A novel view of the syntax and semantics of quantifier scope that argues for a "combinatory" theory of natural language syntax. In Taking Scope, Mark Steedman considers the syntax and semantics of quantifier scope in interaction with negation, polarity, coordination, and pronominal binding, among other constructions. The semantics is "surface compositional," in that there is a direct correspondence between syntactic types and operations of composition and types and compositions at the level of logical form. In that sense, the semantics is in the "natural logic" tradition of Aristotle, Leibniz, Frege, Russell, and others who sought to define a psychologically real logic directly reflecting natural language grammar. The book reunites the generative-transformational tradition initiated by Chomsky--which views the formal syntactic component as entirely autonomous---with the older, strongly lexicalist, construction-based tradition, which has sought to define a more lingistically transparent theory of meaning representation. Steedman offers a logical formalism that relates directly to the surface form of language and to the process of inference and proof that it must support. Such a natural logic, although formal by definition, should be allowed to grow organically from attested language phenomena rather than be axiomatized a priori in terms of any standard logic. Steedman also considers the application of natural semantic interpretations to practical natural language processing tasks, emphasizing throughout the elimination of traditional quantifiers from semantic formalism in favor of devices such as Skolem terms and structure-sharing among representations in processing.

Vol. 49, no. 9 (Sept. 1922) accompanied by a separately paged section entitled ERA: electronic reactions of Abrams.

1 In 1954 Karl Popper published an article attempting to show that the identification of the quantitative concept degree of confirmation with the quantitative concept degree of probability is a serious error. The error was presumably committed by J. M. Keynes, H. Reichen bach and R. Carnap. 2 It was Popper's intention then, to expose the error and to introduce an explicatum for the prescientific concept of degree of confirmation. A few months later Y. Bar-Hillel published an article attempting to show that no serious error had been committed (particularly by Carnap) and that the problem introduced by Popper was simply a "verbal one. "3 Popper replied immediately that "Dr. Bar-Hillel forces me [Popper] now to criticize Carnap's theory further," and he [Popper] introduced further objections which, if accepted, destroy Carnap's theory. 4 About eight years after this exchange took place I was in graduate school at the University of Chicago in search of a topic for a doctoral dissertation. An investigation of the issues involved in this exchange seemed to be ideal for me because I had (and still have) a great ad miration for the work of both Carnap and Popper. A thoroughly revised and I hope improved account of that investigation appears in the first five chapters of this book. Put very briefly, what I found were four main points of contention.

How we can create artificial intelligence with broad, robust common sense rather than narrow, specialized expertise. It’s sometime in the not-so-distant future, and you send your fully autonomous self-driving car to the store to pick up your grocery order. The car is endowed with as much capability as an artificial intelligence agent can have, programmed to drive better than you do. But when the car encounters a traffic light stuck on red, it just sits there—indefinitely. Its obstacle-avoidance, lane-following, and route-calculation capacities are all irrelevant; it fails to act because it lacks the common sense of a human driver, who would quickly figure out what’s happening and find a workaround. In Machines like Us, Ron Brachman and Hector Levesque—both leading experts in AI—consider what it would take to create machines with common sense rather than just the specialized expertise of today’s AI systems. Using the stuck traffic light and other relatable examples, Brachman and Levesque offer an accessible account of how common sense might be built into a machine. They analyze common sense in humans, explain how AI over the years has focused mainly on expertise, and suggest ways to endow an AI system with both common sense and effective reasoning. Finally, they consider the critical issue of how we can trust an autonomous machine to make decisions, identifying two fundamental requirements for trustworthy autonomous AI systems: having reasons for doing what they do, and being able to accept advice. Both in the end are dependent on having common sense.