College and Research Libraries DAN BERGEN Implications of General Systems Theory for Librarianship and Higher Education Systems science has both theoretical and empirical dimensions. General systems theory investigates the isomorphism of concepts, laws, and models in the various domains of human inquiry. General systems theory has implications for both academic librarianship and the teach- ing-learning process in higher education. With respect to the former, it promotes the conceptual reorganization of knowledge for storage and retrieval. With regard to the latter, it seeks to identify concepts whose organizing power transcends the artificial bounds of the aca- demic disciplines. DURING the first quarter of the nine- teenth century, American higher educa- tion was attacked by those who deplored its theological orthodoxy and smug clas- sicism. In defense of the existing order and in an attempt to define what the American college ought to be, the facul- ty of Yale College issued its famous (or, depending upon one's point of view, in- famous) report of 1828. The goals of liberal education were stated early in the report: "The two great points to be gained in intellectual culture, are the discipline and the furniture of the mind; expanding its powers and storing it with knowledge."1 Set in a mid-twentieth- century context, these educational pre- scriptions do not seem entirely irrelevant despite their defensive posture at the time of enunciation. In the 137 years since the issuance of the report, there has 1 Quoted in Frederick Rudolph, The American College ·and University: A History (New York: Knopf, 1962), p. 132. The Yale Report of 1828 was pub- lished as "Original Papers in Relation to a Course of Liberal Education,'' The American Journal of Science and Arts, XV ( 1829), 297-351. It also ap- peared in Reports on the Course of Instruction in Yale College: by a Committee of the Corporation and the Academical Faculty (New Haven [1828] ). Mr. Bergen is Chairman and Associate Professor, Department of Library Science, University of Mississippi. 358 / seemed to develop, the countervailing efforts of this century notwithstanding, an imbalance in American higher educa- tion in favor of one of its two emphases. Concerned with furnishing student minds with knowledge, mainly knowl- edge of a low order of abstraction, Amer- ican institutions of higher education have by and large underemphasized the goal of discipline, that is to say, the task of acquainting students with the means by which the knowledge they have been furnished may be ordered and made meaningful. It is precisely in this realm of discipline, or, put another way, in the imaginative organization of knowl- edge for transmission from one genera- tion to the next, that the major implica- tions of general systems theory for li- brarianship and higher education seem to lie. Only to the extent that these im- plications are realized can Kenneth Boulding' s suggestion of a reorientation in educational thinking-from the exist- ing norm of maximum knowledge trans- mission to a new and more relevant norm of minimum knowledge transmission- be accomplished. 2 It is to that very large 2 See Boulding, The Image: Knowledge in Life ond Society (Ann Arbor: Univ. of Michigan Pr., 1956), pp. 162-63. Implications of General Systems Theory 1359 end that this very modest paper seeks to contribute. The processes underlying the growth of human knowledge are undoubtedly very complex. The very complexity of these processes should not, however, de- ter a search for pattern and order in them. One way of viewing the growth of knowledge historically is through the alternating dominance of raw empiricism and low order empirical theory on one hand, and speculative philosophy and logically and mathematically informed hypothetico-deductive theory on the other. To interpret the development of knowledge within such a cyclic frame- work is to see its growth in terms of the dominant strategies employed over time in its generation. For example, the specu- lative philosophy of the late medieval period can be depicted as hovering above, in its abstractness and formalism, the seeming disorder of the empirical world much like a hovercraft hovers di- rectly above a point on land or water without ever settling on it. The intel- lectual formulations of the fourteenth century have been characterized as "ver- bal, logic-chopping, and abstract" as a consequence of their obsession with ''discipline, method, and artifice." In con- trast, a medieval schoolman reincarnated during the last quarter of the nineteenth century might well have deplored the ex- cessive empiricism of the emergent aca- demic specialties and their emphasis on ''subject matter, experience, and nature," not to mention their inclination toward the "factual, pragmatic, and technical."3 3 Richard P. McKeon, "The Liberating Arts and the Humanizing Arts in Education," in Arthur A. Cohen (ed.), Humanistic Education and Western Civi- lization: Essays for Robert M. Hutchins (New York: Holt, Rinehart and Winston, 1964, pp. 169-70. A modem analogue of the postulates of speculative philosophy are the mathematically formulated deduc- tive postulates of theoretical physics which tend to be judged more on a basis of aesthetic excellence than on any criterion of empirical validation. See Peter Caws, The Philosophy of Science: A Systematic Account (Princeton, N.J.: Van Nostrand, 1965), pp. 283-84. Peter Caws has interpreted the pro- gression of science as a continuing dia- lectic or interplay between its logical and experimental aspects, the former involving ''the rational investigation of connections between concepts, without special regard to the adequacy of the concepts to experience" and the latter "the empirical investigation of connec- tions between events, without special re- gard to the significance of these events in any total scheme of things."4 The re- lationship between logical construction and empirical investigation in the de- velopment of knowledge viewed both holistically and in its man-defined sub- divisions seems indeed to be reflected in a mutual feedback which drives the structure of knowledge, over the long term, toward a kind of homeostatic steady state. It has been suggested that the output of empirical knowledge in the various academic disciplines ( and therefore, by extension, in knowledge as a whole) is controlled by the differen- tial integrative capacities of those disci- plines.5 Thus it follows that where a discipline's logical or theoretic-deduc- tive component is well developed the production of empirical knowledge will be facilitated. This hypothesis also sug- gests, at least for the modern intellectual disciplines, that any imbalance between theoretic and empiric dimensions is ulti- mately corrected by the internal logic of equilibrium. Still, over the short term, the empiric is capable of outrunning the theoretic and vice versa. From the stand- point of organizing knowledge for trans- mission, the seriousness of theoretical knowledge outdistancing empirical knowledge is not so great as that of em- pirical investigation outrunning integra- 4 Caws, op. cit., p. 331. 5 See Norman W. Storer and Talcott Parsons, "The Disciplines as a Differentiating Force," in Dan Bergen ( ed.), The Foundations of Access to Knowledge (Syracuse, N.Y.: School of Library Science, Syracuse University, in Press), (pp. 23-24 in the mimeographed form). 3601 College & Research Libraries • September, 1966 tive capacity.6 It is this latter type of imbalance that can lead to the crisis of a civilization incapable of meaningfully integrating its knowledge. To counter such imbalance would seem to be a most worthy rationale for general sys- tems theory. In an anthropomorphic sense, the growth of knowledge may be viewed in evolutionary perspective as a process of progressive differentiation of the total domain of inquiry. Even though philoso- phy embodied the bulk of human knowl- edge from the classical Greek period to the eighteenth century, some differentia- tion could already be observed during medieval times. If the grammar, rhetoric, and logic of the trivium were not strict- ly subjective matters, the remaining lib- eral arts of the quadrivium (arithmetic, geometry, astronomy, and music) had the character of nascent academic disci- plines.7 There were also, in the medieval universities, separate faculties of law, theology, and medicine. By the early eighteenth century, philosophy itself had developed two identifiable branches- moral philosophy, which remained large- ly speculative and normative in charac- ter, and natural philosophy, which was scientific mainly in the enumerative or taxonomic sense of that term. The mod- em scientific disciplines did not evolve out of natural philosophy until well into the nineteenth century. It is widely held, at the present time that the individual disciplines of the physical, biological, and social sciences are distinguished from one another by the different objects 6 Caws , op. cit., pp. 331-34. Karl W. Deutsch has discerned that the interplay between the empirical- inductive and theoretical-deductive components in sci- ence define a continuum along which the different disciplines may be located in terms of the relative importance of these two dimensions within them. See his "Scientific and Humanistic Knowledge in the Growth of Civilization," in Harcourt Brown { ed. ) , Science and the Creative Spirit: Essays on the Hu- manistic Aspects of Science {Toronto: Univ. of Toron- to Pr., 1958), p. 8. 7 McKeon in Cohen { ed.), op. cit., p. 165. and events which they seek to explain. 8 Russell Ackoff has observed, however, that the intellectual disciplines are sim- ply a reflection of man's effort to or- ganize nature and not necessarily there- fore reflective of the inherent structure of nature. Academic disciplines, as a con- sequence, are differentiated less on the basis of unique subject matters than on a determination of their attention to dif- ferent aspects of reality and by the vari- ous ways in which they explain the same phenomena. 9 This distinction is impor- tant since it calls into question the exis- tence of a correspondence between the structure of knowledge as embodied in the disciplines, and the structure of na- ture. This difference will be discussed in greater detail later in the paper when the distinction between conceptual and concrete systems is considered. It has been intimated above that the development of individual disciplines, indeed the growth of knowledge as a whole, may be viewed cybernetically as the interplay or mutual feedback of theo- retical and empirical components. Saul Gorn has conceptualized this process in terms of information storage and re- trieval using file growth as an analogy: What happens is that there must be a bal- ance, as the file grows, between the retriev- ability of the information and the needed communication flow of the information; the control of the information due to the struc- ture of its arrangement must balance the method by which it is communicated. What has happened, then, is that the continuum of a domain of knowledge when a revolu- tion is due, either splits into distinct fields or changes phase radically by a change in the structure of its arrangement. That is, 8 This is one of the criteria specified by Caws, op. cit., p. 280. This criterion for the differentiation of disciplines is shared by Marc Belth. See his Education as a Discipline: A Study of the Role of Models in Thinking {Boston: Allyn and Bacon, 1965), p. 6. 