Talk:Inductive reasoning

The following discussion is an archived debate of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section.

The result of the debate was move. —Nightstallion (?) 10:30, 13 March 2006 (UTC)

Propose move of this article

Consistent title (cf deductive reasoning and abductive reasoning) -- infinity0 23:22, 8 March 2006 (UTC)

Special:Whatlinkshere/Induction (philosophy) - quite a lot of articles link to inductive reasoning already, which is a redirect to this current page. -- infinity0 23:37, 8 March 2006 (UTC)

• Support Go for it. Simoes 14:48, 9 March 2006 (UTC)

• Support as per nom. David Kernow 17:41, 9 March 2006 (UTC)

• Support, I think. I got confused trying to follow the history of previous relocations and indirections, but it seems like the basic content of the article was once before at a pre-re-directed article Inductive reasoning? Jon Awbrey 17:58, 9 March 2006 (UTC)
  • Yeah, I think the article started off there, but was moved here. I don't know why, but now this article has an orphaned title. -- infinity0 21:10, 10 March 2006 (UTC)

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Introduction

First sentence should be "Induction is a form of reasoning that makes generalizations based on individual instances.[1]" The current introductory sentence is bloated and muddy. Anyone agree? —Preceding unsigned comment added by 128.138.64.101 (talk) 04:30, 8 September 2008 (UTC)

I do. Although the point, that an induction is merely an educated guess, should remain. --Arno Matthias (talk) 12:10, 8 September 2008 (UTC)

Disagreed. The statement that "induction is a form of reasoning that makes generalizations based on individual instances" is not a claim that accurately describes all instances of inductive logic. The sentence makes an inaccurate generalization about induction, at least as far as I can see it, and is misleading in an intro paragraph. For example...

1)90% of humans are right-handed.

2)Joe is a human

Therefore, Joe is right-handed.

This is a very simple, and very common, usage of inductive reasoning that moves from the general to the particular.

Gilesbardsong (talk) 03:06, 24 October 2008 (UTC)

jargon

In the introduction of the article, the term "tokens" is used without explanation. Could the writer please explain it or use a simpler term? Peter Johnson 64.231.45.249 04:04, 5 April 2006 (UTC)

abduction

I find the sentence (in the section ==Validity==)

The writer John Barnes outlined a third method of reasoning, called "abduction", in his book Finity ...

highly questionable. --Arno Matthias 23:23, 24 November 2006 (UTC)

And so it has been removed. Feel free to be a little bold the next time a random anon adds nonsense to an article. Simões (^talk/_contribs) 00:21, 25 November 2006 (UTC)

...well... I haven't read "Finity"... maybe it is pure genius... --Arno Matthias 14:26, 25 November 2006 (UTC)

To my knowledge, abduction was first introduced by the philosopher Charles Saunders Peirce. It is analogous to an "inference to the best explanation", but it is no inductive principle.

References

This article states many facts, of who said what, without proper citations, of where and when. Please help to improve. - 89.247.34.119 20:52, 19 February 2007 (UTC)

Unclear explanation - "strong induction"

The explanation for the following example of "strong induction" seems unclear:

"All observed crows are black. therefore All crows are black. This exemplifies the nature of induction: inducing the universal from the particular. However, the conclusion is not certain. Unless we can systematically verify the possibility of crows of another color, the statement may actually be false."

The problem is that the syntax of the final sentence--which I take to mean that we can legitimately conclude "all crows are black" from the fact that "all observed crows are black" ONLY if "we can systematically verify the [IMPOSSIBILITY] of crows of another color"--makes it difficult to untangle the logic of the resulting statement. If I follow the argument correctly, I believe the sentence should read: "Unless we can systematically verify the IMPOSSIBLITY of crows of another color, the statement may actually be false."

Califgrll 21:42, 24 February 2007 (UTC)

Changed the relevant sentence, it now reads: "falsify the possibility" because I think its more precise than your (nevertheless correct) suggestion. This is due to falsificating the negation of a proposition being the only way to verify a proposition. -Dalailowmo 217.234.81.64 09:39, 25 May 2007 (UTC)

There is also another problem with that section: "A strong induction is thus an argument in which the truth of the premises would make the truth of the conclusion probable, but not definite."

That's not true. Including more inductive cases has absolutely no bearing on the probability of truth or falsity of the statement. With the crows, for example, seeing MORE black crows doesn't mean it's any more probable that only black crows exist. This is actually an important point; it is the foundation of many anti-scientific claims. Famously, it is also the basis of Hume's system -- that we have no real evidence whatsoever to believe that the sun will rise tomorrow.

Furthermore, I think the entire section of strong vs. weak induction is pretty, well -- weak. Once you realize that additional cases don't have any bearing on the truth or falsity of a proposition, you'll see that they are, in reality, no different (unless you're one of the very few that foolishly defends probabilistic induction). -Tris

Citations

I notice there are a whopping 34 citations for the last sentence in the introduction, and none anywhere else. Does this strike anyone else as odd? --Wayne Miller 15:01, 2 August 2007 (UTC)

It's clearly an overkill, just a few of the most significant ones should be left. Reinis^talk 23:19, 28 August 2007 (UTC)

That's approximately the coolest thing I've seen on this site.

