There is a third possibility; a cycle of explanation in which each physical explanation is explained by some other physical explanation.

It's just the old epistemic concern with foundations, regressions and coherence.

There is a third possibility; a cycle of explanation in which each physical explanation is explained by some other physical explanation. — Banno

That cycle of explanations would be #4 in my list.

It's like #1, except cyclical instead of infinite.

But wouldn't it be brute? Such a system cyclically implies itself, but why should any part of it be true.

So #4 would seem to have more in common with #2, a brute-fact, than with #1, an infinite sequence of explanations.

You said something about this matter being a question that philosophers have considered.

That would be interesting, if philosophers have ruled out some of those physics possibilities, #1 through #4.

Have they?

Michael Ossipoff

But wouldn't it be brute? Such a system cyclically implies itself, but why should any part of it be true. — Michael Ossipoff

Each item in the cycle has an explanation. So no item in the cycle is brute, by the assumed definition of "brute".

Further, what sense could we make of asking if the cycle itself has an explanation? We could say that the cycle has no explanation, and hence that it is brute; or that since each element is explained, the cycle explains itself, and hence is not brute.

I'd said:

But wouldn't it be brute? Such a system cyclically implies itself, but why should any part of it be true. — Michael Ossipoff

You replied:

Each item in the cycle has an explanation. So no item in the cycle is brute, by the assumed definition of "brute".
[.quote]

Two bank-robbers are arrested coming out of a bank with the loot. They both deny their guilt. Each one points to the other and says, "I vouch for him. He's telling the truth."

How much does that count for?

The fact that each item in the cycle is explained in terms of the cycle just before it in the cycle means nothing if none of the items in the cycle have any support outside the cycle.

Maybe "brute" needs a more careful definition.

Yes, as you said, the whole cycle is brute, and therefore so is each of its elements, by any meaningful definition of "brute".

Michael Ossipoff

The fact that each item in the cycle is explained in terms of the cycle just before it in the cycle means nothing if none of the items in the cycle have any support outside the cycle.

IF the cycle explains enough, then I don't see a problem. Suppose that every fact were explained by some other fact. Then there would be nothing that is not explained.

You can explain A in terms of B, but that isn't a complete explanation of A unless you can explain B.
.
Each element of the cycle has the next element as is explanation and reason. Each is explained in terms of the next thing in the cycle, which is a complete explanation only if the next thing in the cycle is explained in terms of something. ...in this case,..by the next thing in the cycle. ...and so on.

So, eventually this sequential explanation comes back to the first thing that you wanted to explain: The things that it needs for explanation need for it to be already independently explained, in order for them to be any good for explaining anything.

So,any particular element is explained only because it is explained in terms ot things whose explanations need it to be explained.
-----------------------------------
Try it in the other direction:] . "Obviously this first thing, #1, is explained--we already know the explanation-- and so that next thing, which is explained in terms of
#1, is completely explained. Likewise for the next thing, that needs the previous thing as an explanation....and so on."

Don't you see that the first thing in this chain doesn't have really any explanation? The first explanation-statement in this paragraph can't be made, because the first thing doesn't already have an explanation, to qualify it to explain something else.

The fact that your explanation-cycle doesn't work in the opposite direction should make you suspicious.

When you do it in the opposite direction, you're running a pyramid scheme. The elements are borrowing based on credit that depends on expecting someone else to borrow in order to lend to them. In that way, you can miss the fact that the money isn't there.

To make my bank-robbers example more realistic, make it 3 bank-robbers instead of 2, cyclically vouching for eachother's truthfulness. Meaningless without some outside voucher for at least one of them.

Michael Ossipoff

Circular implication is perfectly valid.

Ask yourself: Which thing in the ring of explanation is not explained?

Circular implication is perfectly valid.

Ask yourself: Which thing in the ring of explanation is not explained? — Banno

Each one is explained...partially. But a complete explanation depends on having an explanation of the facts that are your explanation. Otherwise your explanation is only partial, not complete.

Pyramid schemes work, because it's easy to be deceived by them. . That's why I suggested trying the explanations in the opposite direction. If that doesn't work, that should make you question the matter.

