Information as Thing

Michael K. Buckland
School of Library and Information Studies, University of California, Berkeley, CA 94720

Three meanings of "information" are distinguished: (1) Information-as-process: When someone is in-
'Information-as-process"; "information-as-knowledge"; formed, what they know is changed. In this sense
and "information-as-thing," the attributive use of "in- "information" is "The act of informing.. .; com-
formation" to denote things regarded as informative. munication of the knowledge or 'news' of some
The nature and characteristics of "information-as- fact or occurrence; the action of telling or fact of
thing" are discussed, using an indirect approach ("What
things are informative?"). Varieties of "information-
being told of something" (Oxford English Dic-
as-thing" include data, text, documents, objects, and tionary, 1989, vol. 7 , p. 944).
events. On this view "information" includes but extends ( 2 ) Information-as-knowledge: "Information" is also
beyond communication. Whatever information storage used to denote that which is perceived in "infor-
and retrieval systems store and retrieve is necessarily mation-as-process:" the "knowledge communi-
'information-as-thing." cated concerning some particular fact, subject, or
These three meanings of "information," along with event; that of which one is apprised or told; intelli-
"information processing," offer a basis for classifying gence, news" (Oxford English Dictionary, 1989,
disparate information-related activities (e.g., rhetoric, vol. 7 , p. 944). The notion of information as that
bibliographic retrieval, statistical analysis) and, thereby, which reduces uncertainty could be viewed as
suggest a topography for "information science."
a special case of "information-as-knowledge."
Sometimes information increases uncertainty.
Introduction: The Ambiguity of "Information" (3) Information-as-thing: The term "information" is
also used attributively for objects, such as data and
An exploration of "information" runs into immedi- documents, that are referred to as "information"
ate difficulties. Since information has to do with be- because they are regarded as being informative, as
coming informed, with the reduction of ignorance and "having the quality of imparting knowledge or
of uncertainty, it is ironic that the term "information" is communicating information; instructive." (Oxford
itself ambiguous and used in different ways. (For a con- English Dictionary, 1989, vol. 7 , p. 946).
cise and convenient introduction to varieties of mean-
7
ings of "information" and some related terms see A key characteristic of "information-as-knowledge '
Machlup (1983). See also Braman (1989), NATO (1974, is that it is intangible: one cannot touch it or measure it
1975, 1983); Schrader (1983), Wellisch (1972), Wersig in any direct way. Knowledge, belief, and opinion are
and Neveling (1975)). Faced with the variety of mean- personal, subjective, and conceptual. Therefore, to
ings of "information," we can, at least, take a pragmatic communicate them, they have to be expressed, de-
approach. We can survey the landscape and seeking scribed, or represented in some physical way, as a sig-
to identify groupings of uses of the term "information." nal, text, or communication. Any such expression,
The definitions may not be fully satisfactory, the description, or representation would be "information-
boundaries between these uses may be indistinct, and as-thing." We shall discuss implications of this below.
such an approach could not satisfy anyone determined Some theorists have objected to the attributive use
to establish the one correct meaning of "information." of the term "information" to denote a thing in the third
But if the principal uses can be identified, sorted, and sense above. Wiener asserted that "Information is infor-
characterized, then some progress might be made. mation, not material nor energy." Machlup (1983,
Using this approach we identify three principal uses of p. 642), who restricted information to the context of
the word "information:" communication, was dismissive of this third sense of
information: "The noun 'information' has essentially
two traditional meanings.. .Any meanings other than
Received November 14, 1989; revised March 16, 1990; accepted
March 29, 1990. (1) the telling of something or (2) that which is being told
are either analogies and metaphors or concoctions re-
©1991 by John Wiley & Sons, Inc sulting from the condoned appropriation of a word that

