... I propose to devote my declining years to the composition of a textbook which shall focus the whole art of detection into one volume.
--The Adventure of the Abbey Grange
A logician studies the way we ought to reason; she is interested in the distinction between correct reasoning and incorrect reasoning. Although we all reason and are often interested in whether our reasoning is valid we are not all logicians because we do not make a study of it; that is, we do not reflect deeply enough on this subject.
Now Sherlock Holmes reasons a great deal--Watson calls him the greatest reasoning machine in the world. But he not only reasons he also reflects on how we should reason. Indeed, he was planning on writing a textbook on the art of detection when he retired. When a person is planning on writing on a subject that he knows so well in practice he must have thought deeply about the subject.
Sherlock Holmes' profession is crime detection. His expertise is in finding out what criminals are hiding from the rest of the world. But as he has demonstrated over and over again his kind of reasoning can be applied to all kinds of situations in which we want to uncover knowledge of things to which we do not have direct access. Sherlock Holmes was not present when a crime was committed, but after he had done his work he was able to describe what happened, if not to the last detail, then at least (as he says) in essentials. Watson went out one morning unaccompanied. Afterwards Sherlock Holmes had no difficulty telling him, Watson, what he, Watson, had been doing. Sherlock Holmes is interested in the kind of reasoning that in a textbook on logic would come under the heading 'Scientific Method'. Indeed, in textbooks as well as on other occasions Sherlock Holmes is often brought in to illustrate scientific method. In one of the articles he has published Sherlock Holmes has this to say concerning the power of this kind of reasoning:
From a drop of water ... a logician could infer the possibility of an Atlantic or a Niagara without having seen or heard of one or the other.
--A Study in Scarlet
Things scientists know are not confined to things they have seen or heard. Scientists are able to tell us what happened millions of years ago--just as Sherlock Holmes is able to tell us what happened when a crime was committed. Now how is this done if not by reasoning? There are two ways for us to know: either directly or by inference. Since scientists were not there millions of years ago; since Sherlock Holmes was not there when the crime was committed; they could not have known what they know directly. They have to rely on reasoning. It is through reasoning that they find out what they want to know.
Sherlock Holmes is aware that he engages in a special form of reasoning--special when compared to the kind we ordinarily do. He calls it reasoning backwards.
In solving a problem of this sort, the grand thing is to be able to reason backwards. That is a very useful accomplishment, and a very easy one, but people do not practise it much. In the everyday affairs of life it is more useful to reason forward, and so the other comes to be neglected. There are fifty who can reason synthetically for one who can reason analytically.
--A Study in Scarlet
If Sherlock Holmes had written his textbook on the art of detection it would have been a textbook on reasoning backwards. The writer of such a textbook can rightly be called a logician.
Is Sherlock Holmes a logician or a theseologist? What does Sherlock Holmes himself think?
It is clear that he thinks of himself as a logician. For logic he has great respect; not so for the kind of story-telling that Watson engages in.
If I claim full justice for my art, it is because it is an impersonal thing--a thing beyond myself. Crime is common. Logic is rare. Therefore it is upon the logic rather than upon the crime that you should dwell. You have degraded what should have been a course of lectures into a series of tales.
--The Adventure of the Copper Beeches
Watson is not a logician. Sherlock Holmes is. Sherlock Holmes finds it regretful that Watson has not written a textbook on the art of detection--an omission that he hopes he will make good one day.
What is theseology?
'Theseology' is a word I have coined to stand for the art of finding and following clues. It is obvious there is such an art: Sherlock Holmes practises it, most detectives try to, and most of us know something about it from reading detective stories. But in the past we did not have a name for this art, so I made up this word. It comes from Theseus, who after killing the Minotaur led himself out of the Labyrinth by following a ball of thread which he had unwound when going in.1
Theseology, like logic, is a prescriptive study. Logic tells us how we should reason. This it does by distinguishing between valid and invalid reasoning. Theseology tells us what we should do if we want to find clues, and also how we should follow them when we have found them. Reasoning is not a simple activity. When we reason we often make mistakes, the reason why we need this study we call logic. Finding clues and following clues are not simple either. In this case we need this study we are now calling theseology.
