Laboratory experiments are conducted in which water is issued vertically downward from a finite-length nozzle at a constant speed using a piston. The results of these experiments indicate that the breakup length of the liquid jet is two-valued at Weber numbers greater than unity but less than a certain value, which depends on the nozzle length-to-radius ratio and the Bond number. In addition to a long breakup length, which is consistent with the conventional observation, another shorter breakup length is realized at the same jet issue speed. Each experimental run for a specific jet issue speed begins from the start of liquid issue so that each run is independent of the other runs. Transition between the two breakup lengths seldom occurs in each run. Which of the two breakup lengths occurs is determined at the start of liquid issue, when the capillary wave produced by the liquid jet tip contraction easily reaches the nozzle exit. Unlike the conventional belief, which is based on the Plateau-Rayleigh instability theory, this experimental evidence demonstrates that liquid jet disintegration occurs in a deterministic manner. The previously proposed self-destabilizing mechanism of a liquid jet in microgravity, in which the origin of the unstable wave responsible for the breakups is attributed to the formation of an upstream propagating capillary wave at every breakup, is extended to explore the physics underlying the observed liquid jet disintegration behaviors.