Metalibrarianship. Ch.8: Theories About Knowledge

Metalibrarianship. Ch.8: Theories About Knowledge Nitecki, Joseph Z. 1993. Metalibrarianship : A Model For Intellectual Foundations of Library Information Science. http://twu.edu/library/Nitecki/Metalibrarianship .Volume 1 of The Nitecki Trilogy .Also available as ERIC ED363 346.

CHAPTER 8:
THEORIES ABOUT KNOWLEDGE
8.1 Theories about theory
8.1.1 Introduction

Shaughnessy explains the lack of interest in theory-building by contending that researchers overemphasize form, order, and structure of knowledge at the expense of studying the relationships between the organization of knowledge as knowledge, and as knowledge graphically recorded. His comment that "the knowledge base of librarianship could exist as a field of study whether or not libraries exist"2 reinforces the very basic hypothesis of metalibrarianship as a meta-theory extending b eyond the traditional paradigms of library and information science.

In LIS most of the theoretical explanations offered up until now were borrowed from other disciplines, often developed for different purposes. "We are unaware"-- Boyce and Kraft maintains -- "of a single theory in information science that suggests a testable phenomenon whose successful observation would add to its credence."3

It may be of help to some readers to clarify the distinctions between princip les, laws, theories, systems, and models in terms of their relations to reality. Definitional boundaries between these concepts are vague and are interpreted differently in library literature.

8.1.1.1

A principle is an axiom describing the essence of a given phenomenon. Popper defines it as a "simple, new, and powerful unifying idea about some connection or relation [such as gravitation attraction] between hitherto unconnected things [such as planets and apples] or facts [such as inerti al and gravitational mass] or new theoretical entities [such as field and particles]."4

Ranganathan's five laws are in effect the basic precepts or principles of librarianship, and so are Thompson's seventeen principles. Bierbaum recommends a modified Zipf's principle of Least Effort as a unifying concept for the theory and practice of library and information science. The author links it with the Mooers' Law: "An information retrieval system will tend not to be used whenever it is more painful and troublesome for a customer to have information than for him not to have it!"5 Other examples of library principles include pronouncements such as Ranganathan's 'Save the time of the reader,' or Cutter's 'convenience of the reader.' Bierbaum wonders about the validity of some such principles: Will Least Effort be turned upon itself as the value of information exceeds the cost of the search or exchange?"6

8.1.1.2

Laws are general sta tements about selected aspects of reality. They are experimentally established and operationally defined explanations of facts that describe reality.

Bibliometrics laws in information science are the examples. Bradford's law (1934) describes the productivity of journals determined by a law of diminishing returns. Lotka's law (1926) states that a small number of scientists publish most of the scholarly articles. Zipf's law (1935) provides a formula for determining the frequency of selecting particular words. As Bookstein demonstrated, all these laws are mathematically identical, suggesting "that one reason for the recurrence of these laws is that they are very stable and likely to result from a wide range of different causes."7

Most of library principles were formulated in library science, while library laws were developed in information science. Both principles and laws are usually incorporated in models.

8.1.1.3 In theory, knowledge of a particular aspect of reality is systematically organized around some selected principles and laws. In contrast to experimental or operationally defined laws, in theories at least one term does not refer to observable characteristics and cannot be directly tested. Theoretical statements are always subject to revision if they are not consistent with previously determined laws. The criteria of theory are their coherence, simplicity, and plausibility. As a rule they are more general than laws, covering qualitative aspects of ph enomena, and they refer to ideal rather than concrete situations. Theories are frequently formulated as systems of statements.

There are three major philosophical views on theory: (1) reductionism, focusing on observable sense-data, (2) instrumentalism, considering theory as a tool for deriving predictions from given data, and (3) realism, formulating theories in terms of 'real' or existing entities.

8.1.1.4 Models describe in a variety of ways different patterns of structures, reflecting different aspects of reality.8 Hesse identified five major types of definitions of 'models,' all "essential for prediction and for giving semantic interpretation to theories."9

(1) Logical models consist of entities that interpret sets of axioms, as, e.g., points and straight lines in geometry reproduce the relationships specified by the axioms.
(2) Models of analogy are either structural or isomorphic comparisons between different kinds o f systems, or replication of the parent system. The former analogies are conceptual (e.g., between a swinging pendulum and an oscillating electrical circuit), and the latter express material analogies (e.g., a mechanical part and its copy).
(3) Mathematical probabilistic models provide causal, predictive, and explanatory interpretation of formal systems (e.g., system theory in information science developed as a non-empirical, mathematical concept).
(4) Simplifying models purposefully rep resent an empirical situation in a simplified form to facilitate a research of its application (e.g., models based on statistical approximations).
(5) Theoretical models may be identical with the theory they represent (e.g., a DNA model identical with the theory of the molecular structure). They are used in place of theories that are not yet well established.

8.1.2 Systems

8.1.2.1 Definitions

Library Information Science, as any other science, is a system. In gener al, a system is defined as "a regular or orderly arrangement of components or parts in a connected and interrelated series or whole."10

We isolate certain aspects of reality for a closer examination from a particular viewpoint of the relationships within the selected system. "The concept of the system is such that it is possible under any circumstances to stay within the limits of knowledge. It is not required to know the limits but it is required to stay within the limits."11

Hall and Fagen suggest the following definition of 'system' that is free of colloquial and misleading meanings: "A system is a set of objects together with relationships between the objects and between their attributes." In this explanation (a) 'objects' are physical or abstract parts of the system (e.g., from physical 'atoms' to abstract 'variables'; (b) attributes are the properties of these objects (e.g., atomic weight); and (c) relationships "tie the system together" (e.g., interr elations between parts of the object).12

Mortazavian distinguishes between "philosophy of systems -- a branch of philosophy dealing with problems centered around the notion of system . . . [and] system theory, which is the science of systems."13

