Effing Science pt 2: Realism vs Anti-Realism

[ydoaPs]

There are two main approaches to the fundamental reality of what science does. The first, called Scientific Realism, is that the goal of science to find the least wrong description of reality that it can. The other set of approaches, called Scientific Anti-Realism, is precisely not that. Most of the versions of Anti-Realism make the distinction between believing a theory is true and accepting that it is empirically adequate. The empirical adequacy, I shall argue, leads inevitably to Scientific Realism. This is often called The No Miracle Argument. In presenting it, I shall discuss Larry Laudans critique of Realism.

 

Laudan has an introductory quote to his paper by Hillary Putnam (from which the class of arguments got its name) that I actually like, though Putnam should have tagged his quantifiers: The positive argument for realism is that it is the only philosophy that doesnt make the success of science a miracle (Putnam, 1975). Setting aside the quantifier issue, though Laudan disagrees, I do think it is true that the success of science is indeed an indicator that scientific realism is true. Laudan splits scientific realism into two partstruth and reference. The argument he gives for his opposition is as follows:

• If scientific theories are approximately true, they will typically be empirically successful.
• If the central terms in scientific theories genuinely refer, those theories will generally be empirically successful.
• Scientific theories are empirically successful.
• (Probably) Theories are approximately true and their terms genuinely refer.

Laudan appears to have put it this way to give himself an advantage rather than to charitably present the position of the people he is criticizing. This is readily apparent in his attack on (2) as he describes it. He gives examples of theories realists say truly refer which are empirically unsuccessful. However, if we flip all of the conditionals to more accurately reflect what the realists hold, that issue vanishes:

• If scientific theories are empirically successful, the scientific theories are likely to be approximately true.
• If scientific theories are empirically successful, the central terms in scientific theories probably genuinely refer.
• Scientific Theories are empirically successful.
• Scientific theories are likely to be approximately true and their central terms probably genuinely refer.

Notice that once we have the premises such that it accurately reflects the sentiments given in the Putnam quote, we now have a valid deductive argument rather than an abduction like Laudan gave. But what of the premises? I shall discuss them in order, but first we need to know what Laudan means by empirical success. Laudans definition is: a theory is successful so long as it has worked, i.e., so long as it has functioned in a variety of explanatory contexts, has led to confirmed predictions and has been of broad explanatory scope." So, it should be obvious that when talking about empirical success, were looking at the set of observation statements from prediction e:{e₁, e₂,, e_n} and the likelihood function for the theory P(e_i|h). So, if P(e_i|h)>P(e_i|~h), that is the observation statement is more likely on the theory being true than the theory being false, for a sufficiently large portion of e (which satisfies Laudans definition of empirical success), the theory is more likely to be true (since it is trivial to show that P(h|e)>P(h) iff P(e|h)>P(e|~h)). That gives us the revised premise 1. Premise two is trivially true on any correspondence theory of truth given revised premise (1):

• If scientific theories are empirically successful, they are likely to be approximately true.
• If a statement is true iff it accurately corresponds to reality (that is, it accurately refers).
• By Hypothetical Syllogism, If scientific theories are empirically successful, they likely accurately correspond to reality (that is, they accurately refer)

So, on a correspondence theory, premise (1) gives us premise (2) for free, and it has been shown that premise (1) is true. This means that the truth of scientific realism only relies on the empirical success of science. Since every use of technology is a test of various scientific theories (the computer used to type this paper relies on Electro-Magnetic theory and Quantum Mechanics, for instance), Id say that scientific theories are indeed rather empirically successful. From which it is validly concluded that scientific realism is true.

 

So, we see, that if one accepts a theory as empirically adequate (as the Anti-Realists say to do), one ought to actually believe that the theory is true.

[ydoaPs]

No Anti-Realists here?

[Ben Banana]

Is Realism true?

Is Socialism true?

Is Jeffery McJiggens Sr. true?

Are socks true?


^
Strainful truth-value application. i.e. invalid questions. It's oxymoronic nonsense to apply truth values to nouns. We are capable of operating in a tacit mode, e.g. "Is the Hebrew Bible true?" may be translated to: "Are the historical accounts (stories) of the Hebrew Bible as a whole, including its directly conveyed claims, factual and correct?"

because nouns are often associated with logically malleable units.

Be careful.

Edited April 24, 2013 by Ben Banana

[kristalris]

Is Realism true?

Is Socialism true?

Is Jeffery McJiggens Sr. true?

Are socks true?

 

 

^

Strainful truth-value application. i.e. invalid questions. It's oxymoronic nonsense to apply truth values to nouns. We are capable of operating in a tacit mode, e.g. "Is the Hebrew Bible true?" may be translated to: "Are the historical accounts (stories) of the Hebrew Bible as a whole, including its directly conveyed claims, factual and correct?"