9 See Russell L . Ackoff, Scientific Method: Optimiz- ing Applied Research Decisions {New York: Wiley, 1962), p. 419. Implications of General Systems Theory 1361 once the mass of information got beyond the critical mass, usually either fission or fusion occurred. Fusion is the change in the structure of the individual science itself to make it a much more compact carrier of information; for example, general laws are highly compact bouillon cubes of informa- tion. Fission is the breaking up into various areas of specialization, which we have al- ready mentioned. The fission or fusion oc- curred in each case because the information got beyond our capacity, and its former control began to require too much time to retrieve. In order to be able to retrieve the information, you either have to have separate specialized lumps or have a vio- lent change in structure.lO Viewed macroscopically from the standpoint of knowledge as a whole the discrete disciplines yielded by man-di- rected fission can be re-fused through the imposition of man-inspired constructs which correlate, in some meaningful way, the empirically validated principles of the various academic disciplines. Viewed microscopically, from the stand- point of a particular discipline, the he- gemony of either fission or fusion is de- pendent in very large measure upon that discipline's stage of development. The thrust of fission is strongest during a discipline's initial or taxonomic stage of growth. This first stage of inquiry, called by F. S. C. Northrop the "natural his- tory" stage, is concerned with classifica- tion "whether dealing with pure com- pounds in chemistry, minerals in geolo- gy, species in botany, structures in anat- omy, or specific functions in physiol- ogy."11 The second level of a discipline's development is that of empirically-based theory, what Caws has called the nomo- 10 Saul Com, "Computers, Communications, and Science-Extending Man's Intellect," in Lowell H. Hattery and Edward M. McCormick ( eds.), Informa- tion Retrieval Management (Detroit: American Data Processing, 1962 ), p. 126. 11 R. W. Gerard, "Quantification in Biology," in Harry Woolf ( ed.) Quantification: A History of the Meaning of Measurement in the Natural and Social Sciences (Indianapolis: Bobbs-Merrill, 1961 ), p. 210. logical stage, after the Greek "nomos" or "law." The Pythagorean Theorem, for example, defined a relationship existing between the sides of all of the right tri- angles, regardless of size, which were classified as such during the first, or taxo- nomic, stage of discipline development. In the third and final period of growth, deductive systems based on very broad postulates from which less general prop- ositions can be logically derived are seen to develop. A substantial part of physics has entered this final, often highly math- ematical, stage. Chemistry and that part of biology which has a mathematical orientation are moving gradually into it. Since the late nineteenth century, a vast number of deductive systems have come into existence. Providing great impetus to the proliferation of such systems was, of course, the realization by mathema- ticians of the century just past that the axioms of Euclidean geometry, particu- larly that which proclaimed that through a point not on a given line only one line could be drawn parallel to the given line, were not self-evident truths or giv- en a priori to the intuition, as Kant and scholars in general had believed for some two thousand years. 12 Other de- ductive systems in geometry, as a con- sequence of this discovery, seemed en- tirely possible and were presented as such. Even this last stage of inquiry, where systems of immep.se deductive fer- tility are seen to exist and promote fu- sion, has its serious limitations. The gen- eral faith of mathematical logicians in the viability of axiomatic systems was shaken in the early 1930's by the young Viennese mathematician Kurt Geode! who demonstrated that any consistent system which is strong enough to pro- duce natural numbers and operations 12 While Euclid's axiom still holds for everyday kinds of lines, it is not relevant for Einsteinean space where parallel lines do not exist. See Michael A. Arbib, Brains, Machines, and Mathematics (New York: McGraw-Hill, 1964), p. 120. 362 1 College & Research Libraries • September, 1966 like addition and multiplication can also contain formulae which are meaningful and true even though they cannot be proved with the system itself. Neverthe- less, an academic discipline which has matured to this final stage has made a robust advance. There is a vast differ- ence, as Northrop is wont to point out, between the immediately sensed color "blue" of the natural history stage of in- quiry and the conceptual notion of ''blue" in the final, or deductively for- mulated theory, stage of inquiry where that color is defined by its hue (or wave- length), its intensity ( or the amount of energy it reflects per square centimeter per second), and its saturation, all of these characteristics deriving their mean- ing from the system of postulates of which they are a p.art. 13 It would be a mistake, however, to view all activities within a particular discipline as being at the same level of development and sophistication. Indeed, in the field of biology the taxonomic ac- tivity of the botanist, or entomologist, or zoologist, and the process-oriented studies of the mathematical biologist can prove mutually reinforcing, as the mor- phogenetic studies of Edmund Sinnott on the relationships between form and function seem to suggest. One authority even suggests that physics, much of which is concerned with deductive sys- tems, has its descriptive and nomo- logical areas. And the social sciences are still, despite efforts to formulate a "so- cial physics," by and large restricted in their activity to the descriptive and nomological levels. 14 Although fission and fusion strategies are capable of co- existence (indeed must coexist in some manner), one will usually be dominant at a given point in a historical time. Corn, it has been learned, sees either fis- sion or fusion vis-a-vis the body of knowledge as capable of solving the 13 F. S. C. Northrop, The Logic of the Sciences and the Humanities (New York: Meridian, 1959), pp. 102- 104. 14 Caws, op. cit., pp. 280-81. See also Gerard in Woolf (ed. ), op. cit., p. 210. problem of information storage and re- trieval. From the vantage of one whose primary concern is with meaning and cohesiveness as the compelling reasons underlying any effort to restructure knowledge the strategy of fusion seems far more promising and imperative. 15 The fused character of knowledge and its interconnectivity may be seen his- torically in man's employment of primi- tive models to explain his largely un- differentiated external reality, and later, after it was decided to organize the con- tinuity of knowledge in fundamentally discontinuous disciplines, his adoption of more sophisticated models which, having their genesis in a particular dis- cipline, proved their analogical sugges- tiveness by rapidly penetrating other fields, scholars as diverse as the educa- tional philosopher Marc Belth and the political scientist Karl Deutsch have seen in the study of the development and pervasive diffusion of models for orga- nizing reality a principle which might well underlie the educative process. The earliest models by which man sought to explain the universe of phenomena which impinged upon him were, of course, anthropomorphic, that is to say, an .ascription of human qualities to the elements and behavior of nature. Later in the course of human history the wheel, the balance, the web of thread (with its implication of interaction), the pump and the clock (with their implica- tion of mechanistic determinism), and the classical organism with its infused teleology or goal structure served as mod- els.16 Models, as such, vary in character. 1 5 Sigmund Koch suggests that recent introspection in science, education, and philosophy has reversed the emphasis on an analytico-reductive approach. See his "Psychology and Emerging Conceptions of Knowledge as Unitary," in T. W. Wann (ed.), Behaviorism and Phenomenology: Contrasting Bases for Modern Psy- chology (Chicago: Univ. of Chicago Pr., 1964), pp. 1-4. 16 Karl W. Deutsch, "Higher Education and the Unity of Knowledge: An Operational Approach to the History of Thought," in Lyman Bryson, Louis Finkel- stein, and R. M. Maciver ( eds. ), Goals for American Education (Ninth Symposium of the Conference on Science, Philosophy, and Religion in Their Relation to the Democratic Way of Life [New York: Harper, 1950] ), pp. 102-105. Implications of General Systems Theory 1363 Some are mythic, that is to say, they explain unfamiliar and often frightening phenomena in terms of traditional myths such as that of Isis and Osiris in ancient Egypt, which was used, by analogy, to explain the changes of season.17 Such mythic models were most frequently in- herited and learned by what Deutsch calls mimesis or unanalyzed imitation.18 A typology of models would also include the scientific, the historical, and the ideological. Scientific models, which in theory were open to empirical validation or invalidation, were used to discover, control, and predict the course of natural processes. Newton's mechanism-based model of the motion of large aggregates would be an example of a scientific model. Historical models are, to quote Belth, "concerned with the relation in thought between particular events of one time and particular events of an- other, and with the cumulative and di- rective effect that events have upon one another in the minds of men."19 Like scientific models, historical models may be mathematical (usually, at least in more recent times, probabilistic); ana- logical (like Marx's iron laws of history which find their analogue in both the determinism of classical physics and the psychological determinism of organismic growth in the sense of Teilhard de Chardin); or even theoretical with ana- logical elements (such as the cyclic methahistory of a Toynbee or a Speng- ler). Finally, there are ideological mod- els which "are bound to the systematic image of an unambiguous, perfect, or- derly reality, which serves as the instru- ment for interpreting the imperfections which are all about us as w~ live out our lives."20 An example of an ideological model would be that historic form of utopianism which is the scriptural com- munism of Marx and Lenin. 17 Belth, op. cit., pp. 180-81. 1 8 Deutsch in Bryson et al. ( eds.), op. cit., p. 79. 19 Belth, op. cit., pp. 180-84. See also Henry Mar- genau, Ethics and Science (Princeton , N.J.: Van Nostrand , 1964), pp. 65-66, 89-90. 2 0 Belth, op. cit., pp. 180-84. The analogical suggestiveness of mod- els for disciplines external to those par- ticular disciplines in which such models have their genesis is demonstrated by the models of society currently employed in sociology, almost all of which have been imported from the biological sci- ences, the physical sciences, or mathe- matics. These include such models as the evolutionary, the organismic, the equili- brium, the classical physical science, and the statistical-mathematical.21 The cas- cading quality of analogical suggestive- ness may be seen in the transposition of the notion of entropy from its original locus in thermodynamics to a central position in information theory and, via information theory, into numerous other disciplines. Using Claude Shannon's ar- ticles on information theory in the July and October 1948 issues of the Bell Sys- tem Technical ] ournal as a conceptual unit capable of diffusion, Randall Dahl- ing traced, through the examination of publications, the way in which this con- ceptual innovation penetrated the vari- ous disciplines. In addition to its evoca- tive qualities, Dahling determined that the notion of information theory spread because: ( 1) communication is a com- mon ground for many disciplines; and ( 2) because its mathematical formula- tion and methodology was appealing to natural scientists and the increasingly sizeable _ group of social scientists which is oriented to quantification.22 In the final analysis the adoption of a concept- ual innovation by scholars in disciplines 21 See Alex Inkeles, What Is Sociology? An Intro- duction to the Discipline and Profession (Englewood Cliffs, N.J.: Prentice-Hall, 1964 ), pp. 28-46. 