Concepts are concepts, and equations are equations, and never the twain shall meet; abduction and realism vs anti-realism

As theories and their theoretical concepts in physics are central to the realism vs antirealism debate, I’m going to be mainly referring to theoretical concepts as it relates to physics. Before I begin I also want to clarify how I understand the meanings of the terms I’m mainly concerned with. These will be the working definitions I will be using in this post.

Realism: the view that theoretical concepts are about real entities, or approximations thereof, and somehow have epistemic value, whether by abductive inference or by association with empirically verified mechanics.

Theoretical concepts on realism: any conceptual framework for mechanical equations or abductively inferred explanation generally of which the subject is unobservable.

Antirealism: the view that theoretical concepts are not about real entities, or approximations thereof, and have no epistemic value, whether by abductive inference or by association with empirically verified mechanics, but are purely instrumental in nature.

Theoretical concepts on antirealism: concepts of unobservable and therefor imagined existence that have no epistemic value, but only have value instrumentally as a conceptual framework in formulating mechanical equations.

Mechanical equations: mathematical descriptions in physics of how generally matter behaves.

I’m becoming more and more convinced that the distinction between theoretical concepts on realism and on antirealism is being used in an arbitrary manner. If antirealism is true then apriori theoretical concepts have no epistemic value, and mechanical equations have no bearing on the epistemic value of theoretical concepts since there isn’t any, nor do theoretical concepts have any bearing on whether or not the mechanical equations are correct since they are verified by observation irrelevant of what theory they are attached to.

On realism I think it can be reasonably demonstrated that the use instrumentally of a theoretical concept as a conceptual framework for mechanical equations has no bearing on that theoretical concept’s epistemic value. Nor is there any bearing on how well mechanical equations in physics work based on whether or not the theoretical concept is true on which it’s conceptual framework is based. I think the latter is quite obviously true, and the way I see it, both can be demonstrated through empirical equivalency.

So either way, on both realism and antirealism, mechanical equations have no epistemic bearing on theoretical concepts, nor do theoretical concepts have any bearing on the accuracy of mechanical equations.

As far as I can tell, anti-realism of any stripe at its core says that abductive inferences, and therefore theoretical concepts if derived from abductive inferences, have no epistemic value, whereas realism of any stripe at its core seems to leave open whether or not abductive inferences do have epistemic value.

However the realist has no logical justification to claim that theoretical concepts are based on reality from simply being associated with particular mechanical formulas. It could only find logical justification through abductively inferring so from observable physical evidence.

As a caveat, an anti-realist can take a constructive empiricism view that even though it can be said that a theoretical concept is about something real, there’s just no way to know if it’s a true or false theoretical concept because of the nature of abductive inferences. Or the other view of logical empiricism can be held that whatever can’t be observed or detected is just imagination, and therefore theoretical concept as abductive inferences concerning unobservables have no significance whatsoever as far as reality is concerned.

However, in order to avoid being arbitrary it seems the same position would have to held for all theoretical concept including those where the core subject of the theoretical concept is unobservable, including theoretical concept for things like abiogenesis, evolution, relativity, and big bang cosmology. That is to say that as an antirealist there’s either no way to know that any theoretical concept is either true or false, or at best, that they’re all only of instrumental value with no real hold on reality.

Now I believe in my last post regarding some of the objections that were raised this arbitrary switching back and forth is one thing that was going on. And to address another objection in my last post, it’s understandable that there are those who dismiss my arguments based on my being a layperson and lacking credentials. However, I would point out that the fact that I am a layperson, and that there are plenty of credentialed individuals on this forum, it would seem like my position, if it were unfounded, should be easily challenged and convincingly beyond any reasonable doubt be shown to be so by any of those credentialed individuals.

So it seems to me that if this particular objection is the best objection some can raise to the arguments I’ve made, there must be some pretty significant merit to the arguments I’m making. And in light of that, unless there are some other objections that I haven’t sufficiently responded to that I’m forgetting about, or not aware of, or until someone can present convincing objections that I don’t have any legitimate response to that refute or rebut the arguments I’ve put forth, I think that I’m at the least more than sufficiently warranted to hold to my positions.


That’s not quite right.

Some equations are accurate in limited contexts, but not fully correct.

So equations can represent flawed theories (like equations that work for Newtonian principles) … or they can represent better theories - - we assume that E = mc2 is completely correct…

At least, we assume this until someone discovers another part of that equation that nobody has encountered yet.

One undefined term critical to your post is “observed”. You’re going to need to firm that up.

What exactly do you mean by “represent”?

How do you define it?

That’s just evasion. But I’d say that something is observed if there’s a sufficiently strong inferential chain from sensory input to phenomenon. I, for example, have observed the DNA sequence of the human insulin gene if I read it off the file output of a sequencer. There’s a long chain of inference necessary to come to that conclusion, but it’s a strong one. I have observed that Archaeopteryx lithographica had asymmetrical flight feathers if I examine a photo of the fossil. Another long but strong chain of inference.

Now you.



I don’t think it is a magical word.