Deriving from the word 'theseology' we have the words 'theseologiser' and 'theseologist'. A theseologiser is a person who finds clues and follows clues. A theseologist is a person who studies how we should find clues and how we should follow clues. Most of us are theseologisers but not theseologists, just as most of us reason but are not logicians.
Is Sherlock Holmes a theseologist? There is no question that he is a theseologiser (perhaps the best in the world), but does he also study the ways in which we ought to find clues and follow them?
It seems that Sherlock Holmes has to be a theseologist, since he is known to be reflective of what he does. To solve crimes he has to find clues and to follow them; but he is also the kind of person who will think deeply about how he should go about doing these things.
If Sherlock Holmes had written his textbook on the art of detection it would likely have been a textbook on theseology, that is, a textbook telling us how we should go about looking for clues and how to follow them. Many people can reason but not many can solve crimes. Sherlock Holmes may be the greatest reasoning machine in the world, but he is also the only consulting detective that the world has ever seen. If anyone should write a textbook on theseology he should.
In Watson's memoirs we find innumerable instances in which Sherlock Holmes discourses about the ways in which we should look for clues and in which we should follow them. Since we are here limited by space let us just remind readers of three examples.
Example 1: Sherlock Holmes has a habit of telling a person whom he is meeting for the first time all kinds of things about this person, which this person would not expect him to know. For example, on meeting Watson for the first time one of the things he said after they had been introduced was that Watson had recently come back from Afghanistan. Now how did he know? Watson had not told him, nor had any one else. And why should he cultivate such a habit?
Watson found the explanation in an article Sherlock Holmes published. In this article Sherlock Holmes says this kind of exercise 'sharpens the faculties of observation, and teaches one where to look and what to look for.' It is clear from this article that Sherlock Holmes is talking about clues; for he immediately goes on to write,
By a man's finger-nails, by his coat-sleeve, by his boots, by his trouser-knees, by the callosities of his forefinger and thumb, by his expression, by his shirt-cuff--By each of these things a man's calling is plainly revealed. That all united should fail to enlighten the competent inquirer in any case is almost inconceivable.
--A Study in Scarlet
There are many things about a person which will provide clues as to what that person's trade or profession is. If one knows where to look and what to look for, from these clues one can determine this person's trade or profession. However, to acquire this kind of ability one has to go through a period of training and after that to stay in form through constant practice. Now this is a lesson in theseology, whether Sherlock Holmes uses the word or not (in this article he calls it the Science of Deduction and Analysis). Most of us know about clues, but not every one would have taken the time to provide this kind of advice that Sherlock Holmes is giving us here.
When one knows a subject well, in the things one says about it interesting questions arise. This is how the subject develops. By raising these questions and by answering them the subject advances. Notice now that this advice Sherlock Holmes is giving raises some interesting questions:
Sherlock Holmes says detecting clues and interpreting them require training and practice.
Question: How would training and practice help if the clues one needs in solving a crime, say, are unlike any known in the past?
In detecting clues one has to know what to look for and where to look.
Question: How can one know what to look for and where to look when dealing with a novel case?
In the advice we are considering Sherlock Holmes is saying that even a single clue can tell us something about a person's trade of profession. When we have not one, but many, we obviously can be a great deal more certain.
Question: Intuitively this sounds true, but is it really true? Why? What sort of creature are clues such that the more of them we have pointing to the same conclusion, the more certain we can be of that conclusion?
Example 2: In The Hound of the Baskervilles someone stole one of Sir Henry's boots. Why one boot?! Why, when the other was also there for the taking? Concerning this bewildering incidence Sherlock remarks:
The more outré and grotesque an incident is the more carefully it deserves to be examined, and the very point which appears to complicate a case is, when duly considered and scientifically handled, the one which is most likely to elucidate it.