22 Randall L. Dahling, "Shannon's Information Theory: The Spread of an Idea," in Studies of Innova- tion and of Communication to the Public (Studies in the Utilization of Behavioral Science, XI [Stanford: Institute for Communication Research, Stanford Univ., 1962] ), pp. 119, 136. Chronologically, the diffusion of the information theory concept followed this order: communications engineering, psychology, physiology, physics, linguistics, biology, sociology, optics, statistics, and journalism. It is to be anticipated that more can be learned about the spread of ideas from the cita- tion network studies of Derek Price at Yale, Eugene Garfield at the Institute for Scientific Information, Ben-Ami Lipetz in Carlisle, Mass., and Norman Kaplan at George Washington University. 364 I College & Research Libraries • September, 1966 external to its source seems to rest with some notion of congruence or "fit" be- tween the "culture" ( or set of intellectual commitments) of a potentially receptive discipline at a given point in time, and the nature of the new idea itseH. Incon- gruence between a potentially pervasive concept and the existing W eltanschau- ung shared by the set of potentially re- ceptive scholars in an external discipline can result in a lag in the adoption of the concept whose implications, for a variety of possible reasons, cannot be immedi- ately sensed.23 It is entirely possible, moreover, that the cross-disciplinary utility of concepts increases as such con- cepts undergo an evolutionary process of refinement and reinterpretation in their disciplines of origin. Etymologically, for example, the concept of force had a strongly animistic cast. Its progressive refinement at the hands of Aristotle, Galileo, Kepler, Newton, Kant, Mach, and Hertz has resulted in the present period in a concept of force which is al- most purely mathematical in meaning, that is to say, without immediate physi- cal referents. 24 The capacity to be able to predict the diffusion of an idea into adjacent do- mains, even on a probabilistic basis, could have important implications for the ongoing organization of information in libraries, and for the restructuring of the bibliographic devices which provide access to that information. Even without such predictions, as Don Swanson has suggested, adequate circulation informa- tion and the careful observation of in- library user behavior would permit the association of users with those portions of the graphic record which they use, 23 See Elihu Katz, Margin L. Levin, and Herbert Hamilton, "Traditions of Research in the Diffusion of Innovation," American Sociological Review, XXVIII (April 1963), 240, 249-50. 24 The evolution of the concepts of space, force, and mass have been considered in several books by Max Jammer. See William R. Catton, Jr. , "The Develop- ment of Sociological Thought," in Robert E. L. Faris ( ed. ) , Handbook of Modern Sociology (Chicago: Rand McNally, 1964), p. 935. in such a way that the development of new fields like bionics (or artificial in- telligence) might be anticipated on the basis of the significantly large common group of library users which is making intensive use of the literature of neurol- ogy and information theory. 25 To be · sure, citation analysis might well pro- vide information on patterns of concept diffusion, however, the time differential between the actual acceptance of an externally generated concept by a re- ceptor field and the first publication in that field employing the concept is sub- stantial enough to render such an ap- proach overly a posteriori to be of real utility in the restructuring of biblio- graphic arrangements. Another way of conceptualizing the growth of knowledge is within a frame- work of progressive subordination, via hierarchical ordering, of old concepts, theories, and hypotheses to new ones. Such subordination imparts to the struc- ture of knowledge a "nesting" effect in which newer, more abstract concepts serve as receptacles for the older, less abstract concepts which are their special cases. Another image would be that of intersecting continua. If a vertical con- tinuum ranging from "less abstract" to "more abstract" is developed to intersect with a horizontal continuum moving from "less analogically suggestive" to "more analogically suggestive," the re- lationship defined by the line xy in the following diagram could be expected to develop. The line xy, it can be seen, expresses the direct relationship between abstract- ness and analogical suggestiveness. Lest the foregoing seem overly tautological, it can be speculated that the abstract concepts of "uncertainty" and "comple- mentarity" (or the duality of wave and particle motion) in quantum physics, 2 5 Don R. Swanson, "Dialogues with a Library Ma- chine" (Chicago: Graduate Library School, Univ. of Chicago, n.d. [Mimeographed]), p. 9. Implications of General Systems Theory I 365 More abstract Y / / / Less More ~~o~~llv------------~~-----------~~~o~~lly suggesttve suggestive / / Less abstract not to mention the emerging principles linking matter and anti-matter, will sug- gest more about the nature of social phenomena and, by extension, about the limits of explanation in the social sci- ences than any of the concepts hereto- fore imported into this broad area of in- quiry. Moreover, any general theory which claims to explain behavior in a plurality of phenomenal systems must be sufficiently abstract .and deductively fecund to account not only for already observed ismorphism in a diversity of behavioral realms but also for behavior in systems which have not yet been empirically identified. Returning to the hierarchical ordering of concepts accord- ing to comprehensiveness, Boulding has observed that- The old is almost invariably seen as a spe- cial case of the new. Algebra generalizes the operations of arithmetic. The oalculus generalizes some operations of algebra. The theory of games is a generalization of the theory of simple maximization. In physics, Newtonian mechanics is seen .as a special case of the mechanics of relativity. In eco- nomics, the Keynsian system is easily seen as the generalization of the classical system. 26 The excellent studies of Thomas Kuhn on the revolutionary character on para- digm change in science27 notwithstand- ing, it would seem that there is a con- tinuity in the development of knowledge which is remarkably antithetical to revo- lution. Michael Polanyi suggested this continuity when he characterized the Copernican system as no less anthropo- centric than the system of Ptolemy in that the former satisfied man's need to abstract as well as the latter exploited the enjoyment men find in the use of their senses.28 Indeed, Leonard Nash 2e Boulding op. cit., pp. 77-78. See also Leonard Nash's dist:m'ction between "falsification" and "sub- ordination" (or incorporation) in The Nature of the Natural Sciences (Boston: Little, Brown, 1963 ), pp. 286-96. f s . ·.t:- 27 See Thomas Kuhn, The Structure o c'ent11"" Revolutions (Chicago: Univ. of Chicago Pr., 1962). 2s Michael Polanyi, Personal Knowledge: Toward a Post-Critical Philosophy (London: Routledge and Kegan Paul, 1958), pp. 3-4. It mi?ht not be un- thinkable to suggest that the Ptolemaic system was a rape of precisely those capacities which distinguis~ man from lower orders of living systems, namely, his ability to abstract and theorize. Nash ~as observed that despite the overthrow of the Ptolemaic system by the Copernican, astronomers .continued ~o sp~ak of the regularity of motion, orb1ts, and obJects m the sky. See Nash, op. cit., pp. 284-92. 366 I College & Research Libraries • September, 1966 has used the term evolution to describe the growth of science. Invoking the anal- ogy of organisms evolving toward higher levels of organized complexity, he sup- ports the image, originally proposed by Samuel Sambursky, of science as ad- vancing by "concentric approximations," that is to say, through viewing new, more abstract theory as possessing many of the components of the theory or theo- ries that it has subordinated.29 It may not even be stretching the point to sug- gest that the development of new para- digms in science approximate the emer- gence of new species in biology. The principle of ontogeny recapitulating phylogeny suggests that the life history of an individual organism from concep- tion on takes it through roughly the same stages that evolution took that species of organism up to the time the individual organism came into existence. In much the same way, a new, compre- hensive theory in science typically in- corporates the lower order theoretical structure which had developed in sci- ence up to that time, but represents an improvement on the latter in its orga- nizational and correlative powers.30 There are other ways in which the fundamental continuity of knowledge may be observed, in the humanities as well as the natural and social sciences. A familiar form of identifying continui- ties in literature and other humanistic disciplines is through the analytical iso- lation of those formative structures that J ung called archetypes, or primordial ideas which are rooted in the collective unconscious of the human race.3 1 Jung 29 Nash , op. cit., pp. 286-88. 80 I n a somewhat different but analogous context, Talcott Parsons has defined an "evolutionary universal" as " a complex of structures and associated processes the d evelopment of which so increases the long run adaptive capacity of living systems in a given class that only systems that develop the complex can attain certain higher levels of general adaptive capacity." See his " Evolutionary Universals in Society," American Sociological R eview, XXIX (June 1964), 340-41. 81 These archetypes may be distinguished from Plato's ideas or the categories of Kant in that they are not completely a priori, having had their genesis in man' s conceived archetypes less in terms of inherited ideas about love or religion or, more negatively, about the perversion of love in lust or hatred, but rather as the inherited potential for such ideas, a potential which began to cumulate at that point in historical time when man became conscious of himself as some- how different or set apart from the rest of nature.32 Briefly defined, archetypes are symbols which "carry the same or very similar meanings for a large por- tion, if not all, of mankind."33 In the archetypal mode of literary evaluation, the focus is on the genesis and persist- ence in literature of symbolism such as day and night, spring and autumn, birth and death, innocence and experience, and on such provocative figures as Faust, Odysseus, Satan, and Prometheus. There are dangers, of course, in the inappro- priate use of archetypes in the explana- tion of literature. One is reminded of Douglas Bush's witty comment on arche- types: "Some ideas, such as frustration, became master keys for opening all doors. A crowd of authors and charac- ters were seen trudging along the road back to the womb. Along a parallel road stumbled another crowd driven by the death wish."34 Some archetypes, it may be supposed, can be explained by certain constraints of the natural order. Physically, all men are subject, at least in their natural state, to the laws of gravitation. As a conse- quence, the obstacles to vertical move- own primordial experiences. See Carl Gustav Jung, Collected Works (Vol. IX, The Archetypes and the Collective Unconscious [New York: Pantheon, 1959] ) , pp. 3-4, 16, 23. Jung has referred to archetypes as the " residua of racial experience." See Maud Bodkin, Studies of Type-Ima ges in Poetry, Religion, and Philosophy (London: Oxford University Press, 1951) , · p. 9. See also Jolande Jacobi ( ed. ), Psychological Reflections: An Anthology of the Writings of C. G. ]ung (New York: Harper, 1953) , pp. 38-39. 82 See Jacobi (ed.), op. cit., p. 36. Jung likens arche- types to receptacles which are filled in by the cul- tural experience of a lifetime. sa Philip E. Wheelwright, M etaphor and R eality (Bloomington: Indiana Univ. Pr. , 1962), pp. 111-12. 