Teaching and Research: Beyond the Talking Head

Demonstration Experiments: Beyond the Talking Head

By Daniel M. Siegel
University of Wisconsin, Madison

The author teaches a course in the history of the physical sciences from Newton to Einstein at the University of Wisconsin-Madison in which he incorporates highly popular demonstration experiments. He presented an overview of the course and some of the related philosophical issues at the 1996 CoE workshop, “Beyond Lecture.” The text has been condensed and edited by Marjorie Malley.

Why does one want to do demonstration experiments in a history of science course? First, historical verisimilitude requires it. From the scientific revolution to the early twentieth century experiment was a central part of science. To represent the physical sciences of this period as primarily a literary tradition is a crime against historical authenticity.

Static representation such as pictures, diagrams, and words often leave the student with only the haziest notion of the experiments carried out in a historical situation. Further, verbal and pictorial representations fall short in vividness and pedagogical impact. If one asks what kind of course material students talk about, write home about, and remember for the rest of their lives, demonstration experiments are at or near the top of the list.

A practical way to get started with demonstration experiments is to use demonstration apparatus from science departments. This will mainly be modern equipment, which leads to historiographical questions: Can one avoid an empiricist and presentist bias in this kind of presentation? I think this is possible, given an appropriate general framework.

Most of us have become convinced that there is no universal scientific method that underlies all scientific activity. The practice of science, in particular the role of observation and experiment, varies significantly with time, place, and practitioner. Nevertheless the practice of science has exemplars and traditions, and these can be used as organizing themes. Benjamin Pierce’s analysis of scientific method in the nineteenth century, Ernan McMullin’s use of that analysis in his account of “Conceptions of Science” in the scientific revolution, and my own thinking on science from Newton to Einstein converge on a tripartite classification of approaches to science.

Francis Bacon is exemplar for an empiricist and inductivist approach, which emphasizes primacy of observation and experiment and regards scientific laws and principles as generalizations from observation and experiment. René Descartes (and Euclid) exemplify the rationalist approach, where intuition and deductive logic have primacy, while observation and experiment confirm and refine what is presented by the “inner light.” Christian Huygens and Isaac Newton are exemplars for a hypothetical or hypothetical-deductive approach, in which scientific principles are entertained first as conjectures, then are tested against observation through their consequences.

Newton’s laws of motion provide a paradigmatic example of the hypothetico-inductive approach, and they provides opportunities for some demonstration experiments. The peculiar use of observational data in a hypothetico-inductive argument become quickly apparent in Newton’s presentation of his first law of motion (inertia) in the Principia:

* LAW I: Every body continues in its state of rest or of uniform motion in a straight line unless it is compelled to change that state by forces impressed upon it.
* Projectiles continue in their motions, so far as they are not retarded by the resistance of the air or impelled downward by the force of gravity. [We demonstrate with a dart gun and a ball, and we puzzle over Newton’s first observational example, projectiles, which obviously travel in curved paths, rather than straight lines.]
* A top, whose parts by their cohesion are continually drawn aside from rectilinear motions, does not cease its rotation otherwise than as it is retarded by the air. [We demonstrate with a top and with a gyroscope. Since the particular form of the top is not specified, we are not overly concerned with representing an authentic 17th century top.]

We wonder about Newton’s choice of observational examples. How does the spinning of the top provide support for the first law?

But then we remember that this is a hypothetico-deductive argument. The decisive experimental evidence will come later, after all three laws have been stated and their consequences deduced. Newton is presenting plausibility arguments for the hypothesis. The projectiles and the spinning tops illustrate one aspect of the first law – persistence of motion – but for curvilinear rather than straight line motion. This was as close as Newton could get to direct empirical illustration of the first law, and it was not very close.

Newton introduces his second and third laws of motion with similar plausibility arguments, and then deduces consequences from them. From the first and second laws taken conjointly, he deduces the laws of falling bodies, the parabolic paths of projectiles, and the periods of pendulums. We demonstrate in class with projectiles and with a water jet. Later we take measurements on a bowling ball pendulum. By joining his second and third laws, Newton was able to derive conservation of momentum, which he demonstrated and measured with collisions of pendulums. We demonstrate pendulum collisions, examine elastic and inelastic collisions on an air track, and launch a CO2 rocket (which Newton did not have available).

We conclude that Isaac Newton established his laws of motion from oblique observational evidence brought forth in plausibility arguments, and on indirect evidence obtained when verifying consequences. In a modern account of Newton’s laws, which science students will have received, the laws of motion are presented as generalizations from experience, using air tracks and similar apparatus. By using demonstration experiments in class we can illustrate both the historical and the contemporary situations, and are able to draw the contrasts explicitly.

