Meta Model and theory "patching"

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by Dangus</i>
<br />What differentiates Meta Model? It's an evolving model, all models must be to some degree.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">This is in chapter one of <i>Dark Matter...</i>. MM is deductive from first principles such as "no creation ex nihilo" and "every effect has a proximate, antecedent cause", which follow from logic alone and do not depend on observation or experiment. All other cosmologies are inductive: observe the universe and guess how it got to be that way. So MM evolves only for filling in the details, such as better physical descriptions of elysons and gravitons, which we must observe. MM does not evolve in its fundamentals, which are a framework for descriptions of all physical existence and change.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">if you had to develop a philosophy to keep Meta Model on the course it's currently on, how would you explicitly state what makes it superior to models that must be "patched"? How would you explicitly state the difference in how it evolves relative to such theories?<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">I liked the archery analogy recently used in Science magazine. There are two ways to arrive at an arrow in a bull's eye. (1) Fire an arrow into an existing target. (2) Draw the target around an already fired arrow. Patching conventional theories is like solution (2). Deducing theories and requiring them to work right from first principles is like solution (1). Either looks equally convincing to the casual passerby until you inquire into how they were arrived at. Logical deductions have greater value for true physical understanding than inductive guesswork. -|Tom|-

How does Meta Model's approach avoid creating a deductive line of reasoning that happens to lead to the same outcome, even if it is not actually the way things turned out?

For example: If there is only one road leading into a town, and someone has arrived in that town, you could reason that they must have started at a certain point, and could plot the route it would seem they must hav taken, and the end result would be the same(arriving in the town), but in this case the driver of the car actually decided to offroad over difficult terrain for some unknown reason. Until you go down and investigate very small details(maybe some unexplained dust on the car), the outcome would seem identical, so it would be possible to draw an incorrect line of reasoning.

I don't ask this to be difficult, but rather because when I encounter opponents of Meta Model who are not rabid, foaming-at-the-mouth conventionalists, they most often bring up this point basically. Is there any systemic way to avoid this problem, or does it simply require that those working on the model must be more diligent and try to manually avoid such errors?


As for the archery example, that's a very good one! As mentioned above it doesn't account for everything necessarily, but it certainly shows the weakness of a purely inductive approach.

Using the above example I made, I would wonder how to account for something bizarre, like an arrow fired from an off-angle having it's trajectory changed by the wind, and thus both inductive AND deductive lines would fail if evidence of the wind's interference could be turned up, or if anyone even knew to look for it.

Maybe people are just being too oppositional on this, I don't really know, but as I said, it does seem a common reaction by those who are at least willing to seriously consider Meta Model.

---

"Regret can only change the future" -Me

"Every judgment teeters on the brink of error. To claim absolute knowledge is to become monstrous. Knowledge is an unending adventure at the edge of uncertainty." Frank Herbert, Dune 1965

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"><i>Originally posted by Dangus</i>
<br />How does Meta Model's approach avoid creating a deductive line of reasoning that happens to lead to the same outcome, even if it is not actually the way things turned out?<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">Induction has multiple possibilities, and choosing the right one is a form of educated guesswork. Deduction is unique at every step and has no degrees of freedom. Deduction from a wrong starting point invariably goes quickly off-target. Induction from a correct starting point will generally answer questions and eliminate ambiguities.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">For example: If there is only one road leading into a town, and someone has arrived in that town, you could reason that they must have started at a certain point, and could plot the route it would seem they must have taken, and the end result would be the same(arriving in the town), but in this case the driver of the car actually decided to offroad over difficult terrain for some unknown reason. Until you go down and investigate very small details (maybe some unexplained dust on the car), the outcome would seem identical, so it would be possible to draw an incorrect line of reasoning.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">Your example mixes induction and deduction. If you are in the center of town, there are multiple ways to get there (even if there is only one easy way), and determining how someone got there is just educated guesswork. Deduction requires that you have a valid starting point. Once you have that, nature is not malicious and "God does not play dice", so one can uniquely figure out the course nature must take from that starting point. If it arrives at the correct goal, that validates the starting point; whereas arriving at the wrong goal means the starting point (or something else critical, such as the mode of transportation) was chosen incorrectly.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">I don't ask this to be difficult<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">I did not take it that way. It is a good question.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">Is there any systemic way to avoid this problem, or does it simply require that those working on the model must be more diligent and try to manually avoid such errors?<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">If a mountain lake spills over its banks and a river is formed down the mountainside, and if we know the terrain and the forces acting, we can deduce the exact path of the meandering river infallibly because no free choices are involved. If we attempt this and arrive at a wrong answer, then our assumed starting point is wrong, or we did not include all the important forces, or our knowledge of the terrain was incomplete. We must try again until we get it right. But deducing the path from the source is the only way to understand the result. The chances of guessing it by working backwards from the pool at the base of the mountain are miniscule.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">I would wonder how to account for something bizarre, like an arrow fired from an off-angle having it's trajectory changed by the wind, and thus both inductive AND deductive lines would fail<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">If you do not know all the operative forces, no theory will be successful in explaining the trajectory. At best, you can simply model the trajectory with epicycles, for whatever that is worth.

