Talk:Model-dependent realism

Origin

This article was brought over from Reality and revised. This topic now is introduced there as a sub-topic Reality#Model-dependent realism . John R. Brews 16:40, 12 November 2011 (UTC)

Terminology

The terms "world picture" and "physical theory" are stated by Hawking/Mlodinow to be (math+observations+connecting rules), but they don't always stick to this narrow definition. The term "world picture" is generalized to include any mental construct of reality. They also refer to a "model" as the mathematical part of a theory as distinct from the observational part, but they don't stick to this definition and sometimes use "model" as the same thing as "theory" or "world picture". Hence, quotations from this source can be confusing because one never knows which definition is in use except by reading entire pages for context. John R. Brews 12:26, 13 November 2011 (UTC)

Background

The section Model-dependent realism#Background is very incomplete. Unfortunately, Hawking/Mlodinow make little effort to attach their theory to the very long line of earlier or even contemporary philosophical efforts. The introduction to the book by Cao outlines some of this history and introduces the subject of structural realism. The article on scientific realism the Stanford Encyclopedia of Philosophy lists still more. There is a huge literature, and to place "model-dependent realism" in a proper context requires, in fact, a host of articles on these various topics that can be linked. Otherwise, this article on "model-dependent realism" will become two or three times its present length. John R. Brews 16:35, 13 November 2011 (UTC)

A helpful discussion is found in the article: Mateusz Hohol, Wojciech P. Grygiel. Stephen Hawking's Ontology of Physical Theories. published in: "Philosophy In Science. Methods and Aplications", eds. B. Brożek, J. Mączka, W.P. Grygiel, Copernicus Center Press, Kraków 2011, pp. 105-115.. Retrieved on 2011-11-13. From this article it looks like Hawking has made other efforts to express his views that may be more rigorous and clearer than The Grand Design. John R. Brews 17:08, 13 November 2011 (UTC)

Reviews

The viewpoint of “theory-dependent realism” being espoused in The Grand Design appears to be a kind of half-way house, objective reality being not fully abandoned, but taking different forms depending upon the particular theoretical perspective it is viewed from... Penrose

They advocate "model-dependent realism", which asserts that the "reality" of various elements of nature depends on the model through which one interprets them. This is an interesting approach to ontology, but it won't come as shocking news to philosophers who have thought about the problem. Wall Street Journal

The bulk of reviews focus upon the deliberately controversial elements of this book about God and the validity of M-theory, none of which is important to the concept of "model-dependent realism". One gets the impression that The Grand Design was never a serious attempt at a scholarly position on "reality", but rather a gadfly to get discussion going. John R. Brews 17:08, 13 November 2011 (UTC)

Interpretation of a quote from The Grand Design

A transcription from Talk:Reality. The following quotation is discussed:

There is no picture- or theory-independent concept of reality. Instead we will adopt a view that we will call model-dependent realism: the idea that a physical theory or world picture is a model (generally of a mathematical nature) and a set of rules that connect the elements of the model to observations. This provides a framework with which to interpret modern science.

This is a mixed bag. The first sentence describes their notion of reality. The next sentence is a non-sequitor entirely that introduces the idea of a "physical theory" as a "model and a set of rules that connect the elements of the model to observations". I'd hazard that this statement is an entirely prosaic statement of what most everybody would define as a physical theory. It is simply a definition, and therefore is not attached to the philosopher's "reality". They also introduce here a technical definition for "picture" that is exactly the same thing as a "physical theory". It then defines "model-dependent realism", not as a philosophical position, but as a technical term identical with "physical theory".John R. Brews 16:45, 11 November 2011 (UTC)

Reading it literally, it defines model-dependent realism as "the idea that..." That wording establishes it as "a philosophical position". Anthony.Sebastian 21:09, 12 November 2011 (UTC)

Anthony, that is a misreading. The assertion: "model-dependent realism is the idea that a physical theory or world picture is a model (generally of a mathematical nature) and a set of rules that connect the elements of the model to observations" is not an "idea" at all, whatever they say, because in fact a physical theory is exactly that: "a model (generally of a mathematical nature) and a set of rules that connect the elements of the model to observations". (If you doubt that, consider quantum mechanics as the physical theory consisting of (i) the mathematical model of Hilbert space connected by (ii) rules to the (iii) observations of chemistry.) So their statement is equivalent to the statement "model-dependent realism is the idea that a physical theory is a physical theory", which is an empty remark. To put content into their sentence, it has to be taken as a statement as to what constitutes a physical theory, viz: (model+rules+observations). John R. Brews 05:09, 13 November 2011 (UTC)

