Re-reading ELIZA: Human–machine Interaction as Cognitive Sense-ability

ABSTRACT

This article re-reads ELIZA, the famous computer program of early artificial intelligence created by Joseph Weizenbaum in 1966 to test the possibility of intelligent interaction between humans and machines through language. Given newly emerging intelligent gadgets such as smart-home assistants that operate through natural language, I argue firstly for the timeliness of this program and explore the ways in which ELIZA stimulated debates concerning whether it could be considered an interlocutor for humans or a failure in exactly this regard. Secondly, I develop a feminist technoecological account of interaction at the ELIZA/user interface. In so doing, I seek to challenge the dualisms between body/mind, rationality/affectivity, and subject/object, thereby avoiding a reductionist perspective that calls into question the human subject as an autonomous agent.     

Ethnic differentials: Sri Lanka,Malaysia, and Israel

Studies in Comparative International Development -     

Time served and institutional misconduct: Patterns of involvement in disciplinary infractions among long-term and short-term inmates

This article addresses a number of questions about prison misconduct among long-term and short-term prisoners. Previous studies of institutional misconduct have found an inverse relationship between age and rule breaking and negligible relationships between time served and infraction rates. The data presented here show that the infraction rates of long-term inmates are significantly lower than the rates for short-term prisoners, even in the early years of confinement. However, age is an important variable that intervenes in the relationship, especially in the older-age categories. The types of infractions committed by long-term and short-term inmates are quite similar. The temporal pattern of infraction behavior for short-term inmates provides support for Wheeler's model of an inverted U-shaped curve of prison adjustment, while the pattern for long-term inmates does not fit Wheeler's model. The data suggest that long-term inmates adapt to their situation in ways that are distinguishable from short-term prisoners.