Philosophically, systems theory relates to the complexity created by the relationships between the whole system and its parts. Holists maintain that "we need to study the whole system to see the big picture ,"14 while methodological individualists insist that "the properties of the whole can be deduced from the properties of the parts."15

Another distinction within systems theories, pointed out by Langlois, refers to the causal and goal-oriented approaches. "In the former case, the meaning of a signal is the response it elicits. In the latter, response is also the ultimate criterion of meaning, even if we cannot necessarily understand the meaning of a signal witho ut first knowing the goal that the system is pursuing."16

8.1.2.2 General Systems Theory

Bertalanffy, in formulating his General Systems Theory, was influenced by a pattern of similarity observed in the development of the same or similar viewpoints in different branches of science. In the past, science followed the methodology of breaking up the phenomena into parts, studying each part independently of others. In contemporary science methodology tends to focus on the wholeness of the phenomena. Bertalanffy further noted that there are formally identical or isomorphic laws found in many different fields. They "are characterized by the fact that they hold generally for certain classes of complexes or systems, irrespective of the special kind of entities involved."17 All these considerations led him to postulate General System Theory, defined as "a logico-mathematical discipline, which is in itself purely formal, but is applicable to all scien ces concerned with systems."18

Mortazavian claims that Bertalanffy's theory failed, because it rejected mathematics as a language of science. He distinguishes it from his own definition of a mathematical system. On the other hand, Laszlo maintains that "general system theory is a general theory of systems,"19 differentiated from special system theories such as Shannon's information, Bertalanffy's biological, Wiener's cybernetic, and other systems in mathema tics, sociology, political science, management, communication, and the like. His matrix identified three basic system approaches (scientific, engineering, and philosophical), each subdivided into levels and types. System philosophy, for example, is divided into organic and supraorganic levels, biological/homo and sociocultural types, and are represented by social ethics, social, and political theory, theory of justice, human communication theory, culturology, and technology assessment.20

Philosophically, any attempt to limit general systems theory to computer-related problems would, according to Bahm (1973), restrict the theory to a two-valued system of logic only. It would exclude from the theory any implicit interrelations between systems and would not provide for unity among diverse elements within each system. Furthermore, (a) any system is limited by the capacity of the human mind, which is a part of a more complex reality; and (b) in the global sense, multicultural syste ms will require difficult reconciliation between environmental differences. For example, the Eastern philosophy of reality is based on a priority of unity over plurality while Western philosophy takes an opposite view,

Bahm notes that (a) "polarity is a universal characteristic of existence and hence a kind of structure which needs to be recognized as basic to any adequate general systems theory;" (b) since there "is an element of novelty in every cause-effect situation," its origin should be recogn ized; and (c) general systems theories must consider not only common aspects of all situations but also unique differences between them.21

8.1.2.3 General Systems and Librarianship (Foskett)

Foskett was attracted to Bertalanffy's General System Theory when searching for scientific principle that would help developing an integrative classification system. Bertalanffy formulated principles of 'wholeness' applicable to systems in general. Foskett followed by de fining a system as "a set of entities, all of which have a separate existence, and which can unite to form a single new entity of a higher order of complexity, by establishing certain fixed relations between the elements of the set."22

This approach contradicted the philosophy of reductionism, which was dominant at the time and offered explanations of all phenomena exclusively in physical terms, rejecting the concept of 'wholeness' as a metaphysical, non scientific concept. The reductionistic notion that everything can be explained in terms of abstract symbols led to equating symbols with reality (Foskett, 1974).

Libraries are open systems aiming at the organization of knowledge, constantly reacting to changes in knowledge structure and its environment. Major components of the library, Foskett notes, reflect organizational characteristics, from mechanical arrangements of words in documents to subject arrangements in a bibliography. "Above all, . . . the theory of class ification can claim a truly philosophical basis because it is concerned with the nature and organization of knowledge, a field which has always been central to philosophy." 23

Foskett summarized the major influence of Bertalanffy's theory on librarianship in three groups:
(1) the information function of a librarian is that of a generalist, able to identify for the patron all the subject similarities and differences between individual publications, and their relations t o the whole subject.
(2) The role of technology, and particularly of a computer, is to improve efficiency, but not to substitute it for intellectual functions of librarians.
(3) "The consequence of the systems' philosophical approach is that libraries should never be considered as ends in themselves, as closed systems, but as open systems in constant interaction with their environment." 24

McMahon considers library as "a system for assembling published materials, d eveloping information services, and disseminating information for use by a client and, as such, it is among the particular class of systems which is open to environmental influences and includes human interaction in the context of carrying out defined tasks."25

Library management provides a mechanism for balancing primary objectives, the interaction between different library operations demonstrates its system status, and its holistic approach allows for considering librarianship a s a total system. Bellomy looks to the systems approach for the solution of library problems. He describes a system process which involves (a) assembling interdependent data and ideas that are necessary to achieve planned objectives, and (b) considering planning perspectives, availability of needed resources, external and internal constraints, and the degree of urgency in solving given problems.26

Bergen speculates about the library role in the support of an experimental co llege curriculum based on general systems theory. The library system needed would have to be information- and document-oriented, open, with inductive and deductive capacities, that is capable of providing documents addressing general and specific apects of any issue (Bergen, 1965).