It seems to me they want the reader to understand that physical theories are only models (of reality), and therefore cannot be understood as descriptions of reality, though if they meet their elsewhere-stated standards of a good model they can attribute to them the quality of reality, a tentative reality to be tested for its ability to be fruitful for application and research. That seems part of the idea, or view, of model-dependent realism. Their idea is that models create a reality of their own, one that exists for the model.

Anthony.Sebastian 15:38, 17 November 2011 (UTC)

←outdent Anthony, your wording resembles this quotation (p.7):

When such a model is successful at explaining events, we tend to attribute to it, and to the elements and concepts that constitute it, the quality of reality or absolute truth.

Here the "model" refers to the model-part of a world-picture (model+rules+observations), so it is not a reference to a model of reality, although we are tempted to say it has "the quality of reality or absolute truth".

Hawking/Mlodinow go on to say (same paragraph) that one cannot determine whether one model is more real than another:

If two such physical theories or models accurately predict the same events, one cannot be said to be more real than the other;...

Then (p. 46):

It is pointless to ask whether a model is real, only whether it agrees with observation.

Now it appears that you, Anthony, wish to talk about attribution of "tentative reality", which is not "reality" as "absolute truth". So perhaps you agree that Hawking/Mlodinow do not support "objective reality"?

I'd agree with the notion of a "tentative reality" in the sense that (i) it is conditional upon the selection of world-picture, and (ii) it is transient because world-pictures change, and even though the observations remain largely intact, the concepts of the explanatory model change radically (for example, the Copernican revolution, or general relativity).

I'd also agree with your remark "Their idea is that models create a reality of their own, one that exists for the model." in the sense that each world-picture has its own model and hence its own concepts to organize some set of observations.

I do disagree, however, with your remark "It seems to me they want the reader to understand that physical theories are only models (of reality), and therefore cannot be understood as descriptions of reality"

I'd rephrase to read "they want the reader to understand that world-pictures are only models (of reality), and these world-pictures constitute all we can know about reality according to "model-dependent realism". Hawking/Mlodinow assert (p. 45) that model-dependent realism "short-circuits all this argument and discussion between the realist and the anti-realist schools of thought."

Second, I'd dispute entirely the statement that "if they meet their elsewhere-stated standards of a good model they can attribute to them the quality of reality."

Although Hawking/Mlodinow introduce the idea of selecting between world-pictures this way, they already have stated again and again (and in some of the quotations above) that one cannot distinguish between world-pictures from within the stance of model-dependent realism. This evaluation of competing models is subject to debate, competing lists and interpretations of the list abound, and I do not find Hawking/Mlodinow saying anywhere that this list confers "reality" upon one model more than another. In fact, (p. 52) Hawking/Mlodinow say about these criteria:

"The above criteria are obviously subjective"

Subjectivity is the opposite of what normally is defined as reality, both in Plato's approach, and in Scientific Realism. These criteria are pragmatics one hopes lead to the "ability to be fruitful". However, the justification of the criteria is the success of these pragmatics in beneficially influencing the evolution of science. This justification lies outside model-dependent realism. Historically, the subjective nature of these pragmatics already has led to unending debate among practitioners as to whether theory A fit the criteria better than theory B and over just what the criteria actually should be. Einstein thought it was elegant that the special theory of relativity used only two postulates; Lorentz thought that put a straight-jacket upon the theory. Dirac's formulation of quantum mechanics was elegant, but Von Neumann found it lacked rigor. Feynman thought Dyson's mathematical approach to field theory was myopic, and Dyson thought Feynman's intuitive methods were suspicious. The Standard Model violates all but one of the Hawking/Mlodinow criteria. Today there is debate over whether Popper's criterion of refutability is violated by string theory, making it only another example of the pursuit of abstruse useless logic, sometimes called the "Alice's Restaurant Problem". John R. Brews 21:49, 17 November 2011 (UTC)

Anthony, these remarks present my view of what is meant by "model-dependent realism", which seems to agree only in part with your own. If you do not find my assessment convincing, please provide some quotes of your own that support a different view of Hawking/Mlodinow. Hopefully a deeper assessment will evolve. John R. Brews 18:19, 17 November 2011 (UTC)

Regarding lede sentence as reproduced here

The lede sentence reads:

"In philosophy, model-dependent realism asserts that there is no unique objective reality, but reality consists at least in part of networks of world pictures that connect observations with their explanations or models."

