Scientia Propter Quid: Method for Confirmation
Let T represent some scientific theory or hypothesis. Also, let O represent some derived/deduced consequence of that theory. If T is true, O will be true, as well, where O is an empirically observable phenomenon that we could, in principle, discover. The method for confirming some theory, T, is given as follows:
1. If T, then O.
2. O.
_________________
3. Therefore, probably T.
If the empirically verifiable predictions or entailments of some theory are confirmed to be true, then a degree of merit is conferred upon the theory that makes those predictions.
THEORY CONFIRMATION AS INDUCTION
Notice that the conclusion, (3), says, "probably T." This is because the argument used in theory confirmation is not deductively valid. It is an inductive/probabilistic argument. At most, it can only show that T is probably true (to whatever degree). In a valid deductive argument, if the premises are true, then the conclusion follows necessarily. The argument form above, however, is not of this sort. While T may be a sufficient condition for O's obtaining, it does not follow with necessity that because O has obtained, T likewise has obtained. It's possible that there are other sufficient conditions that would allow O to obtain. For example, T* might also be a sufficient condition for O's obtaining. Thus, merely finding that O is the case does not determine, necessarily, which sufficient condition, whether T or T*, is responsible for its being the case.
However, the above method for theory confirmation can serve as an inductive or probabilistic approach to confirming some theory. Finding that O is the case increases the probability that T is the case.
What's more, using probability theory often allows scientists or philosophers to determine which theories are more probably true, all things considered (more on this in another post).
HYPOTHETICO-DEDUCTIVISM
The above method for theory confirmation (and disconfirmation. See other post) is a form of scientific reasoning called hypothetico-deductivism (H-D). As Richard Dewitt notes, "The basic idea behind the hypothetico-deductive method is that from a hypothesis or set of hypothesis (or theories, broadly speaking) one deduces observational consequences, and then tests to see if those consequences are observed" [1]. On the H-D method, theories are confirmed or disconfirmed depending on whether their observable consequences are found to be the case or found not to be the case.
To cite an historical example of H-D at work, Einstein's theory of gravitation was confirmed when its predicted effects were found to be true. According to Einstein's theory, gravity is the curvature of space-time. Massive objects cause space to curve in and around themselves (think of a bowling ball at the center of a trampoline). If this is true, then light itself should curve around massive objects as it travels by. In 1919, Arthur Eddington set out to capture a solar eclipse. With the sun being eclipsed by the moon, a number of stars on the periphery of the sun were observable. The original location of these stars was known, but the eclipse showed that they appeared to be shifted away from their original locations. This is exactly what was expected if Einsteins theory was true, since massive objects (like the sun) would curve the light that passed by them, making it appear to the observer that the stars had shifted away from their original locations.
T: Einstein's theory of gravitation: space is curved around massive objects.
O: light will curve around massive objects. If a massive object interferes with the trajectory of light coming from a star to observers on earth, it will appear that the star has shifted.
T entailed O. Since O was found to be the case, T was confirmed.
H-D AS INDIRECT CONFIRMATION
An important point to note is that, according to the hypothetic-deductive model of confirmation (and according to the axiomatic approach to theories -- closely related to the H-D model), theories are indirectly confirmed. Thus, when we ask for evidence or confirmation for some theory, we should keep in mind that, given what theories are (more on this in another post), the kind of evidence we receive will generally be indirect in nature. Alex Rosenberg states the matter as follows: "...theories are tested [by] 'Hypothetico-Deductivism,' according to which scientists theorize -- frame hypotheses -- but do not test them directly, because like most theories in science they are typically about processes we cannot directly observe. Rather the scientist deduces testable consequences from these hypotheses. If the tests are borne out by observation, the hypotheses are (indirectly) confirmed" [2].
____________________________________________________
Footnotes:
1. Dewitt, Richard. Worldviews: An Introduction to the History and Philosophy of Science. 2nd ed. N.p.: Blackwell Publishing Ltd, 2003. Print.
