Good point. Defining confluence as an absence of an appropriate self/not-self boundary between one’s self and a part of the environment sees ‘identification with the projection of another’ as qualifying. But, is the person under coercion ‘identifying with the projection’ or ‘responding to the coercion’?
Getting back to the example concerning the man who wished to rid himself of his chronic anxiety and so projected it onto his partner. When talking to his partner, he would leave out vital information and would lead her to become anxious.
Did she ‘identify with his projection’ (which would make projective identification a pretty good name for it, and would also demonstrate, as you have said, someone being confluent with a projection), or did he manipulate the environment so as to facilitate a response in her that aligned itself with his projection? Which would be coercion.
I am inclined to see her resultant state as a personal response to a coercive environment that doesn’t provide her with sufficient information to self-regulate smoothly.
And it really is just ‘my’ inclination - I make no attempt to offer an official Gestalt pronouncement on the matter. The literature seems filled with people talking about slightly, and sometimes vastly, different things when using this term, (as with many other terms). Gestalt literature is noticeably quiet on the matter. One aspect of it that Philip Brownell referred to - controlling the projection once it is planted in the other - remains as lost on me as it does him, and seems to originate from Object Relations concepts that an infant can manipulate the parent to do some of the assimilating of its own affects in order to achieve the previously-too-difficult assimilation himself. More recent ideas of affect attunement, the the failure thereof, between parent and child, seem more to the point, though I have not seen them assimilated into Gestalt theory yet.
Yontef (1993) gives another slant when discussing a patient who projects her intelligence onto her lover and then she becomes confluent with the lover, (the intended inference of the ‘identification’ in ‘projective identification’). The lover’s criticisms match with her disownment of intelligence and her identifying him as intelligent.
I believe these different types of interpersonal process are essential to study by therapists in training. And, I am inclined to describe any such processes in long-hand and shy away from this particular term.
You mention Object Relations, I am not good with Objects Relations language but will offer the following, largely in ‘Gestaltese’, to see if it promotes further discussion. [ I think there are many ingredients in Object Relations, Self Psychology, and particularly Intersubjectivist Theory, that have - and will - inform a “new improved” Gestalt Theory ]
Taking the example of a client who has been abused, Object Relations could speak of the client as having internalised the dyadic relationship of aggressor and victim paving the way for a client who may occupy either role at different times. Equally, the client can now project the “complimentary” role onto the environment. Gestalt would speak of unfinished business and boundary disturbances. I don’t know what Object Relations would say. The Intersubjectivists would speak of organising principles - that the client is modelling perception of self and environment as they were once experienced, (a useable definition of transference).
If projective identification now occurs with such a client, there can be projection of the aggressor and/or projection of the victim role. Projection of the aggressor has a very different dynamic, and flavour, to the man wishing to rid himself of his own anxiety, that is, to project a part of himself and not an internalised other. So different, in fact, that they probably deserve different names, or the earlier mentioned adherence to longhand description. Perhaps someone has given them different names.