✐ ✐ “articoli/castelfranchi” — 2012/12/21 — 13:34 — page 153 — #153 ✐ ✐ ✐ ✐ ✐ ✐ Cognitivizing “Norms”. Norm Internalization and Processing CRISTIANO CASTELFRANCHI * SUMMARY: 1. Our Perspective and Claims – 2. Norm Internalization – 2.1. Goal- adoption – 2.2. Reasons for Goal-adoption – 2.3. Goal-adhesion – 3. Not Only Pre- scribed Behaviors but Expected Mental Attitudes – 3.1. Interpersonal Rights – 4. Nor- mative Adhesion – 4.1. Generalized Goal-adoption – 4.2. Spontaneous Norm Moni- toring for Strong Reciprocity – 5. Against the Reduction of Norms to Sanctions, In- centives, and “Utility” – 6. “Internalization” (and Why It Matters) – 6.1. Conformity and Punishments as Messages – 6.2. Subjects Not Cooperators: The A-technical, Non- rational Nature of the Deontic “Ought” – 6.3. Educating to Norms – 6.4. The “Alien- ated” Nature of Norm Adoption – 7. Influencing Devices in a “Prevention Focus” – 8. Norm Processing from Beliefs to Goals and Intentions – 9. From Goal-adoption, Decision, Intention, . . . to Routines – 10. Norms As Multi-agent Artifacts – 11. Con- cluding Remarks. 1. OUR PERSPECTIVE AND CLAIMS What we present is not an agent-based simulation work 1 ; it is more a the- oretical contribution to normative cognition, the “psychology” of norms in * The Author is research associate at the Institute of Cognitive Sciences and Technolo- gies, National Research Council, “GOAL” Group, Theoretical Psychology Project. This work is the estended text of a talk given at the European University Institute in Fiesole for a WS on “Norm compliance” July 2010, https://sites.google.com/site/normcompliance2010/ program. I thank the participants for the nice discussion. I am also in debt with Rosaria Conte (many years working together or in parallel on these issues), Luca Tummolini, Giulia Andrighetto (for specific contributions on these issues) and the other members of the GOAL group for the general framework and precious feedback. 1 I am sorry to disappoint my reader, but my contribution nor is a discussion of the literature (philosophical, sociological, psychological, and AI) on norm and their working. It is more a restatement of the main issues of our work on norms in the last 15 years, work that has significantly contributed to social simulation studies on norms (see LABSS work http://www.istc.cnr.it/group/labss; R. CONTE, G. ANDRIGHETTO, M. CAMPENNÌ (eds.), Minding Norms. Mechanisms and Dynamics of Social Order in Agent Society, Ox- ford Series on Cognitive Models and Architectures, New York, Oxford University Press, forthcoming, and to the Agent and MAS research (see, for example, NorMAS WS and its community, G. BOELLA,P.NORIEGA, G. PIGOZZI, H. VERHAGEN (eds.), Norma- tive Multi-agent Systems, Dagstuhl Seminar Proceedings 09121, 2009; G. ANDRIGHETTO, G. GOVERNATORI, P. NORIEGA, L. VAN DER TORRE (eds.), Normative Multi-agent Sys- tems, Dagstuhl Seminar Proceedings 12111, 2012; and the Agreement Technology EU Project, http://www.agreement-technologies.eu/wg2, and G. ANDRIGHETTO, C. CASTEL-