This paper is a review of the method for solving quantum field theory in the Heisenberg picture, developed by Abe and Nakanishi since 1991. Starting from field equations and canonical (anti)commutation relations, one sets up a (q-number) Cauchy problem for the totality of d-dimensional (anti)commutators between the fundamental fields, where d is the number of spacetime dimensions. Solving this Cauchy problem, one obtains the operator solution of the theory. Then one calculates all multiple commutators. A representation of the operator solution is obtained by constructing the set of all Wightman functions for the fundamental fields; the truncated Wightman functions are constructed so as to be consistent with all vacuum expectation values of the multiple commutators mentioned above and with the energy-positivity condition. By applying the method described above, exact solutions to various 2-dimensional gauge-theory and quantum-gravity models are found explicitly. The validity of these solutions is confirmed by comparing them with the conventional perturbation-theoretical results. However, a new anomalous feature, called the ``field-equation anomaly'', is often found to appear, and its perturbation-theoretical counterpart, unnoticed previously, is discussed. The conventional notion of an anomaly with respect to symmetry is reconsidered on the basis of the field-equation anomaly, and the derivation of the critical dimension in the BRS-formulated bosonic string theory is criticized. The method outlined above is applied to more realistic theories by expanding everything in powers of the relevant parameter, but this expansion is not equivalent to the conventional perturbative expansion. The new expansion is BRS-invariant at each order, in contrast to that in the conventional perturbation theory. Higher-order calculations are generally extremely laborious to perform explicitly.