…let a number of the proper designations of individual subjects be omitted, so that the assertion becomes a mere blank form for an assertion which can be reconverted into an assertion by filling all the blanks with proper names. I term such a blank form a rheme. If the number of blanks it contains is zero, it may nevertheless be regarded as a rheme, and under this aspect, I term it a medad. A medad is, therefore, merely an assertion regarded in a certain way, namely as subject to the inquiry, How many blanks has it?