--The Hound of the Baskervilles
Clearly again, Sherlock Holmes is speaking here as a theseologist, even though he would not have used the term. Not all clues are the same, he is saying; some are more revealing than others; and sometimes those that are hard to explain are just the ones to reveal the most if you should succeed in deciphering them. Needless to say again, this important observation raises interesting and worthwhile questions, for example, how does one go about deciphering these difficult clues?
Example 3: Clues are often vague; the same clue could have more than one meaning. But in an investigation, in the heat of the chase we often forget about this. Instead of exploring all the possible meanings of a clue we latch on to the first one that comes to our head and regard it as the truth. Of course when we do this we will be paying a price: our investigation will suffer. This happened to Inspector Hopkins on one occasion, prompting Sherlock Holmes to give the following advice to all theseologisers:
One should always look for a possible alternative and provide against it. It is the first rule of criminal investigation.
--The Adventure of Black Peter
Here again one can see that Sherlock Holmes is speaking about clues. Inspector Hopkins had found a notebook at the crime scene bearing the initials of a person the police had picked up. Immediately Inspector Hopkins concluded that this person was guilty. By not looking for a possible alternative for the interpretation of the clue Inspector Hopkins had made a serious mistake: he was charging an innocent person for a crime he did not commit.
We have given three examples to show the kind of interest Sherlock Holmes has in clues. We could have given more, but since this facet of Sherlock Holmes is well known, more would only clutter up our argument. Even though the term 'theseologist' would not have been known to Sherlock Holmes it is nevertheless appropriate to call Sherlock Holmes a theseologist.
Perhaps Sherlock Holmes is both a logician and a theseologist! Our question in this paper is whether he is one or the other, but why can he not be both? It is hard to deny that Sherlock Holmes has an interest in the art of finding and following clues, but if this interest in theseology can co-exist with his interest in logic, then we don't have to say he is one and not the other. For example, this textbook he never wrote on the art of detection could contain one section on theseology and a second section on scientific method (which Sherlock Holmes calls reasoning backwards).
What argument can we provide to support the view that theseology can co-exist with scientific method?
One simple argument is the following and it may be the only argument we need. Whether we investigate nature or crimes we benefit from following clues--this no one would deny. Clues suggest theories--they are valuable for this reason. However, useful though clues are they cannot decide for us whether the theories they suggest are true. Whether a theory is true or false does not depend on where it comes from. We could have arrived at a theory after drinking coffee. Whether the theory is true or false obviously cannot depend on whether we drank coffee. The same with clues. Sometimes clues lead to true theories; sometimes to false. To determine whether a theory is true or false--this is a matter for logic. There is some disagreement whether logic could ever tell us that a theory is true, but this is a debate internal to logic. In any case, many believe that logic can tell us that a theory is false. When a theory is contradicted by a true statement describing some observation logic tells us that the theory is false.
So it seems a person can be both a logician and a theseologist. She can carry out research in both areas, with the two complimenting each other. Perhaps Sherlock Holmes is one such person. Certainly he is interested in whether his theories are true or false when solving crimes. Certainly also he is interested in how clues can suggest theories. To solve crimes he needs theories. If he is not interested in clues; if he is not interested in following them in the proper way; he may have very few theories to propose when solving crimes.
Is it possible for Sherlock Holmes to be a theseologist only and not a logician? This does not mean that he does not reason, only that he has no special contributions to make concerning the way we ought to reason. Sherlock Holmes has much to contribute concerning how we should go about finding and following clues, but these contributions are in theseology, not in logic. Sherlock Holmes does a huge amount of reasoning, and he is careful and rigorous when he does, but things he can teach us about reasoning can likely be found in logic textbooks by other authors. But the same does not hold in the case of theseology. Things that Sherlock Holmes can teach us about theseology cannot be found in textbooks in theseology--for the simple reason that before Sherlock Holmes there was no textbook on theseology. If Sherlock Holmes had written his textbook on the art of detection it would have been the first textbook of this kind. And if he is to write this textbook, he is a theseologist. Notice what he says concerning his future project.
