Art of deduction

[ToddRyler]

I was disappointed, as a child, that Sherlock Holmes was really just a fictional character, and so no such book exists, that was authored by him, and detailed his methods for the general public to read.
I was wondering if you could share your experiences in deduction, where you feel you did a good job at it, or, any curious observations that you have made?

Eg: I find that people tend to swing their dominant hand to a lesser extent while walking, so it might be possible to tell their handedness by just observing them walk. People also tend to keep their dominant foot forward while standing relaxed, and when in presence of an authority, or, when they aren't exactly comfortable, the dominant foot lies behind the non-dominant one.

 

[ToddRyler]

Also, if you happen to know any book which breaks it down for not-so-intelligent people to understand, please share it

 

[QuickTwist]

Deduction is not an art. Its mechanical in nature as opposed to creative.

On Topic (and jokes aside): deduction largely revolves around crystallized intelligence and acute observation as opposed to fluid intelligence and reasoning. Deduction is like putting together a puzzle: you work from a known point and draw conclusions based on that. Fluid intelligence is much more like a priori information that you use to figure things out. Deduction is finding all the pieces and induction is knowing on a deep level how things work (when done correctly).

I think anyone can get better at deduction if they are taught the rules of it.

 

[Hadoblado]

@ToddRyler
By deduction you mean inference (or induction). The difference is that, while inference can work off more general observations to guess at uncertain conclusions, deduction preserves the truth value of a statement to apply to another.

For example, I notice you walking with your right hand more animate than your left. I also notice that the less animate hand of a walker tends to be their dominant. Therefore, I infer that you are left handed. Notice how it's still sort of a guess?

Deduction would be: You are walking with your left hand animated. All inanimate hands are dominant hands. All hands that are not dominant are non-dominant. Therefore, I deduce that your left hand is non-dominant. Notice how, if the premises are true, the conclusion has to be true? This sort of logic is not about truth, it's about the 'form' of an argument.

If you want to learn how to infer stuff, take statistics. If you want to learn how to deduce stuff, learn formal logic.

@QT - I mostly agree with you, sort of
There is art to deduction. You still need to produce the proof, and there's not always a clear-cut way to do this. Logic is a skill, just like any other type of thought. Almost anyone can learn it, but it comes more natually to some than others. It requires both crystal intelligence (the knowledge of logics, the knowledge of the premises you're working with), and fluid (computation, creativity).

 

[AndyC]

Something I read ages ago that teaches formal logic, but I'm not sure how relevant it is to the Art of Deduction: http://people.ucalgary.ca/~rzach/static/open-logic/courses/phil310/phil310-ebook.pdf
Look into relational frame theory (relational frame training), I think that kind of training will help with the computational dimension of the Art. Also practising logic puzzles and doing small online courses on platforms such as Brilliant which teach you to understand things like physics of everyday life should help. All a matter of practise.
For a book on logic puzzles: https://www.amazon.com/Logic-Puzzles-your-reasoning-power/dp/B002CWQDPK
Brilliant: https://brilliant.org/
I would recommend playing chess too.
If you really want to master the art, keep a notebook around and force yourself to analyze a few things around you each day, write it down to reflect on later, then build on this habit of deduction.

 

[Deleted member 1424]

Sherlock Holmes was inspired by a real person.

 

[ToddRyler]

Thank you for your time, QT, Hado, the distinction between inference and deduction i was not aware of
brilliant.org seems to be a website i could get addicted to. Thank you so much for sharing AndyC!
Also, coming to know that SH was based on a real person was pleasantly surprising, Adaire

 

[Cognisant]

Guess Who and 20 Questions are games based on deduction, in both games you use facts to eliminate possibilities until you can deduce the answer.

 

[QuickTwist]

@QT - I mostly agree with you, sort of
There is art to deduction. You still need to produce the proof, and there's not always a clear-cut way to do this. Logic is a skill, just like any other type of thought. Almost anyone can learn it, but it comes more natually to some than others. It requires both crystal intelligence (the knowledge of logics, the knowledge of the premises you're working with), and fluid (computation, creativity).

