William Kent, Data and Reality, 1stBooks, 1998.

Data and Reality was first published by North Holland in 1978. It was republished in 1998 by 1stBooks.

Data and Reality [Excerpts]

What they've said about Data and Reality
Preface to the Second Edition
Preface [to the original edition]
Chapter 12: Philosophy
   Reality and Tools
   Points of View
   A View of Reality

What they’ve said about Data and Reality...

An excellent, philosophical discussion of the problems inherent in describing the real world. There is nothing really similar to this work. I think that all data base researchers should read this document. It might also be assigned as supplementary reading in general graduate and undergraduate courses in data base systems. - Mike Senko (1978)

I expect the book to be one of the most frequently quoted ones for the next few years. It is unique in being an almost exhaustive, condensed rendition of the typical problems encountered. The most striking strong point is its penetration into major data base technology headaches… Many well chosen examples and the lucid style make it easy to read. -Reiner Durcholz (1978)

…highly recommended and even required reading for all DP people… - G.M. Nijssen (1978)

Kent has produced a rather remarkable and highly readable short work… the most important things he has to say are philosophical and go right to the heart of the key concepts that must be understood if a system is to be “successful” (whatever that may mean!)…This is a serious book but not a heavy one. Kent writes easily and without hiding behind the semantics of the data base specialists. The ideas are presented in a straightforward manner with no attempt to preach. -Datamation, March 1979

This excellent study of the problems inherent in describing the real world is unique in (1) being an almost exhaustive, condensed rendition of the typical problems encountered, (2) not offering an own solution as remedy for all evils, and (3) penetrating into the mists of conceptual ambiguity… This book is of important value to all those in the field of data bases and information systems who are concerned with developing a deeper understanding of this matter. It is of equal importance to the systems analyst, to the data base designer, and to the database system designer. -Current Engineering Practice, July 1979

Data and Reality illustrates extensively the pitfalls of any simplistic attempts to capture reality as data in the sense of today’s database systems. The approach taken by the author is one which very logically and carefully delineates the facets of reality being represented in an information system, and also describes the data processing models used in such systems. The linguistic, semantic, and philosophical problems of describing reality are comprehensively examined… The depth of discussion of these concepts, as they impact on information systems, is not likely to be found elsewhere.… the value of this book resides in its critical, probing approach to the difficulties of modeling reality in typical information systems… it is very well written and should prove both enjoyable and enlightening to a careful reader. -ACM Computing Reviews, August 1980

By page eight one has been exposed to an incredible number of philosophical ideas, all cast as concrete data-representation problems… this is basically a book that poses problems and exposes contradictions… A very stimulating read.
-Quantitative Sociology Newsletter, Spring 1981

Kent attacks the pseudo-exactness of existing data models in a very neat and clear (and often humorous) manner… This book is for everyone who thinks about or works on data files and who wants to understand the reasons for his disenchantment.
-European Journal of Operations Research, November 1981

I am using Data and Reality as research material for my current project. It is on my desk right now. -Joe Celko, 1998

The book is still quoted quite often and has a message even - or especially - for today's jaded information scientists.
-Prof. Dr. Robert Meersman, Vrije Universiteit Brussel (1998)

Your book focuses attention on many issues that are still, embarrassingly, not being dealt with in our formalized information systems. It provides an important reference point not only in identifying these problems, but in pointing out origins and the long-standing practice of simply ignoring them. When I reopened your book… I found lots of issues that seem as fresh as ever.
-Roger Burkhart, John Deere (1998)

Preface to the Second Edition

Despite critical acclaim, outside of a small circle of enthusiastic readers this book has been a sleeper for over twenty years. Publishers have recently offered to market and distribute it with more vigor if I would provide a new revised edition, but I’ve resisted. Laziness might be seen as the excuse, but I’m beginning to realize there’s a better reason.

A new revised edition would miss the point of the book. Many texts and reference works are available to keep you on the leading edge of data processing technology. That’s not what this book is about. This book addresses timeless questions about how we as human beings perceive and process information about the world we operate in, and how we struggle to impose that view on our data processing machines. The concerns at this level are the same whether we use hierarchical, relational, or object-oriented information structures; whether we process data via punched-card machines or interactive graphic interfaces; whether we correspond by paper mail or e-mail; whether we shop from paper-based catalogs or the web. No matter what the technology, these underlying issues have to be understood. Failure to address these issues imperils the success of your application regardless of the tools you are using.

That’s not to say the technical matrix of the book is obsolete or antiquated. The data record is still a fundamental component of the way we organize computer information. Sections of the book exploring new models including behavioral elements are precursors of object orientation.

The scope of the book extends beyond computer technology. The questions aren’t so much about how we process data as about how we perceive reality, about the constructs and tactics we use to cope with complexity, ambiguity, incomplete information, mismatched viewpoints, and conflicting objectives.

You can read the book for those reasons, or for other reasons as well. A few years back, almost twenty years after the book was published, I began to notice that the book is also about something else, something far more personal. The scope of the book doesn’t only extend beyond computer data processing into the realm of how we perceive the world. It also extends into our inner domain. I’ve come to recognize that it touches on issues in my own inner life that I, like most of us to some degree or other, have been grappling with for decades.

Consider the key topics: existence, identity, attributes, relationships, behavior, and modeling.