84 As quoted in William C . DeVane, Higher Educa- tion in the Twentieth Century (Cambridge, Mass. : Harvard Univ. Pr., 1965) , p. 111. Implications of General Systems Theory I 367 ment of an upward sort are more for- midable than those confronting one who seeks to go down. Therefore, the basic notions of dominance, achievement, loftiness, and excellence, all of which represent states one must struggle for, are associated with upward movement rather than with the ease of downward motion.35 Most archetypes, however, are elusive and their essence can be only indirectly explained through the use of imperfect metaphors. Nor does continu- ity in archetypal meaning always pre- vail, everywhere and in every period. Symbolically, the wheel as ~rchetype, may have either positive or negative signi£cance. In the East it has the posi- tive connotation of the persistent cycle of birth, death, and rebirth. In the West, it has a more ominous symbolic meaning related to the role of fortune (for ex- ample, the wheel in the game of rou- lette) in determining one's success or failure. 36 Fustel de Coulanges devoted most of his life to studying the elements, particularly religious beliefs informed by archetypes, influencing the growth of Greek .and Roman institutions, particu- larly the life of the city.37 Mircea Eliade has sought to identify the formative con- cepts at work in a number of archaic so- cieties, and found in the course of his investigations that time and the flow of history in such societies is regulated mainly by the operation of archetypes. These archetypes expressed the ontology of primitive societies like those of ancient Mesopotamia or aboriginal Australia- societies which largely lacked the philo- sophical language to express their no- tions of being and reality. The ontology of such societies had to be inferred from their rites, symbols, and myths.ss It was 8 5 Wheelwright, op. cit., pp. 111-12. 86 Ibid., p. 126. 87 See Numa Denis Fustel de Coulanges, The An- cient City: A Study on the Religion, Laws, and In- stitutions of Greece and Rome (Garden City, N.Y.: Doubleday, 1956), pp. 11-14, 142-73. 88 Mircea Eliade, The Myth of the Eternal Return ( ew York: Pantheon, 1954), p. 3. observed that the life of primitive man was fulfilled only to the extent that he performed certain paradigmatic gestures or spoke symbolically powerful words which originated with the celestial gods. The mountains that he climbed, the land that he cultivated, the rivers that he navigated, all of the acts of primitive man, found their meaning in the arche- typal actions primordially performed by transcendent gods. The temples and cities of ancient men were also patterned after such extraterrestial prototypes. Babylon, for example, was surrounded by a vast circular territory bounded by a river (the Euphrates) in precise imi- tation of the Babylonian vision of a celestial heaven. 39 In the sense that these godly actions took place at the beginning of time, the ancient man participated in the beginning of things through their re-enactment. Two of the most fundamental arche- types which recur in philosophy and re- ligion as well as literature are those of saving wisdom and spiritual rebirth.40 Northrop Frye has proposed that these and other very powerful archetypes are embodied in the Greek classics, especially Homer, and in the Bible and have rever- berated through literature to the present period. He sees the myths of creation and fall in Genesis and of the pastoral life as the strongest formative principles in religion and social, political, and phil- osophical thought. 41 Put another way, literature from the end of the first cen- tury A.D. to the present is mainly derived and mimetic, deriving its thematic con- tent from the past and its substantive content from its own historical period in much the same way that a Jungian archetype maintains its form while wear- 89 See ibid., pp. 9-10. 40 Bodkin, op. cit., pp. 35, 175. 41 See Northrop Frye, "The Developing Imagination, .. in Learning in Language and Literature (Cambridge, Mass.: Graduate School of Education, Harvard Univ., 1963 ), pp. 31-60. Also his "Introduction" in Northrop Frye (ed.), Design for Learning (Toronto: U niv. of Toronto Pr., 1962) , pp. 3-17. 368 1 College & Research Libraries • September, 1966 ing the experiential clothing of different eras. In this perspective, literature is the embellishment and elaboration of time- less themes. The literateur, in the final analysis, has no really new stories to tell. This thematic continuity is described by Frye in the following way: "At a point in the narrative, recognition point, or as Aristotle calls it anagnorisis, the reader knows what is going to happen, but he wishes to continue participating in the completion of the design."42 As units of discourse in literary criti- cism, archetypes are conceptually im- portant since they impart to literature its characteristic forms. 43 Even in music, one can sense that "inner unity of sym- bolic forms" that Ernst Cassirer saw as providing a basic homogeneity to human culture.44 Leonard Meyer, bridging be- tween music and psychology, has hy- pothesized that meaning in music may be interpreted in terms of the amount of congruence which prevails between one's generalized expectations about what will come next in a score and what actually does. Skill in composition then rests with the composer's ability to manipulate this tension of expectation which is general for most of mankind. Music which satisfies man's every ex- pectation is, in this theory, highly re- dundant and boring to the musical gour- met. It is from the uncertainty of music and its probabilistic character that man draws his pleasure.45 Great composers, like Beethoven and Mozart, achieve simultaneously a maximum of uncer- tainty or surprise and a maximum of 42 Frye makes many incisive comments on this prob- lem in his "The Road to Excess" in Myth and Symbol: Critical Approaches and Applications (Lincoln: Univ. of Nebraska Pr., 1963 ), pp. 3-38. .a Northrop Frye, Anatomy of Criticism: Four Essays (Princeton, N.J.: Princeton Univ. Pr., 1957), pp. 341- 42. 44 See Cassirer, An Essay on Man: An Introduction to a Philosophy of Human Culture (New Haven, Conn.: Yale Univ. Pr., 1944), pp. 67-68, 70-71, 222. • 5 There are some parallels here with the measures of information transmission in information theory. fulfilled expectations.46 To the continui- ties and discontinuities of the music, man brings his generalized aesthetic ten- sion. From the foregoing it can be seen that both literature and music, though by and large in the stage of taxonomy or naive induction, are capable of sus- taining conceptual frameworks which articulate the parts of the whole. Within this perspective, chronological classifica- tion or grouping by form seem primitive indeed as organizing principles in litera- ture and music. Frye has contended, as a matter of fact, that ultimately litera- ture, like mathematics, has a self-con- tained form and a system of postulates which require no empirical referents. He has even wondered, though partially dis- carding the notion, whether the same formative structures which recur in the history of literature are operative in other fields which are dependent on verbal structures, like metaphysics, the- ology, history, and law.47 At all events, in literature and music, indeed in the humanities generally, there are scholars devoted to Matthew Arnold's precept of "letting the mind play freely around a subject in which there has been much endeavor and little attempt at perspec- tive."48 From the point of view of the writer, the foregoing demonstrates that there are elements of structural con- tinuity in several fields of the humani- ties and that where structure persists over time, even in the midst of seeming uniqueness, the prospect for systemic outlook is radically improved. Corresponding to the archetypal pat- terns which structure the products of artistic activity are what Gerald Holton has called the "themata" of science. It is increasingly sensed in scientific quar- ters that there are elements at work in •o Leonard B. Meyer, Emotion and Meaning in Music (Chicago: Univ. of Chicago Pr., 1956), pp. 1-42. • 7 Frye, Anatomy of Criticism, pp. 350-54. 4 8 Quoted in ibid., p. 3. Implications of General Systems Theory 1369 the process of theory construction, as well as discovery and justification, which are not verbalized or made explicit but which, nevertheless, seem to possess an historic continuity.49 At least one promi- nent scientist sees the primordial formu- lations of science, which persist even to the present day, as deriving mainly from ~e unanalyzed depths of human con- sciousness and as receiving their initial embodiment in religion and philosophy. 50 As a result, so the argument goes, a large part of the history of science can be in- terpreted as a continuing dialectic be- tween those who would emphasize sta- bility and being (as atomism) and those who would emphasize becoming (reality as a continuum), positions which had already received reasonably refined for- mulation in <;reek philosophy. 51 Rene Dubos has noted that . . . the theory of dynamic equilibrium, which was the last word in biochemical sophistication when first enunciated three decades ago, is now being questioned again by a new generation of biochemists. In a very recent series of lectures an eminent biochemist suggested that life might reside in the stability and continuity of nonliving macromolecules within the cells, rather than in the transformation of components which undergo rapid turnover. According to this biochemist, in other words, the course of life might be found in the Cartesian con- cept of "being" rather than in the concept of "becoming" which had its origin in Heraclitus of Ephesus .... 52 49 Koch in Wann ( ed.), op. cit., pp. 21-22. See also Polanyi, op. cit., pp. 15-17. 50 Rene Dubos, The Dreams of Reason: Science and Utopias (New York: Columbia Univ. Pr., 1961 ), p. 123. 51 In his The Edge of Objectivity (Princeton, N.J.: Princeton Univ. Pr., 1960), Charles Gillespie con- tends that atomism (or atomicism) has been the most productive organizing structure in the history of sci- ence. See Dubos, op. cit., pp. 120, 126, 128. 62 Dubos, op. cit., p. 124. Harvey Brooks has ob- served that the stability of dynamic equilibrium, seen macroscopically, may turn out, microscopically, to be a flux of opposing motions which simply cancel each other out. See his .. Scientific Concepts and Cultural Change," Daedalus, XCIV (Winter 1965), 73. Less Jungian in tone and in general less speculative with regard to the pos- sible sources of persisting images in sci- ence is Holton's discussion of themata. 53 He begins by citing the recent discovery by Alexandre Koyre that Newton sup- pressed and failed to publish in Book III of his Principia, a fifth rule of rea- soning in philosophy which read, in part, as follows: "Whatever is not de- rived from things themselves, whether by the external senses or by internal cogitation, is to be taken for hypoth- eses .... And what neither can be demonstrated from the phenomena nor follows from them by argument based on induction, I hold as hypotheses."54 As operative in science, themata are propo- sitions which are "entirely unverifiable and unfalsifiable, yet not quite arbi- trary .... " What may be called the contingent plane of science-that is to say: ( 1) statements which are empirical- ly based; and ( 2) statements deriving from logical or mathematical processes- is thus seen as overly limiting. 55 Holton believes that any complete understand- ing of theory construction in science must take into account a thematic di- mension as supplement to the two-di- mensional contingent plane. He sug- gests, for example, that the thematic component of the concept of force is the masculine principle of potency. This principle of potency has been differently incarnated at various points in human history by scholars as diverse as Aris- 53 Holton would not go so far, for example, as Kant who saw the " categories" as prior to human experience and functioning as stereotyped intellectual equipment common to all men. See Deutsch in Bryson et al. ( eds.) op. cit., p. 58. 