Philosophy Forums
Forums Articles Gallery Links


Register Login Members Calendar New Posts Search Chat Help
Philosophy Forums » Philosophy of Science » Science » Can reason defend itself? » Can reason defend itself?

Can reason defend itself?

PrintPrint


Page: 1 2 3 4 5 6 7 8 9

Can reason defend itself?
noself
Initiate
Avatar

Usergroup: Members
Joined: Aug 07, 2007
Location: Manchester, UK

Total Topics: 3
Total Posts: 10
Posted 08/16/07 - 01:40 AM:
quote post
#71
But surely reason is only needed in a subjective sense, precisely because we lack an objective view? The knife analogy is an example of self-reference. What can be applied to itself without a resulting paradox?
jwdink
Johnnie

Usergroup: Members
Joined: Jul 20, 2007
Location: Maryland

Total Topics: 10
Total Posts: 428
Posted 08/17/07 - 12:38 PM:
quote post
#72
But surely reason is only needed in a subjective sense, precisely because we lack an objective view?




Right, but there's no way to justify reason as the best method towards that objective view.
Taffer
Assistant Professor

Usergroup: Members
Joined: Jan 29, 2007

Total Topics: 5
Total Posts: 254
Posted 08/20/07 - 01:53 AM:
quote post
#73
NoSoul wrote:

Is there anything wrong, in your opinion, with justifying reason on intuitionist grounds? 


I don't know what to do with a question like that, and that is the problem. The whole point of my post was that the notion of "justification" needs to be laid out clearly (i.e. based on what principles?). The unsatisfactory nature of many debates about this type of thing is precisely due to insufficient clearness in this regard. Under a certain interpretation of "on intuitionist grounds", the answer to your question is "trivially, no". But you would have to explain what you mean.
NoSoul
Economic determinist
Avatar

Usergroup: Members
Joined: Feb 05, 2004

Total Topics: 25
Total Posts: 610
Posted 08/21/07 - 12:06 AM:
quote post
#74
Earlier you said that reason is hardwired into our biology. That was your essential defense of it. I agree with you there, but the point is, "phenomenologically" -- mentally, in our minds rationally, as we perceive it -- the mental act of deciding whether to use reason, or not, simply comes naturally to us, appropos of nothing other than...it's biologically hard-wired into us. Phenomenologically, this is equivalent to appealling to simple intuition to justify our use of reason, over other possible mental acts (further intuition, raw guessing, throwing dao de ching sticks, whatever) to make decisions.

The fact that you already admitted, as your base justification, that reason is biologically hardwired into us, is practically the same as admitting there's no further "rational" reason, only intuition, which justifies the use of reason. Please address this fact the next time you respond to me, otherwise please don't bother responding to me. Thank you.
Firmly tie the mind, resembling a mad elephant, to the strong pillar of its perceptual content, with the rope of contemplative inspection, and gradually tame it with the hook of discrimination.

-- Buddhist Handbook, Salamander Press

To the poet and sage, all things are friendly and hallowed, all experiences profitable, all days holy, all men divine. - Nietzsche/Emerson
Taffer
Assistant Professor

Usergroup: Members
Joined: Jan 29, 2007

Total Topics: 5
Total Posts: 254
Posted 08/21/07 - 02:47 AM:
quote post
#75
NoSoul wrote:

The fact that you already admitted, as your base justification, that reason is biologically hardwired into us, is practically the same as admitting there's no further "rational" reason, only intuition, which justifies the use of reason. Please address this fact the next time you respond to me, otherwise please don't bother responding to me. Thank you.


It was a statement of fact. What you can draw from it is up to you.

I have said already, this talk of "justifying reason" doesn't make sense. To make it make sense, you would have to say what you mean by "reason" (I would suggest: the principles of abstract reasoning which are obvious to us, and their consequences), and then explain "justify" (i.e. say what principles one is allowed to make use of).

For example, if you restrict "reason" to "logico-mathematical reasoning" then the matter is easily settled.

If you want to make things more general than that, I don't think I can help you.
Lee Kelly
Ametuer Philosopher
Avatar

Usergroup: Members
Joined: Jun 29, 2007
Location: England

Total Topics: 1
Total Posts: 4
Posted 08/23/07 - 02:39 AM:
quote post
#76
jwdink wrote:
Can reason defend itself?