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">Maybe people are just being too oppositional on this, I don't really know, but as I said, it does seem a common reaction by those who are at least willing to seriously consider Meta Model.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">It took me a quarter of a century to find a valid starting point for MM that led to descriptions resembling reality. Mostly, I had to get rid of all my prejudices in the form of implicit assumptions. But then, it took our entire species a long time to rid itself of the notion that we were at the center of the universe, and we still haven't completely divested from the idea that we live at a special time. But the hardest anthropomorphic notion for me to jettison was the belief that our scale is special.

With inductive hypotheses, one is forever dealing with trying to force them to explain things that are beyond their scope. But what realy sells deductive reasoning is when you finally stumble onto a valid starting point and follow its unique deductive path. With startling quickness, everything drops neatly into place and starts to make good sense. Then the process itself teaches us new things we might never have thought about on our own. I have been fortunate enough to see this "locking" phenomenon several times, and there is nothing to compare with the satisfaction it brings to have all questions on a subject answered and to feel as though one truly understands it.

So the real difference is that, with induction (as for the Big Bang), we must constantly put more work into a hypothesis and maintain it with patches and a supply of free parameters; whereas with deduction from a successful starting point, the hypothesis needs no further maintenance yet continually pays us dividends, and has ready explanations for new phenomena. -|Tom|-

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">Deduction is unique at every step and has no degrees of freedom.
<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote"> I don't agree with this exactly. There are often multiple solutions to equations ... some wrong (by being physically inappropriate) and some right when applied to physical problems. Solving an equation is most certainly deductive, is it not?
<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">
Deduction from a wrong starting point invariably goes quickly off-target.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">
I certainly agree with that. It's almost not debatable. Unfortunately, deduction can also go very far awry after a train of deductions following a mistake in reasoning. It's easy do to even if one is extremely vigilant ... if ... one's logic is incomplete. [As I said some months past, the meta model will founder when it attempts to explain parity non-conservation because the solution must necessarily be non-mechanical and meta model can have no non-mechanical solutions in principle. It will also founder in trying to explain statistics (Fermi-Dirac, Bose-Einstein) which have no mechanical analog ... and the percentage difference in the mass of say, two individual protons, e.g. 100% identical - why? A mechanical model cannot handle such a question - in principle - because a mechanical model allows for only "nearly identical" not "logically identical".]
<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">
Induction from a correct starting point will generally answer questions and eliminate ambiguities. <hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">
You meant to say "deduction" here, didn't you?

---

In general, I get the impression that you eschew induction in favor of deduction? Philosophically speaking, is it your opinion that deduction is an intellectually cleaner or nobler feat than induction?

Would you say that denigrating others, who have "induced" starting points for the universe other than what you have arrived at, is "psychologically defensive" rather than scientific? I get that impression at times. It's to be expected since you are denigrated yourself in many quarters. I think your model is wrong on many counts ... but I commend you for it. It's truly different ... and ... I accept that it might prove out to be correct after all. I may be the one in the wrong, else I wouldn't bother to post here.

---

It would be helpful to the Meta Model if you would reveal some bad starting points that helped you to narrow down the alternatives to

No creation ex nihilo
and
No action without proximate antecedent cause

In my experience, I find no confidence in the above at all. They may be true ... or false ... or the matter may be unknowable (though that course would lead necessarily to inaction and so is not pursued).

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote"> Tom|-It took me a quarter of a century to find a valid starting point for MM that led to descriptions resembling reality. Mostly, I had to get rid of all my prejudices in the form of implicit assumptions. <hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

Similarily, it took me thirty years to discover that I didn't know how to know. Knowing that I didn't know was the first thing I came to know. I had believed everything they taught me, but it did not work out in the end. Now, I am well aware of our conceptual knowledge, which is really, according to eight year old Danielle, "imagination that I can't stop talking to me..." is a topic is well discussed n the literature if you look for it. I found and kept many of their key points, you can find that at www.fixall.org/words.htm if you are interested in language, and how we are used by it.

