Why still no science of logic?

[Yazata]

In short, you're just making the obvious statement that people are observed to reason logically about certain things. I wouldn't have thought that was a very earth-shattering observation; it seems obvious to me. But at least we got there in the end - and only 50 posts into the thread.

If that's Speakpigeon's proposed program, then I can see a potential difficulty.

People think in all sorts of ways. Some of those ways are logically valid and even sound, while others (like the thought process of a schizophrenic) are almost the antithesis of logical. Some ways aren't exactly logical but may have what we might judge are other valuable qualities, such as literary narrativity or poetic allusion.

The point being that there's a normative/prescriptive aspect to logic that simple empirical observation and descriptive modeling of real life thought processes doesn't seem to me to capture.

So ... speculation: you're looking for an empirical science that studies how people reason logically? You're aware that the cognitive sciences already study this, it seems.

I think that cognitive science addresses the problem above pragmatically. It examines cognizers undertaking particular tasks and describes the problem-solving methods that they use, even formalizing those methods in the form of computer algorithms. The additional normative component here is derived from what does and doesn't work.

But you think that the "missing piece" is a "formal model of deductive logic", whatever that means in the context of human cognitive behaviour.

It seems to me that the 'formal model of deductive logic' that Speakpigeon is seeking already exists. There's a variety of them and they are called formal logic, the formal mathematical-style deductive systems that logicians concoct. In other words, logicians spend their time pursuing Speakpigeon's program without his realizing it. Of course they don't base it on scientific observations, they do it from the comfort of their armchairs. But I think that there still is an empirical aspect to it, since it's based on abstracting and formalizing from natural language use.

But there's still the problem of justifying the normative and prescriptive side of logic. The problem of why we should use valid logical forms in our thinking and avoid thinking like a schizophrenic. Cognitive science addresses that by attending to cognitive strategies that are observed to work.

Logicians themselves might take a different tack. They might argue that logic's normative aspect is derived from the fact that logic is truth preserving. 'If A, then B'. Assuming that A is true and assuming that the 'if' is the if of logical deduction, then B must be true too. But one might argue that there's an element of circularity to that argument. It might be assuming the qualities of logic that the argument seeks to justify.

An advantage of the cognitive science approach might be that there could be many effective heuristic strategies that aren't deductively valid or truth preserving. Scientific induction might be an example of that. So there's an argument to be made that these heuristic strategies might need to be empirically discovered, not formally derived as theorems in a formal logical system.

 

[Speakpigeon]

No, the cognitive scientists are producing formal models of real-life empirically-observed reasoning. The logicians themselves are already producing the formal models of deductive reasoning in natural language. (What do you think that formal logic is?)

Derail.
You're talking of logic as "a formal logic". I'm not.
You should know since I made clear in my first post what I was talking about, in plain English, but do you understand English?
So, let's repeat myself:

"By science of logic, I mean a scientific investigation of logic as objective performance and manifest capability of human beings, investigation that would try to develop a formal model of logic which would be accurate and operational."​

See?
I'm uncertain as to why it is so hard for you guys to just read the bloody question. Well, yeah, obviously, reading ain't quite enough in and of itself.

So, learn your English, Sir. It's not what "I think" as you put it so gracefully, but English dictionaries variously defining logic, in the sense I use here, as "valid reasoning", "force or effectiveness in argument or dispute", "reasoned and reasonable judgement" etc.
Me, I say, logic "as objective performance and manifest capability of human beings". Performance such as indeed "valid reasoning", "force or effectiveness in argument or dispute", "reasoned and reasonable judgement" as the dictionaries say. It's not me who think up the stuff, Sir. Please learn to speak your bloody English before commenting.
Still, you can sure derail if you want to.
EB

 

[Speakpigeon]

Because it looks to me that all the aspects of what you are seeking are already there. There's empirical study of real life instances of reasoning, both in natural language and on the hoof, out in the field. Then there's attempts to elucidate the logic seemingly implicit in these instances, ultimately leading to formalization in models, to systems of mathematical logic and/or to computer algorithms. If this is somehow insufficient, why? What else do you want? What do you propose that psychologists, philosophers and logicians do instead?