Wrong question. The question should be: can reason criticise itself?

This is an little piece I wrote recently ago which indirectly address this problem, and a response to another concerning how to solve the problem itself.

I hope this helps some...

The Problem of problematicality

This short essay concerns what I call the problem of problematicality. The problem addresses the standards and rules we employ when identifying problems. In other words, how we answer the question: when is a theory problematic? It is important to note, that by the word 'theory' I mean any theory, hypothesis, statement, observation, proposition, idea, axiom, assumption, guess, presupposition, conjecture, position, thought, presumption, suggestion, concept, notion, definition, intuition, etc.

I believe the problem of problematicality is a central problem of rationology i.e. the study of rationality, and that different theories of rationality can be broadly understood as different responses to this problem. The two to be considered here are justificationism and critical rationalism (hereafter: 'criticalism'). The purpose of a theory of rationality is to provide a context in which rational debate can take place, which allows for the identification of problems and solutions, so that we may demarcate between good theories and bad theories. (It is also worth noting that even irrationalists have a theory of rationality, or else how would they recognise themselves as irrationalists?).

I therefore suggest that every theory submitted for rational discussion is assigned a problem-value, either unproblematic or problematic, but not both, and not neither, nor are there any other problem-values. I think problem-values can be understood to behave very much like truth-values, though are not intended to replace or determine the assignment of truth-values. This however, raises a question. If every theory submitted for rational discussion is assigned a problem-value, then what value do we assign to a theory that has been submitted, but not yet scrutinised? or to put it another way: what is the default problem-value?

How this question is answered has profound consequences for the resulting theory of rationality, for what makes a theory problematic, the goals of rational discussion and the results that follow.

(1) we may consider a theory of rationality which by default assigns the value problematic to every theory submitted into rational discussion. In such a theory of rationality, any theory will be problematic until it is made unproblematic. The goal of rational discussion is to identify some criterion or rule, by which we might make some theories unproblematic, and justify them.
(2) we may consider a theory of rationality which by default assigns the value unproblematic to every theory submitted into rational discussion. In this theory of rationality, any theory will be unproblematic until it is made problematic. The goal of rational discussion is to identify and apply standards of criticism, so that competing theories are made problematic by a survival of the fittest.

The first response to our question can be represented by justificationism, and the second by criticalism. It can be shown, by applying the two theories of rationality to the same statement, different problem-values are returned. In other words, justificationism and criticalism have different answers to the question: when is a theory problematic? To demonstrate, we can take the problem of induction, which originally arose in response to statements like "the sun will rise tomorrow" or "all swans are white."

In the context of justificationism, the statement "the sun will rise tomorrow" is by default problematic, and so must be made unproblematic. To that end a justificationist uses a series of tests, which if passed successfully, provide justification for theories, thus making them unproblematic. Traditionally, the tests used to justify theories have been a check for logical validity and a check for empirical observation.

The statement "the sun will rise tomorrow" passes the check for logical validity, but does not pass the check of empirical observation. No matter how many times we observe the sun rise, it does not follow that the sun will rise tomorrow. In other words, our criterion, or criterions, which were intended to provision justification, thus making an problematic theory (by default) into an unproblematic theory, are logically inadequate. The statement "the sun will rise tomorrow" must remain problematic.

In the context of criticalism, the statement "the sun will rise tomorrow" is by default unproblematic. To that end a criticalists uses a series of tests, which if not passed, successfully criticise theories, thus making them problematic. The tests used to criticise theories are the same as those used to justify, but used differently.

The statement "the sun will rise tomorrow" passes the check of logical validity as in justificationism, but then also passes the check of empirical observation. The difference is that where the justificationist was trying to make the theory unproblematic by deriving it from sense observation, the criticalist is trying to make the theory problematic by falsifying it by sense observation. In consequence, it follows that the statement "the sun will rise tomorrow" is not problematic, and so there is never any need to postulate a logic of induction, or principle of induction in the first place. If there is no need to postulate induction then there is no problem of induction!

It is important to realise that this short essay is not comprehensive. I have omitted many interesting problems and questions which arise here, simply for clarity and brevity. Moreover, the problem of induction is only the tip of the iceberg in terms of the differences which can be exposed between justificationism and criticalism. This essay is intended to simply introduce the problem of problematicality, and its importance to any theory of rationality.