So when you said you rid yourself of "prejudices in the form of implicit assumptions." I know exactly what you mean. And what it took to do that.


<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">But then, it took our entire species a long time to rid itself of the notion that we were at the center of the universe, and we still haven't completely divested from the idea that we live at a special time. But the hardest anthropomorphic notion for me to jettison was the belief that our scale is special.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

Interesting, Banathy writes about how Cromagon and Neanderthal differ, as a species, evidenced by the tools found where they lived, whereas one used the same tools for a thousand years, the other formed social networks, and having the capability of speech, formed tools which would rival that of any hardware store, as one writer put it. Point being that alone, nothing changed, even evolution. But together, cultural evolution blossomed.

(Note: Banathy's premise is that the cultural evolution resulting from the new-found ability to talk trancends mechanical evolution. A point that Loye steadfastly maintains that Darwin himself held in second largely unknown book of love. I can see parallels here, in that gravity is like a mechanical principle of matter evolution, and electromagnetism is like a cultural affair.)

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">With inductive hypotheses, one is forever dealing with trying to force them to explain things that are beyond their scope.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

I'm sorry. but the two of them confuse me. One is top down and the other is bottom up, I don't which is which. I think what you are saying is that top down reasoning does not really explain while bottom up is, well...

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">But what really sells deductive reasoning is when you finally stumble onto a valid starting point and follow its unique deductive path. With startling quickness, everything drops neatly into place and starts to make good sense. Then the process itself teaches us new things we might never have thought about on our own. <hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">


I can relate to that too. I stumbled onto a valid starting point, called it the universal operating principle of the conceptual Universe back then. Now I call it CLAP, as in clap your hands. Did a lot of research looking for others like it. Found a whole Universe of them. The Universe has a system. The Universe is not about things, the Universe is about what things are doing...

And the system does teach us, in a self-instructive way, what we would never thought of on our own. It taught me that there are THREE sides of a coin, not the usual two sides of every coin. The inside holding the two outsides together is a part of the whole too. Such a simple assumption, that a coin has only two sides, and no inbetween.
But I wondered if there had to be more than just that, and there is.


<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">I have been fortunate enough to see this "locking" phenomenon several times, and there is nothing to compare with the satisfaction it brings to have all questions on a subject answered and to feel as though one truly understands it.<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

[]

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">So the real difference is that, with induction (as for the Big Bang), we must constantly put more work into a hypothesis and maintain it with patches and a supply of free parameters;<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

I think I get it now. Isn't what you describe something like they make up a theory, and then try to fit the observations to their theory?

<blockquote id="quote"><font size="2" face="Verdana, Arial, Helvetica" id="quote">quote:<hr height="1" noshade id="quote">whereas with deduction from a successful starting point, the hypothesis needs no further maintenance yet continually pays us dividends, and has ready explanations for new phenomena. -|Tom|-
<hr height="1" noshade id="quote"></blockquote id="quote"></font id="quote">

Oh Yeah. Korzybski of the map is not the territory fame explains this all so well, and especially in the context of this particular letter.

Korzybski says it is unsane to abstract first, and then observe accordingly. He says that this unsane reversal must be reversed, such that the sane way is to observe first, THEN abstract.

It is so clear that the big bang is an abstraction first, then observe. So, the big bang is unsane science.

Korzybski is the author of the book Science and Sanity.

[Tommy &lt;talking about the difference between deductive and inductive logic&gt;] "I'm sorry. but the two of them confuse me. One is top down and the other is bottom up, I don't which is which. I think what you are saying is that top down reasoning does not really explain while bottom up is, well..."


The difference is not so much top-down vs bottom-up as it is forward vs backward.

===

In <u>deductive logic</u> you begin with a set of assumptions (postulates) and reason forward to whatever they require. Mostly this leads to things that don't match reality, at which point you abandon the reasoning and start over with a new set of assumnptions.

With <u>inductive logic</u> you start with a current observation and reason backwards, trying to guess what might have resulted in the observed phenominon. It is fairly easy to come up with a reasonable answer using this approach, and that is its primary weakness. It is so easy to find an answer that <b>if you don't stop with the first one</b> you think of, you can find another. And another, and so on.

===

If you ever do find a set of starting assumptions for the deductive process that don't fly off into la-la land at some point fairly early in the reasoning process, it gets exciting because the farther you push it the more answers it gives you. Answers that match reality.

===

The first three chapters of TVF's book are a case study of how this process leads to the Meta Model. You might want to give it a look.


Regards,
LB

(Excellent question, by the way. It goes to the core of what this site is all about.)