Pragmatics & Cognition - Volume 1, Issue 1, 1993

It is widely believed that there are some emotions (so-called "basic emotions ") which are universally associated with distinctive facial expressions and that one can recognize, universally, an angry face, a happy face, a sad face, and so on. The "basic emotions " are believed to be part of the biological makeup of human species and to be therefore "hardwired". In contrast to this view, Or tony and Turner (1990) have suggested that it is not emotions but some components of emotions which are universally linked with certain facial expressions, or rather with some components of facial expressions. Ortony and Turner have made a good case for this hypothesis. But the theory will be more convincing - and indeed more verifiable - if the postulated "dissociable components of emotions" are formulated in a rigorous, and culture-independent manner. The paper argues that the Natural Semantic Metalanguage, based on universal semantic primitives and devised by the author and colleagues, provides a suitable culture-independent framework.

The two questions that constitute the title of the paper are examined in the context of thought sequences, i.e., progressions of phrase-like expressions that spontaneously run through people 's minds. The analysis of a corpus of such sequences suggests that the articulation of thought in language affords fluidity that makes novelty possible. The articulation makes control possible, it lends momentum to thought, it presents alternative avenues for the further progression of thought, it renders thought into an activity akin to action in the real world, and it results in objectivization that provides compartmentalization and reflection. While the discussion focuses on the medium of language, it is noted that similar patterns hold with other media of articulation, both in natural cognition and in the arts. General implications are proposed and discussed.

The ways in which given and new information are distributed, and the functions associated with the distribution, are examined here in terms of information content of relative and adverbial clauses in oral and written narratives. The conventional view that subordinate clauses tend to code given rather than new information is shown to be inadequate. Moreover, a comparison of oral and written narratives of the same events reveals both extensive differences in the two modalities, and at the same time striking similarities in terms of the information distribution within relative clauses. Relative clauses are found to be far more frequent in oral narratives than in their written analogues. However, when the differences are examined in terms of the relative frequencies of given and new relative clauses, the oral narratives are shown to have far more given relative clauses than the written versions, whereas the frequencies for the new relative clauses is virtually identical in the two modalities. This result is attributed to memory constraints.

Within the controversy between the combinatorial and the connectionist approaches to cognition it has been argued that our semantic and syntactic capacities provide evidence for the combinatorial approach. In this paper I offer a counter-weight to this argument by pointing out that the same type of considerations, when applied to the pragmatics of adjectives, provide evidence for connectionism.

This paper investigates temporal interpretations in Pirahâ, a Muran language spoken in the Brazilian Amazon basin. The analysis assumes the neoReichen-bachian model of tense syntax proposed in Hornstein (1990) and argues that this model provides an elegant account of tense-related facts in Pirahâ, iff it is parametrized. Whereas Hornstein predicts that all tense systems will have a temporal reference point (R), this paper argues that languages may be parametrized as [+R] or [-R] and that this has important implications for their temporal syntax. Moreover, the paper also argues that it is no coincidence that the Pirahâ place little importance on precision time statements, or that the Pirahâ have difficulty translating such statements, since their language does not draw temporal distinctions based on R. The parametrization of R among the Pirahâ is argued to offer anew source of support for Sapir's linguistic relativity hypothesis.

This paper presents a speech act analysis of presumption, using the framework of a dialogue in which two parties reason together. In the speech act of presumption, as opposed to that of assertion, the burden of proof resides not on the proponent to prove, but on the respondent to rebut. Some connections of this account with nonmonotonic reasoning and informal fallacies in argumentation are explored.

Lenat has come to the conclusion that his celebrated computer programs, AM and Eurisko, despite impressive achievements, ultimately f ail as creative discovery systems. The paper discusses Lenaťs various attempts to model sustainable creative intelligence, primarily within the paradigm of symbolic processing. Lenaťs current approach requires a larger knowledge base than that of his previous efforts, and in fact Lenat hopes that his new system, Cyc, will have achieved a human breadth and depth of knowledge in the relatively near future. It is concluded that the Cyc project should be extremely important f or future A.I. research, making a significant contribution to the connectionist/symbolist debate.