Lee Kelly wrote:
That is the great thing about criticalism: it can be applied to itself.

Imagine we submit logic, sense observation, or justificationism itself to rational discussion. In the context of justificationism they are now problematic until made unproblematic, but with what are we to make them unproblematic? If I use logic, sense observation, or justificationism, isn't that just circular?

Though it is rarely expressed in this form, that argument is the root of almost all relativism. For a very common take, check out this thread on PhilosophyForum. If you follow the reasoning of members on that forum, they all assume the justificational theory of rationality, and it shapes the problems that they confront.

If we perform the same procedure for criticalism, and submit logic, sense observation, and criticalism itself to rational discussion, they are by default unproblematic until made problematic. The crucial difference is that we can criticise them without circularity! It is quite possible to refute a philosophy, system of logic, sense experience, or whatever, if it fails to pass its own standards.

The critcal theory of rationality solves a long standing problem of relativism, and the apparent need to commit to some foundational presupposition, even for a rational. In other words, with criticalism we can defeat the charge that we need to have a faith in reason.



Edited by Lee Kelly on 08/27/07 - 02:36 PM
"Every man, wherever he goes, is encompassed by a cloud of comforting convictions, which move with him like flies on a summer day." - Bertrand Russell
Nonblack Raven
Tenured Poster
Avatar

Usergroup: Sponsors
Joined: Jun 19, 2004

Total Topics: 47
Total Posts: 1240
Posted 08/26/07 - 05:52 PM:
quote post
#77
Lee Kelly, one of my personal frustrations with this board is that one can outline a somewhat unusual viewpoint; await responses with interest—and get no response whatsoever.

Of course, another frustration is to get responses that completely fail to comprehend what one is saying. Let me see if I can, at a minimum, move your rather interesting post from the first to second category.

My first question is what exactly are you calling a theory? Is any proposition a theory? Or only propositions across a set of instances, perhaps infinite, not all of which are known?

My second question is why exactly, as I understand it, do you allow only two categories for theories—problematic and unproblematic? It seems to me a lot more categories would be appropriate. For example, there are theories that are unproblematic, theories that are problematic, and theories that are demonstrably false. Alternatively, reflecting my Bayesian sympathies, theories have a wide variety of categories--perhaps a measurable subjective probability, perhaps some less quantitative measure. Why limit theories to problematic or unproblematic?

If I understand you correctly, you seem to be saying a theory in empirically unproblematic if there is no empirical evidence that refutes it. How does this differ from a Popperian idea of falsified and unfalsified theories?

Let us suppose that we grant, if I understand you, that theories that are not refuted by empirical data are unproblematic. Consider theory P and theory Q, where P implies not Q and Q implies not P? (Though there are better examples, consider as an example, if you wish, that P is the theory that all emeralds are green and Q is the theory that all emerald are green first observed before some future date are green and all observed after that date are blue,) Both theory P and theory Q have been proposed—does this not make both theories problematic? Whether the answer is yes or no, how is one to best choose between P and Q?
NBR
jwdink
Johnnie

Usergroup: Members
Joined: Jul 20, 2007
Location: Maryland

Total Topics: 10
Total Posts: 428
Posted 08/27/07 - 09:12 AM:
quote post
#78
Lee Kelly, one of my personal frustrations with this board is that one can outline a somewhat unusual viewpoint; await responses with interest—and get no response whatsoever.


I've noticed that too, and it can be very frustrating, yet now I see myself participating in it. I was planning on responding, and I still am, but I've been pretty busy.

My apologies to Lee Kelly.
Lee Kelly
Ametuer Philosopher
Avatar

Usergroup: Members
Joined: Jun 29, 2007
Location: England

Total Topics: 1
Total Posts: 4
Posted 08/27/07 - 02:30 PM:
quote post
#79
Hi Nonblack Raven,

Thanks for the input. I'll try my best to answer your queries adequately. Though I should note that much of what I will write, and have written, is new even to me. I have only recently began to consider problematicity as a subject of investigation, and my views should be considered tentative--much in need of peer review, so to speak.

Nonblack Raven wrote:
My first question is what exactly are you calling a theory? Is any proposition a theory? Or only propositions across a set of instances, perhaps infinite, not all of which are known?