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE. 42(5):351-360, 1991 CCC 0002-8231/91/050351-10$04.00
had not been meant by earlier users." Fairthorne (1954) event can be filmed. However, the representation is no
objected scornfully to information as "stuff": "informa- more knowledge than the film is the event. Any such
tion is an attribute of the receiver's knowledge and in- representation is necessarily in tangible form (sign, sig-
terpretation of the signal, not of the sender's, nor some nal, data, text, film, etc.) and so representations of
omniscient observer's nor of the signal itself." knowledge (and of events) are necessarily "information-
But language is as it is used and we can hardly dis- as-thing."
miss "information-as-thing" so long as it is a commonly Information-as-thing is of special interest in the
used meaning of the term "information." Indeed, lan- study of information systems. It is with information in
guages evolve and with the expansion of information this sense that information systems deal directly. Li-
technology, the practice of referring to communica- braries deal with books; computer-based information
tions, databases, books, and the like, as "information" systems handle data in the form of physical bits and
appears to be becoming commoner and, perhaps, a sig- bytes; museums deal directly with objects. The inten-
nificant source of confusion as symbols and symbol- tion may be that users will become informed (informa-
bearing objects are easily confused with whatever the tion-as-process) and that there will be an imparting of
symbols denote. Further, "information-as-thing," by knowledge (information-as-knowledge). But the means
whatever name, is of especial interest in relation to provided, what is handled and operated upon, what is
information systems because ultimately information stored and retrieved, is physical information (informa-
systems, including "expert systems" and information re- tion-as-thing). On these definitions, there can be no
trieval systems, can deal directly with information only such thing as a "knowledged-based" expert system or a
in this sense. The development of rules for drawing in- "knowledge access" system, only systems based on
ferences from stored information is an area of theoreti- physical representations of knowledge.
cal and practical interest. But these rules operate upon This introductory discussion can be rounded out by
and only upon information-as-thing. reference to a fourth element: information processing,
The purpose of this examination of the notion of the handling, manipulating, and deriving of new forms
"information-as-thing" is to or versions of information-as-thing. (One could regard
the process of becoming informed as a sort of informa-
(1) Clarify its meaning in relation to other uses of the
term "information;" tion processing, but, to reduce confusion, we prefer to
(2) Affirm the fundamental role of "information-as- separate and exclude mental information-as-process
thing" in information systems; and from the scope of "information processing.")
(3) Speculate on possible use of the notion of "infor- Our discussion thus far can be summarized in terms
mation-as-thing" in bringing theoretical order to of two distinctions (1) between entities and processes;
the heterogeneous, ill-ordered fields associated and (2) between intangibles and tangibles. Taken in
with "information science." conjunction, these two distinctions yield four quite dif-
The distinction between intangibles (knowledge and ferent aspects of information and information systems.
information-as-knowledge) and tangibles (information- See Fig. 1.1.
as-thing) is central to what follows. If you can touch it
or measure it directly, it is not knowledge, but must be
A Reverse Approach: What is Informative?
some physical thing, possibly information-as-thing.
(This distinction may be overstated. Knowledge may Instead of the tedious task of reviewing candidate
well be represented in the brain in some tangible, physi- objects and inquiring whether or not they should be
cal way. However, for present purposes and for the considered to be examples of information-as-thing, we
time being, treating knowledge in the mind as impor- can reverse the process and ask people to identify the
tantly different from artificial stores of information things by or on account of which they came to be in-
seems reasonable and useful. Academic examinations formed. People will say that they are informed by a very
test individuals' ability to answer questions or to solve wide variety of things, such as messages, data, docu-
problems, which is presumed to provide indirect mea- ments, objects, events, the view through the window, by
sures of what they know. But that is not the same.) any kind of evidence. This point was recognized by
Knowledge, however, can be represented, just as an Brookes (1979, p. 14): "In the sciences it has long been

INTANGIBLE TANGIBLE
ENTITY 2. Information-as-knowledge 3. Information-as-thing
Knowledge Data, document

PROCESS 1. Information-as-process 4. Information processing

Becoming informed Data processing

FIG. 1. Four aspects of information.

352 JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991

recognized that the primary source of information is investigation." (Oxford English Dictionary, 1989, vol. 4,
not the literature of the sciences but observation of the p. 469). If something cannot be viewed as having the
relevant natural phenomena. Scientists (and others) find characteristics of evidence, then it is difficult to see
'sermons in stones and books in the running brooks'." how it could be regarded as information. If it has value
How might we best sort out these candidates for being as information concerning something, then it would ap-
regarded as information? (Note we are restricting our pear to have value as evidence of something. "Evi-
attention to physical things and physical events. Some dence" appears to be close enough to the meaning of
people would say that some of their knowledge comes information-as-thing to warrant considering its use as a
from paraphysical sources, notably from divine inspira- synonym when, for example, describing museum objects
tion. Others would deny any such nonphysical source of as "authentic historic pieces of evidence from nature
information, but, to the extent that it may exist, infor- and society." (Schreiner, 1985, p. 27).
mation science would have to be incomplete if it were One area in which the term "evidence" is much used
excluded. Not knowing what to say on the subject we is in law. Much of the concern is with what evidence–
merely note it as a possible area of unusual interest what information-can properly be considered in a legal
within information science.) process. It is not sufficient that information may be
pertinent. It must also have been discovered and made
Information as Evidence available in socially approved ways. However, if we set
One learns from the examination of various sorts of aside the issues of the propriety of the gathering and
things. In order to learn, texts are read, numbers are presentation of evidence and ask what, in law, evidence
tallied, objects and images are inspected, touched, or actually is, we find that it corresponds closely to the
otherwise perceived. In a significant sense information way we are using it here. In English law, evidence can
is used as evidence in learning-as the basis for under- include the performing of experiments and the viewing
standing. One's knowledge and opinions are affected by of places and is defined as: "First, the means, apart
what one sees, reads, hears, and experiences. Textbooks from argument and inference, whereby the court is in-
and encyclopedias provide material for an introduction; formed as to the issues of fact as ascertained by the
literary texts and commentaries provide sources for the pleadings; secondly the subject matter of such means."
study of language and literature; arrays of statistical (Buzzard et al., 1976, p. 6; also Wigmore, 1983).
data provide input for calculations and inference;
statutes and law reports indicate the law; photographs
Types of Information
show what people, places, and events looked like; cita-
tions and sources are verified; and so on. In each case it Pursuing the notion of information as evidence, as
is reasonable to view information-as-thing as evidence, things from which one becomes informed, we can ex-
though without implying that what was read, viewed, amine more specifically what sorts of things this might
listened to, or otherwise perceived or observed was include.
necessarily accurate, useful, or even pertinent to the
user's purposes. Nor need it be assumed that the user
Data
did (or should) believe or agree with what was per-
ceived, "Evidence" is an appropriate term because it "Data," as the plural form of the Latin word "datum,"
denotes something related to understanding, something means "things that have been given." It is, therefore, an
which, if found and correctly understood, could change apt term for the sort of information-as-thing that has
one's knowledge, one's beliefs, concerning some matter. been processed in some way for use. Commonly "data"
Further, the term "evidence" implies passiveness. denotes whatever records are stored in a computer. (See
Evidence, like information-as-thing, does not do any- Machlup (1983, p. 646-649) for a discussion of the use
thing actively. Human beings do things with it or to it. and misuse of the term "data".)
They examine it, describe it, and categorize it. They
understand, misunderstand, interpret, summarize, or
Text and Documents
rebut it. They may even try to fake it, alter it, hide it,
or destroy it. The essence of evidence is precisely that Archives, libraries, and offices are dominated by
perception of it can lead to changes in what people be- texts: papers, letters, forms, books, periodicals, manu-
lieve that they know. scripts, and written records of various kinds, on paper,
Dictionary definitions of "evidence" include: 'An ap- on microform, and in electronic form. The term "docu-
pearance from which inferences can be drawn; an indi- ment" is normally used to denote texts or, more exactly,
cation, mark, sign, token, trace. . . . Ground for belief; text-bearing objects. There seems no reason not to ex-
testimony or facts tending to prove or disprove any con- tend the use of "text" and "document" to include im-
clusion.. . . Information, whether in the form of per- ages, and even sounds intended to convey some sort of
sonal testimony, the language of documents, or the communication, aesthetic, inspirational, instrumental,
production of material objects, that is given in a legal whatever. In this sense, a table of numbers can be con-

