Abductive, Deductive and Inductive Reasoning

Dear reader,

I can't help myself but making this blog entry today a scientific one.

Before I start I want you to know three things

1) I wasn't very scientific in my last post. I forgot to mention the names of the series I mentioned. The BBC production is called "Sherlock". The american series goes under the title of "Elementary".

2) It may surprise some of you that although Holmes was so analytical and scientific, his creator was quite unscientific and gullible. Doyle believed very much in the existence of fairies. It's also difficult to believe that Harry Houdini and Arthur Conan Doyle have been friends for a while. Because their point of view on spiritism was so contrary however, that friendship didn't last long.

3) The producers of "Sherlock" really took great care in creating that show. Sherlock has his own homepage The Science of Deduction. You can also read Dr. John Watson's Blog, which includes comments by Holmes and others!!! Other characters of the series also have their websites: Molly Hooper's blog and the forum of Connie Prince. The last two however may only be of interest to people, who know the series and the persons.

In a way even Sherlock Holmes' homepage is only for people who know the series or fans. Also the title of the page is sort of wrong. Sherlock Holmes is not using deduction in his investigations. This is a mistake not only from the series, but also wrong in Doyle's books. On imdb.com you can find a note on that mistake.

To be honest, each reasoning: abductive, deductive and inductive - are tricky and separating each of them from the others is not quite easy. The differences are very small.

The differences between inductive and deductive reasoning are relatively simple to explain.

In deductive reasoning you set up a general rule. From that rule you set up another rule, of which you can be certain, too. If or rather because both are true, the conclusion will be certain at the end. This kind of reasoning can be found in mathematics, for example in equations with variables:

if x = 2

and if y = 3,

then 2 x + y = 7

Maths is often very much just theory. So let's put it another way:

If chaos is increased in a system, unless you feed it with energy,

and if my flat is a system,

then I should feed my flat with energy and keep it tidy and clean, unless I want to drown in a chaotic mess.

With inductive reasoning you take one single thing and take it to be true. From that you make a general rule that applies to other similar things. A conclusion is likely, but not certain. There is this thought experiment about a white swan. If we see many white swans, we can conclude that there exist white swans. It would be wrong however to conclude that all swans are white, or that there only exist white swans. In science, which is about gathering information, you can find this way of thinking.

Abductive reasoning is about observing something and looking for a possible explanation that would make the observed probable as an outcome. The theorist Charles Sanders Peirce, the founder of abductive reasoning, explained it this way:

"The surprising fact, C, is observed. But if A were true, C would be a matter of course. Hence, there is reason to suspect that A is true."

Finding a conclusion is taking your best shot and not very satisfying. The conclusion you come up with may or may not be true. In medicine you find this way of thinking. The patient tells about his symptoms and the doctor has to think of an illness that would lead to those symptoms, to treat the patient accordingly. Also in court you'll find abductive reasoning: does the prosecution or the defense the better arguments that fit and explain the given situation?

So indeed Holmes doesn't use deduction, but abduction. He cannot be certain to see all the facts of a crime scene that lead to the crime. So Holmes' conclusion are likely to be incomplete and with that nothing more than taking your best shot.

Arthur Conan Doyle used Dr. Joseph Bell as a model for Holmes, as I mentioned already in my last post. Another doctor was very good in observing and making conclusions: Dr. Milton Erickson. Sidney Rosen describes a story in his book "My Voice Will Go with You: The Teaching Tales of Milton H. Erickson", which is a good example to show how good Erickson was in observing and making conclusions. The story is called "The Right Psychiatrist":

A young, beautiful woman came to Erickson. She was very desperate. She wasn't pleased with either of the psychiatrist she had seen so far. So she was uncertain about Erickson and whether he was able to help her. He wrote down some things about the young woman and then said to her that he was the right psychiatrist. He could prove it by asking a question. But the woman won't like that question. The woman wanted to hear the question anyway. So Erickson asked her, "How long have you been wearing women's cloths?" Erickson had seen the woman pick a lint off her sleeve in a straight, direct move, without a "detour" around the breasts, like a woman would.

There's also a video with Tim Minchin, where he talks about the human logic, which addresses another aspect of logic.

Until next blog,

sarah