Justified True Belief

(●) A weak analogy argues from a resemblance that does not hold at the point that matters. + A man shopping for a used car finds the same make, model, year and mileage as the one that ran twelve years for his brother, and buys it on the strength of the likeness. That is an argument from analogy, and a good one. (1) The shape of the move is a likeness pressed past where it holds. - Now suppose the car matched his brother's in colour and nothing else. The likeness is just as real, and it says nothing about the engine. A weak analogy is the colour argument dressed as the mileage argument. - Two words help here. The reasons someone offers are the premises, and what they are meant to support is the conclusion. In an analogy the premises say two things are alike in several ways; the conclusion says they are alike in one more. - Logicians file the failed version among the fallacies of weak induction. Induction is reasoning from cases you have seen to cases you have not. The weak kind offers premises that really do bear on the conclusion, only far too lightly to carry it. - Analogies do two different jobs. Some illustrate. When Jesus says, "The kingdom of heaven is like a grain of mustard seed that a man took and sowed in his field" (Matthew 13:31), nobody objects that kingdoms are not plants, because the seed is a picture, not a proof. - Others argue, the way the used car is offered as a reason to buy. Only an analogy that argues can be a weak one, and pointing out a difference answers only that kind. (2) It tempts because likeness is how we reason nearly all the time. - Drugs are tested on mice before they go to people, and nobody calls that a fallacy, because a mouse is close enough to us for the result to carry over. John Stuart Mill thought every inference from experience rests on some such resemblance. The fallacy is never that a likeness was used, but that nobody asked which likeness was doing the work. - From the inside, a vivid picture feels like a finished thought. "Employees are like children" arrives whole, and the checking never happens. A good analogy and a bad one feel the same at the moment they land. (3) The name is a recent one, and the family has two cousins. - Mill gave analogy its own chapter in A System of Logic (1843) and named the "Fallacy of False Analogies" among his fallacies of generalization. Hurley's textbook prefers weak analogy: the likeness is usually real and merely too slight. Copi's supplies the tests: how many cases, how varied, and above all how relevant the shared features are next to the differences. - False equivalence treats two things as equal because of one shared feature. "Both parties have their extremists, so the parties are equally extreme." - A stretched metaphor, which is nobody's official name for it, is an illustration quietly promoted to an argument. "The economy is an engine, and you don't pour sand in an engine, so no new regulation." Therefore, a weak analogy is not a bad picture. It is a good picture asked to carry a load it was never built for.

(●) Simple example: "Employees are like children, so the office needs a bedtime." + A manager says it at the Monday meeting and then proposes that nobody may log in after nine at night. The rule may be a good one. The reason given for it is not. (1) What is claimed and what is offered. - The claim is that work should stop at a fixed hour. The support is that employees resemble children, and in some ways they do: both need structure, both get short-tempered when tired, both answer to somebody. (2) Why the support does no work. - Ask what makes a bedtime right for a child. She cannot yet judge when she is tired, so somebody decides for her. A grown employee can judge, and the likeness gives out exactly where the conclusion begins. - Try the swap. "Employees are like children, so the office needs a nap room." Same likeness, same air of good sense, and that is the tell: the likeness decorates the conclusion rather than supporting it. (3) What the picture was really doing. - The manager had a picture, and the picture did the deciding. What would bear on the rule is the error rate on work done after nine, and how many people are burning out.

(●) A critic says: "Believing in God is like believing in a teapot orbiting the sun. You can't disprove either one." + In 1952 Bertrand Russell wrote a short piece for a magazine that never printed it. In it he imagined a china teapot circling the sun between Earth and Mars, too small for any telescope to find. Nobody could disprove it, and nobody would believe in it on that account. The teapot deserves a fair hearing. (1) What is being argued. - The claim is that belief in God is unreasonable, or at least unsupported. The support is the likeness: God, like the teapot, cannot be disproved, and nobody thinks that makes the teapot worth believing in. - Grant the strongest form at once. Not being able to disprove a claim is no evidence for it. The Flying Spaghetti Monster, invented in 2005 as a protest against school science standards, is the same analogy in costume. (2) Where the fallacy sits. - The analogy holds at one point. Neither God nor the teapot can be ruled out by a telescope. - It is then asked to carry a second point, that there is nothing to be said for either, and there the likeness gives out. Nobody has ever argued for the teapot; Russell invented it so that nothing could be said for it. God is proposed as the reason there is a universe at all, the reason its physics permits life, and the reason there are moral obligations, each with an argument attached. - So the feature the conclusion needs, having nothing to be said for it, is the one feature the two do not share. - Turn it around: "not believing in God is as sensible as not believing in the teapot." That is just as tidy and settles just as little, because whether the disbelief is sensible depends on the case for God, which is what the teapot leaves out. (3) The sound version. - Underneath the teapot is a point about the burden of proof, and it is correct. Whoever makes a claim carries the job of supporting it, and "you can't disprove it" is never a reason to believe. A Christian should agree without hesitation. The arguments for God are the attempt to carry that burden, and the honest question is whether they do. - If the critic wants an analogy argument that really bites, David Hume supplied one in his Dialogues Concerning Natural Religion (1779). Philo, his skeptic, says the world "resembles more an animal or a vegetable, than it does a watch or a knitting-loom". If the universe's order is to be reasoned about by likeness at all, the likeness points to something grown as readily as to something built. That objection touches the very feature a design argument from likeness depends on. (4) How to answer it. - "You're right that not being able to disprove something is no reason to believe it. But nobody has ever offered you a reason to believe in the teapot, and I'd like to offer you a few for God. Could we look at those instead?" See also: Natural Theology: The Kalam Argument Natural Theology: Leibniz' Contingency Argument Natural Theology: The Moral Argument