So, in other words, you don't understand the question. You should spend more time on trying to understand the question and less time digressing into derails. Please read my replies to other posters. I can't cater for your personal needs.
EB

 

[Speakpigeon]

I have to wonder why SpeakPigeon posts here. He regularly posts questions, then when his ego is not sufficiently stroked, ends up telling people who reply to him how superior his intellect is to theirs. "Your comments have invariably zero value . . .you posted but didn't read properly . . .I'm not sure about what language you speak."

One wonders why he would post pseudoscience to a science forum and then become irate when it is discussed. Surely there is a forum somewhere that will stroke his ego to his satisfaction, and not expose him to the rigors and impediments of science.

And mob mentality now. Excellent.
I didn't post "pseudoscience". This is just slander. I posted a legitimate question. None of you understands the question and you all go into derails. I post here to get possible answers, but not from you, obviously. I don't remember any comment from you that showed you had understood the question. So, please, ignore me.
EB

 

[Speakpigeon]

It's a link to information for you about the field of cognitive science - your other approach of interest, remember?

The information about formal models of deductive logic are in the Wikipedia link, the field of metalogic. You apparently did not know what the word "metalogic" means, but that is explained in the link as well.

You expressed interest in two different approaches to human employment of logic. You now have introductory links to both, from me, and even better links from others once you have some basic familiarity with these fields apparently new to you. Happy reading.

You must be kidding.
EB

 

[Speakpigeon]

Most of your content seems to consist of claims that you're making grand discoveries about things no other human has ever considered before, but without actually providing any details of your "work".

This is pork pie. You're making stuff up. Most of my threads are questions or polls. I never ever talked about any "work" I would be doing, and I never claimed "grand discoveries". This thread is a good example of what I do here and where do you see any claim about "work" and "discoveries"? Where do you get this stuff from?
EB

 

[Speakpigeon]

Right, I'm done.
EB

 

[Yazata]

Derail.

You start threads by asking very interesting questions. I like that a lot. But instead of inviting everyone to do their best to cooperatively explore the questions that you raise, all of your threads seem to consist of you playing 'king-of-the-hill', battling anyone who dares to write anything at all, while insulting all and sundry. I hope that you are capable of seeing why that annoys everyone else. So why do you persist in doing it?

No, the cognitive scientists are producing formal models of real-life empirically-observed reasoning. The logicians themselves are already producing the formal models of deductive reasoning in natural language. (What do you think that formal logic is?)

You're talking of logic as "a formal logic". I'm not.

You should know since I made clear in my first post what I was talking about, in plain English, but do you understand English?

So, let's repeat myself:

"By science of logic, I mean a scientific investigation of logic as objective performance and manifest capability of human beings, investigation that would try to develop a formal model of logic which would be accurate and operational."​

See?

What I wrote was "The logicians themselves are already producing the formal models of deductive reasoning in natural language."

That seems to be precisely what you asked for in the text in which you repeated yourself. The formal logic is the formal models that you seek. What is being modeled is the logic of natural language reasoning.

I'm uncertain as to why it is so hard for you guys to just read the bloody question. Well, yeah, obviously, reading ain't quite enough in and of itself.

So, learn your English, Sir.

I read it and provided what I think is a pretty good response. If you disagree, then you need to do more than bluster. Explain why you disagree. Provide some convincing reason why readers should agree with you rather than me.

Me, I say, logic "as objective performance and manifest capability of human beings". Performance such as indeed "valid reasoning", "force or effectiveness in argument or dispute", "reasoned and reasonable judgement" as the dictionaries say. It's not me who think up the stuff, Sir. Please learn to speak your bloody English before commenting.

If you have ever studied philosophy in a university classroom, you must have been told about the problems inherent in trying to solve philosophical problems with dictionary definitions (which just report on common word usage among the general public).

