Occam’s Razor

[Batman Jr.]

Thanks for the exercise Batman Jr, and please excuse my lack or rigor in the following ...

The 'inside of the box' does exist, because -
We can observe the exterior (or description) of the box.
And from experience we know how boxes are constructed,
and deduce that the box has an interior,
as all other boxes in our experience do have an interior.
Another way of saying this is that, by definition, a box has an interior.
My dictionary tells me this.
Even though it is mandated that we cannot observe its interior, we can still imagine a scenario where it can be observed.
By calling it a 'box', we have decided the matter.
If its interior does NOT exist then this thing is NOT a box.

You can't "deduce" that the box has an inside because the box obviously is not an ordinary box; the normal rules behind boxes don't apply. All other boxes in our experience have had the potential of being broken into. I could, using your logic, "deduce" that since it is a "box," it can also be broken into. The problem is that you are using "induction" and not "deduction." The other problem is that we don't know if you are really inducing that the box has an interior from the inherent properties of boxes or instead from the fact that the existence of interiors can be related to the breaking into of whatever is enclosing them. Since the observations on which the former induction was made can be subsumed into the latter induction, believing there to be an inherent property of boxes that gives them an interior would be unnecessarily complicating inductively derived rules and thus, believe it or not, in violation of Occam's Razor (the real one, not the fake one I've been criticizing).

Our confidence in the existence of the interior is proportional to the match between the observation (or description) of the box, with the other boxes in our experience, ie, how closely it resembles the box described in our dictionary.

If the thought experiment is reworded slightly to read '... whatever is inside the box don't exist?' (this may actually have been the meaning but I did not read it as such) then similarly, the 'whatever is inside of the box' does exist, because -
Whatever is inside the box has the potential to be observed.
Even though the experiment says that it cannot be observed, one can still imagine a scenario where it can be observed.
The experiment mentions nothing else special about this box, except that we cannot observe its interior, so (WRT its ability to have an interior and contain things) it is the same as the box described in our dictionary.
From experience and definition, all boxes have interiors and can contain things.

There is no reason or argument to propose that the contents of any box (including this particular one) does not exist.

(gee - i hope i was allowed to use the terms 'experience' and 'by definition' in this)

Thanks again for the excersise.
PS, I am in strong agreement with Melendwyr's statement.

The "scenario where it could be observed" argument is similar to my contention against Melendwyr's rejection of the existence of the inside of the box in that I posit that it is possible that there are things inside the box interacting with one another, but not the outside world. In that way, they would exist by Melendwyr's "ontological definition of interaction" but would be untestable and unobservable.

 

[Flange Desire]

... so we might even allow it to contain a theoretical perfect vacuum and be truly empty. ...

Harrumphh!
A vacuum is measurable, and so the existance of a vacuum inside a box can be observed.
A vacuum is a little different from 'the inside of the box doesn't exist'.

 

[Flange Desire]

You can't "deduce" that the box has an inside because the box obviously is not an ordinary box; the normal rules behind boxes don't apply.....

Oh OK.
I didn't know that YOUR box did not conform to 'the [other] normal rules behind boxes', as you did not state that.

Thanks again for the excersise.

 

[Batman Jr.]

Oh OK.
I didn't know that YOUR box did not conform to 'the [other] normal rules behind boxes', as you did not state that.

Thanks again for the excersise.

I did state that because the box cannot be broken into. All other boxes have the property of possibly being broken into. Therefore, you don't know if the existence of interiors in other boxes is related to the induced rule that "things which can be broken into have interiors" or "boxes have interiors." Since the first rule explains all observations and doesn't require the second, you are unjustified in saying that there is a property of boxes that gives them interiors. You sounded sort of sarcastic, so I thought I might try to restate in different terms what I said previously.

 

[DevilsAdvocate]

Another way of saying this is that, by definition, a box has an interior.

I agree. I think Batman Jr has made a good point. But the question should not be whether the inside of the box exists, but whether the "contents of the box" exist. We can speculate that the box contains gold, ancient manuscripts, nothing, anti-matter, a pink unicorn, a perpetual motion machine, etc. How do we apply Occam's razor here? Should we conculde that because the box could contain anything or nothing and that because anything would be more complex than nothing that the box therefore contains nothing--that the "contents of the box" do not exist? No. Because any thing would be as likely as no thing. The contents of the box are "unknown".