2. Rosenberg, Alex. Philosophy of Science: A Contemporary Introduction. 3rd ed. New York: Routledge, 2012. Print
1. If T, then O.
2. O.
_________________
3. Therefore, probably T.
If the empirically verifiable predictions or entailments of some theory are confirmed to be true, then a degree of merit is conferred upon the theory that makes those predictions.
THEORY CONFIRMATION AS INDUCTION
Notice that the conclusion, (3), says, "probably T." This is because the argument used in theory confirmation is not deductively valid. It is an inductive/probabilistic argument. At most, it can only show that T is probably true (to whatever degree). In a valid deductive argument, if the premises are true, then the conclusion follows necessarily. The argument form above, however, is not of this sort. While T may be a sufficient condition for O's obtaining, it does not follow with necessity that because O has obtained, T likewise has obtained. It's possible that there are other sufficient conditions that would allow O to obtain. For example, T* might also be a sufficient condition for O's obtaining. Thus, merely finding that O is the case does not determine, necessarily, which sufficient condition, whether T or T*, is responsible for its being the case.
However, the above method for theory confirmation can serve as an inductive or probabilistic approach to confirming some theory. Finding that O is the case increases the probability that T is the case.
What's more, using probability theory often allows scientists or philosophers to determine which theories are more probably true, all things considered (more on this in another post).
HYPOTHETICO-DEDUCTIVISM
The above method for theory confirmation (and disconfirmation. See other post) is a form of scientific reasoning called hypothetico-deductivism (H-D). As Richard Dewitt notes, "The basic idea behind the hypothetico-deductive method is that from a hypothesis or set of hypothesis (or theories, broadly speaking) one deduces observational consequences, and then tests to see if those consequences are observed" [1]. On the H-D method, theories are confirmed or disconfirmed depending on whether their observable consequences are found to be the case or found not to be the case.
To cite an historical example of H-D at work, Einstein's theory of gravitation was confirmed when its predicted effects were found to be true. According to Einstein's theory, gravity is the curvature of space-time. Massive objects cause space to curve in and around themselves (think of a bowling ball at the center of a trampoline). If this is true, then light itself should curve around massive objects as it travels by. In 1919, Arthur Eddington set out to capture a solar eclipse. With the sun being eclipsed by the moon, a number of stars on the periphery of the sun were observable. The original location of these stars was known, but the eclipse showed that they appeared to be shifted away from their original locations. This is exactly what was expected if Einsteins theory was true, since massive objects (like the sun) would curve the light that passed by them, making it appear to the observer that the stars had shifted away from their original locations.
T: Einstein's theory of gravitation: space is curved around massive objects.
O: light will curve around massive objects. If a massive object interferes with the trajectory of light coming from a star to observers on earth, it will appear that the star has shifted.
T entailed O. Since O was found to be the case, T was confirmed.
H-D AS INDIRECT CONFIRMATION
An important point to note is that, according to the hypothetic-deductive model of confirmation (and according to the axiomatic approach to theories -- closely related to the H-D model), theories are indirectly confirmed. Thus, when we ask for evidence or confirmation for some theory, we should keep in mind that, given what theories are (more on this in another post), the kind of evidence we receive will generally be indirect in nature. Alex Rosenberg states the matter as follows: "...theories are tested [by] 'Hypothetico-Deductivism,' according to which scientists theorize -- frame hypotheses -- but do not test them directly, because like most theories in science they are typically about processes we cannot directly observe. Rather the scientist deduces testable consequences from these hypotheses. If the tests are borne out by observation, the hypotheses are (indirectly) confirmed" [2].
____________________________________________________
Footnotes:
1. Dewitt, Richard. Worldviews: An Introduction to the History and Philosophy of Science. 2nd ed. N.p.: Blackwell Publishing Ltd, 2003. Print.
2. Rosenberg, Alex. Philosophy of Science: A Contemporary Introduction. 3rd ed. New York: Routledge, 2012. Print
Comments
Post a Comment