54 Gerald Holton, " Presuppositions in the Construc- tion of Theories," in Harry Woolf ( ed. ), Science as a Cultural Force (Baltimore: Johns Hopkins Pr., 1964) , p. 83. 55 Force, for example, can be "observed" empirically in the deflection of solid bodies as detected by a measuring instrument. The logical or mathematical dimensions of force as it operates in the contingent plane of science are to be found in the mathematical laws of vector calculus, 370 1 College & Research Libraries • September, 1966 totle, Galileo, and Newton, among oth- ers. The persistence of the potency themata can be seen in the early failure of seventeenth-century natural philoso- phers to distinguish between the Aris- totelean notion of force as that which produces motion or velocity and the Galilean concept of force as that which produces a change in velocity or accel- eration, that is to say, a ball to which one imparts motion will continue to roll along a reasonably friction-free hori- zontal plane even after one has ceased to impart motion to it. 56 Other themata are those of ( 1) conservation of energy; ( 2) atomism; ( 3) the probabilistic char- acter of behavior; and ( 4) the behavior of things as a consequence of their shape (for example, the earth rotates because of its near sphericity). A striking exam- ple of themata in action is provided by the contemporary controversy between the evolutionary a11d steady state theo- ries of cosmology and the more ancient Milesian and Parmenidesian cosmolo- gies. The fission of primeval matter and its later fusion in Milesian cosmology has its echo in the expansion-contraction cycles of the more modern pulsating universe of the evolutionary theory. And the steady state hypothesis may, of course, be equated with the ideas of Parmenides. In general, themata are ascendant or obscure in various intel- lectual epochs because of the seeming relevance of their power to structure reality. With the rise and fall of scientific fashions, they are utilized or put into storage but are never really disposed of or discarded with finality. 57 It may perhaps be argued that the foregoing extended discussion of struc- tural continua in the history of the sci- ences and the humanities is overly su- perficial and that any investigator seek- ing to identify historical discontinuities would be, at least, equally rewarded. That may be so, and yet one cannot fail 56 See Northrop, op. cit., pp. 22-34. 57 Holton in Wool£ ( ed. ), Science as a Cultural Force, p. 101. to be somewhat impressed by certain functional parallelisms between themata and archetypes. Indeed, a reasonable ex- tension of general systems theory, how- ever quasi-scientific, might be to seek ismorphism between archetypes and themata as they are employed respective- ly, to organize the reality of the human- ist and the scientist. At all events, the continuities represented by such forma- tive elements might well provide a use- ful historical adjunct to the emerging recognition of general theories of sys- tems. At this point in the course of the paper, it may be appropriate to con- sider the attempts which have been made to view knowledge as a structural unity, particularly as such integrative struc- tures have been supportive of general education in colleges and universities. Historically, efforts to view knowledge holistically begin with the philosophers of classical Greece and run all the way to the general systems theorists of the twentieth century. In the Sophist, Plato directly concerned himself with knowl- edge in its unity and diversity. And the seventeenth and eighteenth centuries were especially characterized by efforts to see knowledge whole. Descartes' para- digm for the unification of knowledge, as explained in his Discourse on Method, was a primitive reduction of all knowl- edge to the relationships at that time, seen as inherent in geometry. Francis Bacon, in his N ovum Organum, sought to unify knowledge through limiting it to the output of those areas of inquiry which were amenable to the application of natural philosophy. Leibnitz and the French Encyclopedists deplored as arti- ficial the arbitrary division of knowledge into subject matters, but differed in their plans for the encyclopedic organi- zation of knowledge. The former, in his projected encyclopedia, would have made the logical (or synthetic-theoretic) ordering of knowledge primary and would have employed an index to pro- vide classification by subject. Diderot Implications of General Systems Theory I 371 and d' Alembert, the French Encyclope- dists, made classification by subject pri- mary and utilized a cross-reference ar- rangement to demonstrate horizontal connections. Kant, in radical disagree- ment with his predecessors, saw each subject matter as a self-contained logical system whose place in the total scheme of things was defined a priori by knowl- edge as an organic whole. For Descartes and Leibnitz, knowledge was simply the sum of its constituents. For Kant, all knowledge was a system and the dis- crete subject matters were logically self- contained subsystems whose articulation within the organic whole was defined by the system itsel£. 58 In Kant's judgment, therefore, the division of knowledge was anything but arbitrary. Throughout the course of history the unification of knowledge has been seen as a desirable goal, at times simply for its own sake, at other times for the sake of its redeem- ing social importance-that is to say, the periodic demand for knowledge codifi- cation as essential to the coordination of human and physical phenomena in the practical sphere. The efforts to unify knowledge have been formally embodied in the formation of special groups (like the Society for General Systems Re- search, and the Foundation for Integra- tive Education); in the construction of encyclopedias (like the International En- cyclopedia of Unified Science); and in the educational efforts of Comenius and the Pansophists or the formulators of the encyclopedic curricula in the eigh- teenth century universities of what is now Germany. 59 58 Caws has suggested that, vis-a-vis the discrete sciences, there is a methodological unity based on the sharing of a single logic. See op. cit., p. 336. 59 See Robert McRae's excellent discussion of efforts to unify science during the seventeenth and eighteenth centuries in his The Problem of the Unity of the Sciences: Bacon to Kant (Toronto: Univ. of Toronto Pr., 1961 ) , especially v-x, 3-23. See also William O'Meara, "Observation, Interpretation, and Integra- , tion," in F. Champion Ward (ed.), The Idea and Practice of General Education (Chicago: U niv. of Chicago Pr., 1950), pp. 235-37, where the strategies employed in attempts to integrate knowledge are classified as encyclopedic, theoretical, and practical. During the two decades preceding 1950, physics (or better, physical state- ments which had been purged of all metaphysical content) was seen as the prime discipline to which all other scien- tific disciplines could be reduced.60 The same kind of disciplinary ethnocentrism which had earlier made theology and geometry the measure of all knowledge now made physics the keystone of the twentieth-century edifice of knowledge. 61 Of late, however, the notion of emerg- ence, which is the inverse of reduction, has been popular. Like reduction, emergence implies the idea of hierar- chical order, but the focus is on the as- cendant rather than the descendant na- ture of vertical order. Caws has pro- vided this description of the proces~ of emergence: We say that one science is emergent with respect to another if there is some event described and explained by the former, whose parts are described and explained by the latter, such that no combination of the explanations of the parts as described by the latter can be made to yield an accepta- ble explanation of the whole as described by the former. Psychology would, accord- ing to this criterion, be emergent with re- spect to physiology if no set of explanations of brain structures could be made to yield an acceptable explanation of, for example, 6 0 Described by Rudolf Camap as quantitative de- scription of space-time-place. See Joergen Joergensen, The D evelopment of Logical Empiricism (International Encyclopedia of Unified Science, Vol. II, No. 9 [Chi- cago: University of Chicago Press , 1951] ), pp. 81-82. The logical empiricists, in general, tried to avoid the ambiguity and metaphysical pregnance of natural ob- servational language. P. W. Bridgman, for example, developed the notion of operations in which he did not deny the possible existence of that which could not be subjected to the operations of measurement, but did deny the scientific validity of concepts which did not possess a corresponding set of operations- op- erations which, contrary to widespread understand- ing, could be either physical or mental. An example of the latter would be mathematics in which one de- sires to determine whether a series of magnitudes is continuous. See Caws, op. cit., pp. 319, 321-22 and Joergenson op. cit., p. 56. 61 Such reduction can be viewed in the image of a chain- from sociology to psychology to biology to chem- istry to physics. See O'Meara in Ward ( ed.), op. cit., p. 233. Caws has suggested that the sciences which formed the links in this chain of reduction had, at least in part, to have reached the nomological (or generalized empirical theory) stage of inquiry. See Caws , op. cit., p. 283. 372 1 College & Research Libraries • September, 1966 some exhibition of problem solving be- havior . . . the defender of emergence would claim that there is something about the whole event as described in the emer- gent science to which the concepts of the science from which it is emergent are not even relevant. The whole, he would say, is greater than the sum of its parts. . . . 62 The historic search for integrating principles in American higher education is interesting to review, especially in light of the efforts at knowledge unifica- tion just discussed. During a substantial part of the nineteenth century, and even before that in what William Cowley calls the classical American literary college, a course in moral philosophy, typically taught to seniors by the president of the college, was regarded as the integrative capstone of the curriculum. Such courses were ordinarily a curious melange of re- ligious orthodoxy and personal opinion. At Williams College, for example, Presi- dent Mark Hopkins, who could without exaggeration be called the worst sort of dilettante, described his course in moral philosophy as covering ··man in his unity and God in his sovereignty."63 In Cath- olic institutions of higher learning, theol- ogy, by virtue of the nobility of its ob- ject (God) and the reliability of its sources (Divine Revelation as interpret- ed within the teaching authority of the Church ), was viewed as the queenly dis- cipline to which all others were ulti- mately linked.64 In the 1930's and 1940's Robert Maynard Hutchins championed the primacy of metaphysics, not simply as the study of first principles, but of 62 Caws, op. cit., pp. 308 and 309. 63 Quoted in Rudolph, op. cit., pp. 140-41. There is considerable evidence that Hopkins, though ap- parently popular with students and alumni, was an anti-intellectual who eschewed Kant and had not even read Darwin or Huxley. He was not, in a phrase, the kind of professor that a really able student would like to have seated opposite him on a log. See Bruce Dearing, "The Myths about the College Teacher," Saturday Review, XLVII (January 1964 ), 65. 