... I propose to devote my declining years to the composition of a textbook which shall focus the whole art of detection into one volume.
--The Adventure of the Abbey Grange
Why does Sherlock Holmes want to write a textbook on the art of detection? Is it because current textbooks on the subject are out of date? Is it because former attempts by other authors have not been successful? Is it because such a textbook is available in other languages but not in English? No; it is none of these. The context is clear: there has been no textbook on this subject! No one has taken the time to collect together all the many things theseologisers have found out about the theseological process, digested them, and put them into a textbook. Sherlock Holmes thinks this can be done and should be done; but he is too busy at the time: he has too much to do developing the subject in his head as well as at a practical level; he will do the writing when he retires.
Many logicians writing about scientific method are troubled by 'the problem of induction', thought to have been discovered by David Hume. When logicians read other logicians writing about scientific method they will want to know what their peers think about the problem of induction. If Sherlock Holmes ever writes a book on logic logicians will want to know what Sherlock Holmes has to say about the problem of induction. But does Sherlock Holmes show any sign that he is interested in the problem of induction? We do not see any. Sherlock Holmes does not seem to think that we need to solve this problem if we are to solve crimes. Let me explain.
Clues determine epistemic value. Earlier we have said that Sherlock Holmes could be both a logician and a theseologist if clues suggest theories and logic determines whether these theories are true. Now it seems Sherlock Holmes would not agree that clues can only suggest theories. He seems to think that if we can find clues we can determine by the aid of these clues that the theories suggested by clues are true (or close to being true). Certainly clues suggest, but it seems Sherlock Holmes would also say, additional clues can determine whether the theories suggested are true. In fact we have already had occasion to report on this view. We have already said that if we have a large number of clues all pointing to the same theory it is very likely that that theory is true, or close to being true. The more clues we have pointing in the same direction, the more likely this is the case. Sherlock Holmes, we have pointed out, shares this view (so would most people). If we have a large number of clues and they all point to the butler, then it is very likely that the butler did it. In such a case the clues do not just suggest the theory that the butler might have done it, but also let us know that this theory suggested is highly probable. If we have only one clue and it suggests that perhaps the butler did it, we would not on this ground say that it is very likely that the butler did it. We will entertain the theory and look for tests to see whether they would confirm or refute the theory. But the situation is very different when we have a large number of clues pointing in the same direction. In this case, as Sherlock Holmes has said, it is very likely that the theory is true. Of course, it is very likely only, not absolutely certain.
Still, why should this be the case? Why is it that if we have a large number of clues pointing to a theory, that theory is likely to be true?
It seems this has to do with the kind of thing clues are, and if Sherlock Holmes were to write his textbook on the art of detection, this would be one of the questions he had to answer. But then he would be writing on theseology, not on logic. It is not a question in logic what kind of thing clues are. Notice that a clue is not just any piece of evidence. At a crime scene there is usually much more evidence than there are clues.
Clues lead to other clues. It is a common phenomenon in following clues that clues can lead to other clues. This is how we can sometimes have a large number of clues all pointing in the same direction. Most investigations begin with a very small number of clues, but because clues can lead to other clues, when we are lucky, by the time we finish we will have accumulated a large number (if we are not lucky; if the one or two clues at the beginning are the only clues we will have; we will not find out anything).
However, clues do not lead to other clues by themselves. If they do lead to other clues, it is the result of what we have done: we develop new clues from the old clues. Thus in an investigation it is not sufficient that we follow clues, we also have to make an effort to develop new clues from old. Now Sherlock Holmes knows this; it has even rubbed off on Watson:
Our only clue lay in the truncated telegram, and with a copy of this in his hand Holmes set forth to find a second link for his chain.
--The Adventure of the Missing Three-Quarter
Watson understood what Sherlock Holmes was doing. Watson knew that one clue was not enough, and not even two; he knew that Sherlock Holmes was looking for a chain.2
How do we develop new clues from old? Sherlock Holmes knows the answer to this question too. We can only develop new clues if we read the old clues aright!
