing to two or more designers.
posted by Alan Rhoda @ 10:00 PM
6 comments
links to this post
6 Comments:
Rhoda:
As William Dembski has pointed out, there are three basic types of explanations we can give for any phenomenon, E:
1. Law: We can posit some nomological regularity L which allows us to predict E as a (probable) consequence of antecedent conditions.
2. Chance: We can say that E had no systematic cause but was simply a coincidence or luck.
3. Design: We can say that E was intentionally brought about by some agent A.
HammsBear:
What about something that has a systematic cause but low probability of happening? (for example, rogue waves)
Would that fall under 'chance' or 'law'?
Good question, HB.
I suppose that would be a sort of intermediate case between Law and Chance. If we think of Law as paradigmatically deterministic, and Chance as paradigmatically random (no systematic tendency in any of the available directions), then stochastic laws partake somewhat of both Law and Chance in a degree proportional to the strength of the propensity they represent.
I’d be interested in your thoughts on how Craig fits in here. He writes (The Case for a Creator):
“There are two types of explanations—scientific and personal. Scientific explanations explain a phenomenon in terms of certain initial conditions and natural laws, which explain how those initial conditions evolved to produce the phenomenon under consideration. By contrast, personal explanations explain things by means of an agent and that agent’s volition and will.”
And illustration:
“Imagine you walked into the kitchen and saw the kettle boiling on the stove. You ask, ‘Why is the kettle boiling?’ Your wife might say, ‘Well, because the kinetic energy of the flame is conducted by the metal bottom of the kettle to the water, causing the water molecules to vibrate faster and faster until they’re thrown off in the form of steam.” That would be a scientific explanation. Or she might say, ‘I put it on to make a cup of tea,’ That would be a personal explanation. Both are legitimate, but they explain the phenomenon in different ways.”
He goes on:
“There cannot be a scientific explanation of the first state of the universe. Since it’s the first state, it cannot be explained in terms of earlier initial conditions and natural laws leading up to it. So if there is an explanation of the first state of the universe, it has to be a personal explanation—that is, an agent who has volition to create it.
“…A second reason is that [the cause of the physical universe (and is the totality of all physical realities)] cannot be a physical reality. It must be nonphysical or immaterial. Well, there are only two types of things that can be…immaterial. One would be abstract objects, like number…. However, abstract objects can’t cause anything to happen. The second kind of immaterial reality would be a mind. A mind can be a cause, and so it makes sense that the universe is the product of an unembodied mind….”
Alan, is it just that his “scientific explanations” include Dembskis’s “law” and “change” (both being scientific) and his “personal” equals Dembski’s “design”? Seems so.
What do you think of Craig's arguments for the personal nature of the designer?
Tom
Hi Tom,
Nice to hear from you. Good questions, as always.
One minor qualm I have is with Craig's demarcation between 'scientific' and 'personal'. 'Scientific' here basically covers Dembski's 'law' and 'chance', whereas 'personal' covers 'design'. For my part, I'd rather not restrict the meaning of 'scientific' in this way. I prefer an open philosophy of science that looks for the best answer to a given question without any prior restrictions on what kinds of answers are admissible. Hence, I think he should contrast 'personal' with 'non-personal' rather than with 'scientific'.
I do think his argument that the first cause must be a personal cause is quite plausible, and that it's probably the best answer given that there is a first temporal moment coinciding with the beginning of the universe and that such beginning needs a cause. It does seem that under those conditions neither 'law', 'chance', or abstract entities could fit the bill. Though perhaps(?) one could respond that there may be other, non-personal, explanatory possibilities even though we don't know what they are. In short, his argument for a personal cause is not a knock-down proof, but I think it's better than any other option on the table that I'm aware of.
In this vein, I think you'd be interested in an ongoing discussion over the Maverick Philosopher's place on Quentin Smith's rejoinder to the kalam.
Hi Alan,
I’ve sometimes found it useful to think of the pair Law( Fully lawful)—Chance( Lawless) as forming a sort of explanatory continuum. Any real world event we are interesting in explaining will find an explanation falling somewhere on that continuum. Phenomena treated by the hard sciences usually can be fairly well approximated by rigorous covering law models, but there always some measure of “noise”( lawlessness) the model can’t eliminate. Leave the hard sciences and in general our models become less able to represent phenomena as highly lawful. Think of economics.
( This point is close to the point about “intermediates” that you suggested to HB.)
A second and often complementary explanatory continuum is Design ( purpose) – No Design ( no purpose ). Many events are obviously well explained by considering the purpose of the people who take part in them.
We want to explain why a missile struck a certain town. One mode of explanatory looks at the laws of physics and engineering involved. It was very likely to hit its target! But why was the town its target ( if it was)? Here we look for purpose, or perhaps mistake or negligence. Both continua offer ways to explain in different ways the event that interests us. No reduction of one mode of explanation is apparent or wanted (for all practical purposes).
Both these kinds of explanation are first order explanations of events, but we can continue the explanatory regress and ask why we have the physical laws we do, or why human operate from the motives that they do. The drive in contemporary physics is very much to get at deeper physical explanations of the strange laws that seem to hold in our universe. Physicists seem pretty unanimous in finding design ( purpose) explanations useless at this point. Second order explanation of the “laws” of human behaviour likewise seem to have long abandoned talk of purpose in the evolution of human psychology. The conclusion seems to be that continuum Design—no design disappears as a useful explanatory schema at the level of second order explanations. The burden or argument at this point seems to be very much on Theists to try to show that invoking “God’s will or purposes” serves any useful purpose at the level of second order explanations.
Hi Mac,
Good stuff. I think you're right that there are two partly overlapping explanatory continua: Law - Chance, and Design - No Design. I think we could represent these by taking an equilateral triangle and labelling its three points Law, Chance, and Design, respectively. Different points in the interior of the triangle will then reflect different combinations of L,C, and D.
As for whether burden of proof lies heavily on theists to show that design explanations ought to be admitted as possible ultimate explanations, I disagree. The possibility of an ultimate design explanation has to be conceded by anyone committed to an open philosophy of science, to not "blocking the road of inquiry", as Charles Peirce put it. Theists do have a burden of proof, however, when it comes to arguing for the actuality of an ultimate design explanation. But even there, I don't think the burden need be particularly onerous. All one has to do at that point is to make a cogent argument that in some particular case neither law, nor chance, nor some combination of the two can plausibly account for the phenomenon in question. Arguments of this sort are, as you know, controversial, and the difficulty here is that "cogency" is not an absolute quality of arguments, like validity or soundness. It varies with the background assumptions of the audience.
Post a Comment
Links to this post:
Create a Link
<< Home