Michael Ossipoff

If that doesn't work, that should make you question the matter. — Michael Ossipoff

Why? If A explains B, it does not follow that B explains A.

Which brings us back to the question of the nature of explanation -

What sort of thing is an explanation? — Banno

What is the relation between an explanation and the thing explained? You seem above to imply that it is material equivalence - that if A explains B then B explains A.

But a complete explanation depends on having an explanation of the facts that are your explanation. — Michael Ossipoff

And again, each and every fact in a cycle of explanation can be explained, including those that form the explanations for other facts.

An explanation explains something.

So, if you hear an explanation which you have not heard before, then you will come to understand some topic better.

Clearly understanding does not rely upon facts alone, though. We are not encyclopedias, but knowers. And knowing is much more than the facts. (though, clearly, knowing is not absent facts either)

A brute fact is something that exists without explanation. That is, there is nothing more to understand. Once you know the fact, there is nothing left to know. To ask, "But why?" is a fruitless endeavor.



***

Something in the exchange that helps me to say why I prefer talking in terms of brute beliefs to brute facts: it seems to me that it's less speculative, at least. There's a lot more positing about how the world must be in order for this or that conclusion to be acceptable when talking in terms of brute facts. But brute beliefs are plausible in that clearly we actually do other things and believe certain ways. That is, there is a terminus in our chain of reasoning (if a chain it could be called), regardless of how the world might be (infinite regress, or no)

Hello, Mollie.

While I am enjoying playing around with the ideas here, my response is that the notion that brute facts are those that need no explanation is not far from those facts of which we are certain. Introducing the notion of brute facts might help some to come to terms with certainty.

One can be certain of whatever one chooses.

One can accept any statement as an explanation of any other.

Certainties, of course, are beliefs. Brute beliefs, if you like. That strikes me as close to the mark.

I think the issue isn't about brute facts as opposed to criteria we use to evaluate whether a fact is brute or not.

You seem above to imply that it is material equivalence - that if A explains B then B explains A.

Certainly not. Two things can't provide complete explanations for eachother. I gave that example with two things because I thought it was obvious that two things can't provide complete explanations of eachother.

In fact, of course neither can three things:

B gets its explanation from A, C gets its explanation from B, and A gets its explanation from C.

If a fact has an explanation, then another way of saying that would be to say that it's true without our accepting it as a brute-fact. So, if we talk about verification of truth, instead of explanation, will that make it more obvious?

Then it's like those 3 bank-robbers. Cyclically, they all verify eachother's truthfulness and honesty.

...and it means nothing.

Here it is with alleged facts instead of bank-robbers:

A says that B is true.B says that C is true. C says that A is true.

Does that mean that all of them are true?

Not unless something brute-true says that at least one of {A,B,C} is true, or an infinite sequence of verifications says so.

(...or my possibility #3, where the sequence of verifications leads to a set of facts that are inevitable)

Michael Ossipoff

But a complete explanation depends on having an explanation of the facts that are your explanation. — Michael Ossipoff


And again, each and every fact in a cycle of explanation can be explained, including those that form the explanations for other facts. — Banno

...and that cycle of explanation, when followed far enough, leads back to the conclusion that the first thing explained, which needs everything else to back up its explanation, it supposed to be the basis for those explanations Making it the basis of its own explanation..

It's called "circular reasoning".

Yes, everything in your cycle has an explanation. But not a complete explanation. An explanation is complete only if it, itself, has explanation, or doesn't need it.

A is explained by B is explained by C is explained by A.

Classic circular reasoning.

Michael Ossipoff

The point remains that circular reasoning is valid. It is rejected usually for aesthetic purposes.

An explanation is complete only if it, itself, has explanation, or doesn't need it. — Michael Ossipoff

Again, suppose each and every item in a cycle of explanation has a complete explanation within the cycle.

If we called it a coherence theory of explanation, would that help?

A is explained by B is explained by C is explained by A. — Michael Ossipoff

Nothing in this sequence is unexplained.