"Systems theory," Foskett concludes, "can be used to stress the importance of studying the relationships into which a library may enter; it can also help in identifying the problems of management within the librar y itself. The role of the library in any organization is the same as that of any other part of the organization, namely, to help the management to achieve its aims . . . [and] to implement the policy through the exploitation of books and other recorded data."27

8.2 Theories in Library Information Science: Examples to illustrate different kinds of theories discussed in LIS.

I selected two dissertations and one thesis, each developing or suggesting a different app roach to the concept of theory. 8.2.1

Bryson's use of the theory is instrumental, aiming at an explanation of the nature of librarianship. "In this view theories are utilized as tools of inquiry, but are not expected necessarily to describe phenomena which can be directly observed; rather, theories, for the instrumentalist, serve to guide one's description and to assist in making judgments regarding the effects of such descriptions."28

Bryson's dissertati on is based on Kaplan's (1964a) logical analysis subject to tests of correspondence, coherence, and function. The theory is postulated on three laws: (1) communication-need, (2) organizational development of efforts in preserving knowledge, and (3) division of labor reflected in management of library operations and scholarship.

Based on the above laws, Bryson derived two postulates, one related to the organizational development as emergence of social organization to design functions and utiliz e needed resources; and the other referring to the division of labor creating tasks to be performed by different assignments. Each of these postulates, in turn, can be further subdivided, into, e.g., teaching in communication, acquisition in organization, or administration in division of labor.

The relationships between the postulates and their subdivisions are represented in the form of a three-dimensional conceptual model, interrelating library functions with individual functions and informat ion need. The theory is validated in terms of correspondence (semantical norms), coherence (syntactical norms), and pragmatism (functional norms). The end result is a demonstration of a unity of librarianship, expressed by "a consistent set of information-needs . . . of library functions which have been developed to meet these needs . . . [and] . . . of individual functions which it asserts are consistent throughout librarianship."29

8.2.2

Bennett in his dis sertation proposes a theory of multiple reality "as being multidimensional, multifocused [with] . . . a multiplicity of viewpoints and horizons."30 This concept of reality is based on Bohm's viewpoint of the universe as a self-consistent whole (Bohm, 1957). Every law formulated about that reality contains errors, because it cannot include all as-yet-unknown qualities and properties of the universe, revealing new contexts and new conditions.31 "There still exists an a bsolute, unique, and objective reality. The essential character of scientific research is, then, that it moves towards the absolute by studying the relative, in its inexhaustible multiplicity and diversity." 32

Consequently, defined reality is, according to Bennett, a fiction, 'a willing suspension of disbelief,' socially constructed to represent the unknown and real universe. The multiple view of reality calls for a methodology that would allow one to move from one interpr etation to another, "from trying to understand the part to trying to understand the whole,"33 and thus increase the perception of both. Bennett selected an interpretative theory of hermeneutics, because it allows him "to move from hermeneutics (interpretation of text) to social contexts (sociology of knowledge), back to texts then to contexts of the development of scientific disciplines (sociology of science) and back to texts."34 Bennett uses hermeneutic 35 method in a form of a fictitious dialogue with well-known library theoreticians to define and explain the concepts of librarianship in all its derivations, from social positions of librarians as scientists to a historical interpretation of the discipline.36

Hayes, in his review of Bennett's essay, called this method a 'content analysis' or a 'discourse analysis' of the texts and their authors. He criticized the method as self-fulfilling, allowing the researche r to find, in the selected passages, his or her own interpretations of the reality (Hayes, 1989).

8.2.3

The main objective of Menzel's master's thesis was to review Barfield's contributions to epistemology as a possible framework for philosophy of librarianship. He endorsed Barfield's rejection of empirical science and its replacement by transcendental methodology. He also questioned the present definition of librarianship and its functions.

In his review of Barfield's philosophy of knowledge, Menzel stresses the importance of the immediacy of knowledge, which is expanded and internalized by using metaphor. "Romanticism emerged as an attempt, through imagination, to do consciously what the ancient or pre-logical mind did unconsciously."37 Barfield identified two levels of knowledge: human consciousness and its transformation into commonplace knowledge. He also divided consciousness into three levels: as an intuitive process (Plato's images), observ ation and acknowledgment of sense perception, and reflective, theoretical thinking. All three levels of consciousness are concerned with appearance: "the first with construction, the second with acceptance, and the third with explanation."38 Plato's fourth level, knowledge, was transformed by Barfield into beta-thinking, 'reflective thinking about thinking,' while Plato's highest level of cognition, intelligence, was designated as an 'advanced form of inspiration.'

Transcenden tal philosophy contradicts the rationalist philosophy of Descartes in its focus on pre-empirical facts and on methods of knowing rather than on the known facts. Librarianship, Menzel implies, should build its theory on the transcendental principles of Barfield's philosophy.

In his concluding chapter, Menzel criticized Shera for not developing the call for epistemological foundations of librarianship. "Shera actually fails to support his position because he does not offer anything which is reall y new . . . his support for a librarianship that is concerned with the 'management of knowledge' betrays the poverty of his position and a weakness in the field of library science as a whole."39 By implication, Menzel suggests that before claiming epistemological foundations, librarianship ought to clarify its position on the preferred epistemological viewpoint.

8.3

Explanation and Understanding of Reality Philosophical and psychological issues related to the distinction between explanation and understanding of the multiple realities are also pertinent to metalibrarianship. I again selected three essays for brief comparison.

8.3.1

In his emphasis on Bohm's concept of multiple realities, Bennett did not directly discuss another of Bohm's notions, the 'implicate order'. This concept, according to Beagle, is important because it provides a new insight into theories about libraries and information.40 Bohm di stinguishes between mechanistic order among entities which are outside each other, and the implicate order among living organisms, in which entities constitute an integral part of the whole organism. This emphasizes Bohm's "central underlying theme . . . [of] unbroken wholeness of the totality of existence as an undivided flowing movement without borders."41

Beagle illustrated the distinction between mechanistic and implicated orders by comparing it to the distinction between volu mes (physical units which are independent of each other) and titles (which are the integral parts of the collection, interrelated by subjects, citations, etc.). Hence, "because one title is represented by many volumes distributed around the world, any library can encompass a very significant portion of the totality of human knowledge."42

Beagle concludes by summarizing Bohm's hypothesis that "libraries are not some negentropic aberration from a fundamental law of cosmic disinte gration, but rather an expression of an integrative law of underlying order."43 8.3.2

According to Howard, the philosophical distinction between an explanation and understanding corresponds to the distinction between the scientific, especially positivistic, and the humanistic, in this case hermeneutic, approach to the theory of understanding. Science argues for neutrality and objectivity, minimizing the role of an individual. The humanities, on the other hand, stress the impact of individuals' purposes and intentions on the way reality is apprehended. Both viewpoints have philosophical merit, and neither should be given a monopoly of interpretation (Howard, 1982).