But more to the point, how is what you wrote inconsistent with what I wrote? Informal logicians already examine natural language rhetoric (your "force or effectiveness in argument or dispute") and then try to abstract logical principles (your "valid reasoning" and "reasoned and reasonable judgement") from the things that people really say (your "objective performance and manifest capability"). Then the formal logicians try to compound these logical principles into formal mathematical-style deductive systems that ideally at least model informal natural language reasoning.

So everything that you think is lacking seems to me to already be there. Hence I think that the thread was begun on a false premise.

I think that you are angry because you once again are treating your thread as a battle of you-against-everyone, and you seem to be getting your ass kicked.

 

[billvon]

None of you understands the question and you all go into derails.

Thank you for proving my point.

So, please, ignore me.

Nope. Your posts - and inevitable angry, condescending rants - amuse me. And it's fun to watch you melt down when it becomes clear to everyone you are getting your ass kicked.

 

[arfa brane]

Why?

Why what?

Why does logic not have to make sense?

If it doesn't have this requirement then it follows that logic need not be empirical.
Meaning would be something humans require, when they're being logical.

p.s. one meaning I can therefore derive is that logic can indeed be all show, and no substance, a bit like this thread

 

[Write4U]

So, let's repeat myself:
"By science of logic, I mean a scientific investigation of logic as objective performance and manifest capability of human beings, investigation that would try to develop a formal model of logic which would be accurate and operational."

Lets begin by looking at a logical data model before trying to describe a formal model of logic.

A logical data model describes the data in as much detail as possible, without regard to how they will be physical implemented in the database.
Features of a logical data model include: all entities and relationships among them. All attributes for each entity are specified.

and

Logical schema,
A logical data model or logical schema is a data model of a specific problem domain expressed independently of a particular database management product or storage technology but in terms of data structures such as relational tables and columns, object-oriented classes, or XML tags.

Wikipedia

It seems to me that Logic is a function, similar to a mathematical function. If so, can Logic even be described other than as a compilation of data with or without a specific format?

 

[Speakpigeon]

But instead of inviting everyone to do their best to cooperatively explore the questions that you raise, all of your threads seem to consist of you playing 'king-of-the-hill', battling anyone who dares to write anything at all, while insulting all and sundry.

This is an egregiously false account. I responded many times, to you in particular but not only you, with a lengthy consideration of your point and a detailed reply.
Ad hominem justified on a very selective memory. Bravo!

I hope that you are capable of seeing why that annoys everyone else. So why do you persist in doing it?

The question rather is why these people persist in making mindlessly irrelevant or downright idiotic comments.
EB

 

[Speakpigeon]

What I wrote was "The logicians themselves are already producing the formal models of deductive reasoning in natural language."
That seems to be precisely what you asked for in the text in which you repeated yourself. The formal logic is the formal models that you seek. What is being modeled is the logic of natural language reasoning.
I read it and provided what I think is a pretty good response. If you disagree, then you need to do more than bluster. Explain why you disagree. Provide some convincing reason why readers should agree with you rather than me.

Perhaps you could try to provide a good response to begin with? Like, relevant?
OK, I do the analysis for you since you don't understand.

Here is a book from an "informal logician": https://books.google.fr/books?id=ZFCaBQAAQBAJ&pg=PA47&lpg=PA47&dq="informal logician"&source=bl&ots=-5aZgvDNeA&sig=ACfU3U33YlBy84vRBJ_QWqt_IuaWwkLjWA&hl=en&sa=X&ved=2ahUKEwi4n_bUpY7jAhVRAWMBHS8gAlwQ6AEwB3oECAcQAQ#v=onepage&q="deduction"&f=false

He contrasts "formal" and "informal" logic:

For the formal logician, validity is an essential element of the goodness of arguments and validity is frequently a function of the argument's form. For the informal logician, the cogency of an argument is not a function of its logical form. - The Rise of Informal Logic: ESSAYS ON ARGUMENTATION, CRITICAL THINKING ... By Ralph H. Johnson​