How?

 

[Hadoblado]

What are you referring to? That's a big quote.

 

[QuickTwist]

What are you referring to? That's a big quote.

I mean, I get that deduction require a tiny bit of fluid intelligence, but its pretty negligible when considering all the concrete data you need to actually draw a conclusion.

Describe how fluid intelligence is used in deduction. As far as I can see it largely plays a secondary role as opposed to a primary role. The hard part about deduction is observation.

Also, did you get the joke?

 

[Hadoblado]

Hmm... Okay I'm going to go and dig up some examples from an intro course I did in first year.

Spoiler

Provide a proof of the validity of the following, or provide an example where these premises necessarily lead to an incorrect conclusion.

Murders are committed and regretted. If murder is not bad, then judgements of regret are incorrect. If determinism is true, then if murders occur and are bad, evil is a necessary feature of the world. If determinism is true, then if murders are regretted and judgements of regret are incorrect, error is a necessary feature of the world. Hence if determinism is correct, either evil or error is a necessary feature of the world.

I don't have time to do this myself, I'm about to head to bed. But you can see how this doesn't require much crystal intelligence at all other than an understanding of formal logic? You come in and just assume all the premises, which means that the operation itself is almost pure fluid intelligence right? Nighto.

 

[QuickTwist]

Hmm... Okay I'm going to go and dig up some examples from an intro course I did in first year.

Spoiler

Provide a proof of the validity of the following, or provide an example where these premises necessarily lead to an incorrect conclusion.

I don't have time to do this myself, I'm about to head to bed. But you can see how this doesn't require much crystal intelligence at all other than an understanding of formal logic? You come in and just assume all the premises, which means that the operation itself is almost pure fluid intelligence right? Nighto.

That's logical extrapolation based on an unproven premise, not deduction. It would be deduction if it gave all the parts of the argument in the premise, but it doesn't do that. Instead, it comes up with its own terms to describe itself logically.

Let me see what deduction I would come up with.

Murders are committed and regretted.

Murder happens more than once.
Murder is done from a person to another.
Out of murder that happens, it is regretted.
Regret happens after the murder.
Murder causes regret.
Murder is done more than once both from a person to another and the regret from the murder takes place after the murder.

 

[Pizzabeak]

The art of deduction is also a science - it can be learned and employed with some degree of skill. The best thing to read is Sherlock Holmes. Doyle said Bell's input into the stories weren't really helpful, ironically. Bell was maybe INFJ. Holmes is probably INTJ. Nonetheless it's a good skill to learn if you were a cop or something.

Doyle didn't really enjoy writing about Holmes, and it showed. If he wanted to make them ridiculous and unrealistic he did but they are still enjoyable for that very reason. So while Bell was a professor and could induce stuff like that is not really wholly out the realm of possibility it would seem. He played the usual sports the kids would play at the usual kind of schools they went to, and got the similar kind of education which I believe would be considered decent and appropriate for that kind of class in my opinion.

 

[EyeSeeCold]

Hmm... Okay I'm going to go and dig up some examples from an intro course I did in first year.

Spoiler

Provide a proof of the validity of the following, or provide an example where these premises necessarily lead to an incorrect conclusion.

I don't have time to do this myself, I'm about to head to bed. But you can see how this doesn't require much crystal intelligence at all other than an understanding of formal logic? You come in and just assume all the premises, which means that the operation itself is almost pure fluid intelligence right? Nighto.

The conclusion is that either evil or error is necessary, but if determinism is true, and murders occur, and murder is good (therefore evil is not considered), and people don't regret, then error isn't necessary. Therefore evil or error is not necessary, therefore the conclusion is false.






I don't agree that deduction involves little fluid intelligence. Because humans do not have a logic sense, i.e. we don't perceive logic directly, the world is a-logical until we analyze it. So reality isn't a collection of congruous and incongruous puzzle pieces to be assembled, and it would be absurd to think so because you can generate an infinite amount of facts about reality and there are infinite ways you can assemble those facts. It is the job of the artist to sculpt a deductive image from the block of reality using tools of logic which is where the fluid intelligence comes in.

