An Autodidact's Bible #19

Reclaiming Your Mind, Part 2, Chapter 6a

This is the nineteenth installment of the serial of my forthcoming book Reclaiming Your Mind: An Autodidact’s Bible. As with other installments, part of it is behind the paywall. Become one of my supporters to get the whole thing.

Catch up on earlier installments here:
#1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12, #13, #14, #15, #16, #17, #18

If your email client chokes on this post, find the whole thing at http://jdanielsawyer.substack.com

Chapter 6
Reasoning

And, finally, we come to the elephant in the room: your ability to think. I’m not talking about your IQ or your penchant for puzzles and riddles. I’m talking about your ability to discipline your thoughts, to use deliberate forms of analysis, to reach conclusions and examine them, to call upon your ever-expanding body of knowledge in an intelligible and useful fashion.

While thinking—i.e. struggling to make sense of things—is a natural human propensity, thinking clearly and reasoning well is not natural. Protocols of thought are, for lack of a better term, the technological fruit of the discipline of philosophy. And “protocols” is the right word. In the same way that there is a particular manner in which one must greet a fellow ambassador (if one were in the diplomatic service) or approach a potentate, and there is a manner in which computers must behave if they are to communicate with each other over a network, so too there are particular manners in which thought must be conducted in order to minimize error, maximize insight, and create intelligible, communicable culture.

Thought is a behavior, and like all behaviors, it has effects, and those effects depend upon the nature of the behavior and the context in which it occurs. And, also like all behaviors, for it to be under your control, it has to be subject to discipline.

Structured thought is called “reasoning,” because the purpose of thought is to adduce the reason behind things.

Reasoning, in other words, is the structured search for meaning. As such, reasoning is an integral part of learning at almost every level (Type-0 learning excepted—at least on a conscious level).

Bear in mind: thinking clearly and thinking well depends on a certain degree of metacognition (i.e. thinking about thinking). You must observe your own thought process, learn to understand how it works, identify characteristic errors, and learn how to overcome them or compensate for them.

Also important to bear in mind, there isn’t just one kind of thinking. Just as the expected behavior of an ambassador is different from that of a fiancee, and just as the Internet runs on a variety of protocols for different purposes (HTTP vs. FTP vs. UDP, etc.), so too are different modes of thought appropriate for different kinds of analysis and contemplation.

Process Trumps Results

“It is better 100 guilty Persons should escape than that one innocent Person should suffer.” -Benjamin Franklin

Also known as “Blackstone’s Formula,” after a 19^th century English legal scholar, this is the fundamental principle of English common law jurisprudence; the moral rationale for “innocent until proven guilty” that governs in our courts and (in eras not characterized by mass psychosis) in the culture at large.

Because of this, both the judicial process and the law itself evolve based upon the usage; as decisions get made, and as litigants challenge those decisions, the entire system shifts. It’s a conversation¹ aimed at eliminating errors and injustices when and where possible—it’s the only system in the world, for example, where a juror can vote to acquit someone based on their belief that a law is unjust.²

As a result of this conversational nature and the guiding philosophy of “innocent until proven guilty,” the judicial process in English common law countries functions according to procedural correctness: If the forms are followed, justice is seen to be served even if, in any given case, justice is not done (or the case is so ambiguous that attaining true justice is a mug’s game).

The wager is that a system characterized by openness and integrity-of-process will, more often than not, serve as a reliable arbiter of justice.

Integrity-of-process is not just the flagship feature of good Common Law thinking, it is the most important feature of thinking full-stop. Thinking is to your mind what weight training is to your body. You’re taking the raw potential you were born with and you’re sharpening it, building it, and giving it definition. Your mind is a tool, and not every stroke of even the finest tool will be flawless—however, a solid tool well-wielded will produce good work more often than bad, and in an imperfect and contingent universe, that kind of “good enough” is the best you can hope for.

And is that so bad? After all, look at what we’ve done with “good enough.” Ten thousand years ago (as near as we can tell), the world was a wilderness peopled by roaming tribal bands. Now, it’s filled to bursting with languages, cultures, inventions, information, art, mythologies, and people who have learned how to control the stuff of life itself and walk on other planets.³

The ability to take the good end of a nearly-even bet is the way that casinos build fortunes, and it’s also the way that humans build civilizations and travel to the stars. Like counting cards in blackjack, if the odds are only slightly in your favor and you stick with it long enough, you’ll come out with one solid win after another.

Good results by a faulty process are always a fluke, and if that process is powerful, the good results are a sign of risk, not of security. Bad results from a reliable process, on the other hand, will eventually wash out. The fundamental value judgment of the trained mind is that process trumps results, every time. It is better to do right than to be right.

