Mario,

Sorry I couldn't get back sooner, I'm finding myself swamped...

***H: "My reply was directed at the principle of justifying scientific realism using Putnam's miracle argument." M: Perhaps, then, I am not a scientific realist. Judging by what you know of my opinions, what would you classify me as? Do you find my views to be internally inconsistent, or do you just disagree with them?***

Your carburetor is misfiring is that metaphor helps.

***H: "You are mixing up 'theories have [pragmatic benefits] in the outside world' and our theories being approximately true. These are two different issues. You can infer all day as to why our theories have pragmatic benefits, but justification doesn't come from explanation." M:
I disagree. I say that if a theory has tangible pragmatic benefits in the outside world that correlate to what that theory predicts what would happen, then it can be inferred to be "approximately true."***

Why? What is it about a primitive homo sapien intelligence that will probably look like an amoeba intelligence by our descendants (assuming we survive) that can be considered as providing 'approximate truth'? Why can't our theories be considered 'approximately false'? Afterall, I think primitive understanding of ether was approximately fasle, right? I think Ptolemy astronomy was approximately false, don't you? I think the Aristotlian physics is approximately false, don't you? What about commonsense understanding of primitive views such as the sun being created anew with every sunrise and self-destructing every sunset, that's approximately false too - don't you think so? What is it about 'modern' views that all of a sudden become 'approximately true'? What trait do they possess that completely false ideas didn't also possess? Maybe our equipment just isn't sensitive enough to show how false are theories are. Theoretical cohesion? I think Aristotle believed he obtained that quality. You see Mario, we cannot clearly demonstrate that are living in a priveleged period where our theories are 'approximately true' whereas the past periods were 'approximately false'.

***Einstein (relativistic physics) is a wonderful example. When Einstein wrote his theory, no one had even considered the possibility of black holes, but yet the theory predicted that they would exist (when his equations were subjected to extreme conditions.) They continued to exist as theoretical objects until observations in the sky showed light bending greatly around what appeared to be empty space. Hence, there was something invisible and incredibly massive there. The most logical conclusion would be that black holes actually do exist, as Einstein's theory predicted. So, I conclude that relativistic physics is superior to describing the universe than classical physics. I correlate this approximate description of the universe to its actual state, independent of our existence (since I am of the opinion that an external reality is neccesary for the functioning of our consciousness, in any context, therefore our observations describe something, no matter how small)***

Unfortunately the problems inherent in Ptolemy astronomy haunt us again. Take a look at this webpage:

http://www.superstringtheory.com/blackh/blackh4.html

Notice a key phrase:

"Note that there is a complication in the relationship between strings and spacetime. String theory does not predict that the Einstein equations are obeyed exactly. String theory adds an infinite series of corrections to the theory of gravity. Under normal circumstances, if we only look at distance scales much larger than a string, then these corrections are not measurable. But as the distance scale gets smaller, these corrections become larger until the Einstein equation no longer adequately describes the result. In fact, when these correction terms become large, there is no spacetime geometry that is guaranteed to describe the result. The equations for determining the spacetime geometry become impossible to solve except under very strict symmetry conditions, such as unbroken supersymmetry, where the large correction terms can be made to vanish or cancel each other out. This is a hint that perhaps spacetime geometry is not something fundamental in string theory, but something that emerges in the theory at large distance scales or weak coupling. This is an idea with enormous philosophical implications."

So, here we are a mere 97 years since Einstein introduced his gravitational theory of relativity, and already it is not exactly the picture. Is it an approximate truth? Or, will we find out in 197 years that it is an approximate falsehood? Time has consistently shown that as our observational and experimental expertise increases, that previous theories explain the discrepancies less and less well. New theories are needed to explain what we observe and those new theories cause us to rethink previous theoretical concepts.

***You said yourself that you consider evidential evidence superior to non-evidential. That is, more accurate. I don't know how you can say that and still claim that pragmatic observations are not valid justifications to approximate truth statements.***

I can say so just fine since I am an (sophisticated) internalist when it comes to my scientific realism. I think you are citing an externalist perspective of explanatory success, and this is the problem with your view. This view is not equipped to account for the discrepancies of science. Here, let me define these terms:

"(i) Externalist, naturalist explanationism. Purely a posteriori methods akin to those in the natural sciences are used to defend realism, and a reliabilist account of why explanatoriness is epistemically relevant is offered: judgements of explanatory superiority are informed by background theoretical considerations, which are, in fact, largely true.
(ii) Internalist, non-naturalist explanationism. Additional constraints on rational belief are revealed by reflecting on the preconditions for the possibility of meaningful inquiry into a subject matter and the cooperative pursuit of understanding. Such preconditions include the acceptance of basic evidential principles without which grasp of relevant concepts would be impossible. Explanatoriness can count as an epistemic virtue when the greater explanatoriness of a theory is grounded in the theory's superior conformity to these truistic commitments." ("The Explanationist Defense of Scientific Reason", Dorit A. Ganson, 2001, p. 71).