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991 353

sidered as text, as a document, or as data. Text that is no reason why what is learned by direct observation
to be analyzed statistically could also be regarded as of the physical environment should not be regarded as
data. There is a tendency to use "data" to denote nu- information just as that which learned by observing the
merical information and to use text to denote natural marks on a document." Wersig (1979) adopted an even
language in any medium. broader view of information as being derived from
Further confusion results from attempting to distin- three sources: (1) "Generated internally" by mental
guish two types of retrieval by making and compound- effort; (2) 'Acquired by sheer perception" of phenom-
ing two unwarranted assumptions about "data" and ena; and (3) 'Acquired by communication." We view
"document": (1) that "data retrieval" should denote the "information-as-thing" as corresponding to Wersig's
retrieval of records that one wishes to inspect and "docu- phenomena (2) and communications (3).
ment retrieval" should denote references to records that Some informative objects, such as people and his-
one may wish to inspect; and (2) that "data retrieval" toric buildings, simply do not lend themselves to being
would be a "known item" search, but that "document collected, stored, and retrieved. But physical relocation
retrieval" would be a "subject search" for an unknown into a collection is not always necessary for continued
item (van Rijsbergen, 1979, p. 2; Blair, 1984). The for- access. Reference to objects in their existing locations
mer assumption imposes an odd definition on both creates, in effect, a "virtual collection." One might also
terms. The second is illogical and contrary to practical create some description or representation of them: a
experience (Buckland, 1988b, pp. 85-87). It is wise not film, a photograph, some measurements, a directory, or
to assume any firm distinction between data, docu- a written description. What one then collects is a docu-
ment, and text. ment describing or representing the person, building, or
other object.
Objects
What is a Document?
The literature on information science has concen-
trated narrowly on data and documents as information We started by using a simple classification of infor-
resources. But this is contrary to common sense. Other mation resources: data, document, and object. But dif-
objects are also potentially informative. How much ficulties arise if we try to be rigorous. What, for
would we know about dinosaurs if no dinosaur fossils example, is a document? A printed book is a document.
had been found? (cf. Orna and Pettit (1980, p. 9), writ- A page of hand-writing is a document. A diagram is a
ing about museums: "In the first stage, the objects them- document. A map is a document. If a map is a docu-
selves are the only repository of information.") Why do ment, why should not a three-dimensional contour map
centers of research assemble many sorts of collections also be a document. Why should not a globe also be
of objects if they do not expect students and researchers considered a document since it is, after all, a physical
to learn something from them? Any established univer- description of something. Early models of locomotives
sity, for example, is likely to have a collection of rocks, were made for informational not recreational purposes
a herbarium of preserved plants, a museum of human (Minns, 1973, p. 5). If a globe, a model of the earth, is
artifacts, a variety of bones, fossils, and skeletons, and a document, why should one not also consider a model
much else besides. The answer is, of course, that ob- of a locomotive or of a ship to be a document? The
jects that are not documents in the normal sense of model is an informative representation of the original.
being texts can nevertheless be information resources, The original locomotive or ship, or even a life-size
information-as-thing. Objects are collected, stored, re- replica, would be even more informative than the
trieved, and examined as information, as a basis for be- model. "The few manuscript remains concerning the
coming informed. One would have to question the three ships that brought the first settlers to Virginia
completeness of any view of information, information have none of the power to represent that experience
science, or information systems that did not extend to that the reconstructed ships have." (Washburn, 1964).
objects as well as documents and data. In this we, like But by now we are rather a long way from customary
Wersig (1979), go further than Machlup (1983, p. 645) notions of what a document is.
who, like Belkin and Robertson (1976), limited infor- The proper meaning of "document" has been of con-
mation to what is intentionally told: "Information takes cern to information scientists in the "documentation"
at least two persons: one who tells (by speaking, writ- movement, seeking to improve information resource
ing, imprinting, signally) and one who listens, reads, management since the beginning of this century. The
watches." Similarly Heilprin (1974, p. 124) stated that documentalist's approach was to use "document" as a
"information science is the science of propagation of generic term to denote any physical information re-
meaningful human messages." Fox (1983) took an even source rather than to limit it to text-bearing objects in
narrower view, examining information and misinforma- specific physical media such as paper, papyrus, vellum,
tion exclusively in terms of propositional sentences. or microform. Otlet and others in the documentation
Brookes (1974), however, was less restrictive: "I see movement affirmed:

354 JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991

 (1) That documentation (i.e., information storage and 1975, p. 12). Meanwhile the semantic problem re-
retrieval) should be concerned with any or all po- mains: What generic term for informative things is
tentially informative objects; wide enough to include, say, museum objects and other
(2) that not all potentially informative objects were scholarly evidence, as well as text-bearing objects? Ob-
documents in the traditional sense of texts on jecting to the use of "information" or of "document" for
paper; and this purpose does not remove the need for a term.
(3) that other informative objects, such as people,
products, events and museum objects generally, Most documents in the conventional usage of the
should not be excluded (Laisiepen, 1980). Even word-letters, books, journals, etc.–are composed of
here, however, except for Wersig's contribution text. One would include diagrams, maps, pictures, and
(Wersig, 1980), the emphasis is, in practice, on sound recordings in an extended sense of the term
forms of communication: data, texts, pictures, in- "text." Perhaps a better term for texts in the general
scriptions. sense of artifacts intended to represent some meaning
would be "discourse." We could also characterize these
Otlet (1934, p. 217), a founder of the documentation texts as "representations" of something or other. How-
movement, stressed the need for the definition of "docu- ever, we could hardly regard an antelope or a ship as
ment" and documentation (i.e., information storage being "discourse." Nor are they representations in any
and retrieval) to include natural objects, artefacts, ob- ordinary sense. Their value as information or evidence
jects bearing traces of human activities, objects such as derives from what they signify about themselves indi-
models designed to represent ideas, and works of art, as vidually or, perhaps, about the class or classes of which
well as texts. The term "document" (or "documentary they are members. In this sense they represent some-
unit") was used as a specialized sense as a generic term thing and, if not a representation, they could be viewed
to denote informative things. Pollard (1944) observed as representative. If an object is not representative of
that "From a scientific or technological point of view something, then it is not clear how far it can signify
the [museum] object itself is of greater value than a anything, i.e., be informative.
written description of it and from the bibliographical One might divide objects into artifacts intended to
point of view it should be regarded therefore as a docu- constitute discourse (such as books), artifacts that were
ment." A French documentalist defined "document" not so intended (such as ships), and objects that are not
as "any concrete or symbolic indication, preserved artifacts at all (such as antelopes). None of this prevents
or recorded, for reconstructing or for proving a phe- any of these from being evidence, from being informa-
nomenon, whether physical or mental." ("Tout indice tive concerning something or other. Nor does it pre-
concret ou symbolique, conserve ou enregistré, aux vent people from making uses different from that which
fins de représenter ou de prouver un phénomène ou may have been intended. A book may be treated as a
physique ou intellectual" (Briet, 1951, p. 7)). On this doorstop. Illuminated initial letters on medieval manu-
view objects are not ordinarily documents but become scripts were intended to be decorative, but have become
so if they are processed for informational purposes. A a major source of information concerning medieval
wild antelope would not be a document, but a captured dress and implements.
specimen of a newly discovered species that was being "Natural sign" is the long-established technical term
studied, described, and exhibited in a zoo would not in philosophy and semiotics for things that are informa-
only have become a document, but "the catalogued an- tive but without communicative intent (Clarke, 1987;
telope is a primary document and other documents are Eco, 1976).
secondary and derived. ("L'antilope catalogueé est un
document initial et les autres documents sont seconds
ou derives." (Briet, 1951, p. 8). Perhaps only a dedicated
Events
documentalist would view an antelope as a document.
But regarding anything informative as a "document" is We also learn from events, but events lend them-
consistent with the origins and early usage of the word, selves even less than objects do to being collected and
which derived from the Latin verb docere, to teach or stored in information systems for future edification.
to inform, with the suffix "-ment" denoting means. How different the study of history would be if they
Hence "document" originally denoted a means of teach- could! Events are (or can be) informative phenomena
ing or informing, whether a lesson, an experience, or a and so should be included in any complete approach to
text. Limitation of "document" to text-bearing objects information science. In practice we find the evidence of
is a later development (Oxford English Dictionary, events is used in three different ways:
1989, vol. 4, p. 916; Sagredo & Izquierdo, 1983, pp. 173-
178). Even among documentalists, however, including (1) Objects, which can be collected or represented,
anything other than text-bearing objects in information may exist as evidence associated with events:
retrieval appears to occur only in theoretical discus- bloodstains on the carpet, perhaps, or a footprint
sions and not always then (Rogalla von Bieberstein, in the sand;