8.3.3

Hollnagel discusses the psychological issue of 'understanding the understanding' processes. Hermeneutics involves a dialectical procedure of repeated interactions between explanation (analysis) and comprehension (synthesis). In cognitive psychology the focus is not on the process of understanding as a whole, but on the perception of its specific instances, the analysis-by-synthesis. Here the meaning of the message is first guessed, then verified by comparing it with the message's wording.

Hollnagel identified two major differences between the two approaches. (1) The method of validation is determined in the hermeneutical approach by the degree of guessing; in the cognitive paradigm it is determined by the correctness, or the end-result of the understanding. (2) The nat ure of the processes itself is conscious in the hermeneutical paradigm and unconscious in the cognitive method of understanding. In the main, however, the two methods are more similar than different: "they essentially point to the same phenomenon in different but compatible ways."44

8.4 Theories about Reality

Most contemporary interpretations of reality are selective (Rower 1974) and fall within three major approaches, roughly corresponding to the three-dimensionality o f procedural, contextual, and conceptual viewpoints, discussed later in this book. (a) In stimulus-response theory, the physical and rational order accounts for patrons' motivations in seeking information. (b) Psychoanalytical theory is based on inner drives, self-gratification, and bibliotheraphy. (c) Cognitive theory offers an explanation of reality by searching for equilibrium between its physical and mental interpretations in terms of the level of individuals' 'cognitive structure,' his or her person al intellectual environment.

Usually, an individual interpretation of reality integrates all three approaches and is based on an 'if-then' proposition. It starts with metaphysical 'if' speculation about reality, which 'then' is verified empirically. The new emerging aspects of that reality are incorporated in the total individual's knowledge; hence understanding of reality is metaphorical.

8.4.1 Empirical Stimulus-response Theories

This approach explains human r esponses to physical stimuli. It is based on British empiricism's notion of mind as a product rather than as the initiator of experience; this view was expanded later by the concept of conditional responses, reinforced by rewards and punishments. Intellectual development is here considered a product of learning, determined by environment and manifested in the behavior of an individual.

The physical dimension of reality is based on the concept of rational order, which explains si milarities between past experiences among individuals, later verified by trial-and-error learning processes. This approach does not, however, account for variations in the motivation for each individual search for information, thus creating a major problem in information retrieval (Swanson, 1977, 1979, 1986) which has never been completely resolved because it does not satisfactorily account for other dimensions of reality. This was also a popular premise of the nineteenth century philosophy of libra rianship which advocated reading as a device for developing desirable behavior in citizens (Wehmeyer, 1984).

The paradigms of empirical approach include didactic logical method, scientific research focusing primarily on artificial intelligence, and information and expert systems. Its technical processes consist of classification of phenomena and falsification of theories about them. Popper's law of falsification (or refutation), expresses the preference in science for a proof of t hat which is false, rather than that which is true, or that which is bad rather than that which is good in moral philosophy (Popper, 1979). It is easier to find faults in a theory than to be sure of its absolute nonrefutable status. The theoreticians of this approach follow directly or indirectly Comte's anti-metaphysical philosophy of positivism; they include Shannon, Bradford, Harlow, and Saracevic. The models representing this approach in LIS include bibliometrics and operational research, developed by documentalists and information scientists.

Library procedural approach resembles the empirical philosophical viewpoint. It focuses on sense experienced reality of librarianship, and thus it describes the epistemological nature of the discipline.

8.4.2 Motivational Psycho-analytical Theories

This philosophy emphasizes inner drives that influence individuals' reactions to the stimuli. It points to a struggle within each individual between the pres sures of physiological stimulations and psychological, intellectually motivated responses. Here reality is perceived in terms of an individual need for self-gratification, which influences learning processes. Thus, for example, the premise of bibliotheraphy is based on the notion that the identification of one's own problems with those of others improves self-understanding (Wehmeyer, 1984).

The psycho-analytical, socio-cultural approach expands the environment of the discipline, by incorporating both the internal (within) and the external (outside of the discipline) stimuli. The discourse concentrates on what is now and what ought to be in the future. It focuses on needs for information and its content on information-related processes and communication, as well as on ethical and aesthetic issues affecting library operations.

The theories and models of this approach, interpreted as contextualism in my model, reflect a humanistic orientation, stressing the importance of r eading and book selection, and include Butler's culturalism, Shera's social epistemology (1964), and Taylor's value-added model (Taylor, 1986).

8.4.3 Rational, Cognitive Theories

This approach is based on an intellectual and deductive (i.e., rational) assertive explanation (i.e., cognitive) of reality. The approach assumes that individuals act rationally, and that thinking is the key element in understanding reality, by searching for underlying assumptions that assert, deny, or question different interpretations of reality.

The approach, called conceptualism in my model, aims at explanations of relations between stimuli and responses to them. It is a logical view, considering recorded thoughts as a key element in understanding the reality of librarianship. It describes the metaphysical meaning or essence of librarianship as a facilitation of a continuous dissemination of knowledge about reality.

