The tense oppositions discussed so far enable us to locate a situation temporally either at, before, or after a reference point, and in turn to locate that reference point at, before, or after the present moment. So far, however, we have not discussed possibilities of locating situations more accurately before or after a given reference point, although in principle the elapsed time between the reference point and the situation could be either very short or very long. Lexically, and by using lexically composite expressions, it is, of course, easy to express such distinctions, as in John arrived five minutes ago; this star went nova five million years ago. In principle, given the use of lexically composite expressions, one can locate the situation with infinitesimal accuracy. The question that now arises is: to what extent is it possible to carry out more accurate time location by means of grammatical categories?

A large number of languages have no grammatical means of carrying out more accurate location in time. In German, for instance, there is no grammatical opposition that will give distinctions in remoteness in the past or in the future. There is also, however, a large number of languages where such distinctions are possible by grammatical means. Indeed, one can find languages from all parts of the world which make such distinctions, including western European languages, although prolific systems distinguishing several different degrees of remoteness tend to be more restricted in occurrence.