Tuesday, January 03, 2012

Part 1 (c): Skeptical Thinking (How to Apply Humbug!) - Humbug! 2nd Edition

Most knowledge is provisional, and over the course of time, many "certainties" prove to be false or misleading. Decision makers (such as bureaucrats), keepers of the "Truth" (such as academics), people who paraphrase press releases (such as journalists), and people who comment on blog posts, often have an "axe to grind", or rather, a view of the world which they hope will prevail. When such people write or speak, they often dress up mere opinions as well founded, research based certainties. Be afraid. Be very afraid of dogmatism masquerading as superior insight (see special pleading). We all need to develop our capacity to identify and challenge humbug.

Our best protection against those who attempt to shape our society and culture through their tendentious bloviation or simple-minded sloganeering is healthy skepticism. Healthy skepticism needs to be underpinned by tools of analysis. Being able to identify humbug provides these tools. The "tools" are the informal fallacies in thinking named and described in the book. Once a reader is sensitised to a range of these fallacies, he or she is able to recognise them and is less vulnerable to... (we have to write it, there's no better word, please forgive if it causes offence)... bullshit. Such intellectual skills retain their utility over the long term, and enable new content to be tested, examined critically and placed on a firm foundation.

In his 1946 book History of Western Philosophy, the philosopher Bertrand Russell makes a clear distinction between scientific beliefs and "other ways of knowing": "It is not what the man of science believes that distinguishes him, but how and why he believes it. His beliefs are tentative, not dogmatic; they are based on evidence, not on authority or intuition."

We believe that Russell's view on what characterizes the epistemology of science can be applied as a general exhortation to the seeker after truth. That is, the proper concern of the seeker after truth - the skeptic - is the soundness or otherwise of the reasoning behind an assertion, rather than the assertion per se. Thus, to be skeptical requires us to:
  1. ask for evidence to support a claim before accepting it as reasonable 
  2. admit to being uncertain when evidence is lacking and
  3. reject a claim as unreasonable when the evidence does not support it.
The following commentary is intended to introduce the reader to some general principles behind sound reasoning that should help one to pick position 1, 2 or 3.

Two deceptively simple questions 
Whether engaged in a discussion, forum, debate, lecture, public talk, seminar, or tutorial – or in reading and writing, the seeker after truth should always keep the following deceptively simple questions in mind.
  1. What are you saying?
  2. Why do you believe that what you are saying is true? 
The first question – what are you saying - is about meaning. A common problem encountered when attempting to understand the reasoning behind an argument is the wording of the argument itself. In debates between antagonists, misunderstandings often lead to fruitless and unnecessary conflict. When proponents and opponents of a position are actually talking at cross-purposes, they are not addressing the same proposition. Similarly, when a reader misinterprets a writer's intention, there is no engagement with the substantive ideas in the article or book.

A necessary condition for a rational argument is therefore clarity of expression. If the proponent cannot put a clear argument, the argument does not merit serious consideration. In some cases, an argument is so badly put that the proponent may be engaging in deliberate obfuscation, or at least careless indifference (see Gibberish). Whether in a formal debate, an informal discussion, or in writing an essay, the question: "What are you saying?" (Or what am I saying?) forces the proponent to clarify an argument. Once the nature and terms of the proposition are clearly established, the second question can now be asked.

The second question – why do you believe what you are saying is true – is about justification. In response to this the proponent should now seek to justify his or her position. Justification is the essential difference between a mere opinion and a rational assertion. An opinion is a belief based on untested grounds, and the foundations of an opinion often do not stand up to careful scrutiny (that's even if they are articulated). While it is possible that a mere opinion could be true, the seeker after truth will remain skeptical so long as the opinion remains unfounded. In contrast to an opinion, a rational assertion is a view that is consistent with the known facts and based on reasoned and sound argument. In short, it is justified (and at the very least, the justifying argument should contain no obvious fallacies).

