Sherlock Holmes almost single-handedly invented forensic science through his legendary application of reasoning to the solution of crimes.
Not too shabby for a fictional character.
Unfortunately, in some ways, Sherlock also managed to confuse generations of would-be thinkers about the nature of reasoning itself. One thing he did was to brag about his use of the science of deduction when he was mostly doing was abductive reasoning as opposed to deductive (or even inductive) reasoning.
In my school days, I remember discussions about two main types of reasoning: deductive and inductive. Nobody bothered to tell me about another kind called abductive reasoning. I only later learned what an important type of reasoning it is.
http://en.wikipedia.org/wiki/Abductive_reasoning
http://www.newworldencyclopedia.org/ent … _reasoning
http://www.cse.ohio-state.edu/~jj/pubs/AbdInfCh1.pdf
The fact of the matter is that what Holmes referred to as deductive reasoning was more typically abductive reasoning. That’s useful to know because what Z-researchers are trying to do more often consists of abductive reasoning than deductive reasoning. Of course, http://www.trutv.com/library/crime/seri … .htmlthere are many wonderful moments when we use deductive reasoning, but it abductive reasoning is what drives most of the research.
The art of inductive reasoning is the art of the explanation. We might think of it as a three step process.
In the first step, we make observations about events that are (presumably) known to have occurred and try to find explanations that would account for those events.
As we do that, we begin to notice anomalies that are hard to account for with the seemingly obvious explanations, so we enter the second phase where we begin to dig deeper for more powerful explanations–meaning explanations that have greater potential explanatory power–that can account for the anomalies.
Then, once we have generated a hypothesis, we enter a third phase in which we evaluate the explanation to assess what virtues it might possess: we try to see if it is consistent with the facts, and if it is internally consistent; we try to evaluate the scope of its explanatory power; we try to evaluate its predictive power; we try to evaluate its verifiability and testability; we try to evaluate it in terms of its economy (think Occam) of entities; we try to evaluate it on esthetic terms–whether it seems elegant, or clumsy; we try to evaluate it in its communication value: whether it has unity, emphasis and coherence; we try to evaluate it in terms of how imaginative it is–too much, just right, or not enough; we try comparing it to other theories to see which seem better; we try to evaluate it in terms of the claims made for it (e.g.: the theorist claims proof rather that conjecture); we try to estimate the level of support it enjoys from actual evidence; we try to see if any known facts can disprove it (okay, now I am going in circles, because this is the same as checking whether it is consistent with the facts).
In the process of evaluating our hypotheses, we often (though not always) introduce deductive logic–when we think of a test for our theory, when we identify an internal consistency, etc. but just as often, there may be no immediate opportunity to leverage deductive reasoning.
This third phase of the abductive reasoning process is also a point where we may make some serious mistakes: we may dismiss a theory off-hand for lack of evidence, while not acknowledging some other virtue(s), such as explanatory or predictive power; we may dismiss it because it is unproven when it should be evaluated is a conjecture, not a proof; we may dismiss it because we it is in competition with an accepted or pet theory–that is, it just doesn’t fit our expectations.
The problem with abductive reasoning is that everybody wants it to be deductive reasoning and it’s not. The virtue of deductive reasoning is that it is tailored to providing certainty. Deductive reasoning can tell you with certainty that if ‘all men are mortal’ and if ‘Socrates is a man’, then ‘Socrates is mortal’. Done! BAM!!
That’s great, but Socrates, to my knowledge, is not a POI in the Zodiac Killer murders–and he has an iron clad alibi.
Deductive reasoning can bring you certainty. It can bring you proof–seemingly the Holy Grail of Z-research–but it has its own fatal weakness. It doesn’t generate new ideas.
And let’s not forget inductive reasoning: it, at least, can help you understand what seems probable, even if it can not deliver certainty and proof. Unfortunately, doesn’t do much in the way of generating new possibilities and explanations for what caused a situation.
For that, we need abductive reasoning. It is surprising that somebody as astute as Sherlock Holmes never knew that. I wonder how one could explain such a lapse.
Best regards,
G
It is surprising that somebody as astute as Sherlock Holmes never knew that. I wonder how one could explain such a lapse.
Inductively?
“I don’t know Chief, he’s very smart or very dumb.“
Hi Daniel,
Upon reading your above comment, I went back to re-read a bit of The Sign of the Four. There was a follow up post that I was writing about Holmes’ process of elimination. The opening scene from The Sign of the Four would be is perfect to make my point, so I will go back and work it in. Might take a while before I am done as I am quite busy lately. Thanks for drawing it to my attention.
Question: I see you sometimes sign your initials D.S.G. or Daniel S. Gillotti. By chance, does the S stand for Stephen?
Regards,
G
Daniel my initials are DGS. When you use yours it always makes me do a double take!
Daniel my initials are DGS. When you use yours it always makes me do a double take!
DGS are you perchance related to JLS?
