In the Attitude-Scenario-Emotion model, social relationships are subpersonnally realized by sentiments: a network of emotions/attitudes representing relational values. We discuss how relational values differ from moral values and raise the issue of their ontogeny from genetic and cultural factors. Because relational values develop early in life, they cannot rely solely on cognition as suggested by the notion of attitude.

Contemporary research in the fields of moral psychology and cognitive philosophy has provided considerable data supporting the claim that there are important similarities in the ways in which different people conceive of morality and produce moral judgments. However, one of the more pressing questions is how to account for the fact that, despite these similarities, moral judgments appear to be highly variable both on a cultural and individual level. This paper addresses this issue by developing a model which is inverted with respect to the one usually embraced by the cognitive literature on morality. Instead of analyzing the problem of moral judgment starting from all the actions that are considered impermissible, this work assumes that people first judge which actions are morally permissible. Permissibility is interpreted in terms of what each subject feels he/she must be free to do. The advantage of this inversion is that it allows us to make a connection between two research lines that are usually considered unrelated concerning on the one hand the processes underlying the production of moral judgments and on the other the problem of determining how people understand ‘freedom’. As for this last issue, the article focusses specifically on George Lakoff’s cognitive analysis on how humans develop their concepts of freedom. The starting point of Lakoff’s analysis is that different groups and different individuals do not have the same understanding of ‘freedom’, even though everybody shares the common empirical core concept. Lakoff puts forward a model trying to explain both the common cognitive ground of the various concepts of freedom and the ways they vary depending on other cognitive elements connected to them. In this work we try to show that Lakoff’s model can provide an explanation of moral judgment that accounts for both the cross-cultural and trans-individual similarities and the cultural, individual and situational differences.

This study examines whether the categories ANIMATE/INANIMATE: might be formed on the basis of information available to the cognitive system. We suggest that the discrimination of percepts according to these categories relies on proprioceptive information, which allows the perceiving subject to know that he is 'animate'. Since other 'objects' in the world exhibit movements, reactions, etc. similar to those that the subject experiences himself, he can 'project' his knowledge onto these objects and recognize them as 'animate' like himself. On this basis we try to corroborate the empricist position in the debate concerning the organization of knowledge as opposed to the nativist view. Furthermore, we argue that the categorical dichotomy ANIMATE: /INANIMATE: is more basic than other analogous ones such as LIVING: /NON-LIVING: , BIOLOGICAL: /NON-BIOLOGICAL: and we sketch a 'categorical stratification' following the line 'humans-animals-plants' based on the hypothesis that humans detect different degrees of 'vitality' according to the degree of similarity they recognise between the considered instance and themselves.

The main aim of this paper is to challenge the implicit formalism in classical cognitive theories of semantic competence. These models describe the human ability to master the words of natural language primarily using a model drawn from artificial intelligence, in which knowing a language means being able to connect linguistic symbols with each other in a proper way. According to the view we put forward in our paper, the capacity to connect words with others according to semantic rules only describes that small part of human semantic competence which can be identified as ‘inferential competence’. However, human semantic competence is also characterized by what is called ‘referential competence’,that is, the capacity to recognize those objects in the world to which words refer. Both these competences, the inferential and the referential, are proposedto be based on the availability of semantic representations, developed by the cognitive system. In the paper we show firstly that this second aspect of semantic competence is the most important one if we aim to understand how humans come to master natural languages. Secondly, a deeper articulation of the constitution of referential competence is put forward. The paper attempts to show that in order to account for referential competence we need to assume that semantic representations are constructed from both information coming from the external world via the sensorium and information produced by the cognitive system itself during the processing of the sensory input coming from the external world via the bodily sensorium (we refer to the latter as “qualitative information”). We argue that this qualitative information plays an essential role in explaining some aspects of referential competence which would otherwise remain obscure such as the capacity to recognize (in a referential sense) internal states corresponding to specific word. Our working hypothesis is presented and discussed in relation to linguistic deficits reported for clinical conditions some examples taken from clinical psychology (Asperger Syndrome and Alexithymia). People affected with these syndromes show specific impairment in their linguistic mastery that can be interpreted as a lack of referential competence with respect to words related to internal states due to problems in processing qualitative information. If so, these diseases would confirm that the achievement of full referential competence requires the availability of qualitative information as we define it.