Banno--

The point remains that circular reasoning is valid. It is rejected usually for aesthetic purposes.

No. Circular reasoning can, and often does, lead to false conclusions.

Recursion gives valid conclusions. Circularity is different from recursion. Circularity can lead to false conclusions.

Again, suppose each and every item in a cycle of explanation has a complete explanation within the cycle.

Then that would be different. But you've been claiming something else, You've been claiming that each item in the cycle is completely explained if merely each item in the cycle has an explanation in terms of the next item in the cycle.

Your original claim didn't pre-suppose that all the items in the cycle are completely explained. That they're all completely explained is the conclusion that you reached from the statement that each item of the cycle has an explanation in terms of the next item in the cycle.

...and I'm saying that that conclusion (that all the items are completely explained) doesn'tt follow from your premise (that each item in the cycle has an explanation in terms of the next item in the cycle).

I know that I can convince you of this.
.
1. Say A is explained by B, and B is explained by C, and C is explained by A.
.
2. Therefore A, B, and C are explained.
.
3. A is completely explained by B if A is explained by B and B is completely explained.
.
4.Otherwise A is merely partly explained.
.
5. A, B, and C are either completely or partly explained.
.
6. From #1 and #3, we can say that A is completely explained by B if B is completely explained. B is completely explained by C if C is completely explained. C is completely explained by A if A is completely explained.
.
7. A, B, and C being only partly, not completely, explained is entirely consistent with statements #1 through #6.
.
7a. Suppose that none of {A,B,C} is completely explained. Statements #1 through #6 could still all be true, and the lack of complete explanation of A, B, and C would be consistent with that.
.
Or look at it particularly in terms of A:
.
Statement #6 says that A is completely explained if A is completely explained. It doesn’t say that A is completely explained.
.
8. Item A could be only partially explained, and that would be consistent with statements #1 through #6.
.
Statements #7 and #8 mean that, even if statements #1 through #6 are true, it’s possible for A, B, and C to be not completely explained, only partially explained.
.
Given statements #1 throught #6, items A, B, and C could be not completely explained, in spite of statements #1 through #6 being true.
.
I've shown why, given that statements #1 through #6 are true, items A, B, and C could be completely explained, or could be only partly explained (explained in terms of something that doesn't have a complete explanation).
.
Michael Ossipoff

I've shown why, given that statements #1 through #6 are true, items A, B, and C could be completely explained, — Michael Ossipoff

Indeed; they could be completely explained by a cycle.

"I've shown why, given that statements #1 through #6 are true, items A, B, and C could be completely explained" — Michael Ossipoff


Indeed; they could be completely explained by a cycle. — Banno

...with a little bit of help from something else, such as one of the cycle-elements being inevitable or self-evident. ...or one of the cycle-elements being explained by something outside the cycle.

The mere fact of a cycle of explanations doesn't mean a cycle of complete explanation.

If some element of that cycle is self-evident, inevitable, or else completely explained from outside the cycle, then yes, everything in the cycle could be completely explained, and the cycle of explanations could be part of why that is.

I must have been tired last night, because I said only that something is completely explained if it's explained in terms of something that's completely explained. That isn't enough to say. It's also necessary to name a few other ways something can be completely explained:

1. Something is completely explained if it's explained in terms of something that's completely explained.

2. ...or if it's self-evident or inevitable

3. ...or if it's explained in terms of something brute.

4. ...or if it's explained by an infinite sequence of explanations.

But, without any of these source of complete explanation, a cycle of explanation isn't enough, because element A's explanation in that cycle ultimately comes back to depending on A being qualified to completely explain the next element of the cycle. How can it be, if it needs that cycle in order to have a complete explanation from something that wouldn't have an explanation if A weren't explained?

Without any of the helps that I mentioned, the cycle, as far as item A is concerned, is just a statement that A is explained if A is already explained.

Of course, if it's an infinite chain of explanations, then physicists, as they probe the physical world more and more deeply, will keep finding things that are consistent with our being here, and with their previous observations.

Michael Ossipoff

...from outside the cycle... — Michael Ossipoff

That bit is no more than an expression of your own aesthetic. But let's leave that as moot.