 

[Hadoblado]

That's logical extrapolation based on an unproven premise, not deduction. It would be deduction if it gave all the parts of the argument in the premise, but it doesn't do that. Instead, it comes up with its own terms to describe itself logically.

Am I missing something?

Logic doesn't require proven premises.

If I were to say:
1) All cats are dogs.
2) All old men are cats
Then it logically follows that all old men are dogs. It doesn't matter whether the premise was proven or not. Logically, we can say 'if the premise is true, then the conclusion must be true'.

Which parts of the argument are not present?

It doesn't feel like we're same paging.

 

[AndyC]

Relational Frame Theory example:
A is the same as B
B is the opposite to C
C is the same as D

Therefore, A is the opposite to D.

A is within B
B is without C
C is within D

A can either be within or without D.

A is less than B
B is more than C
C is less than D

A is either more than or less than D

^^ Syllogistic Logic ^^

Studies have demonstrated that RFT can improve IQ test scores, supposedly it's increasing fluid intelligence, particularly verbal IQ. That causes debate regarding what fluid intelligence is, and if it is changing it.

Formal logic primarily has to do with predicates and implications. (and more when it get's more complicated)

If A then B
If B then C
If C then D

If A then D.

You will see, but the main point of this post was RFT.

 

[AndyC]

I suggest Formal Logic and RFT be put together through some sort of program which also uses hypotheses and false propositions at very high levels to make things more complicated. This will surely make you a master of deduction.

 

[ZenRaiden]

Deductions are easy. What Sherlocks thinking shows is that one can observ things and make deductions based on those observations. What is very important to know is how to think. I dont remember any part of Sherlocks story where deduction was the intristing element. It was his ability to pick up a piece of information that others would always overlook that made him good. Simply put if others dont pay attention to things, they miss things that you dont.

Sherlock was the kind of guy that didnt overlook things. He made sure that each observation he made was worth something in the end. Like that time he observed the scratch marks on a walking stick and deduced stuff about the owner> it was primarily that he noticed it. His deductions were made based on stuff he knew would be true, because of experience he had.

 

[DoIMustHaveAnUsername?]

Am I missing something?

Logic doesn't require proven premises.

If I were to say:
1) All cats are dogs.
2) All old men are cats
Then it logically follows that all old men are dogs. It doesn't matter whether the premise was proven or not. Logically, we can say 'if the premise is true, then the conclusion must be true'.

Which parts of the argument are not present?

It doesn't feel like we're same paging.

Yes, logic doesn't require true premises.
More precisely, " 'if the premise is true, then the conclusion must be true" that's the criteria for 'validity' of an argument.
However 'sound' logical arguments DO NEED premises to be true. For a logical argument to be called 'sound', it must be valid (premises must necessitate the conclusion) and all the premises must be true.

 

[DoIMustHaveAnUsername?]

Also, if you happen to know any book which breaks it down for not-so-intelligent people to understand, please share it

For the basics you can consult these videos:
https://www.youtube.com/user/WirelessPhilosophy/playlists?view=50&shelf_id=8&sort=dd
Since you were talking about detectives, you may be more interested in abduction rather than rigid logical deductions. I am not aware (haven't consulted google) of good sources for developing abductive reasoning. Nevertheless, learning the basic concept of deductive logic, and its difference from induction and abduction, along with some common fallacies and cognitive biases wouldn't hurt.

 

[BurnedOut]

Also, if you happen to know any book which breaks it down for not-so-intelligent people to understand, please share it

If you really are an INTP, by now you would have discovered the gift of thinking spontaneously and not thinking in your head. Making deductions requires imagination and tremendous amounts of Ne and Ni and Ti to filter out unnecessary crap. Nevertheless, its true that you need both high amounts of fluid and crystallised intelligence.


