Tower suggests that countertransference reactions are present for shorter or longer periods in any analysis, and that these are usually well-defined enough to be considered 'countertransference neurosis'. This may be and usually is minor compared to the total treatment situation; yet it is probably very important for two reasons: it may be necessary for the patient to know (unconsciously) that he has been able to make some change in the analyst; and, to cope with the particular patient's transference resistance, a sufficient expansion of ego integrative powers in the analyst may be possible only through interaction of unconscious countertransferences and ego-adaptive responses.

Tower notes group rigidity and resistance to consideration of countertransference, many forbidding references to it, and yet the paradoxical recognition that the unconscious is never eliminated, nor is it subject to conscious control although it can be investigated and brought to awareness.