The focus of this approach is on the purposes, the missio ns, and the interdisciplinary and metascientific character of the discipline. Writers reflecting this viewpoint include Kaplan, Shera, Wright, and the present author, all aiming at the development of a unified theory of librarianship. Crucial here is the way one perceives different aspects of reality in the context of its total 'cognitive structure.' Piaget defines the cognitive structure in terms of a search for an equilibrium between the old and new perceptions of reality. Intellectual development itself is a process of restructuring one's knowledge of the world by a series of changes in one's cognitive structure.

8.4.4 Continuity Principle

I suggest that a unifying principle in all interpretations of reality is the continuity of human effort to understand it. The motives for this persistent search may be a desire to remove uncertainty created by gaps in our knowledge of reality, or a need to be assured or reassured that our understanding is valid.

The continui ty of human constant need for knowledge becomes a substance of metalibrarianship.46 Human's conscious life is a constant search for answers to the unknown, the need for more data, information, or knowledge about themselves and their environment. This need resembles Johnson's 'continuant,' the substance, which is constant while its states or relations may be changing.47 The continuants referred to both physical and psychological aspects of reality, linked together by "causal unity of connection . . . between [their] temporarily or specially separated manifestations."48 That is, reality is not interpreted exclusively as a physical or mental phenomenon, but as a relationship between the two.

8.4.5 Individual Interpretations of Reality

The individual's interpretation of reality consists of several distinct steps. A starting point is a broad, informal, metaphysical speculation. An individual works with some i ntuitive hints about the relations, questioning existing knowledge about a particular issue. The speculation is then restated in broad, known physical principles, statements, or hypotheses as they relate to some observable facts. The restatements are then: (a) verified or refuted, and if accepted, formalized in physical, empirically tested terms of existing theories, (b) further expanded by anticipated new social needs and identification of some possible future consequences of the emergin g theory, and (c) incorporated into an individual's broader understanding, the metaphysical knowledge of reality.

The new formulation of reality is now expanded by identifying metaphysical consequences within the parameters of the theory. Newly acquired understanding is critically evaluated, questioning the anticipated state of understanding. And finally, the fully integrated interpretation of reality becomes the first level of metatheory. The process is repeated, sharpening a t each repetition the metatheoretical formulation of the discipline. Repetitive cycles gradually improve the specificity of the metatheory.

8.5 LIS Theories of Recorded Knowledge

8.5.1 Definitions

To illustrate the variety of model building in Library Information Science, I selected from the library literature three hypothetical approaches to the definition of knowledge as they relate to librarianship, by abstracting from the works of: (a) Kemp's essay on recorded knowledge in library collections, (b) McGarry's sociological and behavioral interpretation of relationships among communication, knowledge, and libraries, and (c) Machlup's discussion of the production of knowledge as an economic process.

8.5.1 Kemp (1976) in his introductory work on the nature and properties of knowledge and its relationships to librarianship and communication, provides a working definition of 'knowledge'. It is an expression of belief, either as a personal , individual, apperceptive knowledge, or as social, collective knowledge that is contained, for example, in library documents. The two are interdependent: "Social knowledge depends for its existence on personal knowledge; much personal knowledge is derived from social knowledge."49

A knowledge statement must be compatible with other statements considered to be true, although none expresses absolute knowledge. Knowledge "is different at different times and in different envi ronments."50 The essential property of knowledge is its organization into a coherent, and social acceptable, whole system. The function of libraries is to facilitate the communication of social knowledge through acquisition, preservation, organization, and dissemination of recorded knowledge.

8.5.1.2 McGarry (1975) reviews the relationships among knowledge, communication, and libraries as a synthesis for a foundation of librarianship. Knowledge is a system of symbolic descriptions of reality which, in turn, is defined as "a social construct, created by the collective perceptions of social man . . . man is . . . a product of his culture and his perceptions are conditioned by a common world picture built up by his ancestors."51

Communication is a mechanism for the transfer of information or knowledge; information refers to "disconnected facts or events or 'patterns of stimuli' and confines the term knowledge to a systematic and ordered s ystem."52

In this approach librarian is assigned "a new social role as 'link(person)'; the (person) with the overview, whose particular expertise is a kind of 'knowledge about knowledge'; who realizes the degree to which the major disciplines of knowledge fragment and overlap each other."53 In his later work McGarry interrelates knowledge with information and language: "both knowledge and information are embodied in language and both involve a conceptual a pparatus that is not derived from the data we receive from the phenomenal world but imposed upon them."54

8.5.1.3 In the words of one of his reviewers, Machlup's essays represent "a great garden - Machlup's Garden of Knowledge."55 In it, he identifies many definitions of knowledge analyzed as an economic activity in the processes of inquiring into knowledge production. I selected here the discussion of definitions related to types, qualities, and products of knowledge.

Machlup identifies five classes of knowledge: (1) practical e.g., professional or political; (2) intellectual, satisfying cultural curiosity; (3) pastime, recreational knowledge; (4) spiritual, religious knowledge; and (5) unwanted knowledge, accidentally acquired and of no immediate interest.56

Qualities of knowledge include the concepts of truth (or belief), beauty, and goodness. Philosophical questioning "whether knowledge is a particu lar kind of belief, whether belief is a particular kind of knowledge, or whether knowledge and belief are different things . . . can be . . . simply interpreted as a fusion of the two meanings of the term's knowledge: the known and the knowing."57

The aesthetic notion of knowledge refers either to the object of knowledge, itself beautiful (e.g., a plot of a poem), or the form in which knowledge is conveyed. Machlup illustrates this notion by citing examples from art and scienc e (e.g., a scientific proposition may be elegant, but its formulation awkward).58 He rejects the notion of 'goodness' in ethical knowledge. It may be interpreted empirically as true or false, but "neither ethical knowledge nor knowledge of ethics (the science of ethics) can be judged to be good or wicked."59 Knowledge as a product is determined by economic considerations of the consumer's choice, entrepreneurial initiative, and political decisions. The m arket mechanism determines only a part of knowledge production (e.g., in book sales and information services); most knowledge is available free. However, many business decisions depend on both the supply and demand for information. Although research is often privately initiated, it is frequently supported by government grants or direct involvement in research and development.60 And finally, "knowledge production can mean producing additional new ideas -- extending the universe of the known -- but also producing a state of knowing in additional minds -- extending the population of knowers."61 Library Information Science assists in the former and is responsible for the latter aspects of knowledge production.