The rational, intuitive Cartesian approach to science is illustrated by the history of the energy concept. Descartes’s notion of conservation of motion and developments from the seventeenth and eighteenth centuries provide the occasion for lecture demonstrations involving colliding pendulums, the bowling ball pendulum swung from the lecturer’s nose, a pegboard pendulum, and so forth. An early nineteenth century set of experiments on electric circuits were also driven by intuitive preconceptions. Hans Christian Oersted firmly believed (on the basis of Naturphilosophie as well as on conservation ideas in the Cartesian tradition) that all natural phenomena must be linked. From this viewpoint, electric current should produce magnetic effects. Oersted’s inability to exhibit these effects did not deter him from believing that they existed. He continued to search obsessively, and was rewarded in 1820 with the discovery of the magnetic effect. But the effect came with a totally unexpected property. The magnetic force was not an attraction or repulsion from the current carrying wire, but instead a lateral force which caused a compass needle to align at right angles to the wire. A demonstration of this experiment carries all the clarity and vividness that one could desire.

The discovery of the reverse effect, the production of electric current by magnetism, also originated in an intuitive belief that all must be connected with all, this time on the part of Michael Faraday. Like Oersted, Faraday was undeterred by his failure to find the effect. Likewise, when he finally succeeded in 1831, the effect showed a surprising aspect: only a changing magnetic situation could produce an electric current. Once again a demonstration of this experiment is historically authentic, clear, and vivid.

The experiments of Oersted and Faraday illustrate various themes. First, we see the power of preconceptions to select the kinds of phenomena that will be revealed. We also see the intractability of these phenomena. The magnetic effect will not exhibit the expected push-pull nature; the electrical effect will not be produced by static magnetism. The combined role of preconceptions and experiential data, illustrated in the investigations of Oersted and of Faraday, is captured in a vector diagram by David Bloor, Knowledge and Social Imagery: “Prior belief” [vector] and “Experience” [vector] combine to give “Resultant belief” [resultant vector].

The method of Baconian empiricism, which emphasizes interrogation of nature and proceeds by generalization from experience, is not often found in the physical sciences from Newton to Einstein. I talk about natural history and taxonomy in the biological part of this history of science course, and these can be characterized, at least in part, as Baconian enterprises. A final example from the course which illustrates the combination of all three approaches to experimental evidence is taken from the nineteenth century work of James Prescott Joule on conservation of energy.

Like Oersted and Faraday, Joule was concerned with the electric circuit and with electric motors. Similarly, Joule saw the electric circuit as the site of transformations of a basic power in the universe whose conservation was intuitively obvious. For demonstration purposes we have a primitive, large, clunky electric motor, which is probably not a bad approximation to the motors Joule used. Stimulated by his observations of relationships between vis viva and heat in electric circuits, Joule carried out a series of experiments on the mechanical equivalent of heat. The repetition of experiments in different ways and under different circumstances produced a triumphant illustration of the Baconian empiricist method, and served to establish Joule’s reputation as a culture hero of that approach.

First, Joule measured the conversion of work to resistive heat in a generator coil (we demonstrate a generator lighting a light bulb, close enough). Then he compressed air and measured the heat evolved (we demonstrate with a fire cylinder, which, though qualitative, is spectacular). Joule agitated water with a paddle wheel arrangement (we use a modern analog, with friction heating water in a cylinder). He repeated the experiment with different fluids and refinements of the apparatus, and obtained a value of 772 foot-pounds of work per British Thermal Unit of heat. “Experience,” wrote Joule, “actual experiment,” leads to a “general rule,” in a very Baconian way. Finally, in a hypothetical-deductive argument, Joule showed how his experiments lent support to, and were in turn explained by, the kinetic theory of heat.

Joule’s work thus displays a combination of the intuitionist, empiricist, and hypothetical approaches. These three are indeed ideal types. Real scientific practice rarely presents pure cases. Rather, we see a variety of combinations and permutations of the three.

The intention of my course in the history of the physical sciences and of the demonstration experiments is to convey and present the variety of uses of experiment in science with the clarity and vividness that only live demonstrations can achieve. The demonstration experiments have the additional great value of keeping the students awake in a late afternoon class. What more can one ask?



1. Laws of Motion: Air table and pucks; air track with elastic and inelastic riders; dart gun; tops and gyroscopes; parabolic water jet; CO2 rocket; colliding pendulums; rubber band and weight
2. Relative Motion: Wind-up toy on cart
3. Optics: Optics board; point light source


1. Clocks: Pendulum and timer; model of verge and foliot
2. Pneumatics and the Steam Engine: Water siphon; water pumps; air pump and mercury column, Magdeburg spheres; crumpling can; suction cups; PA system feedback loop

Conservation Laws

1. Momentum and Vis Viva: Colliding pendulums; air track; bowling-ball pendulum; pegboard pendulum
2. Caloric: Air thermometer; thermal expansion and conduction demos; mixture of Oº and 100º C water samples


1. Electrical circuits: Battery circuits producing heat and light; Oersted experiment; electric motor; electromagentic induction; generator and light bulb; battery circuit with motor and resistor
2. Mechanical Equivalent of Heat: Adiabetic compression (fire cylinder); mechanical equivalent of heat (friction)
3. Engines: 4-stroke engine model


1. Special: Laser with double slit; oscillating weight on cart; cosmic rays in cloud chamber
2. General: Galileo/Stevin experiment

For further information about this course, contact:

Dan Siegel
Department of the History of Science
University of Wisconsin
7131 Social Science Bldg.,
Madison, WI 53706
(608) 262-397