 

because nouns are often associated with logically malleable units.

 

Be careful.

I'd say you are dodging the issue via use of semantics. In short the noun realism points towards the a priori idea that is taken as to be true that there is a real world out there. The use of logic only works when things are taken to be absolutely true BTW. The oxymoronic nature is thus inherent when forced by logic to take something (in part) to be absolutely true in order to investigate how true it is as a probability for instance whereby you take the exact opposite to be absolutely true at the same time when using probabilistic reasoning.

[krash661]

without the use of words from nature,

 

anthropocentric whether we like it or not

Edited April 25, 2013 by krash661

[tar]

There seems to be a "switch" of position required to use the word refer. That is, you have to be already assuming a two position setup.

 

There is the thing as it is.

There is the reference to it.

 

If we all know what it means to refer to something, then the argument is over, and we are all realists.

 

Science's job is to make the references as accurate and consistent as possible.

[ydoaPs]

Generally, people (regardless of how they think it works) use the Fregean distinction in their theory of meaning. Frege used "sense" vs "reference", but there are other words for this distinction. The point is that there is a difference between reference (pointing at a thing) and sense (what about the thing you're pointing at).

 

The standard dead horse example is the morning star and the evening star. They obviously mean different things and, for a long time, people thought they pointed at different things. It turns out that they're both Venus. So, "Morning Star" and "Evening Star" have different senses, but the same reference.

[tar]

ydoaPs,

 

Though, being that we already know how to refer to something, there is an automatic requirement for a referent and a referee.

 

It was the automatic nature of the "idea" of refering that I was alluding to. That this must indicate we have already done a step, before we have anything to think about, or anything to say about a thing, or anything in mind to point another toward.

 

First there is a thing, that causes us to sense it, then we have some analog representation of this thing that we judge against, compare with and otherwise measure in relationship to other stuff we have sensed and remembered in this analog representational way. The "meaning" of a statement, or a thought, is a denoting of the original object. A representation, a symbolic reproduction of the original, "meant" to stand for the original.

 

Frege, in "sense and reference" might be trying to explain in a complicated way, an activity we have already automatically accomplished.

 

Regards, TAR2

[MonDie]

I agree with the arguments given in the OP, but I think ydoaPs has yet to address the issue of how we can refer to things, especially very small things like molecules. Only recently have I taken an interest in philosophy, but I will do the best I can.

 

Nobody can deny that our world follows principles (or laws). I will make the case that, (1) although the fundamental subject of science is the extent to which these principles can be extrapolated, (2) certain principles lead us to judge theories as true or false.

I do not think physical things are separate from the principles they follow. Rather, I think physical things are manifestations of principles. I hold to this sort of mechanical monism only because I find it extraneous to assume things must have non-principle components.

Now, I will give an example of how science studies the extent to which principles can be extrpolated. Kinetic molecular theory is the extrapolation of certain physical laws to molecules. I don't know the history, but I assume the progression went as follows. First, we derived these principles of motion from large scale phenomena. Second, we tested the hypothesis that these same principles apply to molecules.

Yet there is a problem that arises from the above. We judged that large, visible objects and small, invisible molecules follow the same principles of motion, but we did so using different observations in each case. We were able to do this because of Occam's razor. Hypothetically, there are infinitely many explanations one could propose for a given set of observations, making it impossible to falsify every single alternative. However, kinetic molecular theory was preferred because it was the most conservative assumption-wise.

Although Occam's razor tells us what makes one theory more probable than another, it itself is not a theory, it is a principle. Unlike a theory or hypothesis, it has no competing alternatives. You can observe this principle in action yourself by applying the scientific method, and observation (step 1 of the scientific method) is all that is needed for verification.

 

I think this makes me a scientific realist. Does everyone else agree that I am a realist?

Edited April 27, 2013 by Mondays Assignment: Die

[Iggy]

[channeling frank oz] Hello, Yoda. Interesting post, that was. [/channeling frank oz]

 

There are two main approaches to the fundamental reality of what science does. The first, called Scientific Realism, is that the goal of science to find the least wrong description of reality that it can. The other set of approaches, called Scientific Anti-Realism, is precisely not that. Most of the versions of Anti-Realism make the distinction between believing a theory is true and accepting that it is empirically adequate. The empirical adequacy, I shall argue, leads inevitably to Scientific Realism.

 

You have it almost exactly the opposite from how I've ever heard it. Both realism and anti-realism consist of finding the "least wrong description". That doesn't distinguish them. The second half of your paragraph you have the wrong way around. Compare what you say to what Michael Dummett (who coined the term anti-realism) says:

Most of the versions of Anti-Realism make the distinction between believing a theory is true and accepting that it is empirically adequate.

