Books on Amazon
Measuring/Duhem: e.g. Holism: In the beginning it was necessary to calibrate the glass tube, not only to the abstract concepts of arithmetic and geometry, but also to the abstract concept of mass, to the hypotheses of general mechanics and the ones of the sky, which justify the use of the balance for the comparison of mass. The specific gravity of the mercury had to be known at the temperature at which this calibration could be carried out, and for this purpose the specific gravity at 0 °, which cannot be used without knowing the laws of hydrostatics. It was necessary to know the law of the expansion of mercury, which is determined by means of an apparatus in which a magnifying glass is employed, that is, also certain laws of optics.
Measure/Duhem: One does not use any instrument, but a certain one, with characteristic constants.
The experimenter works with two different conceptions of the instrument he uses: the one is the image of the concrete instrument, the other is a schematic type of the same which is constructed with the help of the symbols provided by this theory. It is the ideal and symbolic instrument which he thinks about, to which he applies the laws and formulas of physics.
Accuracy grows not only through improved instruments, but also because the theories provide ever more appropriate methods for establishing the relations between the facts and the schematic concepts which they are intended to represent.
Growing accuracy is bought by growing complexity. At the same time, one must observe secondary facts. The transformations which are made in the directly given facts of the experiment are the corrections.
If the physical experiment were the simple statement of a fact, it would be absurd to make corrections to it.
The experimenter constantly compares two instruments: the ideal and the real. The assistant focuses on the actual, on the second (in which, for example, ideal gases move instead of the concrete ones) the leader applies the laws of hydrostatics. The relationship between the two instruments is the quasi correction.
E.g. the assistant states the height of the mercury column, the leader corrects it. This is due to intellectual operations.
Whoever sees in the experiments only findings of facts will not understand the role of the corrections. Not even the so-called systematic errors.
Systematic errors: if one leaves the cause of a systematic error in an experiment, it means that one has been satisfied with a too simple theoretical picture.
Measuring: E.g. no thermometer can determine the highest temperature. The symbol for absolute temperature does not correspond to any of the measuring methods.
What is called a perfect gas in thermodynamics is only an approximate picture of a real gas.
Ziel und Struktur der physikalischen Theorien Hamburg 1998