moreover,if we use rules (3.8) again,applying them this time to an arbitrary matrix A and using the values for D(E) which have just been determined,we obtain the following
let E be an elementary matrix and let A be arbitrary
recall from (2.19) that every square matrix A can be reduced by elementary row operation to a matrix B which is either the identity I or else has its bottom row zero
we know by (3.5) that D(A)=1 or D(B)=0 according to the case
by (3.2) and induction
we also know D(E) ,by (3.11),and hence we can use this formula to compute D(A)
theorem: axiomatic characterization of the determinant:
the determinant function (3.4) is the only one satisfying rules