3

In the first physics lecture I ever visited, the professor shortly demonstrated the fallibility of our senses with a spinning disc optical illusion to justify the use of measurement devices. He didn't say a word about where these measurement devices “come from”.

If we can't intuit certain propositions by reason alone (i.e. for a simple clock, that identical pendulums will behave in the same way – otherwise, how could a comparison ever be made to get to Galileo's result that the period of a pendulum is independent of the amplitude of its swing?), how can we construct measurement devices? How can we do it, if we assume that scientific observation presupposes that we have and can use measurement devices?

Was there a philosopher who pointed to measurement devices as an argument for rationalism? And how could an empiricist explain that the construction of measurement devices is compatible with strict empiricism?

viuser
  • 4,751
  • 1
  • 17
  • 50
  • Story is that Galileo had a very regular heartbeat and used his pulse to discover his rule. No mathematical theory can be a physical theory unless there's experimental evidence. Of course that proposition is under serious attack from the string theorists and multiverse crowd these days. Not sure if I entirely understand your question but the connection between Galileo's pendulum and his pulse ("The pulse and the pendulum") seemed on point. – user4894 Mar 18 '17 at 04:06
  • @user4894 I don't think any rationalist ever thought that a full-blown physical theory doesn't need empirical evidence. Rationalism only means that in special cases propositions can be gained without empirical evidence. For the rationalist, the bulk of our knowledge is a mixture of empirical evidence and rational intuition & deduction. I know the story about Galileo's heartbeat, but if this counts as empirical evidence, then empiricism is in a very bad shape and states that knowledge is based more on a hope that it will work out allright. What if Galileo didn't have such a regular heartbeat? – viuser Mar 19 '17 at 12:13
  • 1
    This issue was investigated in detail by scientific structuralists, their answer is semantic holism about empirical claims. Bonilla's Meaning and Testability is a good review:"Sneed showed that, in order to avoid a circularity here, it was necessary to interpret the assertion of a scientific theory as a global claim about a set of physical systems (‘partial models’) which can be described without using the concepts dependent on that theory". Regress of justifications generally is an old argument for rationalism, often countered by holism. – Conifold Mar 20 '17 at 19:42
  • cf. Pierre Duhem's Aim & Structure of Physical Theory pt. 2, ch. 1, §2 "Quantity & Measurement," pp. 108-110. – Geremia Apr 01 '17 at 02:55

2 Answers2

4

First, there's an article on measurement in science in the Stanford Encyclopedia.

The sociologist Harry Collins dubbed this problem "the experimenter's regress" in the 1980s. There's a brief discussion in this Stanford Encyclopedia article; for more detail, see his book Changing Order, especially chapters 4 and 5. More recently, the historian and philosopher of science Hasok Chang has written a detailed analysis of the invention of the thermometer, in his book Inventing Temperature.

Briefly, Collins gives a social constructivist/anti-epistemological answer, while Chang's is closer to contemporary social epistemology. (This isn't surprising given the respective times their books were written.) Specifically, Collins points to the internal politics of science, emphasizing how proponents of one approach will recruit allies and marginalize their critics. Chang looks at critical interactions between scientists as more or less epistemically productive debates, which are settled by a complex mix of the better argument winning the day and the kinds of political factors that Collins points to. (Consider how the better argument can win the day by marginalizing people who stick to the worse argument.)

Dan Hicks
  • 2,507
  • 8
  • 25
2

The use of measuring devices (and also observation devices, like microscopes and telescopes) is a known issue in the philosophy of science. It has notably been taken as an argument, not for rationalism, but for a kind of holism.

The point is basically this: when you use a measuring device, and get a result, your device may be in error. Up to a point you can compensate for this by testing and calibrating your device. But the device could also be systematically misleading, in a way that cannot be calibrated away. Or, from another point of view, the interpretation of what the measuring device is actually measuring is often itself depending on a theory, which itself needs empirical support, perhaps by using more instruments. So for example, the interpretation of telescope observations rely on assumptions from optics. When a predicted observation on the moon fails, for example, it could be the that the optical presuppositions were the ones that actually failed, rather than the expectations concerning the moon.

The suggested conclusion is that it is, in general, impossible to test a scientific hypothesis in isolation. You will be testing a whole bunch of theory together with your hypothesis. And you need to take this into account when interpreting the results.

This idea of scientific/empiricist holism is known as the Duhem-Quine thesis. It is named after Pierre Duhem, who offered such a thesis for physics specifically, in The Aim and Structure of a Physical Theory (1906), and after W.V.O Quine, who offered a more general thesis as part of a reform of empiricism in general, in Two Dogmas of Empiricism (1951).

Ram Tobolski
  • 7,371
  • 1
  • 13
  • 21