When Boswell told Dr. Johnson that while in Italy he had several times seen the experiment tried of placing a scorpion within a circle of burning coals, and that in every instance the scorpion, after trying to break through the fiery circle, retired to the centre and committed suicide by darting its sting into its head, the doctor showed the true scientific spirit by demanding further proof of the fact. The mere testimony of the eye under such circumstances was not enough; appearances are often deceptive. "If the great anatomist Morgagni," said the doctor, "after dissecting a scorpion on which the experiment had been tried, should certify that its sting had penetrated its head, that would be convincing." For almost the only time in his life the superstitious doctor showed himself, I say, a true scientist, a man refusing to accept the truth of appearances.
But this frame of mind was not habitual to him, for the next moment he said that swallows sleep all winter in the bed of a river or pond, "conglobulated" into a ball. The scientific spirit would have required him to insist upon the proof of the alleged fact in this case the same as in the other. Has any competent observer verified this statement? Have swallows been taken out of the mud, or been seen to throw themselves into the water?
Albertus Magnus (1193-1280), in his book on animals, says that the eel leaves the water in the nighttime, and invades the fields and gardens to feed upon peas and lentils. A scientific man makes this statement, and probably upon no stronger proof than that some eels dropped by poachers in their hasty retreat had been found in a pea patch. If peas had been found, and found in many cases, in the stomachs of eels, that would have been pretty conclusive proof that eels eat peas.
The great thing in observation is not to be influenced by our preconceived notions, or by what we want to be true, or by our fears, hopes, or any personal element, and to see the thing just as it is. A person who believes in ghosts and apparitions cannot be depended upon to investigate an alleged phenomenon of this sort, because he will not press his inquiry far enough, and will take for granted the very fact we want proof of.