More central is the issue of what an explanation consists in. I'm not convinced it can be much beyond a belief that leads to one holding another belief.

Anyone?

"...from outside the cycle.".. — Michael Ossipoff


That bit is no more than an expression of your own aesthetic. But let's leave that as moot. — Banno

Incorrect. lt isn't an aesthetic question.

I slightly mis-spoke when I said that your {A,B,C} cycle of explanations amounts to saying that A is explained because A is explained.

Actually (as I've already admitted), every element of your explanation-cycle is explained...partially.

If you explain A in terms of B, but you can't explain B, then I call that a partial, not complete,. explanation of jA.

So let me re-word my statement:

Your claim for A being completely-explained depends on B being completely explained, by B is explained by C, which is completely explained by C being explained by A, which is completely-explained.

Therefore, given your {A,B,C} cycle of explanations: Saying that A, B, & C are completely explained amounts to slaying that A is completely explained because A is completely explained.

Do you see why that's circular reasoning? And why it isn't a useful conclusion?

What if A isn't completely-explained? Then A isn't completely explained.

Do you see that that's possible?

Now, if A is brute, self-evident, inevitable, or explained by something outside the cycle that is brute, inevitable, self-evident or completely-explained, then yes, A is completely-explained, and that makes B and C completely-explained too.

If not, if no member of the cycle is completely-explained as described in the above paragraph, then the above way of saying that A, B, and C are completely explained aren't true.

That's my best effort to explain it to you. If I still haven't reached you, then I accept that it isn't possible..

As discussed in another recent discussion-thread, people don't readily change their beliefs, and it's often or usually futile to get people to reconsider their beliefs.

Michael Ossipoff

Now, if A is brute, self-evident, inevitable, or explained by something outside the cycle that is brute, inevitable, self-evident or completely-explained, then yes, A is completely-explained, and that makes B and C completely-explained too. — Michael Ossipoff

I should have said: "...then yes, A is completely explained or brute, and that makes B and C completely explained too."

Michael Ossipoff

Do you see why that's circular reasoning? And why it isn't a useful concl — Michael Ossipoff

Of course it is circular. But that does not make it useless.

It's coherentism; and it is a serious epistemic theory.

As I said at the outset:

Further, what sense could we make of asking if the cycle itself has an explanation? We could say that the cycle has no explanation, and hence that it is brute; or that since each element is explained, the cycle explains itself, and hence is not brute. — Banno

I explicitly recognised the two possibilities.

It is a moot point, so far as I am concerned. More interesting is the so far unanswered question of the nature of explanation.

Again, I posit that "A explains B" means that belief in B is implied by belief in A. I conclude that discussion of explanation reduces to discussion of belief, and repeat my main point, that it is clearer to talking about certainty than brute facts.

I'd said:

That's my best effort to explain it to you. If I still haven't reached you, then I accept that it isn't possible. — Michael Ossipoff

...but I'm going to contradict myself now, on that matter, because maybe you are reachable, with the clearer, undeniable, demonstration which follows, after this paragraph (after which I'll answer the comments in your most recent post on this topic):

As before, I'll number the statements:

1. Suppose A, B, and C are in a cycle of explanations. Each one is explained and verified in terms of the one before it in that list.

2. An item can be explained and verified in either one of the following two ways:

2a) The item is self-evident or inevitable.

2b) The item is explained and verified in terms of something else which is, itself, completely explained (by either 2a or 2b).

3. Suppose that neither A, nor B, nor C is self evident, inevitable, or explained or verified by anything outside the cycle.

4. Therefore, A has no complete explanation or verification other than that which it might get from C (because, by statements #1 and #2, A is completely explained and verified if C is completely explained and verified, and because, by statement #3, A has no other source of complete explanation or verification.

So A is completely explained if and only if C is completely explained and verified. And of course, likewise, C is explained and verified only if B is explained and verified. And B is completely explained and verified only if A is completely explained and verified.

From the above, then, A is completely explained if and only if A is completely explained and verified.