Yet, in reading Hawking/Mlodinow, who introduce 'model-dependent realism', I cannot find such an assertion. I agree they never mention there existing a unique objective reality, but that's not the same.

They do say there can be no picture- or theory independent concept of reality.

They do say, also, that realism is difficult to defend, but that is not their saying there is no unique objective reality. They seem to argue for a model of reality that comprises a network of models, without arguing that model is reality. Anthony.Sebastian 05:02, 17 November 2011 (UTC)

Hi Anthony. I am ready to believe that there are some subtleties that I need some help with. I hope you have the patience to deal with them. Of course, we do not have to come to any resolution of "what is reality" but only to agree upon what Hawking/Mlodinow say about it. I am, unfortunately, unimpressed by these authors ability to use terms consistently or to stick to the same formulation everywhere, so this investigation may lead only to identifying some ambiguities in their presentation. The ultimate objective here is an Introduction that is at least consistent with some form of the Hawking/Mlodinow term "model-dependent realism".

Here is the quote you mention from Hawking/Mlodinow, the lead sentence in the quote above:

There is no picture- or theory-independent concept of reality.

Therefore, in their view, "objective" reality, being a concept of reality, must be a picture- or theory-dependent concept of reality. I would take "picture" to refer to "world picture" and a synonym for "theory", namely the combination (model + rules + observations) in which the "observations" are connected by the "rules" to the "model", the last being perhaps a mathematical structure.

Hawking/Mlodinow also say:

We make models in science, but we also make them in everyday life. model-dependent realism applies not only to scientific models but also to the conscious and subconscious mental models we all create in order to interpret and understand the everyday world. Model-dependent realism corresponds to the way we perceive objects.

I take this as saying that "world pictures" are allowed to encompass as their model component, models more general than scientific models, but they all have the three part structure (model + rules + observations). Combined with the first statement, I am led to understand that "objective" reality, to be allowed as a concept of reality by Hawking/Mlodinow, also is of this form (model + rules + observations).

They say:

We form mental concepts of our home, trees, other people, the electricity that flows from wall sockets, atoms, molecules, and other universes. These mental concepts are the only reality we can know. There is no model-independent test of reality.

Thus, "objective" reality, if it is a concept for Hawking/Mlodinow, can be tested only using a model. Which seems to say that the concept of "objective" reality, if it is a concept for Hawking/Mlodinow, can be approached only through some approximating "world picture" or "theory", because "world pictures" include all constructs that consist of (model+rules+observations).

So I am left with this position on "objective" reality: Our observations, models, and rules connecting them are the only "reality" we can know. If one wishes, one can imagine an objective reality that lies outside this picture, perhaps as some limiting form toward which our "world-pictures" are progressing, but that imagining can never be tested, and all we can know about this "objective" reality gradually emerging from the haze of uncertainty is contained in the best "world picture" we can muster in our particular lifetimes. Inasmuch as the best "world picture" today does not encompass all our observations, and we are forced to use multiple world pictures in some arenas of observation, our best notion of "objective" reality contains ambiguities where our world pictures overlap, and we cannot know how, when, or if these ambiguities will someday be resolved.

This idea of positing "objective" reality as a limiting case, like the limit of a series in mathematics in which each finite sum of terms gets ever closer to the sum to infinity, appears to me to be consistent with Hawking/Mlodinow's model-dependent realism, but to go beyond anything they say. I think they would scoff at this idea because we cannot know anything more about this reality than what is contained in our present best guess at its content, so where the whole process of evolving world-pictures is going is just conjecture, and the process actually may never converge.