So, here we already move away from "validity", to "cogency". Cogency is a broad synonym of validity but not of logical validity. So, in effect, we don't know what "cogency" is supposed to mean here. The author also doesn't define "cogency" as they use the term so I have to assume they mean the ordinary dictionary definition, which is "convincing". However, convincing is not necessarily, and indeed most often not, deductive.
In effect, these guys are not even interested in deduction. The book contains only five times the word "deduction", and never to discuss it in the context of informal logic! Deduction appears all but irrelevant here.
They are only interested in the effectiveness, or "cogency", of "arguments" in general. For example, the same author says:

Argument schemes make reference to such substantive relations as analogy and causality, which are not definable in formal terms.​

And here we all told that informal logic, not surprisingly given the label, doesn't take form to be an operational criterion for "cogency". In other word, they think you can't generally use form to assess the "cogency" of the argument. And I agree of course since we're not talking about deductive arguments only but all sorts of arguments, some relying on analogy and causality for example. And this is apt for what they do because, whatever it may be, "the cogency of an argument is not a function of its logical form".
Somewhat confusingly, they are also only interested in logic as it manifests itself in language. However, their assessment of "cogency" takes into account the pragmatic of language, which definitely take us far away from deduction.
So, overall, just irrelevant. Interesting, probably, but irrelevant here.
Frankly, I'm not sure why you can't see that all by yourself. You provided a very definite and assertive reply. But it's just wrong. So, clearly, you haven't done the homework necessary to support your assertion.
Is that good enough, do you think, as a considerate reply?
You should try that.
EB

 

[Speakpigeon]

But more to the point, how is what you wrote inconsistent with what I wrote? Informal logicians already examine natural language rhetoric (your "force or effectiveness in argument or dispute") and then try to abstract logical principles (your "valid reasoning" and "reasoned and reasonable judgement") from the things that people really say (your "objective performance and manifest capability"). Then the formal logicians try to compound these logical principles into formal mathematical-style deductive systems that ideally at least model informal natural language reasoning.

Valid reasoning.
Not "cogent argument", whatever that is.
EB

 

[Speakpigeon]

Thank you for proving my point.

Nope. Your posts - and inevitable angry, condescending rants - amuse me. And it's fun to watch you melt down when it becomes clear to everyone you are getting your ass kicked.

Remind me when I'm melting down. That's surely something interesting to see with my own eyes.
You like to tell stories...
To yourself?!
EB

 

[exchemist]

Right, I'm done.
EB

Oh, really?

 

[billvon]

Remind me when I'm melting down.

When your replies become more than 50% boasts and insults, and no longer address the topic. Usually happens within about 10 posts.

 

[Yazata]

Perhaps you could try to provide a good response to begin with? Like, relevant?
OK, I do the analysis for you since you don't understand.

You aren't going to convince other people to agree with you unless you make them want to agree. Turning all of your threads into a zero-sum game, into an ego-contest in which you feel that you have to win and everyone else is demeaned, will only make your task much harder. It's psychologically disfunctional behavior.

You start out by asking interesting and thought-provoking philosophical questions. I like that a lot and thank you for doing it. But then you immediately force your own threads off the rails by attacking anyone else who posts to them. You should be trying to encourage friendly and cooperative exploration of philosophical problems, where everyone is encouraged to try their hand in contributing (even if they have no formal education in the subject) and where criticism of other people's ideas is gentle and sympathetic.

Here is a book from an "informal logician": https://books.google.fr/books?id=ZFCaBQAAQBAJ&pg=PA47&lpg=PA47&dq="informal logician"&source=bl&ots=-5aZgvDNeA&sig=ACfU3U33YlBy84vRBJ_QWqt_IuaWwkLjWA&hl=en&sa=X&ved=2ahUKEwi4n_bUpY7jAhVRAWMBHS8gAlwQ6AEwB3oECAcQAQ#v=onepage&q="deduction"&f=false

Ralph Johnson is a proponent of what we might call the 'Windsor program' (referring to the U. of Windsor in Canada) which takes a rather strong view of what 'informal logic' is and should be. It's controversial.

