The author deals with the problem of containment in situations of attack on the analytic link, paying most attention to two points: 1) the interaction between the two participants in the analytic encounter; 2) the relation between containment and interpretation.

Regarding the first point, in dealing with a relationship that is not working well—when the analyst is openly or subtly deprived of his or her specific function of interpreting but the two participants continue to interact despite impaired verbal communication—Ponsi chooses an approach emphasizing the interactive aspects of the analytic relationship. To this aim, she makes use of the distinction between content messages and relational messages, fundamental in the pragmatics of human communication of the Palo Alto school, and considers that, during the attack on the analytic link and the stalemate in interpretative activity, the