(●) A Christian says: "The universe is like a watch, so it must have a watchmaker." + William Paley opened Natural Theology (1802) with a man crossing a heath who kicks a stone and then finds a watch. The stone might always have lain there; the watch, with its spring and wheels fitted to tell the time, had a maker. It is the most famous analogy in apologetics, and the one a Christian most needs to handle with care. (1) What is being argued. - The claim is that the universe has a designer. The support is the likeness: a watch shows parts fitted to a purpose and watches come from watchmakers, and the eye and the wing show the same fitting, so they come from a maker as well. - Paley was no fool, and most of his book is anatomy rather than watches. Philosophers still argue about whether he meant a strict analogy or something closer to an inference to the best explanation, the reasoning a doctor uses when she picks the diagnosis that would make the symptoms least surprising. Said across a kitchen table, though, it is an analogy, and that is the version that gets into trouble. (2) Where the fallacy sits. - Hume's Philo had the differences in place before Paley wrote, and Paley's first chapter anticipates some of them. We have watched watches being made and never a universe. The universe is one case, with nothing to set beside it. And the world resembles a growing thing at least as much as a built thing. - So the feature the conclusion needs, a mind behind the making, is precisely the feature we have never observed the universe to share with the watch. - The junkyard tornado is the same mistake. Fred Hoyle wrote in The Intelligent Universe (1983) that the chance assembly of a living cell was like a whirlwind blowing through a scrapyard and leaving a Boeing 747 behind. Whatever that is worth about the first cell, it fails when it is carried over to evolution, because natural selection keeps every small gain and a tornado keeps nothing. Richard Dawkins's Climbing Mount Improbable (1996) is a book about that one difference: a cliff nobody could leap, and a long gentle slope up the back. (3) The sound version. - Modern design arguments need not run on analogy. Robin Collins's "The Teleological Argument" (2009) builds the fine-tuning case as a comparison of likelihoods, which means setting two questions side by side. How probable is a universe whose physical constants, the fixed numbers built into its laws, permit life if there is a designer? How probable is it if there is only one undesigned universe? - Nothing in that argument claims the universe resembles a watch, so no difference between the two can touch it. Paley's observation survives inside it, as the thing to be explained rather than as a likeness. (4) How to answer a fellow Christian who says it. - "Hume answered the watch before Paley wrote it, and the answer is fair. We don't need a likeness. The fine-tuning argument doesn't use one, and it is stronger for that." See also: Natural Theology: Cosmic Fine-Tuning Natural Theology: Paley (The Watchmaker)

(●) A charge that misses: "The Kalam's first premise is just an analogy from things inside the universe to the universe itself, and the analogy is weak." + The Kalam is the argument that whatever begins to exist has a cause, that the universe began to exist, and that the universe therefore has a cause. The charge goes after the first line and says it generalizes from tables and babies to the cosmos. (1) What the charge claims. - It reads the premise as an argument from analogy: everything we know of that begins has a cause, the universe began, so the universe is like those things in having a cause. - The differences it lists are real. Tables sit inside space and time, with causes waiting nearby. The universe is the whole of space and time, one case rather than a sample, and its beginning is the beginning of time rather than an event within it. (2) Why the charge misses. - The premise is not offered as a generalization from cases. William Lane Craig, its best-known defender, gives it three grounds in Reasonable Faith. The first is that something cannot come from nothing, a claim about what is possible rather than about what resembles what. The second is that if things could come into being uncaused, there is no explaining why anything and everything does not. The third is that experience confirms the premise without exception. - Only the third ground is an induction, and the differences the critic lists bite only there. - Mill and Copi draw the line the same way. An argument from a likeness is legitimate when the shared feature is the one doing the causal work, and the differences count only if they touch that feature. A used car's mileage matters; its colour does not. - The test to carry away: does the difference touch the feature the conclusion needs, or only the picture? - The same test keeps an illustration safe. "Kingdoms are not plants" touches only the picture, and so does "the universe is not a table", unless the premise rested on tables, which it does not. - One thing the charge gets right should be granted. The third ground really is an induction, and the universe's beginning is the one case outside every sample anyone has drawn. How far experience of things beginning within time reaches toward the beginning of time is a fair question, and it is the question the Kalam debate is actually about. (3) What the charge would need to be true. - It would need to show that the premise rests on the resemblance alone, and that means answering the first two grounds on their own terms, starting with what it would even be for something to come from nothing. A charge of weak analogy speaks about a likeness, and it says nothing to a principle. (4) How to answer it. - "If the premise rested on tables and babies, you'd be right. It rests on the claim that nothing comes from nothing. Do you think something can? That is the argument worth having." See also: Natural Theology: The Kalam Argument