A final paper for PHIL 362: Topics in the Philosophy of Science.
Karl Popper had an overriding desire to distinguish between authentic science and pseudoscience. For Popper, Einstein had developed a theory that was scientifically falsifiable – unlike Marxist economic theory, or Freudian and Adlerian psychology. Einstein’s theory could be tested to determine whether it was false, which, accordingly, increased its value as actual scientific theory.
Desiring to establish a principle of falsification, Popper relied upon modus tollens, which is demonstrated by the following logical argument:
If P then Q; not Q; therefore not P.
Here is a simple example: If opera composer Richard Wagner was born in Italy (P), then he was an Italian opera composer (Q). Wagner was not an Italian opera composer (Q); therefore, he was not born in Italy (P).
Challenged by David Hume’s stance regarding the problem of induction, Popper attempted to establish that scientific conclusions can be deductive. According to Hume, all scientific conclusions are necessarily inductive; one can never be certain that today’s result will also be the same result tomorrow. Convinced that scientists propose bold hypotheses, Popper argued that scientists will concentrate on testing these hypotheses by using the aforementioned modus tollens. If the hypothesis fails the modus tollens test, the scientist will abandon the hypothesis. However, if a hypothesis withstands the falsification process, with “… detailed and severe tests and is not superseded by another theory…,” it may therefore be considered deductive.
It must also be stated that while a theory may not pass the falsification test, Popper insists that any such non-scientific theory may possess some truth, but that they will remain non-scientific.
Carl Hempel referenced Ignaz Semmelweis’ work with childbed fever to demonstrate the merits of his own theory, which emphasized auxiliary assumptions. Semmelweis initially tried multiple approaches to curb an epidemic of childbed fever raging in one of his hospital’s two maternity wards, but all proved ineffectual. He also noted that medical students did wash their hands with soap before handling patients, but the childbed fever continued unabated. However, the accidental death of one of his colleagues led to the notion that putrid matter (namely corpses handled without sufficient disinfection after) caused the fever. Soap and water did not suffice for disinfection, but chlorinated lime did. This is an example of Hempel’s auxiliary assumptions in action. Although Semmelweis’ instructing his colleagues to use chlorinated solution did not follow from his original hypothesis, and while the original auxiliary assumption (soap and water) proved inadequate, the additional auxiliary assumption (chlorinated lime) was successful.
The following conveys Hempel’s theory:
If both H and A are true, then so is I.
I is not true.
—————-
H and A are not true.
In short, if H (the hypothesis) and A (the auxiliary assumption) are true (all things being equal), then I (the prediction) is also true. If I is not true, then H and A in conjunction are not true. However, it is possible that H is actually true, but that A is inadequate – which may lead to an adjustment or replacement of A. As Hempel states, “Semmelweis’ hypothesis might still have been true: the negative test result might have been due to inefficacy of the chloride…”. In addition, if there are additional auxiliary assumptions with the hypothesis (H and A1, A2, A3…AN), but conclusion I is false (Not I), then either H or A1, A2, A3…AN are problematic. Furthermore, Hempel argued that “initial conditions” play a vital role, which are signified by C, whereas the predicted event is signified by E. Therefore, I (the prediction) has become: If C; therefore, E. However, it is essential to keep in mind that, for instance, if testing equipment is inadequate, then E may fail to happen, even if H and A are correct.
To demonstrate: Tycho Brahe rejected Copernicus’ theory that the earth revolved around the sun. Brahe, without a telescope, observed that fixed stars did not appear to gradually change in appearance or placement in the sky (which, if the earth revolves around the sun, they would presumably do), and developed an auxiliary assumption that the fixed stars are close enough to detect such changes. Accordingly, he rejected Copernicus’ theory. However, Brahe’s auxiliary assumption was incorrect, as fixed stars are far more remote than he assumed.
In conclusion, while Popper’s falsification framework is logically valid, it is inadequate. Challenging Popper’s theory, Hempel indicated that the modus tollens framework must be enriched and extended with “auxiliary assumptions” in order to ensure a sound scientific method. Popper’s use of “If P then Q; not Q; therefore, not P” was deemed too simplistic, in that it did not consider contingencies, faulty equipment, or other auxiliary assumptions. Hempel argued that while testing scientific hypotheses, auxiliary assumptions are “…the rule rather than the exception…,” and that they provide a more reliable framework in which to test hypotheses. Hempel also indicated that a scientist may add one or more additional variables to tease out whether the hypothesis itself is faulty, or to possibly determine if whether one or more auxiliary assumptions are incorrect, yet the hypothesis remain correct.
Consequently, Hempel’s theory lacks the rigidity of Popper’s falsification framework, and allows for creative innovation and contingencies.