Before we can even apply Occams razor, we would have to know "somehting" about the contents of the box in order to formulate hypothesis that could be cut by the razor. We have no evidence of any properties of what is in the box. Therfore, it could be "any thing" (which includes "no thing"). Because the contents of the box (if any) do not affect anything outside the box, there is no phenomena that can be measured, therfore it does not require any explaination. Therefore you have no set of explaination to which Occam's raazor could be applied. Something may or may not exist in the box, but it doesn't matter because the box can't be opened and the contents of the box don't affect anything.

 

[Batman Jr.]

I agree. I think Batman Jr has made a good point. But the question should not be whether the inside of the box exists, but whether the "contents of the box" exist. We can speculate that the box contains gold, ancient manuscripts, nothing, anti-matter, a pink unicorn, a perpetual motion machine, etc. How do we apply Occam's razor here? Should we conculde that because the box could contain anything or nothing and that because anything would be more complex than nothing that the box therefore contains nothing--that the "contents of the box" do not exist? No. Because any thing would be as likely as no thing. The contents of the box are "unknown".

Before we can even apply Occams razor, we would have to know "somehting" about the contents of the box in order to formulate hypothesis that could be cut by the razor. We have no evidence of any properties of what is in the box. Therfore, it could be "any thing" (which includes "no thing"). Because the contents of the box (if any) do not affect anything outside the box, there is no phenomena that can be measured, therfore it does not require any explaination. Therefore you have no set of explaination to which Occam's raazor could be applied. Something may or may not exist in the box, but it doesn't matter because the box can't be opened and the contents of the box don't affect anything.

This is an important point; it's also a more apt analogue for actual uses of Occam's Razor because I was originally attacking how people use Occam's Razor and then default to the contingency of "nothingness" for the unknown variable. The problem is that "nothingness" is also a value for the unknown variable. We don't know that "nothingness" is more likely than any other description of the unknown, so it's best to say that we just don't know the state of the indefinable variable.

 

[DevilsAdvocate]

Exactly. Which goes back to what you were saying about genes.

Principles of heredity were established before the discovery of genes, so before heredity knew of the double-helix, heredity theory didn't incorporate into its explanations the role genes play in the passing on of characteristics; it only noted that the passing on of characteristics did happen.

The explanation of heredity was unknown, and therefore should be stated as unknown and not that there is "nothing".

I was trying to say before (perhaps not very well) that to use Occam's razor, you must first have something to explain, and multiple possible explanations which can't be proven or disproven.

If you say "dinosaurs used tools", it is not Occam's razor that dismisses this, but rather lack of evidence. But if you say "petrified wood has unexplained notches", then you could conclude that the notches were caused by dinosaur tools, caused by natural formations from petrification, caused by methods of tree growth, caused by natural teeth/horns of dinosaurs, caused by aliens, caused by insects of fungus, caused by God, etc. If the notches COULD be caused by any of these, then we apply Occam's razor. If we accept the explanation of "dinosaur tools", we would have to "create" dinosaur tools because none are know to exist and no evidence suggests that dinosaurs could have used tools. Similarly, we have no evidence of aliens or God, so the razor cuts those possibilities out for the explanation.

Note that Occam's razor does not mean that there is no dinosaur tools, aliens, or God. It just means that inventing these things as an explanation for notches in petrified wood when other explanations are available is not necessary. Or even improbable and unacceptable.

 

[sphenisc]

Then are you a solipsist? Subjective experience cannot be scientifically observed, so that must constitute proof to me that you're a p-zombie, right? If something can't be observed, it can't be observed. What is your explicit rationale for believing that this means the thing is nonexistent? Here is a thought experiment: Consider a box that cannot be broken into. On top of that, it is made of a material which doesn't permit us to detect the inside of it in any other way either. Does this mean to you that the inside of the box doesn't exist?

Forget the hypothetical box - a standard black hole will do. Does the inside of a black hole exist?

 

[MRC_Hans]

We can observe the inside of a black hole indirectly, that is, we can observe the effect of it (gravity). There are a lot of things in physics that we can ever only observe indirectly.