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991 355

 There may well be representations of the event it- ward a more ecumenical approach to information and
self: photos, newspaper reports, memoirs. Such information systems (Bearman, 1989).
documents can be stored and retrieved; and, also,
Events can, to some extent, be created or recre-
ated. In experimental sciences, it is regarded as When is Information not Information?
being of great importance that an experiment-an
event-be designed and described in such a way Even if we dismiss the argument that untrue infor-
that it can be replicated subsequently by others. mation is not information, we could still ask what could
Since an event cannot be stored and since ac- not be information? Since being evidence, being infor-
counts of the results are no more than hearsay evi- mation, is a quality attributed to things, we may well
dence, the feasibility of reenacting the experiment ask what limits there might be to what could or could
so that the validity of the evidence, of the infor- not be information. The question has to be rephrased as
mation, can be verified is highly desirable. "What things could not be regarded as informative?"
Regarding events as informative and noting that, al- We have already noted that a great variety of things
though events themselves cannot be retrieved, there is can be regarded as informative so the range is clearly
some scope for recreating them, adds another element very large.
to the full range of information resource management. We might say that objects of which nobody is aware
If the recreated event is a source of evidence, of infor- cannot be information, while hastening to add that they
mation, then it is not unreasonable to regard the labora- might well become so when someone does become
tory (or other) equipment used to reenact the event as aware of them. It is not uncommon to infer that some
being somehow analogous to the objects and documents sort of evidence, of which we are not aware, ought to or
that are usually regarded as information sources. In might exist and, if found, would be of particular impor-
what senses does it matter whether the answer to an tance as evidence, as when detectives search, more or
inquiry derives from records stored in a data base or less systematically, for clues.
from reenacting an experiment? What significant dif- Determining what might be informative is a difficult
ference is there for the user of logarithms between a task. Trees, for example, provide wood, as lumber for
logarithmic value read from a table of logarithms and a building and as firewood for heating. One does not
logarithmic value newly calculated as and when needed? normally think of trees as information, but trees are
The inquirer might be wise to compare the two, but informative in at least two ways. Obviously, as repre-
would surely regard both as being equally information. sentative trees they are informative about trees. Less
Indeed it would be a logical development of current obviously, differences in the thickness of tree rings are
trends in the use of computers to expect a blurring of caused by, and so are evidence of, variations in the
the distinction between the retrieval of the results of weather. Patterns reflecting a specific cycle of years
old analyses and the presentation of the results of a constitute valuable information for archaeologists seek-
fresh analysis. ing to date old beams (Ottaway, 1983). But if lumber
To include objects and events, as well as data and and firewood can be information, one hesitates to state
documents, as species of information is to adopt a categorically of any object that it could not, in any cir-
broader concept than is common. However, if we are to cumstances, be information or evidence. We conclude
define information in terms of the potential for the pro- that we are unable to say confidently of anything that it
cess of informing, i.e., as evidence, there would seem could not be information.
no adequate ground for restricting what is included to This leads us to an unhelpful conclusion: If anything
processed data and documents as some would prefer, is, or might be, informative, then everything is, or might
e.g., by defining information as "Data processed and well be, information. In which case calling something
assembled into a meaningful form." (Meadows, 1984, "information" does little or nothing to define it. If
p. 105). There are two difficulties with such a restricted everything is information, then being information is
definition: Firstly, it leaves unanswered the question of nothing special.
what to call other informative things, such as fossils,
footprints, and screams of terror. Secondly, it adds the
Being Information is Situational
additional question of how much processing and/or as-
sembling is needed for data to be called information. In Information-as-process is situational. Therefore, evi-
addition to these two specific difficulties there is the dence involved in information-as-process is so situa-
more general criterion that, all things being equal, a tionally also. Hence, whether any particular object,
simpler solution is to be preferred to a more compli- document, data, or event is going to be informative de-
cated one. Therefore we retain our simpler view of "in- pends on the circumstances, just as the "relevance" of
formation-as-thing" as being tantamount to physical a document or a fact is situational depending on the
evidence: Whatever thing one might learn from (Orna inquiry and on the expertise of the inquirer (Wilson,
& Pettit, 1980, p. 3). Fortunately there are moves in the 1973). It follows from this that the capability of "being
English-language literature of information retrieval to- informative," the essential characteristic of informa-