E=mc2 can be said to REPRESENT the theory of relativity… do you have problems using that word?
If you do, give me a better one and I’ll be happy to use it.

Jim, the problem is not your lack of credentials per se. If you had no formalized training but actually knew some science at a technical level (e.g. like @structureoftruth), it would be no problem. The issue I see in your post is that there are a lot of undefined or vague terms:

  • Observed vs. unobserved (as @John_Harshman has noted)
  • Epistemic vs. instrumental value
  • “Based in reality”
  • Abductive inference vs. mechanical equation - your stated definition is very vague.
  • Constructive vs. logical empiricism

Some of these terms I’ve encountered before by casually reading philosophy of science literature, but you seem use them in a non-standard or unclear way, so I’m left guessing what you exactly mean. You also don’t give any examples of “observed” vs. “unobserved” quantities, abductive inference vs. “mechanical equation”, nor how a particular stance (constructive vs. logical empiricism) would view a specific theory (e.g. special relativity). Instead, you speak in broad generalities, casually citing “abiogenesis, evolution, relativity, and big bang cosmology” as examples of “theories where the core subject is unobservable” - these are 4 big separate domains of science, each with their own methodologies and standards of evidence!

Because of this, I have basically no idea what you mean by the following sentence:

What does it mean for a mechanical equation to have an “epistemic bearing” on a theory? I’m not even sure how you differentiate mechanical equations from a theory.

This is why specific knowledge of the science is important - not just to prove or disprove your case, but just to communicate what you’re thinking. It’s very tempting to try to interpret your post as it is and dispute the parts that I “disagree” with. But I would just be disagreeing with my own interpretation of what you wrote and we would just be wasting time and electricity.


@jim would you download and read a book if I suggest one to you? I think it would help you formulate your thoughts.


What kind of inference is being made? Deductive, inductive, abductive?

I would define observation as it’s generally understood, and that for specific observations that are in question due to the nature of how they were observed they would need to be judged on a case by case basis. And if any abductive inferences are involved it would be arguable whether or not an observation could be considered a confirmed observation.

However, for the purposes of the op, all that is required to engage with the argument is to acknowledge that generally there are entities that are observable without dispute, and others which are not observable without dispute. Sorting out the gray areas can be bracketed off for the moment as they are not central for engaging with the argument.

I would say a better way to express it is that the theory of relativity is the conceptual framework from which the equation E=mc2 was formulated. And in regards to the op the claim is that there is no attributable epistemological value for the theory itself in regards to that specific conceptual framework relationship.

What is the relevance of the “technical level” you are referring to with the issues I’m raising? I think I do have the relevant philosophical knowledge required for the nature of the issues I’m raising which center around how certain inferences are being made. I don’t see how a “technical level” of science would make any difference for issues that are philosophical in nature. Maybe you can explain to me how that works?

  • Observed vs. unobserved (as @John_Harshman has noted). See answer to John above.
  • Epistemic value means it provides some degree of knowledge. “Instrumental” is a term coined by antirealists for what I understand as the value of a theory “instrumentally” as a conceptual framework for mechanical formulas of physics while at the same time having no epistemic value of it’s own.
  • Reality generally the world or the state of things as they actually exist.
  • Abductive inference vs mechanical equation. I’m not aware of what definition you are referring to but maybe the answer I give below to the next part of your post is what you had in mind?
  • Constructive vs. logical empiricism. Crucially, unlike logical empiricism, constructive empiricism interprets theories in precisely the same manner as realism. The antirealism of the position is due entirely to its epistemology—it recommends belief in our best theories only insofar as they describe observable phenomena, and is satisfied with an agnostic attitude regarding anything unobservable. The constructive empiricist thus recognizes claims about unobservables as true or false, but feels no need to believe or disbelieve them. (I meant to provide this link in the op, but forgot. :slight_smile:)

These are my initial attempts at defining them. If you have a better way to define them, I’m all ears. :slight_smile:

Regardless of what the methodology or standard of evidence, that doesn’t change the fact that on realism they are all abductive inferences. That is the point that was being made.

Mechanical equations are descriptive and have deductive answers where only one answer is possible, and are inductively verified by observation where only one observational result is possible. Theories, as I’m using the term, on the other hand, being generally directed at unobserved phenomena are abductively inferred explanations with more than 1 answer possible.

So to answer your question with a question, since abductive inferences in science are informed from scientific evidence, i.e., objective and verifiable observations, how is it that equations as abstractions can inform, or have any “bearing epistemically” on an abductively inferred theory?

I think the I have adequate knowledge for the issues I’m raising. However, my experience is that scientists in general have at times difficulty agreeing on terms, or sometimes use terms that are somewhat confusing. Seems to me that there might be a double standard here if I’m expected to do better than what scientists themselves, and academics in general, seem to have a notable degree of difficulty doing.

I always welcome new input. I would ask, though, in what way is it relevant to the op?

I will admit that I’m not quite sure of the difference between abduction and induction. Science is very seldom deductive; at the very least the premises must be arrived at by means other than deduction. Why do you ask?

By the way, I notice that you never answer my questions, though I answer yours.

