4.8 TAXONOMY OF SCIENTIFIC MEASUREMENTS
Scientific measurement ultimately always involves a human observer manipulating deliberately organized sets of objects of experience generally referred to as the measuring apparatus. However, the organizing principle involved in each case is different according to the intention of the observer. This allows us to produce the following taxonomy of interactions between an observer armed with appropriate apparatuses and the object he observes, according to their intended purpose. First, there is ordinary measurement in the service of quantifying the behavior of an object of experience such as Galileo conducted in his desire to quantify the trajectory of a freely falling body. In that measurement, he used a yardstick to ascertain the location of the body by noting down the coincidence between the instantaneous position occupied by that body and markings he had previously made on that yardstick, and he used a rudimentary clock consisting of water flowing out through a hole in the side of a container to ascertain the moment when that coincidence had occurred. The purpose for which Galileo set-up this arrangement of objects of experience—the body, the yardstick, and the clock—was to quantify the behavior of the body as it freely fell to earth. As we know, he found that the distance traveled along the yardstick was proportional to the square of the time it took the body to traverse that distance and was independent of the nature of that body. This mode of quantifying the behavior of an object of experience in terms of comparisons made between an indicator body, here the body itself, and a set of calibrated bodies, here the yardstick and the clock, operates in all ordinary measurements and is made possible by the axiomatic project which replaced contextual location with a point in a homogeneous space, and replaced personal time with the uniform ticking of a clock. Similarly, the measurement of the intensity of an electric current, for instance, involves the indicator needle of an ammeter and the measurement of a temperature involves the indicator mercury column of a thermometer.
Next, there is ordinary measurement in the service of testing alternative explanations for what had been observed to be the case in the ordinary measurements which were conducted previously in the service of quantification. Galileo himself expressly desired to avoid such an undertaking as he was then wisely concerned with the question of how the body gravitated to the earth, not with the question of why it did so. Normally, we refer to such measurements as experiments. Experiments are designed to verify the serviceability of one of a set of alternative explanations. As we argued before, explanations which are being tested in this way consist of conceiving in the mind of some mechanism which appears to involve the relevant objects of experience with each other in a way which is qualitatively consistent with the behavior observed. The experimenter then arranges the objects in an appropriate combination designed to verify that the qualitative expectation does translate into a quantitative match. No attempt is made to focus attention on the object of explanation we have introduced in order to mediate the mechanism under consideration; all we want to know is whether that mechanism, and the object mediating it, well serve or not to explain the experience we are examining.
Finally, there is predicated measurement in the service of quantifying the observed behavior of an object of explanation. It is designed to deal with the quite different situation in which the object subjected to measurement is not given to us but must be first manufactured; because the measurement is then predicated upon that manufacturing having taken place, we have called it a predicated measurement. In such a measurement there are two separate apparatuses, one designed to produce the object, another designed to observe it. The first one is a replica of the apparatus that quantified the behavior which was once explained by the object of explanation we are trying to manufacture; its inclusion in the predicated measurement enacts the manufacturing process in the same way in which a re-assembled J.J. Thomson’s apparatus became an electron gun in the Davisson-Germer experimental arrangement. The second apparatus is designed to observe the behavior of the object of explanation thus manufactured in the same way in which the nickel crystal and the electron collector were used to observe electron scattering in that experiment.
For completeness, one might have been inclined to include into this taxonomy the equivalent of an ordinary experiment, that is an undertaking which would have been used to test alternative explanatory mechanisms for the way in which the object of explanation under observation had behaved, and which should therefore have been called a predicated experiment. However, because self-consistency rather than serviceability is the corner stone of a predicated measurement, we do not have the freedom to select amongst alternative objects of explanation the one which would best serve to explain what we have observed; all we can do is ascertain which modifications to the object of explanation we had originally given to ourselves for observation naturally recommend themselves to us. Therefore no such entry as a predicated experiment properly belongs in the taxonomy of scientific measurements.