356 JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991

tion-as-thing, must also be situational. We may say of formation systems, in archives, data bases, libraries,
some object or document that in such-and-such a com- museums, and office files. But because these decisions
bination of circumstances, in such-and-such a situation, are based on a compounding of different judgements, as
it would be informative, it would be information, i.e., noted above, it is not surprising that there should be
information-as-thing. disagreement. Nevertheless, it is on this basis that data
But, as noted above, we could in principle say that of are collected and fed into databases, librarians select
any object or document: One just has to be imaginative books, museums collect objects, and publishers issue
enough in surmising the situation in which it could be books. It is a very reasonable prediction that copies of
informative. And if one can describe anything this way, the San Francisco telephone directory will be informa-
we are making little progress in distinguishing what in- tive, though there is no guarantee that each and every
formation-as-thing is. Further, it is a matter of individ- copy will necessarily be used.
ual judgement, of opinion "Information-as-thing", then, is meaningful in two
(1) whether some particular thing would be pertinent;
senses: (1) At quite specific situations and points in time
and, if so, an object or event may actually be informative, i.e.,
(2) whether the probability of it being used as evi- constitute evidence that is used in a way that affects
dence would be significant; and, if so, someone's beliefs; and (2) Since the use of evidence is
(3) whether its use as evidence would be important. predictable, albeit imperfectly, the term "information"
(The issue might be trivial or, even if important, is commonly and reasonably used to denote some popu-
this particular evidence might be redundant, unre- lation of objects to which some significant probability
liable, or otherwise problematic.) And, if so, of being usefully informative in the future has been at-
(4) whether the importance of the issue, the impor- tributed. It is in this sense that collection development
tance of the evidence, and the probability of its is concerned with collections of information.
being used-in combination-warrant the preser-
vation of this particular evidence. If all of these
are viewed positively, then one would regard the
thing-event, object, text, or document-as likely Copies of Information and Representations
to be useful information and, presumably, take
steps to preserve it or, at least, a representa-
tion of it. Copies: Type and Token
In the provision of access to information by means
Information by Consensus of formal information systems, the question of whether
or not two pieces of information are the same (or, at
We have shown that (1) the virtue of being informa- least, equivalent) is important. When copies are identi-
tion-as-thing is situational and that (2) determining that cal one would speak formally of types and tokens. Ex-
any thing is likely to be useful information depends on amples that are not the same as each other are referred
a compounding of subjective judgements. Progress be- to as different types; identical copies are referred to as
yond an anarchy of individual opinions concerning tokens. If only one example exists, then one would say
what is or is not reasonably treated as information de- that there is only one "token" of that "type."
pends on agreement, or on at least some consensus. We The creation of identical, equally authentic copies is
can use an historical example to illustrate this point. It the result of particular technologies of mass production,
used to be considered important to know whether a such as printing. If you want to reread a particular title
woman was a witch or not. One source of evidence was (type), you would want to read some copy (token) of it,
trial by water. The unfortunate woman would be put in but you would not insist on rereading the exact same
a pond. If she floated she was a witch. If she sank she copy as before. Similarly, if you had read a book on
was not. This event, the outcome of the experiment, some subject and wanted to know more, you would or-
was, by consensus, the information-as-thing needed for dinarily move on to reading a copy of another different
the identification of a witch. Nowadays it would be de- title in preference to reading a different copy of the
nied, by consensus, that the exact same event consti- same title.
tuted the information that it had previously been This feature of equally acceptable copies can be
accepted, by consensus, as being. found in other examples of information systems. Some
Where there is a consensus of judgement, the con- sorts of museum objects are mass-produced, such as
sensus is sometimes so strong that the status of objects, telephones. With telephones as with printed books, one
especially documents, being information is unques- example is as acceptable as any other from the same
tioned, e.g., telephone directories, airline timetables, production run. There is, however, a major qualifica-
and textbooks. In these cases arguments are only over tion. In archival practice, as in museums, two physi-
niceties such as accuracy, currency, completeness, and cally identical documents are regarded as different if
cost. As a practical matter some consensus is needed to they occur in different places in the original order of
agree on what to collect and store in retrieval-based in- the files. The rationale is that their unique positioning

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE- June 1991 357
in relation to other documents makes them unique by and texts to data. Exceptions to this, such as from
association and, thereby, different. object to object or from document back to object
In electronic data bases the situation is a little less (physical replicas and models) can also be found
clear. One can have copies of two sorts: There can be (Schlebecker, 1977).
temporary, virtual copies displayed on a screen; or one (4) Additional details related to the object but not evi-
dent from it might be added to the representation,
can make copies of a longer lasting form on paper or
either to inform or to misinform.
other storage medium. These copies might not, from
(5) Representation can continue indefinitely. There
some engineering error, be quite the same as the origi- can be representations of representations of repre-
nal. However, it is ordinarily assumed that either the sentations.
copy is authentic or that errors will be so marked as to (6) For practical reasons representations are com-
be self-evident. There may be difficulty in knowing monly (but not necessarily) briefer or smaller than
whether the copy is a copy of the latest, official version whatever is being represented, concentrating on
of the database, but that is a different issue. With hand- the features expected to be most significant. A
written, manuscript texts, one should expect each ex- summary, almost by definition, is an incomplete
ample to be at least slightly different, even if it purports description.
to be a copy. The person making a copy is likely to Progress in information technology continually per-
omit, add, and change parts of the text. A significant mits improvements in our ability to make physical
feature of medieval studies is the necessity of examin- descriptions, examples of information-as-thing. Photo-
ing closely all copies of related manuscripts not only to graphs improve on drawings; digital images improve on
identify the differences, but also to infer which might photographs. The voice of the nineteenth century
be the more correct versions where they do differ. singer, Jenny Lind, was described by Queen Victoria as
In general, then, the existence of identical, equally "a most exquisite, powerful and really quite peculiar
informative, equally authoritative copies is unusual. voice, so round, so soft and flexible.. ." (Sadie, 1980,
Printed materials in libraries are a notable exception. v. 10, p. 865). Although this description is better than
More general is the case where copies are not altogether none, we could learn much more from a phonograph
identical, though they may be equally acceptable for recording.
most purposes. Reproductions of works of art and of museum arti-
facts may suffice for some purposes and have the ad-
Interpretations and Summaries of Evidence vantages that they can provide much increased physical
access without wear and tear on the originals. Yet they
Progress in information technology increases the will always be deficient in some ways as representations
scope for creating and using information-as-thing. of the original, even though, as in the case of works of
Much of the information in information systems has art and museum objects, even experts cannot always
been processed by being coded, interpreted, summa- identify which is an original and which is a copy (Mills
rized, or otherwise transformed. Books are a good ex- & Mansfield, 1979).
ample. Virtually all of the books in the collections are
based, at least in part, on earlier evidence, both texts
and other forms of information. Scholarship is perme- Information, Information Systems, Information
ated with descriptions and summaries, or, as we prefer Science
to call them, representations. We started with two academically respectable usages
Representations have important characteristics: of the term "information" ("information-as-knowledge"
(1) Every representation can be expected to be more and "information-as-process")and we noted that infor-
or less incomplete in some regard. A photograph mation systems can deal directly only with "information-
does not indicate movement and may not depict as-thing." Stating this paradox differently, information
the color. Even a color photograph will generally systems handle information only in a sense of informa-
show colors imperfectly-and fade with time. A tion dismissed by leading theorists of information. We
written narrative will reflect the viewpoint of the also concluded that anything might be information-as-
writer and the limitations of the language. Films thing. Small wonder that progress in the development
and photographs usually show only one perspec- of paradigms for describing and explaining phenomena
tive. Something of the original is always lost. in the shapeless, ill-defined reaches of "information
There is always some distortion, even if only science" has been slow. But, perhaps, "information-as-
through incompleteness.
thing" could be used to provide some order or arrange-
(2) Representations are made for convenience, which
in this context tends to mean easier to store, to ment with respect t o information-related activities,
understand, and/or to search. along with the two more respectable definitions.
(3) Because of the quest for convenience, representa- First, although all information systems deal directly
tions are normally a shift from event or object to with "information-as-thing," we might create some
text, from one text to another text, or from objects order within this area if we could identify a subset of