The choice of the word 'theory' may be inappropriate. I realise that it is not usually meant in the way that I intend here. I literally use it as a variable, which can stand for any position, proposition, presupposition, presumption, assumption, axiom, theory, hypothesis, conjecture, statement, observation, suggestion, speculation, concept, definition, intuition, guess, idea, notion, instinct, habit, conviction, etc. In other words, any idea whatever.

Unfortunately, I do not understand what you mean by "only propositions across a set of instances, perhaps infinite, not all of which are known," though am inclined to say yes, since I am trying to use the word 'theory' very broadly, to capture as much as I can within its net. As I have said, the word 'theory' may be inappropriate, and the choice is tentative. If you believe a different word would be more appropriate, I'd be interested to hear your suggestion.

My second question is why exactly, as I understand it, do you allow only two categories for theories—problematic and unproblematic? It seems to me a lot more categories would be appropriate. For example, there are theories that are unproblematic, theories that are problematic, and theories that are demonstrably false. Alternatively, reflecting my Bayesian sympathies, theories have a wide variety of categories--perhaps a measurable subjective probability, perhaps some less quantitative measure. Why limit theories to problematic or unproblematic?

Good question. I limit theories to being problematic or unproblematic because I am trying to cut across differences in philosophies, to identify presuppositions that they all share. It occurs to me that a philosophy might designate a theory as problematic becuase it is untrue, unjustified, untestable, unclear, etc. and any given philosophy may identify different theories as problemtic for different reasons, such as you suggested with the Bayesian.

However, it is precisely those different reasons to identify a theory as problematic, (and by implication, what would count as a solution) which forms the traditional battleground for philosophic debate. In other words, philosophers, even where they disagree most strongly on what makes a theory problematic, and what kind of solution counts, implicitly accept that theories are either problematic or unproblematic.

I hope that given this context, you might preempt my response to your "demonstratably false" category. In short, I consider such a category to be a reason to identify a theory as problematic, not as an alternative category to 'problematic.' If a theory does not satisfy the test of logic i.e. is inconistent, then it is problematic.

I realise that I use the word 'problematic' in a way which diverges from ordinary usage, but I could not think of an alternative which captured the meaning I was after, so I hope this brief response has helped clarify my intention.

If I understand you correctly, you seem to be saying a theory in empirically unproblematic if there is no empirical evidence that refutes it. How does this differ from a Popperian idea of falsified and unfalsified theories?

I am tempted to respond "it doesn't," but should be cautious, I have learned that what I mean by the 'Popperian falsification" and what others mean by "Popperian falsification" are rarely the same. In fact, I'd prefer to avoid all mention of Popper himself, since it tends to detract from the problem which I am adressing, and become bogged down in discussion regarding what various philosophers did and did not think--a usually fruitless and irrelevent task.

Let us suppose that we grant, if I understand you, that theories that are not refuted by empirical data are unproblematic. Consider theory P and theory Q, where P implies not Q and Q implies not P? (Though there are better examples, consider as an example, if you wish, that P is the theory that all emeralds are green and Q is the theory that all emerald are green first observed before some future date are green and all observed after that date are blue,) Both theory P and theory Q have been proposed—does this not make both theories problematic? Whether the answer is yes or no, how is one to best choose between P and Q?

smiling face

I ended my short essay like this:

"It is important to realise that this short essay is not comprehensive. I have omitted many interesting problems and questions which arise here, simply for clarity and brevity. Moreover, the problem of induction is only the tip of the iceberg in terms of the differences which can be exposed between justificationism and criticalism."

The question you pose, (the grue-problem of induction), is one such "interesting problem" which I am referring to. I also believe it can be solved within the context of criticalism, though to explain how would require a more in-depth discussion regarding criticism, and how criticism is conducted. The essay I posted was not intended to dive to such depths.

However, I will briefly outline the solution here, (I do not have time for a throrough explanation). In short, criticism can take many forms. I usually distinguish between four kinds: logical, empirical, methodological, and rhetorical. It is methodological criticism which is important here, since it allows us to assess theories in the context of background knowledge i.e. solutions to old problems.

If the rationalist holds to the rule that only problemtic theories are in need to revision, then he may entertain the theory "all emeralds are grue," he may even test it when the time comes, but it should not cause him any problem or difficulty by itself. There are always an infinite number of logically consistent theories which fit the facts, but the rationalist need not be peturbed by that, since the mere possibility that his preferred theory might be false does not constitute a criticism of that theory, and therefore does not render it problematic.