Does a thing we cannot observe at all exist? The only sensible answer is: "We cannot know". However, the existene of something does not depend on our ability to observe it (unless you are a solipsist ).

Hans

 

[sphenisc]

Drkitten,

"Things that, by their very nature, cannot potentially be the subject of scientific inquiry, are unreal." -Melendwyr

.

Festus' remaining arguments attempt to disprove that any examples given by others, are things which "cannot be the subject of scientific inquiry".

If carried to its logical conclusion Festus' will demonstrate that there is nothing which, by its very nature, cannot potentially be the subject of scientific inquiry.

Combine with Melendwyr's statement this leads use to the conclusion that

Nothing is unreal.

I thank you!

 

[bruto]

Harrumphh!
A vacuum is measurable, and so the existance of a vacuum inside a box can be observed.
A vacuum is a little different from 'the inside of the box doesn't exist'.

Certainly, and as I said, I think we must admit that if it is a box at all in any meaningful sense of the term, it must have an "inside" or it is not a box. We cannot, however, count on the inside containing anything other than space or extension. As with anything else measurable, we can only measure a vacuum if we can get at it, but in this case we cannot. For the purpose of this thought experiment the box is perfectly sealed. Perhaps I was being ontologically careless in considering a vacuum "nothing at all." It might be necessary to conclude that if a box is a box and not something else, then no matter how it is redefined to be impervious to investigation, it must be assumed to contain "something" if we conclude that total emptiness is itself "something" because it can be measured relative to everything else.

 

[bruto]

I did state that because the box cannot be broken into. All other boxes have the property of possibly being broken into. Therefore, you don't know if the existence of interiors in other boxes is related to the induced rule that "things which can be broken into have interiors" or "boxes have interiors." Since the first rule explains all observations and doesn't require the second, you are unjustified in saying that there is a property of boxes that gives them interiors. You sounded sort of sarcastic, so I thought I might try to restate in different terms what I said previously.

I have to disagree. Accessibility is not what gives a box its interior. Extension is. The only way you can redefine a box as not having an interior is to redefine it as not occupying space, which essentially redefines it as entirely nonexistent. Now certainly a box that is defined as not existing can be authoritatively stated as not having any properties at all, but it's kind of an isometric exercise, isn't it?

You can conclude that for all intents and purposes the interior is irrelevant, and that from the pragmatic point of view it might as well contain nothing at all, since no interaction with the known world is allowed by definition, but you cannot conclude that it has no interior, because if it is a box, it occupies space, and that space is interior to it. If it is a box of any dimension, then in theory at least there is a location in the universe into which you cannot place anything else, because the space that is interior to the box already is occupied by it. However remote and inaccessible the box itself is, this is a conceivable consequence that does not call for violation of physical laws, and therefore, it seems to me, it allows the idea of the box's interior to have meaning even if the nature of that interior is entirely immune to scientific investigation.

 

[drkitten]

I read better than you write apparently. I addressed this issue in my last post. If you didn't want your example evaluated in it's fictional context then you should not have used a fictional example.

So address the non-fictional example of the colour patterns of [I[Archeopteryx[/i].

You seem to regard time as some sort of universal solvent that destroys everything utterly with no trace.

Not everything. Just some things. And those things that are destoryed -- or more accurately, irrevocably transformed beyond recognition -- cannot be recovered.

Yet the first law of thermodynamics states that matter cannot be created nor destroyed, only transformed. It stands to reason then that if matter, say an Archaeopteryx, is enclosed in a system, such as trapped in a mudslide, frozen in ice, or sunk in a pond and layered in sand, then all we need is an understanding of how the matter that made up the Archaeopteryx would have changed and reverse the process either physically or through a detailed model.

Unfortunately, irrevocable transformations exist. This is the second law of thermodynamics.

Is it too complicated to sort out?

Yes.

Physicists at the Japan Marine Science and Technology Center have built the Earth Simulator in part to model atmospheric conditions on a planet-wide scale. Could the decay and dissemination of the matter of anything be similarly modeled?

No.

The potential exists.

No.

Could such a technology be expanded to ever larger scales?

No.

How about genetics?

No.

If evolution is in fact the manner which all that lives came to be in the form it is then could the process be reversed?