Saying that A is completely explained and verified only if A is completely explained and verified doesn't constitute a meaningful or reliable explanation and verification of A.

Even if you say that A could (somehow) be explained and verified in that manner (A is completely explained and verified if A is completely explained and verified, you wouldn't have anything like a reliable explanation and verification.

How do I know you're not a liar? Because i say I'm not.

What's the explanation for how or why that machine works? It works because it does.

That example brings out the fact that, even if you could say that A might be, completely explained and verified, from the statement that A is completely explained and verified if it is--the fact is, that A still might not be true. Or if we're talking about explanations, A might not be explained.

And that's if we generously let you say that A even might be true, when none of A, B, or C have any explanation or verification other than cyclically from eachother.

So even if we allow you the most generous latitude and benefit of the doubt, item A might not be true or explained.

I don't call that a useful or meaningful verification or explanation. It isn't any explanation or verification at all. A maybe complete explanation or verification isn't a complete explanation or verification at all.
.
------------------------------------------------------------------------

As I said at the outset:

Further, what sense could we make of asking if the cycle itself has an explanation? We could say that the cycle has no explanation, and hence that it is brute; or that since each element is explained, the cycle explains itself, and hence is not brute. — Banno


I explicitly recognised the two possibilities.
— Banno

If you said that, I missed it..

But, in any case, if you recognize both possibilities, then you're admitting that item A is only maybe explained or verified.

I suggest that maybe being explained or verified, isn't worth anything, isn't any explanation or verification at all.

It is a moot point,

A moot explanation or verification isn't any explanation or verification.

Michael Ossipoff

For now I'm staying focused on the topic of explanation. I think that'll suss out differences better than on the topic of certainty, mostly because I couldn't think of a reply (i.e. it didn't seem to me we were disagreeing much):


If a belief that leads to one holding another belief is all an explanation is, then that's a pretty permissive standard for an explanation.

I believe the medicine will cure me. This, in turn, makes me believe the doctor can be trusted. Is the former belief really an explanation of the latter belief? Psychologically, perhaps, it makes sense of the latter belief. But it's not like any sort of thing I'd call an explanation extemporaneously. And surely you can see how there's a difference between what an explanation consists in, and how people, psychologically, can be drawn to a belief from another belief they hold without it being an explanation of the content of the belief, even if it happens to (given such and such an understanding of human psychology) count as an explanation for the belief itself.

This description also doesn't really tell us what explanations consist of, either. It leaves out the process, which is an important element in understanding what explanation consists of. An explanation brings one to a deeper place of knowing about some topic. Explanation is about knowledge, and in particular how it changes a person's ability (understanding) and relationship to the topic at hand, not just about changing beliefs.

I believe the medicine will cure me. This, in turn, makes me believe the doctor can be trusted. Is the former belief really an explanation of the latter belief? — Moliere

Yep, it is. The Doctor gave me a sensible prescription, hence I conclude that she can be trusted. Can I explain why I trust the doctor? She gave me the right medicine.

An explanation brings one to a deeper place of knowing about some topic. Explanation is about knowledge, and in particular how it changes a person's ability (understanding) and relationship to the topic at hand, not just about changing beliefs. — Moliere

Is it? What counts as evidence here? Or can we make up any old shit?

Here's some evidence:

https://www.wolframalpha.com/input/?i=Explanation&rawformassumption=%7B%22C%22,+%22Explanation%22%7D+-%3E+%7B%22Word%22%7D

Exhibit two:


http://www.thesaurus.com/browse/explanation?s=t

Yes, we can make up any old shit and it is all explanation of a kind. We can explain the origin of a flower by reference to a nymph who misbehaved. But we are interested in what makes a good explanation. Why should we reject some explanations and accept others?

I submit that a good explanation is one that is not vulnerable to any of the reasons why we reject explanations. That is, we should first look at what makes an explanation a bad one. For example: flowers exist and nymphs don't. Doctors are trustworthy and do not deliberately try to harm us. (The exceptions don't entirely vitiate the explanations, but they do mean that there is never a complete knock-down explanation of anything at all.)