We can now consider the sequence 'TN', which is substitutable for 'N': e.g. 'a cheese', 'some cheese', for 'cheese', 'cheeses', in 'We can use — in place of meat'. However, 'TN' cannot replace the 'N' in some of the preceding equations: we cannot substitute 'TN' for 'N' in 'A N³ = N³' for we would derive a non-extant 'A TN³' ('Swiss some cheeses'). Therefore the resultant 'N' must have a new sub-class numbering which will preclude its substitution in the preceding equations: 'TN³ = N⁴'. We can now say that the morpheme class equals this 'N⁴': thus it is substitutable for 'freedom', the long grind in '— will be re-established'. Among the later equations we will have ones like 'N⁴'ˌ 'N⁴ V = N⁴' : 'the books my various friends borrowed', or 'men I have known replacing fish' in '— will be discussed later'. Each higher numbered symbol represents all lower numbered identical symbols, but not vice versa. Therefore, the higher numbered symbols have a more inclusive representation, and are of greater importance in any compact classification of the morpheme sequences of a corpus.
- Harris (1951), Pag. 266