8.5.2 Models

The models of knowledge discussed in this section are again arbitrarily extracted from the works of Shera, Wilson, and Dretske, none necessarily reflecting the main thrust of their essays.

8.5.2.1 Sher a acknowledged that the model of social epistemology, introduced by him and Egan, was an extension of Butler's argument for an epistemological base of librarianship. Butler maintained that knowledge about knowledge is basic to librarianship, as expressed in his concept of 'history of scholarship.' Shera agrees. In order to fulfill library responsibility for the dissemination of knowledge, Shera claims that two questions must be answered first: "What is knowledge; or more specifically, what are the charac teristics of recorded knowledge? and How it is put to work?"62

In defining knowledge, Shera follows Boulding's distinction between information and knowledge: "information is the input of knowledge . . . knowledge is everything an organism has learned or assimilated -- values as well as facts or information -- organized according to whatever concepts, images, or relations it has been able to master."63 Knowledge is disseminated through communication, with the library performing an important role in the communication network. 64 This role is not only to provide a link in a system but also to transmit knowledge.65

In the past, the epistemological studies of the nature of knowledge were individual-oriented. They bypassed the differentiation and integration of knowledge within society. To remedy the omission, Shera and Egan proposed social epistemology, or social cognition, as a framework for analyzing intellect ual processes in the society. They defined it as "a study of the ways in which society as a whole achieves a perceptive relation to its total environment . . . [with the focus] upon the production, flow, integration, and consumption of all forms of communicated thought throughout the social fabric."66

Shera considered this new approach essential in the theory of librarianship since it clarifies the role of library in the dissemination of society's records. As the management co mponent of knowledge, librarianship's goal is "to maximize the social utility of graphic records."67 This can be accomplished only if the librarians understand the role of knowledge in the society.68

8.5.2.2 Patrick Wilson extended Shera's concept of social epistemology to the study of bibliography and the issue of cognitive authority.69 He first makes a distinction between firsthand knowledge based on direct, personal experi ences, and secondhand knowledge obtained from other people's experiences. The purpose of Wilson study is to "explore not what is known about the world, but rather what people think about knowledge -- how they decide who knows what about what. This is a question about cognitive authority."70

Cognitive or epistemic authority is described by Wilson as a kind of influence a person has on another in terms of his or her recognized expertise. The authority is relative to a specific fi eld of interest. It involves credibility, based on competence and trustworthiness, and it is consciously recognized as proper by the person accepting the authority of another. Cognitive authority applies to any sphere of knowledge or opinion, not imposing on others what to think, but suggesting, or advising, on issues they are competent in.71 Wilson restates the above definition in an abstract form: "Person A is a cognitive authority for person B with respect to sphere of interest S to t he degree that what A says about questions falling within the sphere S carries weight for B."72

The cognitive authority of a librarian is to make a judgment about cognitive authority of authors and their published work. "Librarians cannot simultaneously deny competence to judge the quality of texts and assert competence to answer questions by finding the answers in books. Library reference service appears to be based on a contradiction: the simultaneous assertion and denial of competence to evaluate texts."73 This contradiction is a core of contention in the discussions of librarians' neutrality. Wilson makes a distinction between the position of a librarian (a) in a didactic library, in which the content of book collection is predetermined by the official policy, e.g., in the case of religious or specialized governmental libraries, and (b) in a liberal library which aims at a collection that includes all opinions. In the former case the librarian cannot be neutral; he or she must either have, or pretend to have, an opinion on the subjects represented in the collection. In the latter case, the librarian must be neutral, by not influencing the patron in any way. "The contrast between didactic and liberal could hardly be more extreme. In the one, cognitive authority is the dominant consideration; in the other, consumer demand is the dominant consideration."74 Wilson recommends, as a way out of the dilemma, a skeptical "attitude of one who neither asserted nor denied the possibility of knowledge, but continued to inquire, though always unsatisfied that knowledge had yet been found."75

8.5.2.3 Dretske (1981) took a diametrically opposite view, based on the materialistic metaphysical position that knowledge is absolute and identified with information-caused belief, rather than with traditional epistemological notion of knowledge as a 'form of justified true belief.' Dretske defines perceptual knowledge in terms of information and its component, belief; "we already understand what knowledge is in its use of the concept information."76 He dissociates the concept of information from that of meaning, considering communication as a process of providing information, not meaning. "There is no reason to think that every meaningful sign must carry information or, if it does, that the information it carries must be identical to its meaning."77 Although normal exchange of information involves the use of meaningful signs, the meaning of these symbols is not a synonym with the information they carry.

He defined informational content as follows:

Dretske illustrates the dependence of knowledge on available information by describing the lottery paradox: the chances of winning by someone are 100%, but almost none for a particular individual participating in the lottery. Yet no one knows whether he or she will loose, although somebody will win.79 That is, assuming that the signal r (e.g., an advertisement of the lottery) says that 'someone will win the lottery' (i.e., 's is F'), then the probability that I may win th at ticket (i.e., I am 's') cannot be high unless I know something about the way the lottery operates in selecting my ticket, which of course, is not the case in an honestly run lottery system.