358 JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991

information-handling activities that are concerned with Cognitive psychology, rhetoric, and other studies of in-
information only in this sense. As examples we might terpersonal communication and persuasion would be
choose information theory (in the sense of the mathe- examples. Alternative means, i.e., alternative physical
matical theory of signal transmission associated with media, might be equally acceptable. Indeed, inasmuch
Shannon and Weaver and that has nothing to do with as the primary interest is on cognition and persuasion,
semantic content (Bar-Hillel, 1964); historical bibliogra- the actual information-as-knowledge, also a neces-
phy (the study of books as physical objects); and statisti- sary ingredient, may also be of little direct interest.
cal analysis (identifying and defining patterns in The focus could well be more on how beliefs change
populations of objects and/or events). Each of these than on which beliefs are changed or which knowledge
fields has refined techniques for developing and formal- is represented.
ized ways of describing concise and effective repre- It is not asserted that sorting areas of information
sentations of their particular kind of information-as- science with respect to their relationship to informa-
thing. The findings of these useful arts may well be of tion-as-thing would produce clearly distinct popula-
great significance, but their concern is primarily with tions. Nor is any hierarchy of scholarly respectability
the evidence itself. An analysis of a channel, a book, or intended. The point is rather that examination of "in-
a population would cease to be valid if the physical formation-as-thing" might be useful in bringing shape
characteristics of the channel, book, or population to this amorphous field and in avoiding simplistic, ex-
were changed. clusive boundaries based on past academic traditions.
Second, information storage and retrieval systems
can deal directly only with "information-as-thing," but Summary
the things that can be stored for retrieval in actual or
virtual collections vary in significant ways. Historic Numerous definitions have been proposed for "in-
buildings, films, printed books, and coded data impose formation." One important use of "information" is to
different constraints on the tasks associated with infor- denote knowledge imparted; another is to denote the
mation retrieval systems: selection, collection, storage, process of informing. Some leading theorists have dis-
representation, identification, location, and physical ac- missed the attributive use of "information" to refer to
cess. Put simply, a museum, an archive, library of things that are informative. However, "information-as-
printed books, an online bibliographic database, and a thing? deserves careful examination, partly because it
corporate management information system of numeric is the only form of information with which information
data can all validly be regarded as species of informa- systems can deal directly. People are informed not only
tion retrieval system. But differences in their physical by intentional communications, but by a wide variety of
attributes affect how the stored items can be handled objects and events. Being "informative" is situational
(Buckland, 1988a). These differences provide one basis and it would be rash to state of any thing that it might
for the comparative analysis of information storage and not be informative, hence information, in some conceiv-
retrieval systems. able situation. Varieties of "information-as-thing" vary
Third, representations of knowledge form a distin- in their physical characteristics and so are not equally
guishable subset of information-as-thing and so could, suited for storage and retrieval. There is, however, con-
in principle, be used to identify and define another siderable scope for using representations instead.
class of information systems in which the primary con-
cern is based on the knowledge represented. This is the Acknowledgment
conventional area of information storage and retrieval,
This work was partially supported by a Fulbright
subject bibliography, and "knowledge bases" for expert
Research Scholarship at Graz University of Technol-
systems. In these cases the information-as-thing is un-
ogy, Austria, during sabbatical leave from the Univer-
avoidably of concern, but only a means for dealing with
sity of California at Berkeley. The helpful comments of
information-as-knowledge and, being merely a means,
William S. Cooper, Brian Peaslee, W. Boyd Rayward,
considerable latitude is imaginable. In providing an in-
and Patrick Wilson are gratefully acknowledged.
formation service different physical forms of informa-
tion and different text-bearing media (texts on paper, on
microform, or displayed on a terminal) may be equally References
acceptable. Further, a wide variety of variant texts Bar-Hillel, Y. (1964). Language and information. London: Addison-
could be more or less substitutable-in English or in Wesley. Chap. 16: 'An examination of Information Theory,"
French, lengthy or concise, recent or old-if they repre- 275-297.
sented the same knowledge to an acceptable degree. Bearman, D. (1989). Archives and manuscript control with biblio-
Fourth, information-as-process could also be the graphic utilities: Challenges and opportunities. American
Archivist 52, 26-39.
basis for defining a class of information-related stud- Belkin, N; Robertson, S. E. (1976). Information science and the
ies. Here again, information-as-thing cannot be ig- phenomena of information. Journal of the American Society for
nored, but is, again, of secondary interest as a means. Information Science, 27, 197-204.

JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE-June 1991 359

Blair, D. C. (1984). The data-document distinction in information NATO Advanced Study Institute in Information Science, Crete,
retrieval. Communications of the association for Computing Ma- 1978. (1983). Information science in action: Systems design.
chinery, 27 369-374. Boston: Martinus Nijhoff.
Braman, S. (1989). Defining information. Telecommunications Orna, E., & Pettit, C. (1980). Information handing systems in mu-
Policy, 13, 233-242. seums. New York: Saur.
Briet, S. (1951). Qu'est que la documentation? Paris: Editions Docu- Otlet, P. (1934). Traité de documentation. Brussels: Editiones
mentaires Industrielles et Techniques. Mundaneum.
Brookes, B. C. (1974). Robert Fairthorne and the scope of informa- Ottaway, B. S. (Ed.). (1983). Archaeology, dendrochronology, and
tion science. Journal of Documentation 30, 139-152. the radiocarbon calibrating curve. Edinburgh: Edinburgh Univer-
Brookes, B. C. (1979). Theoretical informatics in stage II mecha- sity, Dept of Archaeology.
nization of IR processes. In International Federation for Docu- Oxford English Dictionary (1989) 2nd ed. Oxford: Clarendon Press.
mentation, Theoretical problems of informatics: New trends in Pollard, A. F. C. (1944). British Societyfor International Bibliography
informatics and its terminology. F I D 568. (pp. 10-18). Moscow: Proceedings 6, 54.
VINITI. Rogalla von Bieberstein, J. (1975). Archiv, Bibliothek und Museum
Buckland, M. K. (1988a). Library materials: Paper, microform, als Dokumentationsbereiche: Einheit und gegenseitige Abgrenzung.
database. College and Research libraries 49, 117-122. Pullach bei München: Verlag Dokumentation.
Buckland, M. K. (1988b). Library services in theory and context. 2nd Sadie, S. (Ed.). (1980). The new Grove dictionary of music and musi-
ed. New York: Pergamon. cians. Vol. 10, p. 865. London: Macmillan.
Clarke, D. S. (1987). Principles of semiotic. London: Routledge & Sagredo, F., & Izquierdo, J. M. (1983). Concepcion lógico-lingüístico
Kegan Paul. de la Documentacion. Madrid: Ibercom-Red Comnet de la
Buzzard, J. H. et al. (1976). Phipson on Evidence. 12th ed. (The Unesco.
Common Law Library, 10). London: Sweet & Maxwell. Schlebecker, J.T. (1977). T h e use of objects in historical research.
Eco, U. (1976). A theory of semiotics. Bloomington, Ind.: Indiana Agricultural History 51, 200-208.
University Press. Schrader, A. (1984). In search of a name: Information science and
Fairthorne, R . A. (1954). The theory of communication. Aslib Pro- its conceptual antecedents. Library and Information Science Re-
ceedings 6, 255-267. (Reprinted in Fairthorne, R. A. (1961) To- search 6, 227-271.
wards Information Retrieval. London: Butterworths. pp. 64-79. Schreiner, K. (1985). Fundamentals of museology: On the theory and
Fox, C. J. (1983). Information and misinformation. Westport, CT: methodology of collecting, preserving, decoding, and utilizing
Greenwood. musealia Waren, GDR: n.p.
Heilprin, L , (1974). Operational definitions. In: A. Debons (Ed.), van Rijsbergen, C. R . (1979). Information retrieval. 2nd ed. London:
NATO Advanced Study Institute in Information Science, Cham- Butterworths.
pion, 1972. Information science: Search for identity, (pp. 115-38). Washburn, W. E. (1964). Manuscripts and manufacts. American
New York: Marcel Dekker. Archivist 27, 245-250.
Laisiepen, K. et al. (1980). Grundlagen derpraktischen Information Wellisch, H. (1972). From information science to informatics. Jour-
und Dokumentation: Eine Einfuhrung. 2. Aufl. München: Saur. nal of Librarianship 4, 157-187.
Machlup, F. (1983). Semantic quirks in studies of information. In Wersig, G.; Neveling, U. (1975). The phenomena of interest to in-
F. Machlup, & U. Mansfield, (Eds), The study of information: formation science. Information Scientist 9, 127-140.
Interdisciplinary messages. (pp. 641-671). New York: Wiley. Wersig, G. (1979). The Problematic Situation as a Basic Concept of
Meadows, A. J. et al. (1984). Dictionary of computing and new infor- Information Science in the Framework of the Social Sciences: A
mation technology. London: Kegan Paul. Reply to Belkin. In International Federation for Documentation.
Mills, J. F., & Mansfield, J. M. (1979). The genuine article. London: Theoretical Problems of Informatics: New Trends in Informatics
British Broadcasting Corporation. and its Terminology. F I D 568. (pp. 48-57). Moscow: VINITI.
Minns, J. E. (1973). Model railway engines. London: Octopus Books. Wersig, G. (1980). Informationstatigkeit. In K. Laisiepen, et al.
NATO Advanced Study Institute in Information Science, Cham- (Eds.), Grundlagen der praktischen Information und Dokumenta-
pion, 1972. (1974). Information science: Search for identity. tion: Eine Einführung. 2. Aufl. (pp. 161-192). München: Saur.
(pp. 115-38). New York: Marcel Dekker. Wigmore, J. H. (1983). Evidence in trials at common law Vol. 1, re-
NATO Advanced Study Institute in Information Science, Aberyst- vised by P. Tillers. (2-12). Boston: Little Brown, 1983.
wy th, 1974. (1975) Perspectives in information science. Leyden: Wilson, P. G. (1973). Situational relevance. Information Storage and
Noordhoff. Retrieval 9, 457-471.

360 JOURNAL OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE- June 1991