Would you? Then why won’t you? It’s annoying of you to demand something from me, which I then delivered, that you’re unwilling to do yourself.

There is no answer. You merely allude to the possible existence of an answer and suggest that there is ambiguity.

1 Like

It is relevant in that it will give you a good overview of some of the topics you are trying to address. I agree very much with @dga471’s critique. I don’t think you are speaking the same language as us. The issue isn’t merely how much you know about science, but also what you know of philosophy and epistemology.

This is the book I think you should read, before posting much more:

I want to be clear that this is not a put down. Earlier this year, a philosopher I respect was frustrated by some things I was saying, and he asked me to read this book. It was not a waste of time. It provides a very good overview of

Now, the particular reason I was asked to read this book, I don’t know if it was valid. In the end, I think we were using a term in legitimately different ways, partly due to disciplinary differences. Still, it built up credibility to actually follow through and learn more about his field. Moreover, I actually learned something. In the end, I am not an expert in every field. I need to be humble enough to slow down and learn before I attempt to argue for my position.

@jim I think this is a topic you really care about. Before trying to press your case, I hope you stop. Pause. Take a breadth, and read this book. Think about it, and discuss what you learned from it that might benefit others. And then after all that, if it still makes sense, see how to frame some of your ideas in light of that new knowledge.


How’s this?

The distinction here between the observable and the unobservable reflects human sensory capabilities: the observable is that which can, under favorable conditions, be perceived using the unaided senses (for example, planets and platypuses); the unobservable is that which cannot be detected this way (for example, proteins and protons). This is to privilege vision merely for terminological convenience, and differs from scientific conceptions of observability, which generally extend to things that are detectable using instruments (Shapere 1982). The distinction itself has been problematized (Maxwell 1962; Churchland 1985; Musgrave 1985; Dicken & Lipton 2006) and defended (Muller 2004, 2005; cf. Turner 2007 regarding the distant past). If it is problematic, this is arguably a concern primarily for certain forms of antirealism, which adopt an epistemically positive attitude only with respect to the observable. It is not ultimately a concern for scientific realism, which does not discriminate epistemically between observables and unobservables per se .

OK, I’m going to give more examples of how your definitions are still not clear, and why you’re not communicating effectively. Let’s take your answer to the question of “observed” vs. “unobserved”.

Let’s start out with the following:

I would define observation as it’s generally understood

People ask you to give a specific definition, and you reply by saying “it’s just the common definition.” Do you realize this is not clear at all? As Josh said, we’re not speaking a common language, so we don’t have a common definition. I don’t know what you’re talking about.

and that for specific observations that are in question due to the nature of how they were observed they would need to be judged on a case by case basis

Judging on a case-by-case basis is useless if you don’t tell us what principles you are using to judge them. It could easily turn into “it’s an observation if Jim says so”.

to acknowledge that generally there are entities that are observable without dispute, and others which are not observable without dispute.

What do you mean by dispute? Disputed among whom? Scientists? Philosophers of science? Creationists? Flat earthers?

Even among physicists, there are disputes about whether, say, Feynman diagrams have physical reality or are just calculational shorthands. Did Hulse and Taylor already observe gravitational waves in 1974? How many links in the chain of inference are allowed before it becomes “disputed”?

Finally, I’d like to repeat again: as @John_Harshman mentioned, scientists generally don’t distinguish between abductive and inductive inferences. If you don’t want to carefully define these terms, then you will get nowhere.

Well, this is a website run by a scientist, filled with majority scientists. If you want to talk about philosophy with scientists, then you have to learn how to speak their language.

1 Like

It seems to contradict your earlier claims, and also seems to be an unattributed quote from some unidentified source, which you should avoid.

Ah, I see there’s a link. The new color for links doesn’t make it very visible. Apologies. But it still contradicts what you said previously.

How so?

@swamidass So far I like what I’ve read from this book you recommended. :slight_smile:

The philosophy of natural science is basically the study of what natural science is, what it does, how it works, why it works and how far it works. A reasonable place to begin would be with a definition of natural science. However, the term has no standard, accepted definition. That might seem to be an insurmountable difficulty. How can we investigate the nature of science if we do not, strictly speaking, know what we are talking about? But such problems are not insurmountable in comparable situations. For instance, it is almost a cliché that no one can define love. But that does not stop us from proclaiming (often correctly) our undying version of love to select persons on Valentine’s Day, and it does not keep us from marrying for love. We can often recognize instances of and characteristics of a concept even if we are unable to formulate an ironclad definition of it, and we often have a good general idea even if we cannot specify all of the details. Such is the case with the general concept of science.


Great. Thanks for picking it up.

So take a pause from the forum to focus on read that maybe? I’ll be curious to hear what you report back after getting through that book. :slight_smile:

1 Like

If it wasn’t you who said atoms were observable, I apologize.

Now of course there is some overlap between what we can observe and what we can know to be true, but we can observe things that aren’t true, and we can know things are true without observing them. So I’m not sure what your major point is supposed to be.

1 Like

This topic was automatically closed after 7 hours. New replies are no longer allowed.