JLS was a particular favorite of mine.
I probably read it a hundred times.
G
I don’t think so! 
I thought for a second you meant this JLS lol. Only a second mind. http://www.jlsofficial.com/gb/home/
“I don’t know Chief, he’s very smart or very dumb.“
The G stands for state!
I have been meaning to write more on this topic. This is a long one:
I would like to discuss a principle of logic made famous by Sherlock through his adage that "once you eliminate the impossible, whatever remains, however improbable, must be the truth".
Before you go any further, however, I need to warn you that what I am about to say will likely sound VERY pedantic. I apologize if that is the case, but in my mind it is a topic worth discussing as there is a principle of logic that I see being misapplied frequently.
Holmes’ process of elimination is a very useful principle of deductive reasoning–but under some circumstances only. It works extremely well in the world of fictional detectives, but in the real world it can be an extremely slippery tool.
I would like to quote a dialogue between Dr Watson and Sherlock Holmes where the detective showcases and explains his methods. He states that there is a significant distinction between observation and deduction, then goes on to demonstrate both. What is so fascinating about this exchange is how Holmes’ expert application of the process of elimination leads to such decisive results.
The quote begins with Watson speaking:
"Not at all," I answered, earnestly. "It is of the greatest interest to me, especially since I have had the opportunity of observing your practical application of it. But you spoke just now of observation and deduction. Surely the one to some extent implies the other."
"Why, hardly," he answered, leaning back luxuriously in his arm-chair, and sending up thick blue wreaths from his pipe. "For example, observation shows me that you have been to the Wigmore Street Post-Office this morning, but deduction lets me know that when there you dispatched a telegram."
"Right!" said I. "Right on both points! But I confess that I don’t see how you arrived at it. It was a sudden impulse upon my part, and I have mentioned it to no one."
"It is simplicity itself," he remarked, chuckling at my surprise,—"so absurdly simple that an explanation is superfluous; and yet it may serve to define the limits of observation and of deduction. Observation tells me that you have a little reddish mould adhering to your instep. Just opposite the Seymour Street Office they have taken up the pavement and thrown up some earth which lies in such a way that it is difficult to avoid treading in it in entering. The earth is of this peculiar reddish tint which is found, as far as I know, nowhere else in the neighborhood. So much is observation. The rest is deduction."
"How, then, did you deduce the telegram?"
"Why, of course I knew that you had not written a letter, since I sat opposite to you all morning. I see also in your open desk there that you have a sheet of stamps and a thick bundle of post-cards. What could you go into the post-office for, then, but to send a wire? Eliminate all other factors, and the one which remains must be the truth."
Sherlock’s favorite principle, as explained above, is a perfect example of deductive logic applicable in cases where you are dealing with a finite and accurately known set of options.
In the quote above, Sherlock explains how the reddish mould on the instep of Watson’s shoe indicates that he must have gone to the Wigmore Street Post-Office. That is observation, he says, as contrasted with deduction.
Then he goes on to the deductive part of his argument, in the form of the process of elimination. Here he implicitly itemizes the possible reasons for Watson’s visit to the post office and those that he was able to eliminate:
1) Watson might have gone to post a letter, but Holmes can eliminate that possibility because he was with Watson all morning and did not see Watson writing one.
2) Watson might have gone to buy stamps, but Holmes eliminates this possibility on the grounds that there are already many stamps on his desk.
3) Watson might have gone to buy post cards, but that can be eliminated for the same reason: there are already plenty of post cards on Watson’s desk.
With those three options eliminated, we are left to understand that ONLY other possibility remains: namely, that Watson must have gone to send a telegram.
The argument certainly takes the logical form of a deductive argument, and as long as we allow that this is Sherlock Holmes’ universe, we can feel satisfied that there were precisely 4 possible reasons for Watson to go to the post office:
1) To mail a letter,
2) To buy stamps,
3) To buy post cards, or
4) To send a telegram
As long as these are the only possible reasons that Watson could have had for going to the post office, then Holmes’ did truly perform a feat of logical deduction, and his conclusion is as good as proven.
Nice work, Sherlock!!
This type of reasoning is child’s play for SH, but how well does it work in the real world?
Well, before I answer that question, suppose Watson stopped acting like a character in a book who simply reads the lines assigned to him by the author and decided he’s tired of having no other purpose but to make Holmes look like a genius. Suppose he decided to behave less like a loyal sidekick and and more like a perverse troublemaker and said to Holmes:
"That’s terrific, Sherlock old-boy, but there is just one hitch! Your exposition, brilliant though it was, was predicated on the proposition that there were precisely four possible reasons why I might have gone to the post office. As it happens, there is a fifth reason."
"Extraordinary, Watson! You intrigue me. Do tell me what I missed."
"You have to guess!"
"What?"
"I said, you gave to guess!"
"Absolutely not! I am Sherlock Holmes, the world’s only consulting detective. I never guess!"
