Search

The design of codes and formats for information storage must be guided by the requirements of the processes in which the stored information is to be used. In the case of medical data, a code must be designed to facilitate several different processes, including case management, administration, evaluation and research. Design of such a system can be facilitated by an examination of the theory, in which knowledge is represented by a formal model consisting of nodes and arcs connecting the nodes. Arcs are of five types: paradigmatic, syntagmatic, discursive, attitudinal and metalingual. Metalingual arcs, which allow for shifts in level of abstraction, provide the only natural means of passage among the several languages of medicine. The technical problems raised by these shifts have been neglected in the philosophy of science, in linguistics and in information retrieval. Any cognitive network needs to record paradigmatic, syntagmatic, discursive and metalingual arcs; attitudinal arcs are also sometimes needed. Medical record files must be able to keep all types of linkages to provide accurate histories, and medical authority files need them to record the structure of the science. The SNOMed system provides some, but not all, of these linkages. (Author/SL)

This paper explains the main concepts of communicative grammar and provides a detailed view of how communicative grammar analyses language at various levels. Language is discussed in terms of communication; the central elements in the analysis are those that carry information. Communicative grammar seeks to describe the process of the linguistic formulation of a thought. Details for analysis are provided here along with examples. Implications for theoretical, applied, and computational linguistics, language instruction, and translation are noted. A bibliography is included. (VM)

The author suggests that mathematical formulas could provide some direct and easily understandable frameworks for analogies in literary criticism. Most studies of textual problems, he points out, either have failed to use full mathematical models or have been reckless with the inherent limits of these techniques. Enumeration of linguistic traits is very common to literary analysis, and discussions of genre and text style based on linguistic data can be very informative. There are, however, serious dangers to be considered in the choice and size of text samples, and oversimplification of qualitative features should be avoided. He suggests that the most successful studies of this type involve a consideration of syllable and word structure rather than frequency of a literary device. In the text of this paper, the author examines various treatises on language usage in literary works, and comments on their value. His discussion of the applicability of mathematical formulas to literary analysis is quite technical. [Not available in hard copy due to marginal legibility of original document.] (FB)

This annotated bibliography lists approximately 100 books, reports, Rand memoranda, and papers, written between 1961 and 1971, in the fields of linguistic theory, language research, computational linguistics, and formal linguistics. Many documents involve the use of computers in problems of language analysis; many concern the analysis of Russian. Several works on machine translation are also listed. (VM)

One of man's most distinctive characteristics is the manner in which he stores and communicates information. Language has always been an important part of this process, but recently machines have begun to share the spotlight. This book presents a look at the role of language in the process of communication and man's relation, present and future, to his communications technology. The author discusses the strengths and limits of memory and evaluates the human factor in communications systems. He then surveys the surprising range of near-human activities that computers can perform. The book also includes an introduction to psycholinguistics; an unbiased look at developments in the field of physical research; and a discussion of Project Grammarama: a study of how people learn the grammatical rules underlying artificial languages. (Author/AMM)

Analyzes the jargon of surgeons (with a high density of concrete items of information) and of logicians (emphasizing the relationships between statements), and discusses how these studies may be used in information retrieval. (AL)

THIS PAPER DISCUSSES THE THEORY THAT CONTENT ANALYSIS OF MESSAGES IS A CONTRAL TOPIC IN ALL SCIENCES DEALING WITH MAN. CONTENT ANALYSIS IS THE DETERMINATION OF CHARACTERISTICS OF A SOURCE FROM THE NATURAL LANGUAGE UTTERANCES IT EMITS. ASSIGNING PROPERTIES TO WORDS AND THEN COUNTING THE APPEARANCES OF THE PROPERTIES IS INSUFFICIENT--THE CONTENT OF A MESSAGE SHOULD BE ANALYZED IN TERMS OF WHAT HAS GONE BEFORE AND ITS RELATIONSHIP TO WHAT THE SENDER AND THE RECEIVER ALREADY KNOW. THE CONTRIBUTION OF LINGUISTICS IS TO EXPLICATE THE RELATIONSHIP BETWEEN OVERT EXPRESSION AND CONTENT. THE REMARKABLY COMPLICATED RELATIONSHIPS BETWEEN ACOUSTIC SIGNALS AND CONTENT ARE SIMPLIFIED BY SYNTACTIC ANALYSIS, PREFERABLY THROUGH THE USE OF COMPUTERS. MUCH ABOUT LANGUAGE CANNOT BE EXPLAINED WITHOUT GRAND THEORY, AND THE UNEXPLAINED PART IS WHAT IS NEEDED FOR CONTENT ANALYSIS. WHILE LINGUISTICS CAN FURNISH A PARTIAL MODEL OF THE COMMUNICATOR, MODELS OF COGNITIVE PROCESSING OF ATTITUDE FORMATION AND CHANGE ARE ALSO NEEDED. SERIOUS WORK ON CONTENT ANALYSIS COULD ALTER THE SOCIAL AND BEHAVIORAL SCIENCES IN FUNDAMENTAL WAYS. THIS DOCUMENT (RAND PAPER P-3714) IS AVAILABLE FOR $1.00 (CHECK OR MONEY ORDER) FROM THE REPORTS DEPARTMENT, THE RAND CORPORATION, 1700 MAIN STREET, SANTA MONICA, CALIFORNIA 90406. CALIFORNIA RESIDENTS SHOULD ADD FIVE PERCENT SALES TAX. (AUTHOR/JD)

SESSION 63 OF THE 1967 INSTITUTE OF ELECTRICAL AND ELECTRONIC ENGINEERS INTERNATIONAL CONVENTION BROUGHT TOGETHER SEVEN DISTINGUISHED MEN WORKING IN FIELDS RELEVANT TO LANGUAGE. THEIR TOPICS INCLUDED ORIGIN AND EVOLUTION OF SPEECH AND LANGUAGE, LANGUAGE AND CULTURE, MAN'S PHYSIOLOGICAL MECHANISMS FOR SPEECH, LINGUISTICS, AND TECHNOLOGY AND LANGUAGE. AS THIS SUMMARY OF THEIR REMARKS AND SUBSEQUENT DISCUSSION SHOWS, THE SESSION BROUGHT UP A DIVERSITY OF IDEAS, SPECULATIONS, AND QUESTIONS, ESPECIALLY CONCERNING THE INTERACTION OF ENGINEERING, TECHNOLOGY, AND SPEECH. THE PANELISTS BRIEFLY SUMMARIZED RESEARCH IN THEIR FIELDS, EMPHASIZING THOSE PARTICULAR QUESTIONS WHICH WOULD BE OF GREATEST INTEREST TO ENGINEERS. MR. HOMER DUDLEY, INVENTOR OF THE VOCODER, DID THE ORGANIZING WORK AND POSED THE QUESTIONS FOR THE PANEL (WHICH INCLUDED GEORGE A. MILLER, JOSEPH BRAM, FRANKLIN S. COOPER, JOHN LOTZ, AND MANFRED R. SCHROEDER). CHAIRMAN OF THE SESSION WAS LEO L. BERANEK, AND THE MODERATOR WAS J.C.R. LICKLIDER. THIS ARTICLE WAS PUBLISHED IN THE "IEEE SPECTRUM," JUNE 1967, BY THE INSTITUTE OF ELECTRICAL AND ELECTRONIC ENGINEERS, 345 EAST 47TH STREET, NEW YORK, NEW YORK 10017. (JD)