On the Loadstone and Magnetic Bodies

Overview

De Magnete is three hundred pages in six books, published in London in 1600 in Latin, its full title running to a sentence: on the magnet, magnetic bodies, and that great magnet the earth, a new physiology demonstrated by many arguments and experiments.

William Gilbert was a physician, president of the College of Physicians and eventually doctor to Elizabeth I, and he spent nearly two decades and, he says, a good deal of his own money on this. He died of plague in 1603.

The book’s reputation rests on how it proceeds. Sixteenth-century natural philosophy largely consisted of commentary on Aristotle and the compilation of reported marvels; Gilbert does experiments, describes the apparatus, reports what happened, and treats a claim that fails a test as false regardless of who made it. He is also rude about his predecessors in a way that is entertaining to read.

Two things temper the picture. Gilbert believed the lodestone had a soul, and Book VI is speculative cosmology in which magnetism explains the motions of the heavens. And the Latin is technical and the translations are Victorian. Most readers should take Book I, the opening of Book II, and Book V, and leave the rest.

The argument

Book I: what a lodestone is, and the terrella. Gilbert begins with a history of what has been written about magnets, most of which he rejects, and then with the facts as he has established them: that a lodestone has poles, that like poles repel and unlike attract, that iron touched by one becomes one, that a magnet broken in two yields two complete magnets each with both poles.

Then the experiment the book is built on. He had lodestones ground into spheres, which he called terrellae, little earths. Move a small compass needle over the surface of one and it behaves exactly as a compass behaves on the earth: it points along the meridian toward the pole, it dips, and the angle of dip increases as it approaches the pole and is vertical there.

This is a model in the full modern sense, and the inference is the one that makes the book famous. The earth is not a magnet because something inside it is; the earth is a magnet.

Book II: electricity, and the demolition. Here Gilbert separates two phenomena that had always been lumped together. Amber rubbed will pick up chaff; a lodestone will pick up iron. He shows that these are different: the amber effect requires rubbing, works on many light substances, is blocked by a sheet of paper or by moisture, and disappears; the magnetic effect is permanent, specific to iron, and passes through obstacles. He tested a long list of materials and found many that behave like amber, and named the whole class after the Greek word for amber, electrum. That is where the word electricity comes from.

The rest of the book is a series of received claims put to the test and found false. That garlic destroys a lodestone’s virtue, so that helmsmen were forbidden to eat it near the compass. That a diamond will magnetise iron. That a lodestone will not attract iron in the presence of a diamond. Gilbert tried them, sometimes repeatedly and with several stones, and says so.

Books III and IV: direction, variation and dip. How iron takes up magnetism from the earth, and why a compass does not point at true north. Variation, the deviation between magnetic and true north, had been a practical problem for navigators for a century; Gilbert attributes it to the unequal distribution of land and sea on the globe, which is wrong but is a physical explanation rather than a mystery. Dip, the tendency of a freely suspended needle to point downward, had been discovered by Robert Norman, a compass maker, and Gilbert gives him credit and explains it from the terrella.

Book V: the great magnet. The argument gathered: the earth’s magnetism, its poles, and the consequences for navigation. Gilbert hoped dip could be used to determine latitude at sea, which did not work out.

Book VI: the earth moves. The most speculative book and the one that interested Kepler. Gilbert argues that the earth rotates daily, and that its magnetism is what keeps its axis steady, giving the diurnal rotation a physical cause rather than leaving it as a bare assertion. He is scornful of the idea that the entire sphere of stars whirls around the earth every day, calling it monstrous to suppose so vast a thing moves for the sake of so small a one.

He does not accept the full Copernican system, and he attributes a kind of soul to the magnet, which sits oddly beside the experimental method. Kepler took the physical part and ran with it: his account of the sun driving the planets by a force that weakens with distance is built on Gilbert’s magnetic philosophy.

How to read it

Read the preface first. It is three pages and it is a statement of method: sure experiments and demonstrated arguments against probable conjectures and the opinions of speculators.

Then Book I for the terrella and Book II for the electrics and the demolitions. Together that is perhaps eighty pages and it contains everything a general reader needs.

Read the failed claims carefully. Gilbert testing whether garlic spoils a compass is funny and it is also the point: somebody finally checked.

Use the Mottelay translation, which is free and complete, and be prepared for Victorian scientific English. There is no modern critical translation in general circulation.

Keep the terrella in mind as the book’s real contribution to method. Gilbert could not experiment on the earth, so he made a small one and argued from it. Everything about modern modelling, from wind tunnels to climate simulations, rests on the move he makes in Book I.

And notice the mixture. This is a book that founds experimental physics and also holds that the lodestone has a soul and that the earth’s magnetism is a kind of animate power. The two were not, in 1600, obviously in tension.

Questions it raises

• The terrella behaves like the earth, so the earth is a magnet. How strong is an argument from a model, and what would make one fail?
• Gilbert separates electric from magnetic attraction by listing differences. Is that a discovery of two phenomena or a decision to classify them apart?
• He tests and rejects claims that had circulated for centuries. Why had nobody tried them, and what does that suggest about how knowledge was transmitted?
• Magnetism is offered as the physical cause of the earth’s rotation. Is a wrong mechanism better than none?
• Gilbert attributes a soul to the lodestone in the same book as the experiments. Does that undermine the method, or show what the method had not yet excluded?
• Kepler took the physics and made planetary theory out of it. What is it about a wrong idea that makes it fruitful?