The point made is that "we must accept the view (built into our characterization of knowledge) that the knowledge that s is F requires (because it is required as a cause of belief) the information that s is F."80

Dretske also discusses another distinction between perceptua l and cognitive knowledge that is relevant to our study. Perceptual knowledge refers to the sensory experience of, e.g., "seeing x'. Cognitive knowledge stands for the recognition of things seen as x. Perceptual knowledge is a cognitively neutral experience of 'seeing'. Cognitive knowledge implies a belief, an internal state of mind, affecting the way, e.g., we see x. 81

All signals contain information in both forms. The information is carried in digital form if a signal car ries only the information that is already contained in 's is F.' If the signal carries additional information about the statement, 's is F,' then that information is carried in analog form.82

Specific information is extracted from the analog form by the analog-to-digital conversion. The information in analog form that is irrelevant to a particular search is excluded, and thus lost, thus reducing its informational content. The process allows for classification, generalization, a nd recognition of a specific input as a part of a more inclusive category.83 "To summarize . . . our perceptual experience . . . is being identified with an information-carrying structure -- a structure in which information about a source is coded in analog form and made available to something like a digital converter . . . for cognitive utilization."84

8.5.3 Limitations in Structuring Knowledge

In this section I focus on two kinds of limitations: (a) between human memory and the organization of knowledge discussed by Harmon (1973), and (b) limitations created by theoretical, deductive, and empirical, inductive, components of a discipline, reviewed by Storer and Parsons (1968).

8.5.3.1 Harmon is interested in "the extent to which fields of recorded knowledge, particularly information science, have been formed, structured, and circumscribed in accordance with the necessiti es and constraints imposed by a relatively constant and limited span of human memory."85

Limited human memory lead to the use of records; the increased number of records, in turn, initiates specialization and subsequent fragmentation of knowledge into different subsystems, systems, and interdisciplinary systems of knowledge. A subsystem is defined by Harmon as a component of a larger system. It is based on a synthesis of prior contributions. Euclidean geometry, calculus, a nd universal gravitation are the examples of such subsystems. System is a culmination of a number of relevant subsystems organized for cognitive manageability, such as physics and chemistry. The resulting disciplinary limitations of scope and depth are minimized by integrative interdisciplinary movement between the disciplines, especially between communication and behavioral sciences.

Suprasystem combines several systems into a comprehensive and encyclopedic system of knowledge. Presentl y four such suprasystems are identified: humanities, physical, biological, and social sciences; and according to Harmon, a fifth suprasystem, information science, is now emerging. Harmon proposes some hypotheses about the manner in which the changing structure of knowledge takes place.
(1) Scientists synthesize their findings into cognitive, restricted configurations, thus making information-processing easier.
(2) The smaller systems, through synthesis, merge into larger systems of kno wledge.
(3) The merge is done within the limits of short-term memory of the scientists.
(4) The process of synthesizing can be done simultaneously by different scientists, thus accounting for simultaneous discoveries.
(5) The "basic pattern underlying the formation of systematic knowledge is one of synthesis . . . larger supradisciplinary organizations of knowledge will tend to consist of approximately seven systems."86

The overriding hypothesis is the notion that the limitations of human memory are major factor in specialization and differentiation among departments of knowledge, resulting in development of a classificatory structure based on a progressive "encyclopedic reduction, synthesis, and systematization of knowledge.87 A chronological review of the development of different fields of knowledge confirms the above hypotheses.

Harmon suggests that information science may become a major suprasystem. "A potential long-range r ole for information science involves active participation in forming a complete suprasystem of knowledge that could unify the arts, sciences, and professions . . . [it could] strive to overcome the limitations of human memory and thereby increase the scope of human comprehension." 88

8.5.3.2 Storer's and Parsons' essay (1968) provides an analysis of factors responsible for the differentiation between academic disciplines by focusing on the academic profession. The author s define the academic profession in terms of its "responsibility for the maintenance, transmission, and employment of a specialized body of knowledge, and often for its extension in both empirical and theoretical directions."89

Internal differentiation among academic professions is reflected in the division of knowledge into separate disciplines. For reasons of professional autonomy, the tendency is to emphasize the differences rather than similarities between these subdivisions. The differentiation of knowledge into disciplines is discussed in terms of the validity of disciplinary separateness in making significant contributions to knowledge.

The validity test for natural sciences is in their ability to provide logical predictions based on empirical data, and in relative independence from cultural factors. In humanities, the contributions are mainly in the area of 'cultural products' such as books or works of art, analyzed in the context of established values and m eaning. Social sciences combine empirical validity in assessing sociological data and humanistic, socially determined standards.

The significance of contributions made by each discipline is determined by the intrinsic phenomena studied. Natural sciences, through a series of reductionism and antireductionisms, established open systems of generalizations and integration, which provides for the emergence of new phenomena with mathematical precision. In the humanities, focusing on the meaning rath er than on description of phenomena, the significance of their contribution is limited by a conflict between the social acceptance and scientific precision of their pursuit of knowledge. Social sciences again share the advantages and disadvantages of the previous two disciplines. Because of the lack of a satisfactory 'grand theory,' there is a considerable struggle between the subdivisions within social sciences, with a tendency to break away rather than to unify.

Important empirical factors af fecting differentiation among the disciplines are monographic publications, accessible through libraries, and professional journals. Both perform "the essential function of making available to scholars materials relevant to their interest of which they were previously unaware or which are new and could not have been known before." 90 Usually scholars subscribe to the journals in their field, and use the library 'on demand' only. The importance of journals varies with the discipline; the ir rate of growth, scope, and cost to individual scientists creates a problem of access to information. An increased quantity of recorded information limits the discipline's theoretical ability to integrate and results in a steady-state in each discipline "where the quantity of material added to the literature is limited by a discipline's ability to organize this information. As organizational or integrative ability increases, the output of information facilitated by it also increases, so that the degree of 'strain' may remain relatively constant."91 The authors express a hope that closer future cooperation between academic theorists and information specialists may contribute to a better organization of knowledge and access to it.