Types of Reasoning

Deductive Reasoning

First, we have the most basic of the disciplined thinking protocols: logic. There are a number of forms of logic, from the formal to the symbolic to forms mathematical, but since a proper exploration of those would take at least half the length of this book (and would be of deep interest mostly to logicians and philosophers), you’ll have to chase down that particular rabbit trail on your own. For our purposes, we’re concerned here with deductive reasoning, which most people call “logic,” and which is properly a subset of logic called “Aristotelian logic.”

Why “Aristotelian?” Because Aristotle invented it, more or less. Deductive reasoning is actually a conversational implementation of formal logic that is properly expressed in mathematical theorums.⁴ Way back in Chapter 3 of Part 1 (in the section Of Silos and Spider Webs) talked about the labor shortage caused by the Black Death, the printing press invented by Gutenberg to settle a business debt, and the rest of the strange winding commerce-driven path that created the modern world. That whole strange sequence of events might never have given us the modern world if it hadn’t been for this long-dead Greek and the way that the Islamic world fell in love with him. Because of that love affair (and the remarkable fruits it produced), when Moorish Toledo fell to a band of mercenaries led by El Cid (a.k.a. Rodrigo Díaz de Vivar) in the 11^th century, the European mercenaries discovered libraries full of mathematics, astronomy, physics, and, especially, philosophy from the ancient world. The first to make its way into Northern Europe and into the popular consciousness (among people who could read) was Aristotle’s logic. It proved to be the intellectual ember that would light the fuse of Reformation, Renaissance, and Enlightenment.

Okay, then, how does it work?

In its most straightforward, basic form, logic is deductive,⁵ meaning that you look at what you know, and think about it in a way that reveals what else is implied.

In isolation, a deductive theorem is a very simple if/then statement:

If A, then B. (If all birds are blue, then this bird I hear outside is blue.).

When formulated in a syllogism, which is the proper structure for a deductive argument, you marry a first premise to a second premise, then you draw your conclusion based on the interaction of the parts.

Thus:

Premise 1: All birds are blue.

Premise 2: I hear a bird.

Conclusion: Therefore that bird that I hear is blue.

In a syllogism, if the conclusion follows from the premises, the syllogism is said to be valid. If the premises are true, then the logic is sound.

Notice how the validity of the syllogism doesn’t actually bear on how true it is. It is possible for reasoning to be absolutely airtight, and yet for the conclusions to be erroneous. In a syllogism, as in all reasoning, the process is more important than the conclusion. The wager a thinker makes is the same as the English Common Law system makes, and the same as a card-counting blackjack player:

The thinker is betting that a good process will yield dependable results more reliably than a faulty process. It is understood from the get-go that, no matter how reliable a reasoning method, it is always possible for mistakes and errors to creep in.

With that said, let’s look at our sample syllogism. It is obviously, demonstrably unsound (i.e. false)—but it’s not false in its conclusion. The conclusion is correct:

Therefore that bird that I hear is blue.

Note how, just because the conclusion is correct, this does not mean that the bird is actually blue.

How is this possible?

Note the “therefore” in the conclusion. This is the key clause. It is the “then” in an if/then statement. If we are to grant that premise 1 and 2 are true, then the conclusion must be true. It is a logical certainty (literally). That’s how validity works. So long as the logic is valid (i.e. if the conclusion really does follow necessarily from the premises), you can’t debunk an erroneous syllogism by attacking its conclusion. You have to look at the premises.

In the case of our sample, Premise 2 must be taken as true at least for the sake of argument⁶ because it is a testimonial (i.e. a statement of personal experience). It is possible, of course, that the the person making the argument could be lying (and does not actually hear a bird) or hallucinating (and is thus hearing phantom birds), but that’s a very long trek to take when we have a demonstrable error staring straight at us out of Premise 1.

All birds are not, after all, blue. Anyone who has spent a little time walking in the woods knows this. Indeed, if ornithologists and The Audubon Society are to be believed, out of all the birds in the world, only a very few are blue.

So, as you see, deduction isn’t quite what Sherlock Holmes would lead you to expect. His style of reasoning is actually abduction, which we’ll get to in a little bit. For now, remember:

• Deduction is an elementary form of Aristotelian logic.

• It is formulated in arguments called “syllogisms” which form proofs.

• Such proofs are not true or false, but valid or invalid.

• Valid syllogisms are vulnerable to criticism and disproof by finding flaws in the premises (an argumentation method known as “undermining”). If they are not undermined, their conclusions are certain, and the logic is sound.

As mentioned earlier when we talked about deduction in First Principles back in Chapter 3 of Part 2, deductive logic is not a good method for discovering new knowledge, but it is very good for teasing out the implications of knowledge you already possess. It is a foundational thinking tool.

Finally, bear in mind that deduction is a branch of formal logic. Informal logic is a different beast, and we’ll deal with it a bit later on in this chapter.

Inductive Reasoning

Despite surface similarities, induction is not a form of deduction—and the best way to understand it is to contrast it with deduction. Induction differs from deduction in both method and result.

