Milgram's own studies support the role of the agenticstate in obedience
Most of Milgram'sppresisted giving the shocks at somepoint and often asked the Experimenterquestions about the procedure
One of these was 'Whoisresponsible if Mr Wallace (the Learner) is harmed?'
When the Experimenterreplied I'm responsible, the pp often went through the procedurequickly with nofurtherobjections
This shows that once pp perceived they were nolongerresponsible for their ownbehaviour, they acted more easily as the Experimenter'sagent, as Milgramsuggested.
LIMITATION:
doesn't explain many researchfindings about obedience
For example, it does not explain the findings of Rank and Jacobson's1977 study
They found that 16 out of 18hospitalnursesdisobeyed orders from a doctor to administer an excessive drugdose to a patient
The doctor was an obviousauthorityfigure
But almostall the nurses remained autonomous, as did many of Milgram'sparticipants.
This suggests that, at best, the agenticshift can only account for some situations of obedience.