Logic and reason 
There are two kinds of reasoning processes by which we establish what we know – deductive and inductive. Deductive reasoning is a guarantee of truth, so long as the premises are true. By way of illustration, consider the following statements: "Socrates is a human" (premise) and "all humans are mortal" (premise). If we accept that these premises are true then we must agree that "Socrates is mortal" (conclusion). Deductive reasoning often leads to trivial truths of this sort.

By way of contrast, inductive reasoning leads to the generation of a general law or principle after numerous specific observations have been made. For example, it has been observed that every single time an object is dropped near the surface of the Earth, the object has fallen towards the centre of the Earth (the observation). By inductive reasoning, we conclude that all dropped objects will always fall towards the centre of the Earth (the principle). Given a large number of observations of X, and if all known instances of X lead to Y, then all (known and unknown) Xs will lead to Ys as well. In terms of "watertight" formal logic, inductive reasoning is difficult (some would argue impossible) to establish. After all, we cannot know that all objects will always fall to the centre of the earth unless we personally witness every occurrence of a falling object (and even then, how can we know with certainty?).

This essentially pragmatic book does not therefore concern itself with formal logic. Most beliefs are inductive. Believing the sun will rise tomorrow morning is inductive – but it is not, strictly speaking, logical. It is, however, rational. There is a clear difference between the concepts "logical" and "rational" even though these two words are often used interchangeably. Logical reasoning, in its strictest sense is valid because of the tautological nature of the statements considered. If you are introduced to a bachelor, it follows that he is an unmarried man. Asserting that red is not a colour is logically false, because by definition red is a colour. A rational explanation on the other hand, is one that is justified by a reasoned and plausible argument that is not logically false (self-contradictory).

Identifying unreason and making your case 
If an argument is not illogical (i.e., not internally inconsistent) it does not follow that it is rational. A rational argument should not contain any fallacies. Attacking the quality of reasoning underlying a fallacy is at times a commonsense task, given that many fallacies are obvious non-sequiturs (non sequitur is a Latin term which literally means "does not follow"). However a heightened sensitivity to fallacies can be cultivated and developed.

The short descriptions at the beginning of the book allow fallacies to be identified quickly. Once a fallacy is identified it needs to be isolated and nullified. This can be done by demonstrating the error or errors in reasoning. The fallacies which constitute the body of this book collectively provide a large number of "worked examples" for consideration by the reader. Once the reader has considered a significant number of these examples, he or she should be well equipped with an armoury of debunking techniques and skeptical strategies which can be built upon in the future.

How should you go about improving your humbug hunting skills? When you are ready to get serious about improving your hunting skills, you should find a piece of writing (say an article or blog post) which you suspect if rubbish. You should re-read it more carefully, with the Humbug! short descriptions in front of you. When you read a "dodgy" statement, see if you can find a name for it. (That is, does it appear to be one of the fallacies named and described in Humbug!?) Make preliminary notes at first, and then check your initial impressions about the nature of the fallacies you have spotted by reading the more extensive descriptions in the body of Humbug! (It is dangerous to rely on the brief descriptions only. They are necessarily brief and therefore somewhat ambiguous.)

Here is a grossly oversimplified example. Suppose an author (Bonehead, 2007) makes a statement like this: "There is only one reason a student ever comes late to class - a bad attitude".

You could critique this statement by writing: "That's just ****." While Bonehead's statement might in fact be *****, your criticism is disastrously weak. You might "beef up" your criticism by giving a reason for your view. For example, you could write: "This statement is  **** because Bonehead is a ****." Unfortunately, this is not much of an improvement on your original response (even if Bonehead is in fact a ****).

An effective critique of Bonehead's statement might be worded as follows:
Bonehead's statement is in error because he is employing the single cause fallacy. In particular, his claim that there is only one reason (my emphasis) is clearly far too extreme. In their book Humbug!, Clark and Clark (2012) describe the single cause fallacy in the following terms: 'Single cause fallacies occur when a person assumes that there is only one cause of a complex problem.' Bonehead fails to recognise that in reality, there are many possible reasons why a particular student may be late to class on a particular occasion.