He contrasts "formal" and "informal" logic:

For the formal logician, validity is an essential element of the goodness of arguments and validity is frequently a function of the argument's form. For the informal logician, the cogency of an argument is not a function of its logical form. - The Rise of Informal Logic: ESSAYS ON ARGUMENTATION, CRITICAL THINKING ... By Ralph H. Johnson​

The Windsor program seems to me to want to think of informal logic as an entirely separate kind of logic, alongside but distinct from deductive logic. I don't totally disagree, but think that idea is probably too strong. I think that informal reasoning in natural language needn't mean reasoning entirely without logical form or where logical form is irrelevant. I believe that informal logic might in principle be formalized in a formal language, even if it isn't delivered that way originally in ordinary language. Successfully performing that formalization might require extensions and modifications to the formal language such a new logical operators. (We've seen that with things like modal, tense or epistemic logics.)

I think that I would prefer to think of informal logic as (Johnson p.11) "a focus on the actual natural language arguments used in public discourse" (your "objective performance and manifest capability of human beings"). Real life examples of human reasoning can obviously be rational, but that logic isn't formalized like a proof in an artificial formal-mathematical language. Focusing on it is precisely what you complain in your OP that nobody is doing. So you seem to me to be wrong about that.

I even agree with Johnson (p.11) in sharing "serious doubt about whether deductive and standard inductive logic approaches are sufficient to model all, or even the major, forms of legitimate argument". But in my opinion, perhaps that should read 'insufficient as things currently stand'. I don't want to insert an unjustified hidden premise that they will never be sufficient, or that they are somehow insufficient in principle. That's why I emphasized extensions to classical formal logic above and in earlier posts. The history of 20th century logic has been in some large part the history of exactly this.

I also agree with Johnson (p.11) that informal logic embodies "an interest in expanding the study of reasoning to include, besides argumentation even broadly conceived, such activities as critical thinking, problem solving and decision making". That's where it starts to approach cognitive science.

I think that this short quote captures the motivation behind informal logic quite well (Perelman and Albrechts-Tytica The New Rhetoric - Treatise on Argumentation, 1969 U. Notre Dame Press p. 10)...just think Frege and Russell and their interests. Think of mathematical logic as it's taught today.

"Logic underwent a brilliant development during the last century when... it set out to analyze the methods of proof effectively used by mathematicians. Modern formal logic became, in this way, the study of the methods of demonstration used in the mathematical sciences... Logicians owe it to themselves to complete the theory of demonstration obtained in this way by a theory of argumentation."

It's simply a fact that the vast majority of real life reasoning that we all engage in constantly doesn't take the form of chains of formally valid deductions like we would encounter in a mathematical proof. Yet we typically believe that what we are doing might nevertheless constitute examples of good reasoning. Whatever is happening, it deserves the study that you claim nobody is giving it.

So, here we already move away from "validity", to "cogency". Cogency is a broad synonym of validity but not of logical validity.

It's broader, but it still probably includes logical validity as one of its most important components. There's going to be other things besides, such as induction, abduction, analogical reasoning and whatever else they find in there.

And I agree of course since we're not talking about deductive arguments only but all sorts of arguments, some relying on analogy and causality for example. And this is apt for what they do because, whatever it may be, "the cogency of an argument is not a function of its logical form".

Ok. But I don't want to exclude logical form from the things that constitute cogency.

Somewhat confusingly, they are also only interested in logic as it manifests itself in language. However, their assessment of "cogency" takes into account the pragmatic of language, which definitely take us far away from deduction.

So, overall, just irrelevant. Interesting, probably, but irrelevant here.

How is it irrelevant? You started this thread complaining what "science" is ignoring real life reasoning as an "objective performance or manifest capability of human beings". Now you are complaining that although they are in fact considering it, they aren't doing that in the way you would prefer (which you have yet to explain and try to defend) and that unless they do things your way, it's all irrelevant.

Frankly, I'm not sure why you can't see that all by yourself.

It still looks to me like you began this thread on a false premise and now you're scrambling to move the goalpost.

 

[arfa brane]

I've tried, but failed to manage to exclude computational science as a science of logic.

I mean, duh.