I was already firmly convinced, Watson, that there were not three separate mysteries here, but one only, and that if I could read the Musgrave Ritual aright I should hold in my hand the clue which would lead me to the truth concerning both the butler Brunton and the maid Howells.
--The Musgrave Ritual
In The Musgrave Ritual the Ritual was a clue. By interpreting this ritual correctly Sherlock Holmes developed a whole series of clues which enabled him to solve not just one mystery, but a number of mysteries. We have pointed out just a moment ago that clues do not just suggest theories, they can help us evaluate theories, to determine their epistemic worth. We have here another instance. When clues lead to new clues they enable us to evaluate those theories proposed in answer to the old clues. To repeat again, old clues cannot lead to new ones unless they have been correctly interpreted.
Why should this be the case? Why can old clues not lead to new ones unless they have been correctly interpreted? We do not find an answer to this question in Watson's memoirs, but it is reasonable to expect such an answer from any one who writes a textbook on the art of detection.
There is a qualification that has to be entered here before we continue. Old clues cannot lead to new clues unless they have been correctly interpreted, but there are exceptions. Occasionally, old clues wrongly interpreted can lead to false clues by accident. But false clues peter out, so that these exceptions are not really all that serious. Still, we have to be aware of them. (It is well known among theseologisers that false clues peter out. Why should they? This is another question for theseology.)
From what has been said, we can see that theseology has the means within itself to determine whether theories suggested by clues are true or false. Now Sherlock Holmes is interested in uncovering knowledge of things to which he does not have direct access. To explain to himself and others why this kind of knowledge is possible there is no need for him to engage in any special kinds of research in logic, particularly the kind that would enable him 'to solve the problem of induction'. Rather, what he has to do is to concentrate on theseology, and see if he can find within theseology the reason why this kind of knowledge is possible. Sherlock Holmes is a theseologist; he cannot be Sherlock Holmes without being a theseologist; but he does not have to be a logician.
Some will ask, what about reasoning backwards? Sherlock Holmes is good at reasoning backwards. If he had written his textbook he most certainly would have written about reasoning backwards. And if he writes about reasoning backwards he is a logician.
Not so. For what is reasoning backwards? From examining footprints Sherlock Holmes can estimate a person's height. This to him is reasoning backwards. Tall people make large strides--this is reasoning forwards, and this we do all the time: we notice tall people and we expect them to make large strides. But ordinary people do not usually go in the opposite direction. Ordinary people do not usually go about measuring footprints and thereby estimate people's heights. But Sherlock Holmes does--not as an idle exercise, but as a way to track down criminals. So what is reasoning backwards but in reality following clues?
Does this mean theseology is a branch of logic--that branch that deals with reasoning backwards? Theseology is the art of finding and following clues. Since following clues is the same as reasoning backwards, theseology (it seems) should be a branch of logic.
But it is not. Theseology and logic are separate disciplines. Earlier we have pointed out some of the questions theseology has to answer. These are not the kind of questions we ask in logic (we have pointed out). For example, why is it that old clues can only lead to new clues when they have been correctly interpreted? In reasoning, whether premises are true or false, conclusions can follow (validly). But in the theseological process old clues incorrectly interpreted will have no offsprings (in the form of new clues). Why? Why do clues behave in this peculiar fashion? As we have pointed out, this is not the kind of question we ask in logic.
Theseology is not a branch of logic. In theseology we ask a different kind of questions. In theseology the concept of a clue is central. If we know what a clue is, we know how to find clues and why by following them we can uncover knowledge of things to which we do not have direct access.3
Memorial University of Newfoundland
1 I would like to thank Dr. John Scott for help in finding this name. Of course only I am responsible for adopting it.
2 A chain of clues is different from a chain of reasoning. We can produce a chain of reasoning by locking ourselves up in our study, but to find the second link to his chain of clues Sherlock Holmes had to go to a specific telegraph office.
3 For more on theseology see my book The Art of Detection--The textbook Sherlock Holmes did not write.