Let's take an example : What can you deduce from a finger which is slightly fatty, has some rugged skin beneath the lunula, small grown nails, some hair on the middle phalanx?

Most probable answer after a lot of Ti filtering : Shy Introvert, faces high amounts of social and general anxiety, more likely to have a sedative personality, less friends, isolation, faces from depression, does not care too much about sex appeal and has a smaller/blander wardrobe (all this only applies to girls)

 

[Hadoblado]

How is it that you think:

+ any INTP would have learned this

and

+ this requires tremendous amounts of Ni

when INTPs are specifically known for not having huge amounts of Ni? You claim to be some sort of exception, but then generalise your ability to be normative?

Again not picking fights. I just notice some difference in our outlooks and seek resolution.

 

[BurnedOut]

How is it that you think:

+ any INTP would have learned this

and

+ this requires tremendous amounts of Ni

when INTPs are specifically known for not having huge amounts of Ni? You claim to be some sort of exception, but then generalise your ability to be normative?

Again not picking fights. I just notice some difference in our outlooks and seek resolution.

Oh, I usually assume that INTPs are naturally good at spontaneous thinking, the ones I've been around with are terrific brainstormers. Moreover many INTPs have their Ni evolved, you can even search for threads. My point is INTPs can produce deductions quite naturally along with ENTPs but obviously for spontaneous brainstorming, a good Ni is required anyway. What do you deduce from my deductions ? Which functions were i using ?

 

[Minute Squirrel]

Oh, I usually assume that INTPs are naturally good at spontaneous thinking, the ones I've been around with are terrific brainstormers. Moreover many INTPs have their Ni evolved, you can even search for threads. My point is INTPs can produce deductions quite naturally along with ENTPs but obviously for spontaneous brainstorming, a good Ni is required anyway. What do you deduce from my deductions ? Which functions were i using ?

I'm not sure but I don't think you understand jungian functions very well, or maybe it's just me. But I'm fairly certain that to be able to use Ni and Ne interchangeably and competently would be extremely difficult if not impossible. This is because one has an introverted attitude while the other is extroverted. To be able to use the introverted attitude of a function the extroverted attitude must be repressed. I would assume it would take a lot of time and practice and even then I'm not sure if both could be used in a competent way.

Honestly I don't even think Ni would be all that helpful when making deductions. Ni is good at finding uncouncious patterns and contains great insight when it comes to ideas, but I'm not sure what any of that by itself has to do with deduction. (Except knowing what to look for when making a real time deduction....maybe though I would argue then Se would be just as important if you want it to be correct)

 

[BurnedOut]

I'm not sure but I don't think you understand jungian functions very well, or maybe it's just me. But I'm fairly certain that to be able to use Ni and Ne interchangeably and competently would be extremely difficult if not impossible. This is because one has an introverted attitude while the other is extroverted. To be able to use the introverted attitude of a function the extroverted attitude must be repressed. I would assume it would take a lot of time and practice and even then I'm not sure if both could be used in a competent way.

Honestly I don't even think Ni would be all that helpful when making deductions. Ni is good at finding uncouncious patterns and contains great insight when it comes to ideas, but I'm not sure what any of that by itself has to do with deduction. (Except knowing what to look for when making a real time deduction....maybe though I would argue then Se would be just as important if you want it to be correct)

The function hierarchy for every type does not even have a scientific basis whatsoever. So i guess its safe to deviate from the model because there can be a total of 8P4 (permutations) in total. So just imagine the number of personality types out there in reality

 

[Cogitant]

The function hierarchy for every type does not even have a scientific basis whatsoever. So i guess its safe to deviate from the model because there can be a total of 8P4 (permutations) in total. So just imagine the number of personality types out there in reality

FYI, and I'm sure it will be of benefit, here is the PDF of C.G Jung's Personality Types (1946 version). Perhaps that might shed light on the theory behind the models.

https://monoskop.org/images/8/8d/Jung_Gustav_Carl_Psychological_Types_1946.Pdf

 

[cosmicflow]

13 ways to mentalism is the best book