Let’s take those two items in reverse order:

• Result While deduction can only reveal what you already know (and can not be used to discover new things), induction can, and does, yield novel results and create new knowledge when employed correctly.

• Method The inductive method is similar to the deductive method in that one begins with premises and reaches a conclusion that follows from the premises, but it differs in that the conclusion is drawn, rather than consequential. This means that, where deductive arguments are necessarily true if the premises are sound and the reasoning is valid, inductive arguments are not. Inductive arguments can be strong (meaning there is a tight connection between premises and conclusions) and the reasoning can be cogent (meaning that the conclusions really do follow from the premises), yet the conclusion can still be erroneous. In a sense, this uncertainty is the price you pay for the power of the inductive method to discover new truths.

Why is this the case?

Deduction is a special formulation of a mathematical concept called “set theory,” and embodies its power and shortcomings. Basically, it’s a game of definitions. Bob is mortal because the word “mortal” has a definition which means “things which die.” Since Bob is alive, he will eventually die, and thus he is part of the set of “mortal things.” This is why deduction is certain; it only deals in closed systems.

Induction, on the other hand, operates in an open landscape of possibilities. The method might sound familiar:

You take the evidence at hand and, by noting the connections between the evidence, you sketch out an area of uncertainty defined thereby. Then, you make a guess at what might fit in that area of uncertainty.

Induction is the method you use for solving puzzles. It’s police work (at least, as practiced by ethical detectives). It’s also conditional (the technical term is “Baysean” after statistician and minister Thomas Bayes). It takes the known facts of reality and makes predictions as to what you should expect based on what you already know. The result is a conclusion based on conditional probability—your best guess could wind up being wrong, but that doesn’t mean you are wrong to make the guess. In fact, with induction, it is better to use cogent reasoning to arrive at an incorrect conclusion than it is to arrive at a correct conclusion by uncogent reasoning.

But! But! But! I hear you say from your side of the page. You’re overlooking a huge problem!!! And, you would be right, there is a huge hole in the scheme I’ve just laid out:

If induction is conditioned upon known facts, that means it’s conditioned upon your prior knowledge. And if your priors are even slightly wrong…well, the usefulness of induction is shot, isn’t it? You’re going to follow your chain of reasoning up a long, winding, fascinating road to nowhere.

This was the problem with Scholasticism,⁷ and why, as philosophically fertile as it was, it eventually fell out of fashion. It led to a multiplication of closely-related fields of study and argumentation separated by fierce disputes over small differences, which in turn led to longstanding intellectual deadlocks that were only ever eventually resolved by either practicality (in the case of the great Copernican dispute) or with violence (as in the case of the Reformation).

The usefulness of practicality in settling scholarly disputes is what eventually led to the scientific revolution, and the more-or-less formal integration of Type-1 learning into the process of induction created the scientific method:

1. Induction

2. Testing for disproof

3. Using conclusions that have survived testing to feed the next round of inquiry

The difference between scholarship and science is whether you have confirmation or disconfirmation as your selection gate in the signal flow.

Confirmation is not without its charms or advantages. When you’re chasing an inspired reasoning chain to a tantalizing conclusion, you look for confirmation at every step the same way you check for the dash marks at the edge of a lane on a rainy night when you’re speeding toward a grand destination and the promise of new adventures. Each dash is a little nugget of confirmation which reassures you that you’re still on the right path (and unlikely to plow into oncoming traffic). It’s the way that you can quickly parlay inspiration into grand insights.

But do those insights stand to reason?⁸ Are they correct?

If your thinking is good, your insights will stand to reason by the time you’re done. As far as being correct...well, that’s trickier. To be correct they have to withstand the discipline of reality.

This is where disconfirmation comes in. Remember via negativa? It is easier, always, to figure out what is not true than what is true. To complete the trial-and-error loop with your inductive reasoning, you must eventually become critical of your own most dearly won conclusions.

When you subject your inductive conclusions to disconfirmatory testing, you make up for induction’s great weakness without undercutting any of its power—and you also reduce the amount of uncertainty inherent in any induction.

Next time we will delve into the methods of Sherlock Holmes and other forms of reasoning in the second installment of Part 2, Chapter 6: Reasoning.

1. Otherwise known as a “dialectic.” More on this later.

2. A phenomenon known as “jury nullification.”

3. Or “planetoids” if you want to get pedantic about what kind of heavenly body Luna is.

4. Deductive reasoning is also the mother-discipline containing just about every kind of calculus, probability theory, set theory, etc. that has been invented since Aristotle.

5. You will recognize this method from the section on First Principles Reasoning, in Part 2, Chapter 3

6. Again, literally. This sort of exercise is where the expression “for the sake of argument” comes from.

7. See Part 1, Chapter 3

8. This is also the literal use of the idiom. “Stands to reason” means “Survives the test of analysis,” and originally comes from logic.