"Fine then. Do you still think I went to the post office to send a telegram?"
"Yes. Absolutely!"
"Are you sure?"
"Yes, of course! Well, I’m pretty sure."
"Oh really? How sure?"
"Well, pretty sure…mostly sure, I mean."
Now, that probably wouldn’t turn out to be the kind of story that would have given birth to the legend of Sherlock Holmes, but in some respects, it would have put the onus on him to actually solve puzzles that are a bit more like the ones that happen in the real world.
In the real world, it is extremely common to make claims based on the principle of elimination. I would venture a guess that it is one of the most frequently employed principles of reasoning that humans use–a tool that they use frequently, often without even noticing.
Even Holmes doesn’t always seem to notice every time he uses the principle of elimination.
Remember when Watson suggests that observation and deduction must tend to imply each other. Sherlock answers, "Why, hardly", then goes on to explain how observation showed him that Watson had been to the Wigmore Street Post-Office this morning, but deduction lets him know that when there he dispatched a telegram.
He then points out the reddish mould on Watson’s instep, and how that appeared to be the same as the reddish-tinted earth that just happens to be exposed in the street near the post-office.
"Just opposite the Seymour Street Office they have taken up the pavement and thrown up some earth which lies in such a way that it is difficult to avoid treading in it in entering. The earth is of this peculiar reddish tint which is found, as far as I know, nowhere else in the neighborhood…"
He then reiterates that this conclusion came by way of observation, as opposed to deduction.
But what Holmes seems to ignore is that his conclusion that Watson visited the post office made use of the principle of elimination just as much as his conclusion as to what Watson did after he got there.
The only difference is that Holmes did not talk as much about his process of elimination. Nonetheless, he does imply it when he says, "…The earth is of this peculiar reddish tint which is found, as far as I know, nowhere else in the neighborhood…"
In other words, Holmes is applying the process of elimination in order to justify his reasoning, because, if he did not, we would not be prepared to accept his conclusion. But then he proceeds to mislabel his feat as observation only, not deduction.
But it was deduction, just as much as observation. In order for Sherlock to be right in his argument that the red dirt on Watson’s shoe leads to the conclusion that he went to the post office, Sherlock needs to eliminate at least two sets of possibilities:
1) That there is not similar dirt elsewhere that Watson might have stepped in, and
2) There is no other possible consequence to stepping in the dirt near the post office than to go into the post office itself. For example, to be justified in his conclusion, Sherlock would need to have eliminated the possibility that Watson could not have stepped in the red dirt as he walked past the post office, or as he walked into the store across the street from the post office.
When I re-read this chapter, and thought about it, it became evident that Watson was closer to the truth than Holmes when he said, "…But you spoke just now of observation and deduction. Surely the one to some extent implies the other." He is right. We rarely make good use of one without the other.
Contrary to what he tells Watson, Sherlock used observation and deduction in equal parts in order to justify both conclusions:
a) that Watson went to the post office, and
b) that, when there, he dispatched a telegram.
And the subtext is that Sherlock, like so many of us, was apparently oblivious of his own application of the principle of elimination in arriving at his first conclusion.
And that leads me to the most important point of all: like the infallible Holmes, most of us are often oblivious when we are applying deductive reasoning–especially when it comes to applying the process of elimination.
Even though Sherlock often presents his reasoning–or more specifically, his use of the process of elimination–in the form of a deductive logical argument, the true underlying substance of that reasoning is not deductive at all. In the examples above, he is using a disguised form of inductive (probabilistic) reasoning to arrive at his conclusions, then packaging them in the form of deductive reasoning for explanation purposes.
It is worth understanding this, imo, because this is what most people really do, much of the time.
Why do I say it is probabilistic? That is because when we have to come up with a set of possible explanations that might explain a particular behavior–or effect of any kind–it will very often be very difficult, if not impossible, to account for an exhaustive set of possible reasonable explanations.
But, in order to apply the process of elimination deductively, it is absolutely necessary to be able to account for a precise and finite subset of explanations. If a counterargument can be made that at least one unaccounted for option exists that cannot be eliminated, then the conclusion fails the litmus test of deductive reasoning.
The argument may still be useful, if it can be defended on grounds of probability. The problem is that we often present–disguise–such arguments in deductive form, so it seems, even to ourselves, that we have proven something that is by no means proven.
But you said we are oblivious to this behavior. How do we know that a person–or we, ourselves–are doing such a thing?
Vigilance, is the best defense. There are often clues in our choice of words, but the question we must ask ourselves is whether our words imply that we have applied deductive reasoning, and, having considered ALL possible alternatives, have eliminated all but one. Whenever we suspect that is happening, we can help ourselves by challenging the conclusion being offered, by trying to think of a single alternative that cannot be eliminated. Very often we will see that the only thing that prevented us from coming up with another alternative, is a failure of our imagination.
As always, I apologize for talking so long.
G