Article Lacks Clarity

This article could use a general edit I think - though I am unqualified to do it. the following paragraph was particulary opaque to me:

It is however possible to derive a true statement using inductive reasoning if you know the conclusion. The only way to have an efficient argument by induction is for the known conclusion to be able to be true only if an unstated external conclusion is true, from which the initial conclusion was built and has certain criteria to be met in order to be true (separate from the stated conclusion). By substitution of one conclusion for the other, you can inductively find out what evidence you need in order for your induction to be true. For example, you have a window that opens only one way, but not the other. Assuming that you know that the only way for that to happen is that the hinges are faulty, inductively you can postulate that the only way for that window to be fixed would be to apply oil (whatever will fix the unstated conclusion). From there on you can successfully build your case. However, if your unstated conclusion is false, which can only be proven by deductive reasoning, then your whole argument by induction collapses. Thus ultimately, pure inductive reasoning does not exist.

Thanks! Cyclopsface 21:58, 27 August 2007 (UTC)

The intro, especially the sentence with 34 citations, is currently an attack. It lacks a balance which John Awbrey's edits once provided, as this is only one type of reasoning cataloged by C. S. Peirce. --Ancheta Wis 10:14, 20 September 2007 (UTC)

Inconsistent Treatment, Factual, and Technical Errors

There are a number problems here, some debatable, some not.

There are two main types of induction. One is mathematical and produces certain results, and the other basically says (for example) that since the sun has not failed to rise every 24 hours or so, it will not fail to rise within the next 24 hours. In Principles of Mathematics (1903), Bertrand Russell calls mathematical induction "disguised deduction" and the other kind of induction he calls a method of making guesses. The first characterization is debatable, the second is not.

So the editorial claim, "Inductive reasoning is the complement of deductive reasoning." is at least imprecise.

Mathematical induction has been used when a set is constructed from an initial state and a generative principle, and it can be shown that all members of the set must have a particular property. Also proofs using mathematical induction use the natural numbers for the incremental and ordinal properties of the set, but mathematical induction is not a statement about the natural numbers per se (as stated in the entry on that subject).

The other kind of induction has been used when we project a trend. Neither mathematical induction nor the other kind of induction has anything to do with syllogisms or argument by authority. I haven't read the (what 5 with virtually the same title?) Popper articles, but I suspect some of the things being classed as induction are there because someone wants to declare that only deduction is logically valid and therefore lumps anything they consider not logical (i.e. the complement) into "induction".

Returning to syllogisms, deductive reasoning has syllogism as a result, but syllogism is not deductive reasoning per se (in the sense that Sherlock Holmes uses the phrase). Literal deduction consists in eliminating all impossible conclusions, leaving the set of possible conclusions.

Abductive reasoning is much closer to what is called statistical syllogism (though not as statistical syllogism is represented on this page). It consists of interpreting a large number of facts to infer a consistent general principle. Abduction usually appears in the context of diagnosis; so it is usually concerned with discerning causal principles.

Umberto Eco (possibly originally from Charles Peirce. Sorry, I don't recall the citation - probably Semiotics and the Philosophy of Language) described the three thus (I'm paraphrasing and augmenting):

Deduction proceeds from certain general qualities to certain logical cases to specific factual inferences. Induction proceeds from specific facts to categorical cases to inferring general principles. Abduction proceeds from specific facts to consistent general principles without inferring cases.

All this differentiation is subject to the same debate that Russell introduced (if he was the one who originally introduced it - I don't know), but it is almost certainly true that calling induction the complement of deduction is misleading and organizes the entries incorrectly.

Kikilamb 20:57, 11 October 2007 (UTC)

Regarding the section on Inductive strength, I thought Strength just meant how much new information the conclusion was telling you. For example (Weak)"Tomarrow the sun will rise." to (Strong)"Tomarrow , on the East Coast, the sun will rise at 6:51 am". The Weak Induction section refers to overgeneralization, which is topic in Inductive probability. I'm confused. —Preceding unsigned comment added by 69.124.128.196 (talk) 14:36, 2 February 2008 (UTC)

Argument by Authority

"Authority" in this context is a social status. There are many rationales for designating an authority, such as "Religious text X said so.", "My father said so, and my father wouldn't lie to me.", or the given rationale "The source has been truthful in the past."

It is one thing for us to say that a source has been truthful in the past and therefore they will probably be truthful in the future, and it is another to say that a source has been truthful in the past and therefore they are probably being truthful now. The latter is not a type of reasoning; in fact, it is an abdication of reasoning. On that basis, all designators of authority/oraclehood in this context, including ones that seem inductive, I call rationalizations rather than reasons. Since this is not reasoning, it cannot be some subtype of reasoning; and specifically, it is not inductive reasoning.

