7 Are instrumental measurements valid observations?

How instruments are reliable extensions of our senses.

This is the eighth post in this series. If you haven’t read the Introduction titled, The Myth of Scientific Uncertainty, and posts numbered 1-6, you may want to do those first.

Since the invention of the telescope, thinkers have raised the question that if there is nothing in science we can be sure of, on what basis can we trust the results produced by scientific instruments? This is understandable because for many people, the way complex technological devices like computers and cell phones work can seem truly mysterious (and sometimes frustrating). Scientists call instrumental results “observations” even though they were not made with our senses. But are they valid “observations”?

To answer that, we must connect the operation of instruments to that which we have determined in earlier sections to be trustworthy. In this section we will see that, broken down, all devices and instruments are just combinations of individual, readily understood, bits whose function is based on verified laws. For a logical argument for why this is so, we will start with some basic measurement concepts.

Instruments designed to measure quantities produce numbers that are related in known ways to the properties to be determined. Unless we can visually count them, there are only a few things we can quantify directly, like the dimensions of an object or the angle between two lines. We compare the length or angle of the object with the numbered marks on a measuring tape or a protractor. But things like temperature, time, and weight can’t be measured with a scale printed on a tape or semicircle. To measure such non-visible quantities, we need devices that will convert the things we want to measure into something we can see. The goal is to produce a number by which we can assess the brightness of a star or the temperature at which a material melts. So, we use a photometer to measure the intensity of the starlight or a thermometer to observe the temperature at which melting occurs.

Here is one way we can measure the intensity of light. A light sensor produces an electrical current related to the photon flux at the sensor. This current moves the pointer in a current meter that has a scale like the one pictured.

The position of the pointer against the scale gives us a number related to the light intensity. The higher the number, the brighter the light.

The light meter described above employs three conversion devices: a light intensity-to-current converter, a current-to-needle position converter, and our eyes which convert the needle position to a number. Instruments that count things we cannot see like photons, aerosol particles, and red blood cells also require conversion devices. A sensor converts each item or event into an electrical pulse. Another device accumulates the counts and another displays or stores the result.

All electronic measurement devices from those in our automobile dash to those in advanced labs depend on combinations of conversion devices that convert the quantity we want to measure into one that we can see^[1].

This principle of measurement has several important consequences concerning the reliability of scientific data. The accuracy and precision of the final measurement depends on the reproducibility and stability of each of the conversion devices and the accuracy of the instrument calibration. Each conversion device in an instrument depends on verified laws that relate its input and output quantities. The assumption that the equations underlying the conversion devices work consistently is valid if the instrument is working within the tested limits of the laws its conversion devices rely on.

For those who argue that we should only believe what we perceive with our senses, I would point out that the sense system in our body uses the same conversion device concept as our instruments. In vision, our photo sensory cells convert photons of light to impulses of charge that our nerves carry to our brain for interpretation. The same is true for sound, taste, smell, and touch. At least in instruments, we know and control how the signals from the sensors are being processed and interpreted. In our bodies and brains, it’s more complicated.

The use of instrumentation in scientific measurements is a key example of how science builds on itself. We use valid equations to devise novel conversion devices so we can observe new phenomena in reliable ways. For example adapting a Michelson interferometer to measure gravitational waves.

If scientists who used sophisticated instruments, like computer-controlled telescopes or mass spectrometers, listed all the laws their observations relied on, it could easily run into dozens. When a measurement result challenges accepted models of the system studied, we need to examine all aspects of the instrumentation employed like they did with the OPERA experiment at CERN as described above.

Next time, we will explore whether a scientific explanation, once conjectural, can become a truth.

---

[1] Enke, C. G. Anal. Chem. 43: 69A-80A (1971)

---