-You

The conflict between realism and anti-realism is a conflict about the kind of meaning possessed by statements of the disputed class. For the anti-realist, an understanding of such a statement consists in knowing what counts as evidence adequate for the assertion of the statement, and the truth of the statement can consist only in the existence of such [adequate] evidence...

-Dummett

 

Or, compare to the first two lines of the SEP article on constructive empiricism:

 

Constructive empiricism is the version of scientific anti-realism promulgated by Bas van Fraassen in his famous book The Scientific Image (1980). Van Fraassen defines the view as follows:

 

Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate. (1980, 12)

 

 

 

So... for the anti-realist, the truth of a statement consists *only* of the evidence of the statement. To be empirically adequate is to be true. They are one and the same. Put in math terms: the proof of the statement is the validity of the statement.

 

The empirical adequacy, I shall argue, leads inevitably to Scientific Realism.

 

Scientific realism requires more than empirical adequacy. Dummett's quote continues to explain,

...For the realist, the notion of truth plays a more crucial role in the manner of determining the meaning of the statement. To know the meaning of the statement is to know what it is for the statement to be true: we may in the first place derive such knowledge from learning what is counted as evidence for its truth, but in this case we do so in such a way as to have a conception of the statement’s being true even in the absence of such evidence.

-Michael Dummett, Realism, 1959

The anti-realist considers empirical support to be the truth of a thing, while the realist finds empirical support indicative of the truth of a thing, but there is an additional aspect of truth for the realist. Truth involves something like a Platonic form -- a quality independent of evidence.

 

 

• If scientific theories are empirically successful, the scientific theories are likely to be approximately true.
• If scientific theories are empirically successful, the central terms in scientific theories probably genuinely refer.

<...>

Premise two is trivially true on any correspondence theory of truth given revised premise (1):

 

Your conclusion assumes the truth of the premise. A correspondence theory of truth obviously corresponds. A coherence theory of truth wouldn't necessarily. I don't see a basis for picking the former.

[tar]

Monday's Assignment Die,

 

But a principle is of our design. There is a thing, causing us to derive the principle, but as you say, there could be other explanations. The thing itself remains extant, and complicated, and extremely capable of continuing to be, without us writing the ultimate equatiion for it.

 

How we can refer to molecules is a matter of grain size, taking what we know works, and applying it on another scale. There do seem to be analogies that carry through...but not all components operate exactly as imagined. We find descrepencies, things that are happening differently, because the scale is different. Brownian motion for instance happens under a high powered microscope, and is a reality on that scale that doesn't exhibit itself at the grain size we normally operate under.

 

On our scale of size and duration, a rock is rather hard and solid. On the molecular level, it turns out to be mostly space and fields of magnetic and nuclear forces. Same principles that operate here, don't apply there. Except here is there, and there, here. So the principles are more of our design, put into place by us, to describe what applies, IF one was to shift to a particular grain size, and define the size and duration under consideration.

 

Doubtful that the principle itself has any way to be real, without the manifested reality from which we derive it.

 

Regards, TAR2

Edited April 28, 2013 by tar

[ydoaPs]

[channeling frank oz] Hello, Yoda. Interesting post, that was. [/channeling frank oz]

 

 

You have it almost exactly the opposite from how I've ever heard it. Both realism and anti-realism consist of finding the "least wrong description". That doesn't distinguish them. The second half of your paragraph you have the wrong way around. Compare what you say to what Michael Dummett (who coined the term anti-realism) says:

Most of the versions of Anti-Realism make the distinction between believing a theory is true and accepting that it is empirically adequate.

 

-You

The conflict between realism and anti-realism is a conflict about the kind of meaning possessed by statements of the disputed class. For the anti-realist, an understanding of such a statement consists in knowing what counts as evidence adequate for the assertion of the statement, and the truth of the statement can consist only in the existence of such [adequate] evidence...

 

-Dummett

Or, compare to the first two lines of the SEP article on constructive empiricism:

Constructive empiricism is the version of scientific anti-realism promulgated by Bas van Fraassen in his famous book The Scientific Image (1980). Van Fraassen defines the view as follows:

 

Science aims to give us theories which are empirically adequate; and acceptance of a theory involves as belief only that it is empirically adequate. (1980, 12)

 

 

Anti-realism isn't about reality; it's about how good a theory is at predicting the position of a needle. It's not saying science is a description of reality. I can assure you that I did not get them nor the rest of the paragraph flipped around.

 

So... for the anti-realist, the truth of a statement consists *only* of the evidence of the statement.

Which is precisely contrary to when you said "Both realism and anti-realism consist of finding the 'least wrong description'".