If the theory "all emeralds are green" is not problematic, then the theory "all emeralds are grue" does not solve a problem. It is an interesting possibility, to be sure. However, a rationalist (as conceived by criticalism) will not be compelled to see this alternative possibility as a problem. I hope that makes sense, though it is jumping the gun a bit with respect to the topic of my original post.

Regards,
Lee
"Every man, wherever he goes, is encompassed by a cloud of comforting convictions, which move with him like flies on a summer day." - Bertrand Russell
Nonblack Raven
Tenured Poster
Avatar

Usergroup: Sponsors
Joined: Jun 19, 2004

Total Topics: 47
Total Posts: 1240
Posted 08/27/07 - 06:47 PM:
quote post
#80
To jwdink and lee Kelly, I am enormously fascinated with epistemological questions, but I always turn to, what seems to me, a pragmatic criterion for evaluating epistemological approaches—if I have to make a decision, what view, or set of uncertainties abut various views, should I adopt.

This is partly my problem with jdwink’s views:

jwdink wrote:
NBR, I guess I see where our misunderstanding took place. We just have very different things in mind when we speak of "reason". To yours, faith and axioms just aren't very good descriptors. I'm just not sure whose idea of reason is more accurate.

Reason as "a strategy in approaching a situation" might not conflict too badly with my conclusion; it just subverts my argument for that conclusion. After all, if this strategy can vary so much, its not particularly objective, is it? Then again, I think you're really arguing that it just depends on the situation, but in each situation, there is an optimal approach. In this case, however, I can continue to be stubborn and insist that there's no way of being certain that a reasoning strategy will be optimal in a particular situation. Thus, how is the reasoning strategy you choose for a particular situation justified? What could we base this justification on?

Eh?


So let me put this as simply as possible, let us suppose we actually have to make a decision. Shall we make it based upon reason, or on—what exactly? If something other than reason, why exactly do you think this is superior to reason? And what is your basis for thinking this?

No, I do not think that there are optimal positions. Instead, I subscribe to what, for lack of a better term, I will call incremental rationalism.

Given I have to make a decision; I will use the best decision making systems available, as tested by the best means available. What exactly is your alternative that is demonstrably superior?

Note the question for pragmatist is not is decision making method D subject to doubt, but rather is decision making method D demonstrably inferior to some other decision making method D’. You seem to be happy to assert decision making D is subject to doubt, but fail to assert that anything else is demonstrably superior.

But this totally fails to answer the utterly rational pragmatic question—given I have to make a decision, what is the best method to use?

Now, jwdink, you have several times in this thread stated that an argument that would render my points irrelevant—e.g that you are incapable of being anything other than reasonable. Whatever the case for you personally, there is a vast psychological literature showing that most persons, even those with courses in probability and statistics, or in propositional and predicate logic, routinely make ridiculous mistakes in both. In short, nobody is stuck with either or incapable of denying either. So my question to you is: Do you endorse these mistakes as just as likely to be correct as reasoning that does not make these mistakes? One way or the other, but not, it doesn’t matter, or they are not provable. We all have to make decisions—what methods shall we use and why?

Now let me turn to Lee Kelly’s views. Now as nearly as I can tell, you view a vast varieties of theories as falling into one of two categories, unproblematic or problematic. Now let me ask my pragmatic questions—

1. Are all unproblematic theories equivalent for the purposes of possible actions? Would you really be prepared to act equally happily on any unproblematic theory? If not, why are they equivalent?

2. Are all problematic theories equivalent for the purposes of possible actions? Would you really be prepared to act equally happily on any problematic theory? If not, why are they equivalent?

3. Are all unproblematic theories inferior, for purposes of actions to problematic theories? Would you always prefer an unproblematic theory to a problematic theory?

If the answer to any of these questions are no, why exactly is your theory, whatever its epistemological advantages a sound epistemological theory?

In short, an epistemological theory, whatever problems it solves, that yields absurd results as to what theories one should adopt for decision making purposes, is, to my mind, a really bad epistemological theory.
NBR
Download thread as

Page: 1 2 3 4 5 6 7 8 9



Sorry, you don't have permission to post. Log in, or register if you haven't yet.

Powered by WSN Forum

The time is now 11/22/09 - 05:56 PM.