114 See Vincent E. Smith, The School Examined: An Essay on the Curriculum (Milwaukee: Bruce, 1960), pp. 70-94. Christoper Dawson, the noted historian, has more recently suggested the study of Christian civilization as an integrating principle in higher edu- cation. all that such principles imply for be- havior in human and nonhuman reality.65 Yet in the late 1940's and early 1950's, the innovating college at the University of Chicago was by no means oriented to the Great Books or even the metaphysics of classical antiquity. To be sure, philos- ophy and history did serve as the disci- plines of integration in the curriculum, but the selections from original works, syllabi of ccGreat Snippets" as they were called, the content of which was ordered by philosophical and historical princi- ples, were quite modern in origin and the medieval period was, on the whole, rather underrepresented. The objective of this curriculum was to develop in students the capacity to think critically and judge wisely. 66 There are many in- deed who view the ecOid College" at Chi- cago, with its symbol of erudition and intellectual high play-Aristotle Schwartz -as a remarkably successful experiment in higher education. Those who imple- mented the college curriculum were con- cerned that the students grasped ideas and methods and that they learned a little nl!: about the different ways in which knowledge might be structured. They were not interested in stuffing stu- dents with subject matter and low-order empirical content. 67 It is curious that in the latest reorganization of the college, adopted within the last few months, there is the suggestion that the older emphasis on acquainting students with the structure of knowledge, despite the intervening i;ncrease in low-order em- pirical knowledge, is no longer viable. 65 Robert Maynard Hutchins, The Higher Learning in America (New Haven, Conn.: Yale Univ. Pr., 1936), pp. 107-108. 86 F. Champion Ward, "Principles and Particulars in Liberal Education," in Cohen ( ed.), op. cit., pp. 122-23, 127-28. 67 Richard McKeon has given an account of the shift away from an emphasis on discipline and method, which prevailed in the medieval period, to a concern, in the Renaissance, with the facts and data of individual subject matters. One of the fundamental objectives of the pre-1953 college curriculum at the University of Chicago was to sensitize the student to how bits and pieces of knowledge might be articu- lated in knowledge viewed as a seamless, but com- plex, whole. See McKeon in Cohen ( ed.), op. cit., p. 168. Implications of General Systems Theory I 373 Apparently the older ideal will survive only in the so-called "collegiate division" (or Division of Integrated Studies), one of the five divisions in which a student may now concentrate.6s Prior to 1942, the integration course in the college of Chicago was called .. Methods, Values, and Concepts." After that year, it was known as Observation, Interpretation, and Integration and aimed at giving the student an "oppor- tunity to acquire the knowledge and ca- pacity necessary to work out for himself an intelligent theory on the interrelation- ships of the fields of knowledge."69 A similar course, called .. Organization, Methods, and Principles of Knowledge," still exists at Shimer College, a small experimental institution located at Mount Carroll, Illinois, which adopted the Chi- cago College Plan of general education in the early 1950's and has promoted its evolution. 70 In the main, however, the various concepts of general education which were formulated earlier in the century are encountering rough sled- ding. Substantive innovations like the sequence of courses in contemporary civ- ilization adopted by Columbia College in 1919, the broad humanities courses developed at Reed College in 1921, and the earlier attempt at Chicago to formu- late a course on the Nature of the World and Man ( 1924) 71 have, it appears, few imaginative parallels in the present pe- riod. It is true, of course, that in portions of the country-for example, New Eng- land with its Goddard and Marlboro- there are colleges which continue to em- 68 The others are Physical Science, Biological Science, Social Science, and the Humanities. See Wayne C. Booth, " The New College," The University of Chicago Magazine, LVIII (November 1965), 8-10. It is freely admitted that a major rationale of the reorganization was to link the college more closely to the interests and commitments of the graduate divisions and to check the flow of faculty away from a commitment to undergraduate teaching and towards an interest in the discipline-oriented research and graduate teaching of the subject departments. 69 O'Meara in Ward ( ed.), op. cit., p. 234. 1o See the Shimer College Catalogue, 1964-1965 (Mount Carroll, Illinois: Shimer College, December 1963), pp. 3, 39. 7 1 See Rudolph, op. cit., pp. 69, 76. phasize those principles upon which in- stitutions . like Bennington and Sarah Lawrence were founded earlier in the century, namely, Freudian psychology and the educational philosophy of Dew- ey. It is also true that there are new colleges being based on technological ·innovations, or on what this writer would call secondary innovations when contrasted with a primary innovation like the restructuring of knowledge- even though, for example, the construc- tion of programed educational materials does result in the rethinking of how sub- ject material can be most adequately presented. There is also the call for col- leges with a concern for their impact on the values of students and with the pos- ing of relevant sets of value alternatives for student choice. 72 There are even colleges, like Shimer and Monteith Col- lege of Wayne State University, which are devoted to constant review of the way in which knowledge is organized for transmission to students. By and large, however, there is relatively little effort in the universe of higher education at or- ganizing knowledge within frameworks which are fundamentally new, and it almost seems as if those engaged in high- er learning are waiting for a final theory of learning based on neurophysiology or an epistemologically perfect conceptua- lization of "reality." Perhaps this assess- ment of things as they now stand is over- drawn and unfair, however, since there are many factors constraining the de- velopment and implementation of schemes of general education. 73 72 Such a concern has been reflected recently in the presidential remarks of both Nathan Pusey of Harvard and John Sloan Dickey of Dartmouth. 7 3 Chicago of course has substantially modified its pre-1953 program of prescribed studies in general education and now permits students to elect major concentrations. Columbia College has also, in recent years, cut back on its two-year sequence of integrated social science studies called "Contemporary Civiliza- tion." Both actions, while obviously motivated by a multiplicity of factors, suggest that the prescription of general content for all may be less viable as a strategy than to modify the presentation of general content so as to fit the cognitive structure of the individual learner. 374 1 College & Research Libraries • September:~l966 Not the least of the constraints ef- fecting general education at any level of higher education are the subject organi- zation and narrow emphases of graduate study. Scholars who develop a general orientation do so, one suspects, almost in spite of what they have been exposed to, with minor exceptions, during their upper division undergraduate and grad- uate preparation. In addition, there is the widespread assumption that general education is being squeezed out by the percolation of specialized interests from the graduate schools standing above and the appropriation of its very content by the high schools positioned below. As the recent dialogue between David Tru- man and Jacques Barzun revealed, how- ever, there is by no means general agree- ment on the nature of this squeeze-in- deed, on whether or not any squeeze ac- tually exists. 74 If Freud, Dostoevsky, and Marx were traditionally read in the good colleges during a student's lower division years, it is not now precisely clear what having "had" these authors in high school really means. It may, of course, mean that the high school student has genuinely grappled with the ideas con- tained in those words. It may also mean, on the contrary, that the acquaintance which he developed with these authors was superficial and that he has no sense of the importance of their writings in the larger context of knowledge. With respect to the articulated structure of education and the ways in which higher 74 David Riesman, reviewing a report on Princeton men made by Roy Heath in the mid-1950's, noted that those students who prepared senior honors theses severely limited the scope of their topics "in that mysterious way in which a college senior will say of himself, 'I'm a biophysicist with an interest in the morphology of the cell,' or 'I'm in eighteenth-century literature! " See Roy Heath, The Reasonable Adven- turer: A Study of the Development of Thirty-six Un- dergraduates at Princeton (Pittsburgh: Univ. of Pitts- burgh Pr., 1964), xiii. Robert Wolff has even more recently observed that upper division undergraduates are, in many of the strong autonomous and university- based liberal arts colleges, taking courses which are in reality baby graduate courses and writing senior honors theses which are in effect baby dissertations. See his "The Race to College," Atlantic, CCXVI (November 1965), 147. levels influence lower levels, Robert Wolff comments: Once more, the education of the present, for which the student gave up so much in high school, is sacrificed to the demands of the future .... Each present was sacri- ficed to the future, until the presents were all pasts, and the future an empty present. It is a familiar enough story in our society. We call it prudence, or deferral of gratifi- cation, depending on our tastes in moral discourse. 75 Commitment to general education, or the view that knowledge should be looked at as a coherent and integrated whole, appears to provide limited pay- offs-at least on the surface-for faculty members as well as students. After all, most institutions of higher education are discipline oriented, and rewards accrue to both students and faculty on the basis of their intradisciplinary performance. The marginality of general education, particularly with respect to the ongoing educational activities of the large uni- versities, lies in its very definition as an approach which is nonspecialized, in- deed those who espouse it often have a genuine aversion to academic speciali- zation and its ultimate usefulness. 76 It cannot be denied, of course, that gen- eral education attracts the best and worst kinds of men, those who are capable of transcending specialization and those who, failing to make the grade in a par- ticular discipline, try it as an alterna- tive.77 It is interesting to observe, never- theless, that general education, in partial eclipse at the undergraduate level, is beginning to make inroads at the gradu- ate level. At Chicago, for example, a number of those faculty members who were active in the "Old College" as pro- fessors or students now serve on one or 7 5 Wolff, op. cit., pp. 147-48. 76 Joseph R. Gusfield, "General Education as a Career: A Sociological Analysis," Journal of General Education, X (January 1957), 38. 