No.

Given sufficient computing power could all possible DNA combinations be computed? Could then these sequences be modeled to show what sort of creature would result?

No.

Deny all you like this is not the question.Niether I nor anyone else has to have the ability to investigate however many spurious examples you come up with. All I have to show is a means of inquiry could potentially exist.[/I]

Which you have not done.

Perhaps you could come up with a logical fallacy in Melendwyr's assertion?

Certainly. Non sequitor -- "it does not follow." The same argument that applies when someone makes an argument of the form "I have shoes, therefore it will rain tomorrow."

An object subject to an irreversible transformation cannot be recovered for study. We know that irreversible transformations exist from elementary thermodynamics. Therefore, there exist object that cannot be recovered for study. And thus is Melendwyr's assertion conclusively refuted.

 

[Melendwyr]

An awful lot of the leading scholars in the field of the philosophy of mind. I've explained why many, including myself, have come to this conclusion too often and am sick of having arguments over it. I'd suggest reading proposed answers to Chalmers' "hard problem of consciousness" to find out more.

My degree is in cognitive psychology, Batman Jr.

Don't teach a daw to suck eggs.

What if there are things inside the box that interact with one another, but not anything outside the box?

So? What if there are? They would constitute a universe (albeit a small and pretty uninteresting one) completely separate from our own. They wouldn't exist relative to us, nor us to them.

 

[Melendwyr]

Festus' remaining arguments attempt to disprove that any examples given by others, are things which "cannot be the subject of scientific inquiry".

If carried to its logical conclusion Festus' will demonstrate that there is nothing which, by its very nature, cannot potentially be the subject of scientific inquiry.

Combine with Melendwyr's statement this leads use to the conclusion that

Nothing is unreal.

I hope this argument is a joke, as the conclusion doesn't follow from the premises.

 

[Melendwyr]

An object subject to an irreversible transformation cannot be recovered for study. We know that irreversible transformations exist from elementary thermodynamics. Therefore, there exist object that cannot be recovered for study. And thus is Melendwyr's assertion conclusively refuted.

The question of what the object was like before it was irreversibly transformed is an empirical, historical one that is potentially open to scientific examination.

Of course, if you rule out "past" and "future" from the realm of things that can be scientifically examined, your point would hold. Doing so, however, requires declaring those concepts to be without real referrents - that is, they describe unreal things.

 

[drkitten]

The question of what the object was like before it was irreversibly transformed is an empirical, historical one that is potentially open to scientific examination.

No, by definition of "irreversible." Any scientific examination would implicitly involve the reversal of the transformation.

Of course, if you rule out "past" and "future" from the realm of things that can be scientifically examined, your point would hold.

The inaccessible past, I do rule out. That's my central point.

But the other half of my central point is that the following statement is simply untrue.

Doing so, however, requires declaring those concepts to be without real referrents - that is, they describe unreal things.

The inaccessible past is not unreal. I specifically declare that the inaccessible past is not part of the scientifically examinable, while still being real. Your second sentence is simply a restatement of your original claim that anything unexaminable is unreal. As such, it's still wrong.

 

[Melendwyr]

No, by definition of "irreversible." Any scientific examination would implicitly involve the reversal of the transformation.

No, it wouldn't. Or is archaeology necessarily unscientific, in your mind.

The inaccessible past is not unreal. I specifically declare that the inaccessible past is not part of the scientifically examinable, while still being real.

You can redefine the word to mean whatever you like. but the inaccessible past isn't real. There are no possible events in the world that require that past to explain - therefore, it's not part of our universe.

Assuming that the Big Bang cosmology is correct, and further granting that the expansion-contraction hypothesis is valid, in what way could we say that anything before our Big Bang is real?

 

[drkitten]

You can redefine the word to mean whatever you like. but the inaccessible past isn't real.

The same challenge to you, then. Tell me about the colouration or mating behavior of Archeopteryx.

 

[RichardR]

Re this “box” discussion:

I would say, if you can’t measure something directly or indirectly, then there is no difference between a universe where this “thing” exists and a universe where this “thing” does not exist. So essentially the “thing” might as well not exist, for all the difference it makes.

That is not quite the same as saying if you can’t measure it, it doesn’t exist.