In this discipline, there isn't really any notion of "performance"; as I stated, logic doesn't perform.

However, if you want an efficient algorithm, or a computer that doesn't waste a lot of energy (things that make sense), then perhaps you want to invoke a design logic which is about performance, in which context "efficient logic" makes sense.

And what kind of statistic is there in giving a group of programmers the job of designing and implementing the same algorithm, then comparing the results?

 

[Speakpigeon]

I think that I would prefer to think of informal logic as (Johnson p.11) "a focus on the actual natural language arguments used in public discourse" (your "objective performance and manifest capability of human beings"). Real life examples of human reasoning can obviously be rational, but that logic isn't formalized like a proof in an artificial formal-mathematical language. Focusing on it is precisely what you complain in your OP that nobody is doing. So you seem to me to be wrong about that.

I used the phrase "objective performance and manifest capability of human beings" to explain which sense of the word "logic" I was using, not to suggest public discourse was the place to look to investigate logic.
I seems clear to me that if you go into "the actual natural language arguments used in public discourse", you're not going to focus on deduction and on any formal model of deduction. It's all good and well that they should do it, but that's not reason not to do the science of deduction.
Language is not essential to deductive logic. In fact, using language is lumbering our ability to make deduction, which explains why many people are apparently bad at logic. They are not in fact bad at logic, they are bad at doing deduction using language. So, investigating deduction by looking at language would require an awareness of this fundamental difficulty.
Any argument whose conclusion follows necessarily from the premises, even left implicit, is deductive. Any argument whose conclusion doesn't follow necessarily from the premises is not deductive, and as such, I don't see why we would call that a logical argument at all. So, "cogency" is a good term to refer to something these people think has not much to do with logic. Something you are trying to glaze over.

I even agree with Johnson (p.11) in sharing "serious doubt about whether deductive and standard inductive logic approaches are sufficient to model all, or even the major, forms of legitimate argument". But in my opinion, perhaps that should read 'insufficient as things currently stand'. I don't want to insert an unjustified hidden premise that they will never be sufficient, or that they are somehow insufficient in principle. That's why I emphasized extensions to classical formal logic above and in earlier posts. The history of 20th century logic has been in some large part the history of exactly this.

Sure, logic is absolutely necessary to any argument whatsoever but these people are not focusing on that. To the extent that they will need any formal logic at all, they will be happy to use mathematical logic without making any effort to validate it on empirical basis.

I think that this short quote captures the motivation behind informal logic quite well.

I agree but we are obviously not reading the same interpretation into it.

"Logic underwent a brilliant development during the last century when... it set out to analyze the methods of proof effectively used by mathematicians. Modern formal logic became, in this way, the study of the methods of demonstration used in the mathematical sciences... Logicians owe it to themselves to complete the theory of demonstration obtained in this way by a theory of argumentation."

And this put the last nail into the coffin of your claim. This shows they accept mathematical logic to be a proper model of logic, and thus they will have no motivation to make any effort to look into whether this model agrees with the empirical evidence.

It's simply a fact that the vast majority of real life reasoning that we all engage in constantly doesn't take the form of chains of formally valid deductions like we would encounter in a mathematical proof. Yet we typically believe that what we are doing might nevertheless constitute examples of good reasoning. Whatever is happening, it deserves the study that you claim nobody is giving it.

Again, I'm not talking about logic as formal model. I mean logic as performance of humans and capability of the human mind. The formal model will come as a result of investigating logic in this sense.
Any effective linguistic performance requires deductive capability, so even when people produce bad argument, they are still using their deductive capacity efficiently. The result may be bad, but we already know that deduction doesn't guaranty the truth of all conclusions.

It's broader, but it still probably includes logical validity as one of its most important components. There's going to be other things besides, such as induction, abduction, analogical reasoning and whatever else they find in there.

Sure, but you won't find any indication that these people are going to do any scientific investigation of our deductive performance to produce a formal model of logic. They will defer to mathematical logic instead.

It still looks to me like you began this thread on a false premise and now you're scrambling to move the goalpost.

If that's what you think, why are you even bothering?
EB