77 See Joseph Gusfield and David Riesman, " Faculty Culture and Academic Careers: Some Sources of In- novation in Higher Education,'' Sociology of Education, XXXVII (Summer 1964), 305. Implications of General Systems Theory I 375 more of the three interdisciplinary com- mittees which award doctorates, one might .almost say, in general education- the Committee on Social Thought ( orig- inally founded in the early 1940's with Hutchins' active assistance, as the Com- mittee on Interrelation), the Committee on the Analysis of Ideas and the Study of Methods, and the Committee on the History of Culture. The program of the Committee on the Analysis of Ideas and the Study of Methods resembles, in its objectives, the .. Old College" course on Observation, Interpretation, and In- tegration; indeed, Richard McKeon, who taught that course, is the chairman of the committee. The graduate program in the History of Ideas at Brandeis Uni- versity parallels the Chicago efforts in its attempt «to examine the interrelations of ideas in the various disciplines, the interconnections between theoretical .and practical activities, and the reciprocal in- fluence of ideas and historical events."78 The existence of these programs at the graduate level and the anticipated de- velopment of similar activities in other universities would seem to augur well for undergraduate programs of general education which need a steady supply of generalists. Two problems which are likely to bear upon efforts to reorganize knowledge are: ( 1 ) investigations into the nature of the relationship between the observer and what is observed in the natural and social sciences; and ( 2) investigations into the correlation of concepts, as men- tal constructs, with their neurophysio- logical bases. With regard to the former, it can be suggested that the mechanism which science has developed to cope with uncertainty in the subatomic realin -that is to say, the forecasting of aver- ages in ensembles-will play an increas- ing role in the social sciences whose ob- jects of inquiry-individuals, individuals 78 See the Bulletin of the Graduate School of Arts and Sciences, Brandeis University, 1965/196 ; (Walt- ham, Mass.: Brandeis Univ., 1965), pp. 108-18. in groups, and groups themselves-are also subject to behavioral deflection by the very act of observation. 79 It is as if the grossest observations of the social scientist affect those macro systems which he investigates in the same way that the finite quanta of energy which physicists employ when studying the microcosmos affect the behavior of ele- mentary particles. The observations of both the social scientists and the physi- cist are, in deterministic perspective, what Henry Margenau calls .. epistemo- logically destructive."80 In his judgment, the statistical devices which explain sub- atomic behavior in probabilistic terms can also be applied to the kind of free behavior observed with respect to so- cial groups. 81 The problem of predicting individual human action, even on a probabilistic basis, poses far more diffi- culty than the statistical explanations of social group behavior. While the suicide rate for a sizable population can be pre- dicted within the limits of probability, a variety of factors carry adequate pre- dictions of individual human action well beyond the capacity of existing statistical tools. 82 Nevertheless, science and social 79 Margenau, op. cit., pp. 54-55. The analogue in the social sciences of Heisenberg's principle of un- certainty (or indeterminacy) in physics is the "Haw- thorne Effect." In the late 1930's, investigators at the Hawthorne Plant of the Western Electric Company noted substantial variations in output during those periods when they were observing the production workers. Like the physicist dealing with elementary particles, the social scientist often encounters situations in which he can neither exactly know the present state of a system nor the precise nature of the cor- relation or causal relationship between the existing state of affairs and some future state of a system. The criteria for demonstrating determinism are to be found in Caws, op. cit., pp. 300-301. 8o Margenau, op. cit., pp. 62-63. Karl Popper sees the solution (or better, resolution) of such fundamental epistemological problems in Alfred Tarski's theory of objective truth as synonymous with simple compliance with the facts. See his discussion of Tarski's thesis in Coniectures and R efutations: The Growth of Scientific Knowledge (London: Routledge and Kegan Paul, 1963), pp. 223-26. 81 Margenau , op. cit., pp. 93-97. 82 Caws, op. cit., p. 303. One reason why man's knowledge of the nature of nonhuman organic and inorganic entities may always exceed that which he has of himself is due to his anti-life or anti-nature posture. Such a posture has its origin in the very process of hominization and sets man epistemologically at a distance from nonhuman nahue, thus providing, 376 1 College & Research Libraries • September, 1966 science appear to share a statistical meth- odology which permits them to fathom, in like fashion, the complexity of social and subatomic life. Another problem which has implica- tions for the organization of knowledge is the relationship between conventional or nonexperimental epistemology and the newer empirical epistemology which is based on a variety of disciplines in- cluding neurophysiology and information theory. Needless to say, physiological theories of knowing have a much briefer history than their speculative counter- parts. 83 The following remarks by some Oxford dons, attributable, it is my under- standing, to Ralph Gerard, exemplify the difficulty of precisely correlating con- cepts with the structures and processes in the central nervous system on which they are based. The dons are reputed to have seen a group of women walking just ahead and obviously plying the trade. Said one, "I see ahead of us a jam of tarts." The second picked it up, "I should describe them as a volume of Trollop's." The third rose to the occasion, "To me they are a flourish of strumpets." The fourth was still abreast of the punning and said, "No, I think we have an anthol- ogy of prose." The field of experimental epistemology it is suspected, a kind of built-in objectivity vis-a-vis such phenomena. Such objectivity is probably denied the infant and young child who perceives a basic continuity between man and nonhuman life prior to that time when the limits of his ego and nonego be- come more sharply d elimited. See Harold D. Lasswell, "The Major Trends in World Politics," in Harold D. Lasswell and Harlan Cleveland ( eds. ) , The Ethic of Power: The Interplay of R eligion, Philosophy, and Politics (New York: Harper, 1962) , p. 349. 83 Warren S. McCulloch, " A Historical Introduction to the Postulational Foundations of Experimental Epis- temology," in F. S. C. Northrop and Helen H. Liv- ingston ( eds.), Cross-Cultural Understanding: Epis- temology in Anthropology (New York: Harper, 1964), p. 183. Boulding has suggested that there may be a " Heisenberg principle" at work which will prevent in- vestigators from knowing in any deterministic kind of way the nature of the physiological structures corre- sponding to man's knowledge structures. See Boulding, op. cit., pp. 16-17. Whether this is so remains to be seen as Koch in Wann (ed.), op. cit., pp. 4-5, has suggested. received great impetus when Warren McCulloch first enunciated his postulate regarding the circular configuration of sets of neurons in the human nervous system. It is important to realize that for McCulloch the notion of neuron circuits was abstract and postulated and that within the bounds of his theory vacuum tubes or transistors might have served just as well as neurons. In this theory, circularly arranged neurons are set off through the firing of one neuron along the circuit which is in some way linked to a stimulus coming in over the sensory organs. After the initial firing, these neural rings continue to reverberate in much the same way that a circle of dominoes might continue to reverberate if the domino which originally set off the circuit by falling were stood up, along with the other fallen dominoes, before the reverberation had completed its circuit and reached again its starting point. In a neural circuit the recovery, or metabolic process in the individual neuron, must take less time than a com- plete reverberation of the circuit or the circuit will cease to reverberate. There are many such neural circuits present in the human nervous system, each ca- pable, it is hypothesized, of trapping uni- versals (like shape regardless of size) from environmental stimuli. It seems pos- sible, moreover, that neural circuits of trapped impulses corresponding to en- vironmental stimuli may be permuted and associated in the cortex and thus employed to define other, more fully abstract notions which will reverberate around their own neural circuits. Thus it is possible to postulate a cortical heir- archy containing neural circuits corre- sponding to philosophical concepts as well as inductive facts. 84 The source of concepts which are postulated by the imagination or intellect, that is to say- 84 F. S. C. Northrop, The Complexity of L egal- Ethical Experience: Studies in the Method of Norma- tive Subjects (Boston: Little, Brown, 1959), pp. 109- 10. Implications of General Systems Theory 1377 not given in immediate sense experience, is a puzzle of far greater complexity. It is my understanding that experimental epistemologists know considerably less about the neurophysiological foundations of imagined or intellected concepts than about those of concepts which have, at some point in time, been abstracted from reality by a process of induction. With regard to the latter, it is believed that the regenerative activity in a neural circuit permits a trapped stimulus to reverberate initially for about half an hour. At the end of that time, if the re- verberation has not been interrupted, the neurons along the circuit seem to contain an increased amount of ribonu- cleic acid and protein synthesis has got- ten under way. This activity results in local changes in the electrical character- istics of the neurons themselves and leads, in some way, to the anatomical embodiment of the impulse trapped in the reverberating circuit. 85 Child psy- chologists see . this ability to trap univer- sals as developing in humans between the ages of three and eleven. From the third year on, for instance, children de- velop the capacity of "conservation," that is to say, they are able to conserve the notion of circularity despite changes in the area or color or background of the circles presented to them. Sometimes after developing the capacity to con- serve, children begin to classify objects according to their configuration and to develop the power of seriation-that is to say, the ability to order objects hier- archically in terms of the increasing or decreasing value of some attribute. The principles of classification and hierar- chical ordering (or grouping and seria- tion) are, in the judgment of Jean Pia get and his colleagues in genetic psychology 85 See McCulloch in Northrop and Livingston ( eds.), op. cit., pp. 191-92. Reverberation may also result in lowered resistance along a given circuit which can be easily reactivated by fresh perceptual stimulus (that is to say, recognition after long memory) or even by internal stimulus in the absence of perception (that is to say, in response to representation). at the University of Geneva, the basi for concept formation in adult life.86 The experimental epistemology which has developed up to now makes no claim to full physiological realism. 87 But the theoretical model developed by McCul- . loch does suggest that man does not bring a tabula rasa or Lockean blank tablet to the data provided him by his senses. And the notion of circular rather than linear ordering of neurons pro- vides excellent support for this sugges- tion. In terms of concept formation and retention, there is a world of difference between closed loops and linear trains. The McCulloch theory also accords with Dean Wooldridge's hypothesis of atten- tion-focusing mechanisms which not only select out those circuits which will be permitted to reverberate for the full half hour but also those which will be em- bodied in the brain over the long term. 88 At all events, it seems that man brings to his experience on an a priori structure rooted in the nervous system, through 88 Michael Wallach has described the child's ability to conserve as "the understanding that no change has occurred regarding one or more aspects of an object or a relationship, despite change in other perceivable features." See his "Research on Children's Thinking," in Harold W. Stevenson ( ed. ) , Child Psychology ( The Sixty-second Yearbook of the National Society for the Study of Education [Chicago: Univ. of Chicago Pr., 1963] ), p. 246, also pp. 255, 259, 262, 264. 