To be empirically adequate is to be true. They are one and the same.

Not to the Constructive Empiricist and the other anti-realists. They're wrong, and I've proven it mathematically.

Your conclusion assumes the truth of the premise. A correspondence theory of truth obviously corresponds. A coherence theory of truth wouldn't necessarily. I don't see a basis for picking the former.

There is no defensible coherence theory of truth. None. Not one.

 

There are two main things philosophy does when doing something like coming up with a theory of a concept. One is conceptual analysis (teasing out intricacies of everyday concepts) and the other is explication (creating more precise counterparts to everyday concepts which are often (hopefully most of the time if not all of the time) able to replace the ordinary concept.

 

Any attempt at a coherence theory of truth fails on both counts as it is a necessary condition of the everyday concept of truth that a true statement describes reality as it is. That's completely counter to any coherence theory and as such is not a conceptual analysis of the ordinary concept and cannot be interchanged with the ordinary concept which means it is also not an explication of the ordinary concept.

 

Coherence theories of how we can be justified in believing something is true however, are easily justifiable.

[EdEarl]

I voted other, because the dictionary definition of realism was ambiguous; though, it seemed that realism was related to the way we sense reality. However, we know from physics that our senses do not perceive the quantum world, and thereby lie to us. Things are almost entirely empty space, yet we sense solids, liquids and gasses.

 

Basically, I do not know what real is true means or what anti-real is true means, but neither one seems right.

 

Other seemed to be the best choice, because I think we do not know.

[Ophiolite]

The standard dead horse example is the morning star and the evening star. They obviously mean different things and, for a long time, people thought they pointed at different things. It turns out that they're both Venus. So, "Morning Star" and "Evening Star" have different senses, but the same reference.

The same reference? I think not. Since a real thing cannot be divorced from the context in which it is real then the Morning Star and Evening Star are different.

[ydoaPs]

They're not both Venus?

[Ophiolite]

They're not both Venus?

They are both Venus, but you are restricting your interpretation to the physical expression of the planet while ignoring the context in which it appears. Context is hugely important.

[ydoaPs]

They are both Venus, but you are restricting your interpretation to the physical expression of the planet while ignoring the context in which it appears. Context is hugely important.

Which is the distinction I was pointing out. The words point to the same thing, but they point to different things about it.

[MonDie]

Language was meant to interact with reality. Through this interaction, concepts that are different in the abstract can be identical for empirical purposes.

[Science&Society]

The “two main approaches” to the fundamental reality of what science does, as asserted in the OP, depends very much upon an initial assumption about Scientific Realism in relation to all other metaphysical positions in the philosophy of science.

 

Although Scientific Realism is a popular view of science, it is not the only philosophical approach that explains what science does. Other positions such as Instrumentalism, Epistemological Anarchism, Constructivism, Empirical Falsificationism and more explain scientific activity or elements of scientific activity through different analytical techniques and metaphysics. What stands Scientific Realism apart is the focus on “truth” as an empirical qualifier in the relationship between scientific theory and physical phenomena. It is on the basis of this metaphysical difference that any philosophy not appealing to a doctrine of truth, and instead focuses on development of scientific theory in different ways, is classified as “anti-realist”.

 

The distinction between realism and anti-realism is really a political distinction, rather than a rational or scientific distinction. By classifying other metaphysical positions as “anti”, Scientific Realism is normalized as something positive against which other positions must be considered. But as often the case, non-realist positions are based, and developed, on other metaphysical prerogatives in their attempts to explain scientific activity. Any comparison between realist and anti-realist position is inappropriate because the doctrine of truth has no metaphysical significance in "anti-realist" theory. (It could even be argued that Scientific Realism’s adherence and demand to view knowledge in terms of truth is essentialist and anti-scientific. But that is a topic for a different debate.)

 

The assumption that I alluded to which the OP makes is that Scientific Realism is normal, standard, or natural; and incorrectly assumes other positions as hostile, challenging, and adverse, and therefore inherently opposed to realism. For example, the OP states:

 

“Most of the versions of Anti-Realism make the distinction between believing a theory is true and accepting that it is empirically adequate.”

 

But if you read the “anti-realists” such as Feyerabend or Kuhn, they make no such distinction, nor explore the difference between realism or anti-realism. Only through a Scientific Realist interpretation of the issues in their work do their respective theses appear to be anti-realist. The perception that numerous alternate positions are anti-realist ultimately requires reading and translating “anti-realist” ideas within the interests of realist metaphysics, especially truth evaluation (e.g. "Is antirealism true?" is a realist, not an anti-realist question). "Anti-realist" positions are respectively based on different metaphysical groundings.

Edited July 26, 2013 by Science&Society