On that basis I have deleted the subentry. —Preceding unsigned comment added by Kikilamb (talk • contribs) 21:26, 13 March 2008 (UTC)

Deductive reasoning vs experimental evidence

Under "Validity" "By substitution of one conclusion for the other, you can inductively find out what evidence you need in order for your induction to be true. For example, you have a window that opens only one way, but not the other. Assuming that you know that the only way for that to happen is that the hinges are faulty, inductively you can postulate that the only way for that window to be fixed would be to apply oil (whatever will fix the unstated conclusion). From there on you can successfully build your case. However, if your unstated conclusion is false, which can only be proven by deductive reasoning, then your whole argument by induction collapses." - I assume it's possible that the falseness of the unstated conclusion could also be determined thru experimental evidence (direct sensual knowing), or is evidence considered part of deductive reasoning? You try oiling the hinge but the window still isn't fixed, obviously (before formal reasoning can take place) oiling didn't fix the window, or, inductively, an alternate hypothesis as to what will fix the window is generated - "hitting the window hinge with a hammer will fix this". --69.243.168.118 (talk) 03:38, 23 March 2008 (UTC)

?"truth of the universal"?

re:

Sextus Empiricus questioned how the truth of the universal can be established by examining some of the particulars. Examining all the particulars is difficult as they are infinite in number.

Can anybody explain what is meant by the term "the universal" in the above sentence? As the term is usually used a universal is not something which could be either true or false. The definition of universal provided at universals seems fair enough:-

In metaphysics, a universal is a what particular things have in common, namely characteristics or qualities. In other words, universals are repeatable or recurrent entities that can be instantiated or exemplified by many particular things.

On that definition a universal is not a truth-bearer/not the sort of thing that can be true or false. --Philogo 12:39, 29 July 2008 (UTC)

Sextus Empiricus is referring to things that are univerally true (true in all cases) rather than things that are only true in particular cases (under some circumstances, such as those one has observed). --Llewdor (talk) 20:13, 12 August 2008 (UTC)

Science does not require induction

The article asserts, "Scientists still rely on induction nevertheless." This is patently false. Science involves certainty with regard to disproving hypotheses (this using deduction), and then responds with new hypotheses (supposition). Never is induction used to form unsupported conclusions.

Applied science (engineering) relies on induction because it needs to act as if it knows what things are true, but scientists only ever know what things are false, and thus induction plays no role whatever in their work.

The claim that science relies on induction could be used to define unscientific claims as science. --Llewdor (talk) 20:19, 12 August 2008 (UTC)

How could it do that? It is defining science as a subset of inductive practice, not inductive practice as a subset of science. You are making this flawed inference: Some induction is done unscientifically. Therefore, all induction is unscientific.

That isn't right. Science uses induction and deduction. (Any postulated "Law of Nature" is an induction, for example.) Pjwerner (talk) 20:28, 24 October 2008 (UTC)

That's nonsense, science does not use induction. Induction does not exist. (Not only) Science (but every valid inference humans ever make) is purely deductive. A Law of Nature is a Law of Nature, not an induction. It would be allegedly an induction to say that this Law of Nature is valid because of this evidence and that evidence, but it's not really induction, it's merely an invalid argument. Engineering does not rely on some allegedly existing induction either. Engineering exploits physical laws, it doesn't rely on any Induction in relation to them. --rtc (talk) 12:28, 25 October 2008 (UTC)

If there is no such thing as induction, all talk of "Laws" would be meaningless. That clearly isn't the case. I don't want to get in a huge philosophical dispute here, because I understand your point. But your point is a controversial one, and as such, both sides should be aired. Pjwerner (talk) 12:35, 25 October 2008 (UTC)

Refs need fixing

The refs section of this article is terrible. It has tons of references about inductive probability, and little to nothing on anything else. I don't have the background to select good references on inductive reasoning, but I strongly feel we should cut all but a few important refs on inductive probability and add refs on inductive reasoning in general. 99.233.26.121 (talk) 13:27, 26 August 2008 (UTC)

Odd and even example

I'm no logician, but I can't help but notice that "odd + odd = even" is given as an example of inductive reasoning, but this can be mathematically proven. Mathematical proofs are deductive, so it would seem to me this example is flawed. 12.216.1.235 (talk) 05:50, 29 September 2008 (UTC)

If the argument was that "one or more pairs of odd numbers added together make an even number therefore all pairs of numbers added together make an even number" then it would be an inductive argument, even if there were a deductive argument that proved the result. There may be cases in the history of math where a result was known by repeated instances and later proved by deduction, e.g. I think Pythagoras' theorem. I believe Goldbach's conjecture is "known" by induction and remains to be proven. Nevertheless this is a poor example of an inductive argument since (A) it proceeds from just one instance (b) there is a deductive proof for the result. The example therefore is bound to cause confusion. There are less confusion well-worn examples that can be given, e.g. "Repeated observations show that if a price of iron is heated then it expands, therefore iron expands when heated". The traditional example for frailty of induction is "The dog observed that whenever he woke up early in the dark and narked in the end the sun would rise; therefore, the dog concluded, barking makes the sun rise"

PS Whether mathematical arguments are purely logical or empirical is a hot topic: see Philosophy of Mathematics.--Philogo 12:40, 29 September 2008 (UTC)