8.5.4 Cognitive models. I conclude this chapter with a brief review of the most recent research interests, especially among the information scientists, to relate the concept of perception of reality to the psychological notion of cognit ion.

Cognition, a concept of particular interest to contemporary information scientists, is a generic term for all processes involved in knowing. It is triggered by the interest in the perception of an external world, and it is extended to the analyses of relations in terms of more specific metaphorical categories. Cognition differs from knowledge by being limited to perceptive experiences, while knowledge is the relations discovered in cognitive and other processes. Perceptions are specific ex periences of external objects based on sensory stimulation. Metaphorically speaking, they reflect what we see.92

The categories are more complex, imperfect reflections of reality. Perception relates to existing information, and categorizations extend information to the beliefs about the perceived reality. This is well expressed by an old proverb, that 'seeing and thinking' are not synonymous experiences. However, the two experiences are intertwined; at basic levels of analys is we see directly which are the things analyzed and what are the theories about them.93

Thus the importance of theories in categorization, called by Lakoff 'idealized cognitive models,' is established. "Our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature."94 Accordingly, "the essence of metaphor is understanding and experiencing one kind of thing in terms of another."95 The cognitive aspects of a metaphor relate to the ways knowledge is acquired, interpreted, and communicated. Therefore, the metaphor is a transfer of meaning. Of interest to information and library scientists is the way that meaning is transferred from the originator of the message or its carrier to the receiver.

The recent work in cognitive communication and psychology begun with the study of concept formation processes in the 1960s and 70s. In 1973, Rosch redirected the research by focusing on an ecological approach of linguistic and cultural categories used in everyday situations. A distinction was made between a given category, called by Rosch a 'prototype,' and other, derivative 'across categories.' Categories are defined in relation to specific models, in which perceptual similarity between them plays a critical role in the process of categorization. Neisser distinguishes between (a) processes by which individual objects are categorized, (b) the gradients of 'typicality' that appears i n the category structures, and (c) the meaning of the category itself.96 "To categorize is to treat a set of things as somehow equivalent: to put them in the same pile, or call them by the same name, or respond to them in the same way."97 "To categorize an object . . . is to assert that it bears a particular relation to a particular set of ideas."98

As Beardsley pointed out,99 theories of metaphor address two basic issues: (a) the tension created by the implied meaning in the metaphor, and (b) the intelligibility of the meaning contained in it. The Emotive theory considers metaphors unverifiable, and hence meaningless. The Comparison theory, on the other hand, considers them as meaningful, explaining the tension within metaphors by the remoteness of the relationships suggested by them, or in terms of the degree of the metaphoric relation itself. The Iconic Signification theory extends that analogy by pointing out to the similarity between the iconic representation of the meaning that the metaphor stands for and the actual meaning that it signifies. The Verbal-Opposition theory interprets metaphor in terms of two levels of meaning: the central (literary, dictionary) and marginal (suggestive) meanings (Beardsley, 1967).

McCauley defined a classical theory of the category by stating that "something is a member of a particular category because it satisfies the set of necessary and sufficient conditions which c onstitute the category's defining properties, and any particular thing either has or has not these characteristics."100 Neisser distinguishes between functional, perceptive categories based on interchangeability, and thematic, narrative categories based on common membership. In the former, different categories play similar roles; in the latter, different objects play a part in a combined activity.

To summarize, cognitive processes consist of three basic components: the p erception itself, its analogical reasoning, and the categorizations of the results of the analyzes. Analogical reasoning is metaphorical, distinguishing among similarities, differences, and identities in perceptive processes. The categorization may be functional, thematic, or relational.

In the late 1970s I related metaphors to library models, by making a distinction between three kinds of metaphorical categorizations. The functional categories are based on procedural, empirical perceptions whi le the thematic, contextual categories are based on culturally influenced conventions. I added a third, relational, conceptual category, based on similarities and differences between various perceptions. The 'difference' is not merely an opposite of 'similarity'; each category manifests some unique perceptive characteristics of its own (Nitecki, 1979).

Since then, a number of more detailed theories were proposed. Taylor's 'user-driven model,' focused on the environment of the users of information, distinguishing it from the content and technology-driven models (Taylor, 1986). Dervin emphasized a qualitative approach in her 'sense-making theory,' in which the individual searches for missing information, evaluated in terms of the information-seeker's objectives (Dervin, 1992). Schon (1979) based his analyses of social-policy development on the metaphorical description of the social setting made by people perceiving the problems. Reddy's 'conduit-metaphor' explained how we i nterpret ideas communicated to us by others. Based on the proposition that "no one receives anyone else's thoughts directly in their minds," the conduit metaphor precedes a framework, within which "the listener's task must be one of extraction. He must find the meaning 'in the words' and take it out of them, so that it gets 'into his head.'" 101

Recently, several studies by information scientists were focused on the application of the metaphorical theories to information and library practice. Green's application of the conduit metaphor to an analysis of the use of the word 'information' demonstrated the significance of the linguistic predisposition in defining library concepts such as, e.g., information retrieval.102 Carbonell's model of 'invariance hierarchy' pointed out the invariable, constant hierarchy of relationships between the source of the metaphor (i.e., its metaphorical meaning) and its target. The stronger the variance, the more convincing the m etaphor.103 And finally, D. Nitecki utilized all the above metaphorical concepts in designing a study for analyzing the view of the library among university faculty, administrators, and librarians. Using a qualitative method, she identified a number of metaphors describing the similarities and differences in conceptual perception of the roles and functions of a library by different members of the university community. Here, "metaphors serve not only a role in tracking what is said a bout libraries, but also serve as a vehicle to describe what may be unknown, confusing, and perhaps even elusive concepts of the future libraries."104