Sorry the philosophical jargon, but I think it is important to distinguish these two different forms of realism (as well as use the terminology being used in the philosophy of science). Your attempts to justify scientific truth using externalist means are bound to fail. In my opinion (and the opinion of Ganson, et al) is that only an internalist approach to 'approximate truth' will work. That is, the antirealists have succeeded in disabling the externalist view of truth.

***H: "How do humans justify that they possess truth or even an approximation of truth. There is no means that I know of to obtain this kind of result. If there were such a justification, it should never be wrong. But, theories have been wrong (not just approximately true - but just plainly wrong). Obviously a theory can't be approximately true one moment and fully wrong the next." M: Well, ANY statement about reality is 'true' within a certain margin of error. That is, every truth statement (other than one that is absolutely true, which we can never know for sure) is an approximation. By approximation I mean the best approximation. Saying that the universe is made of earth, air, fire, and water was an approximation, as is relativistic physics. But since relativism is better justified through pragmatic knowledge, I consider it to be more accurate, hence, closer to truth.***

Was Aristotle's view an 'approximate truth'? How about Ptolemy's? What you have here are demonstrations of invented truths. They are truths that satisfy us with only what we see. Improve our vision, increase our sound sensitivity, give us instruments, etc and we are no longer satisfied with those invented truths. We toss them aside and say that they no longer satisfy our classification of an approximate truth. The same can be said of modern science, increase the sensitivity of our instruments, and these new observations lead us to completely replace our old theories to the point to where they no longer look approximately true.

Your objection can still be taken to mean that some theories are always true for that range of observations which they were meant to satisfy. But, you have the natural selection of theories problem where we select the theories which happen to explain something otherwise we do not select them as approximate truths. We keep holding those theories until the theory begins to fail, and which point we label them 'approximate truths for those range of observations with a certain range of error'. Do you see the problem with this? If it predicts badly, we toss the theory aside right away as an incorrect hypothesis. If it holds for a little while we call it a model and then toss it. If it holds for a little longer still, we call it a theory, and an approximation of truth valid for a certain range of phenomena. And, if it holds for a great, great while, we call it a law of physics. This is no more than natural selection of theories that fill a particular niche until 'environmental pressures' cause the theory to evolve or die. To say a theory is approximately true is no better than saying a species is approximately correct. The correctness of a species is only relative to the environment (i.e., human sensitivity to observation) and when the environment changes to apply pressure to that species (i.e., we become aware of new observations that contradict our theories), the species will either die out or acquire further adaptations (i.e., a theory will either adapt to fit the new observables or a new theory will take over). Perhaps us primitive humans are no better than nature in how it selects species based on best fit. In that case, Einstein's GR isn't approximately true, it is currently the most fit to meet our observational requirements. Isn't philosophy fun!

***H: "You cannot justify a position which thinks that our axioms of reality (truth statements) are justified, if that were the case, then we would have absolute knowledge." M: I do not attempt to justify truth statements. I try to avoid it as much as possible. I only try to find out which use statements best fit the observations. I then assign those use statements to a 'proxy' truth statement for the time being, but with full knowledge that it is only an approximation,***

Mario, I think you are trying to live on both sides of realism and antirealism. You want theories to express basic truths (albeit approximate) of the universe, but on the other end you don't want to commit to their truthfulness but rather only want to assert their applicability to being true in a certain range. The problem is that you need to justify the latter. How do you distinguish that a theory is 'approximately true' from a theory is 'appropriately selected'? If theory's are selected by means of some selective criteria, that doesn't make them approximately true. It only makes them appropriately selected based on the current empirical evidence to date. In such a case you cannot state that some theoretical entity likely exists is a derived use statement. In order for something to be a use statement it must be a justified position. However, as we see with the natural selection of theories, you cannot be justified in saying something likely exists when you can just as easily justly say that something likely doesn't exist since this is a mere selected theory based on current phenomological evidence.

Warm regards, Harv