87 Neurophysiological models of brain functioning can still be constructed, for, as McCulloch has put it, with the birth of cybernetics "teleology had its proper mechanistic base in engineering and biology." See McCulloch, in Northrop, op. cit., pp 185-86. 88 Dean Wooldridge has hypothesized a three-stage concept of memory: ( 1 ) The Sensory Input Stage (in which the datum of the senses is sorted for special attention and reinforcement by a focusing mechanism ) ; ( 2) The Medium Term Memory Stage (corresponding to McCulloch's notion of reverberating circuits of neurons); and ( 3) The Permanent Memory Stage (in which a reverberating neural circuit is incorporated into the memory system, it is suggested, via some kind of change in the efficiency of the synapses which connect the neurons in the circuit involved in the final storage of a memory trace. W. Ritchie Russell has proposed that the low resistance neural circuits where a memory has already been established are constantly reinforced by the action of circuits which have been randomly generated through the occasional firing of neurons in the absence of some external stimulus. This means that the memory traces which have been stored in the storage mechanism of the brain grow progres- sively stronger over time through the action of these randomly generated circuits. See Dean Wooldridge, The Machinery of the Brain (New York: McGraw- Hill, 1964). 378 1 College & Research Libraries • September, 1966 which incoming data is interpreted in much the same way that a program in- structs a computer to perform certain operations on input data. This suggests furth er that the brain has a mechanical causality which is teleological in nature, that i to say, that the brain manipulates incoming information in terms of a hier- archical complex of goals and subgoals. Donald MacKay has speculated that the term in which a given individual per- ceive some aspect of the empirical world will depend, in the final analysis, on th e extent to which such a perception is supportive, nonsupportive, or capable of modifying the individual's hierarchy of goals. In the «internal matching" of stimulus .and goal complex, the stimu- lus which generates no response in the goal apparatus will simply :riot be per- ceived or conceived in any meaningful way. These goal complexes, originally developed in response to what MacKay calls .. patterns of demand" which the world presents to the organism, serve as organizing systems. Complex structures are thus perceived in terms of the in- ternal organizers which previous goal seeking has developed in the individual. It is possible, MacKay suggests, for in- dividuals to have «epistemological blind spots" in which no complex of organizers defined by the individual's goal com- plex corresponds to incoming sensory in- formation. In other words, conceptual in- novation at certain levels is out of the question for some individuals who-in order to react meaningfully to certain concepts-would need to demolish and recon truct their go.al complexes, a task of great formidability. 89 Herbert Simon and Allen Newell have sought to under- stand the operation of the organizers specified by goal complexes independent- ly of the biology of the brain, by inferring from the behavioral response to a known input the kind of information processing that took place in the brain. In this way, 89 See Donald M. MacKay, "Communication and Meaning-A Functional Approach," in Northrop and Livingston (eds.), op. cit., pp. 168-70. a computer program can be developed which will permit a simulation of the problem-solving processes in a sopho- more's mind. 90 Related to ongoing research in the area of human cognition is the recent work of the experimental psychologist, J. C. R. Licklider, on the design of an advanced system which would intervene between a given corpus of recorded in- formation and the cognitive structure or map which an individual seeking knowl- edge brings with him to the interaction with a body of knowledge. The thesis seems to be that it is easier to organize information for transmission to indi- viduals with unique cognitive structures than to ·attempt to reorient cognitive structures to a statically organized body of information. Such a system, called by Licklider a procognitive system, would permit a process of negotiation between the cognitive structure of the informa- tion seeker, as target, and a body of in- formation with a given initial organiza- tion. In all likelihood, such a system would consist of some linkage between a computer with a very large random access storage capacity91 and a teaching machine and, of course, the sources of knowledge generation. 92 Unlike conven- tional library and information systems, the units of discourse in this design would be conceptual and factual knowl- edge, not the physical artifacts or docu- ments within whose bounds knowledge is constrained and arbitrarily organized only in terms of the cognitive system of the author and those who are on the 90 Herbert A. Simon and Allen ewell, "Information Processing in Computer and Man," American Scientist, LIT ( 1964), 281-82. 91 Serial access computer memories have storage capacities which are roughly one thousand times great- er than random access memories. In terins of access speed, however, random access memories are about ten thousand times faster than serial access memories. 02 The console of the procognitive system would, therefore, have buttons labeled respectively "Where am I?" and "What should I do next?" These buttons would enable the knowledge seeker to negotiate a rapprochement with the body of recorded knowledge through the conceptual good offices of the interme- diate procognitive system itself. See J. C. R. Licklider, Libraries of the Future (Cambridge: MIT Press, 1965), p. 127. r . Implications of General Systems Theory 1 319 same wavelength with him.93 Defining solid information as that which excludes popularizations, ephemera, and unquali- fied contributions, Licklider has predict- ed that the increasing size of high speed random access computer memories by 1985 will permit, · despite prodigious growth in the body of recorded knowl- edge, a procognitive system user to inter- act simultaneously with the solid litera- ture of a subfield of science and tech- nology, with a scientific or technical dis- cipline by 1988 or 1989, and with all of science and technology by 1996.94 This assumes, of course, that an adequate mechanism for knowledge organization can be developed with sufficient versa- tility by those dates. It is interesting to note, however, that a subcommittee of the Federal Council for Science and 93 Compare, for example, the imaginative specula- tions of Licklider, ibid. , pp. 21-69, with the less sophis- ticated extrapolations of John Kemeny in his " A Li- brary for 2000 A.D. ," in Martin Greenberger ( ed.), Management and the Computer of the Future (New York: MIT Press and Wiley, 1962) , pp. 134-177, where the library is seen as continuing to contain mainly physical artifacts (like books and periodicals) ; where the library is conventionally classified; and where information access is assured, by and large, through refinements in presently existing technology. It certainly can be speculated that scientists operating on the leading edges of theoretical research prefer informal or nonbibliographical channels of commu- nication for r easons other than the time lag between the germination of an idea for an article in the mind of a scholar and the formal entry into the public domain of science via the published article ultimately embodying the idea. The feedback incorporated in interpersonal interaction ( whether orally or through correspondence) cannot be duplicated in the inter- action between a scholar and the formal publications of a colleague. Scholars are likely, for example, to be more speculative, to risk more, and to be less circum- spect with regard to the dimensions of their thought in an informal situation. That is why there is a need to formalize the informal means of communication without, at the same time, eliminating the informal which makes these channels so viable. Russel Ackoff has in mind, as I understand it, a project which would formalize certain aspects of the intellectual interaction in the "invisible colleges" identified by Derek DeS. Price in his Little Science, Big Science (New York: Columbia Univ. Pr., 1963). See also the discussions of informal communication in William D. Garvey and Belver C. Griffith, ' 'Informal Channels of Communi- cation in the Behavioral Sciences : Their Relevance in the Structuring of Formal or Bibliographic Commu- nication," in Bergen ( ed.), op. cit. and Richard H. Orr and Associates, "Communication Problems in Biomedical Research: Report of a Study," Proceedings of the Federation of American Societies for Experi- mental Biology, XXIII (September-October 1964) , 1117-32. 9 4 See Lickli4er, op. cit., pp. 13-20. Technology, known as the Committee on Scientific and Technical Information, is planning a national information system which embodies two distinct but related subsystems. There is to be a library sub- system (consisting of a complex of li- braries) which is document-oriented, and an information subsystem (directed to the evaluation, storage, and retrieval of information per se). The library sys- tem would satisfy the requirements of those who seek humanistic and some forms of social science knowledge and who are interested in nonscientific ideas. The information system would provide natural scientists and most social scien- tists with conceptual .and factual infor- mation.95 Another development of in- terest in the organization of knowledge (as recorded knowledge) is the SYNTOL ( Syntagmatic Organization Language), a general system developed by Jean Gardin in France between 1960 and 1962 and currently applied to a set of brief abstracts in the field of cultural anthropology. In this system, there is a paradigmatic organization in which relations are established a priori be- tween index terms which have been de- veloped out of a general surveillance of the fields involved, but which are not de- rived from any specific document. The strategy employed in developing the paradigmatic dimension of the system is not unlike that used in constructing an a priori classification scheme. The syn- tagmatic dimension of the system, unlike the paradigmatic, is a posteriori-that is to say, based on relations between terms which have actually been found in the documents of the fields involved.96 The 9 5 William T . Knox, " The Changing Role of Li- braries," ALA Bulletin, LIX (September 1965 ), 720, 724. us As I understand it, the syntagmatic dimension of SYNTOL bears at least some resemblance to the con- cept of associative indexing suggested by L a uren Doyle in his "Semantic Road Maps for Litera ture Searchers," Journal of the Association for Computing Machinery, VIII ( 1961 ), 553-78. It also seems r elated to the concept of faceted classification favored b y the Classification Research Group in England and by the Indian classificationist Ranganathan. 3801 College & Research Libraries • September, 1966 SYNTOL is, therefore, both precoordi- nate and postcoordinate in character. My understanding of this system is limited and imperfect, but it is supposed to be capable of concept coordination. 97 In discussing the system, Victor Yngve has foreseen the possibility of revising, re- organizing, and updating the paradig- matic component on the basis of the on- going development of the syntagmatic component, linked as it is with the on- going development of knowledge as in- corporated in actual documents. Draw- ing an analogy between the paradigmatic dimension and innate knowledge and the syntagmatic dimension and acquired knowledge, he suggests that "one can imagine a learning method by which syntagma that are collected-acquired from new documents, in this case docu- ment representations or surrogates- would be abstractly incorporated in the paradigmatic, innate structure, subject to certain conditions of acceptability. In this way the