In Antiquity people told time by the Sun and stars. A boy and girl might arrange to meet during the evening when Venus falls below the horizon. In the event of clouds, they could guess at the meeting time. Serious consequences if they guessed wrong seldom occurred. Other methods of keeping time could be used as backups, however. Water clocks (clepsydras) had been known since ancient Egyptian times. These depended on the relatively constant lowering of the level of water in a vessel with a deliberately made leak. Clepsydras were not very accurate, not easy to read in a poor light, and needed frequent refilling to be of any use. Burning tapers and lamps might also show the passage of time, but time varied with the specific taper or lamp.

In the Middle Ages, however, members of religious orders were expected to pray at definite times. Failure to maintain godly habits because of cloudiness or variable flames was not acceptable. There was an answer, however. Sources from Antiquity (and probably rumors of Chinese inventions) referred to devices that could imitate the Sun and stars. Such devices were powered by water clocks, but the Chinese rumors may have told of a clever device, which we call an escapement, that could convert the smooth motion of water flowing into a series of short rotary motions. These motions could give a more accurate representation of the movement of astronomical bodies. In duplicating this concept, an unknown European inventor recognized that with an escapement to slow the fall, a weight could be substituted for water. If the weight were lifted by hand at regular intervals, there would be no need to deal with the continual addition of water to operate the mechanism. The falling-weight idea, however, required something to make it more regular, since a weight accelerates, or falls faster, as it goes along. If the weight could be made to fall the same short distance over and over, the motion would be more regular.

This was accomplished by adding a sort of dumbbell, called a foliot, to the escapement. The weights at the end of the foliot fall a short distance with each tick of the clock, each one balanced by the other one. The combined fall of the weight that powers the clock and the short stroke of the foliot gave the time off by only an hour or so each day. The earliest mechanisms of this type, around the end of the 13th and beginning of the 14th century, were used to power representations of the heavens. Thus, if the clock was out of whack when the weight was pulled up, it could be reset on any clear day or night by comparing the representation with the actual positions of heavenly bodies. The monks and nuns were summoned to prayer by a bell.

Soon someone realized that the elaborate astronomical model was not needed; a system of striking the hour with a series of rings of the bell was sufficient. Sometime after that, people added a dial to show the hours with a pointer (hand). A similar pointer for minutes was not needed until clocks greatly improved in accuracy. Although the first clocks were installed for use in religion, within a few years people began to keep time by the hours, since the ringing of the bell often could be heard or the dial seen all over a village. The manufacture of clocks became a thriving industry in the 14th and 15th centuries. Every town soon had to have its own. Although the basic mechanism did not change during this period, the development of better ways to cut gears and to make other metal parts was an important precursor of the Industrial Revolution. No other artifact of this period required such careful workmanship.

Despite this, clocks were not nearly accurate enough to bother with minutes, much less seconds. When Galileo needed to time short intervals around the turn of the 17th century, he used his pulse because clocks were not sufficiently accurate.