However, I may be more skeptical than Hawking/Mlodinow, who say (p. 7):

In the history of science we have discovered a sequence of better and better theories or models...Will this sequence eventually reach an end point,...or will we continue forever finding better theories, but never one that cannot be improved upon? We do not yet have a definitive answer to this question, but we now have a candidate for the ultimate theory of everything, if indeed one exists, called M-theory. John R. Brews 19:43, 17 November 2011 (UTC)

Assuming some agreement about this, what is to be done with the sentence:

"In philosophy, model-dependent realism asserts that there is no unique objective reality, but reality consists at least in part of networks of world pictures that connect observations with their explanations or models."

How about this:

"In philosophy, model-dependent realism asserts that all we can know about "objective" reality consists of networks of world pictures that connect observations with their explanations or models."

John R. Brews 12:45, 17 November 2011 (UTC)

I revised the leading sentence to read:

In philosophy, model-dependent realism asserts that all we can know about "objective reality" consists of networks of world pictures that explain observations by connecting them by rules to concepts defined in models.

John R. Brews 21:58, 17 November 2011 (UTC)

←Outdent -

John, you have given the issue careful and insightful thought. Your two lede paragraphs now read:

In philosophy, model-dependent realism asserts that all we can know about "objective reality" consists of networks of world pictures that explain observations by connecting them by rules to concepts defined in models. A world picture consists of the combination of a set of observations accompanied by a conceptual model and by rules connecting the model concepts to the observations. Different world pictures that describe particular data equally well all have equal claims to be valid. There is no requirement that a world picture be unique, or even that the data selected include all available observations. The universe of all observations possibly may be covered by a network of overlapping world pictures and, where overlap occurs; multiple, equally valid, world pictures exist.

I think you've captured the essence of model-dependent realism, though I have one comment.

• As written, the first sentence may give the reader the impression that H/M believe there exists an “objective reality”, else how could they, in your words, qualify it as “…consists of networks of world pictures…”. If they do so believe in an objective reality, they never explicitly admit it. Their somewhat cryptic remark, “But as we ponder the manner in which we observe and form concepts about our surroundings, we bump into the question, do we really have reason to believe that an objective reality exists?”, makes one at least wonder. They do not define ‘objective’, even in their Glossary. Why not just go with:

In philosophy, model-dependent realism asserts that all we can know about "reality" consists of networks of world pictures that explain observations by connecting them by rules to concepts defined in models.

Anthony.Sebastian 01:27, 18 November 2011 (UTC)

Good, I'll make that change. John R. Brews 06:42, 18 November 2011 (UTC)

Duhem & mechanical models

The Note stating:

" Duhem seems particularly to address mechanical models, as "aids to memory". What we would call mathematical models Duhem seems not to have subsumed under the rubric of 'model'. "

is too readily misinterpreted as a summary of Duhem's work, which it is not. There is no doubt that Duhem discusses mathematical formulations, and it is really neither here nor there whether he refers to mathematical formulations specifically by the name "models".

On p. 21 is said:

The definition we have just outlined distinguishes four fundamental operations in a physical theory: (1) the definition and measurement of physical magnitudes; (2) the selection of hypothesis; (3) the mathematical development of the theory; (4) the comparison of the theory with experiment.

There follows a discussion of the merits of theory as a succinct classification system to mentally organize and summarize an otherwise extended miscellany of experimental results.

Duhem introduces mechanical models only to describe and contrast the approaches of scientists in various countries, and to discredit such approaches as "a failure of abstraction". Mechanical models are not the thrust of his thought. On p. 73 Duhem says:

No doubt, wherever mechanical theories have been cultivated, they have owed their birth and progress to a lapse in the faculty of abstracting, that is, to a victory of imagination over reason.

The pertinent chapter is entitled Abstract theories and mechanical models.

John R. Brews 15:07, 18 November 2011 (UTC)

I have modified this note and added a quotation from Duhem. John R. Brews 16:40, 18 November 2011 (UTC)

John, Duhem's thesis reminds me of 'weak-minded' Richard Feynman, who said he always had to visualize his abstractions, or something to that effect.

Do you think the average reader of CZ has a "strong and narrow" mind or a "broad and weak" mind, to use the unfortunate terminology of Pascal/Duhem? Anthony.Sebastian 03:59, 19 November 2011 (UTC)

Hi Anthony: I'll take the easy way on this and say that the number involved in CZ is so small that no statistical assessment is possible because the variance exceeds the measure. ☺ John R. Brews 06:52, 19 November 2011 (UTC)