Definitions

[W]hatever may be the practical or aesthetic advantages of turning scientific laws into logically necessary truths, it does not advance our knowledge, or in any way add to the security of our beliefs. For what we gain in one way, we lose in another. If we make it a matter of definition that there are just so many million molecules in every gram of hydrogen, then we can indeed be certain that every gram of hydrogen will contain that number of molecules: but we must become correspondingly more doubtful, in any given case, whether what we take to be a gram of hydrogen really is so. The more we put into our definitions, the more uncertain it becomes whether anything satisfies them.

Alfred Jules Ayer, What is a law of nature?, Revue internationale de philosophie, vol. 36, no. 2, 1956, pp. 144–165, p. 151

Omnis definitio est periculosa. Certain writers make such a point of repeating the statement, “It is dangerous to define,” that at times one wonders whether defining might not be especially dangerous for their own thinking.

Giovanni Sartori, Democratic theory, Westport, Conn, 1962, p. 6