Existence: Is cogito ergo sum sufficient? To what extent am I really present and engaged in the process of life around me? How real are the physical things I experience? To what extent do I exist in some spiritual realm independent of the physical context?

Identity: The old “Who am I?” bit. What is the true nature of the kind of person I am? What sorts of needs, goals, outlooks define who I really am?

Attributes: What kind of person am I? What are my values, my assets, my limitations?

Relationships: This is the core of it all. What is the quality of my interaction with parents, lovers, spouses, children, siblings, friends, colleagues, and other acquaintances? What are my connections with things material, social, spiritual, and otherwise? What are my needs here? What are the issues and problems? How can they be improved?

Behavior: What should I plan to do in various situations? How? What might be the consequences, both intended and otherwise? What contingencies need to be anticipated?

Modeling: How accurate and useful are the constructs I use to explain all these things? How effective are these kinds of explanations in helping me change what needs to be changed?

This book certainly shouldn’t be classified in the social sciences, but it is remarkable to observe how technology issues can resonate as metaphors for our inner lives. This perspective seems to explain why I’ve engaged so intimately with these ideas, why I’ve argued so passionately about them at standards committee meetings and in the hallways at conferences.

I repeat the invitation, made in the book’s original preface, to discover for yourself what you might think the book is about. It just might be about you. But if that’s too much pop psychology for your comfort, if that’s too invasive of your personal space, then just read it for its insights into data processing and reality.

Preface [to the original edition]

A message to mapmakers: highways are not painted red, rivers don't have county lines running down the middle, and you can't see contour lines on a mountain.

For some time now my work has concerned the representation of information in computers. The work has involved such things as file organizations, indexes, hierarchical structures, network structures, relational models, and so on. After a while it dawned on me that these are all just maps, being poor artificial approximations of some real underlying terrain.

These structures give us useful ways to deal with information, but they don't always fit naturally, and sometimes not at all. Like different kinds of maps, each kind of structure has its strengths and weaknesses, serving different purposes, and appealing to different people in different situations. Data structures are artificial formalisms. They differ from information in the same sense that grammars don't describe the language we really use, and formal logical systems don't describe the way we think. "The map is not the territory" [Hayakawa].

What is the territory really like? How can I describe it to you? Any description I give you is just another map. But we do need some language (and I mean natural language) in order to discuss this subject, and to articulate concepts. Such constructs as "entities", "categories", "names", "relationships", and "attributes" seem to be useful. They give us at least one way to organize our perceptions and discussions of information. In a sense, such terms represent the basis of my "data structure", or "model", for perceiving real information. Later chapters discuss these constructs and their central characteristics -- especially the difficulties involved in trying to define or apply them precisely.

Along the way, we implicitly suggest a hypothesis (by sheer weight of examples, rather than any kind of proof -- such a hypothesis is beyond proof): there is probably no adequate formal modelling system. Information in its "real" essence is probably too amorphous, too ambiguous, too subjective, too slippery and elusive, to ever be pinned down precisely by the objective and deterministic processes embodied in a computer. (At least in the conventional uses of computers as we see them today; future developments in artificial intelligence may endow these machines with more of our capacity to cope.) This follows a path pointed out by Zemanek, connecting data processing with certain philosophical observations about the real world, especially the aspects of human judgement on which semantics ultimately depend [Zemanek 72].

In spite of such difficulties (and because I see no alternative), we also begin to explore the extent and manner in which such constructs can and have been incorporated into various data models. We are looking at real information, as it occurs in the interactions among people, but always with a view toward modelling that information in a computer based system. The questions are these: What is a useful way to perceive information for that purpose? What constructs are useful for organizing the way we think about information? Might those same constructs be employed in a computer based model of the information? How successfully are they reflected in current modelling systems? How badly oversimplified is the view of information in currently used data models? Are there limits to the effectiveness of any system of constructs for modelling information?

In spite of my conjecture about the inherent limits of formal modelling, we do need models in order to go about our business of processing information. So, undaunted, I have assimilated some of my own ideas about a "good" modelling system, and these appear toward the end.

Keep in mind that I am not talking about "information" in a very broad sense. I am not talking about very ambitious information systems. We are not in the domain of artificial intelligence, where the effort is to match the intellectual capabilities of the human mind (reasoning, inference, value judgements, etc.). We are not even trying to process prose text; we are not attempting to understand natural language, analyze grammar, or retrieve information from documents. We are primarily concerned with that kind of information which is managed in most current files and data bases. We are looking at information that occurs in large quantities, is permanently maintained, and has some simplistic structure and format to it. Examples include personnel files, bank records, and inventory records.

Even this modest bit of territory offers ample opportunity for misunderstanding the semantics of the information being represented.

Within these bounds, we focus on describing the information content of some system. The system involved might be one or more files, a data base, a system catalog, a data dictionary, or perhaps something else. We are limiting ourselves to the information content of such systems, excluding such concerns as:

Real implementations, representation techniques, performance.
Manipulation and use of the data.
Work flow, transactions, scheduling, message handling.
Integrity, recovery, security.

A caution to the lay reader in search of a tutorial: this book is not about data processing as it is. As obvious as these concepts may seem, they are not reflected in, or are just dimly understood in, the current state of data processing systems. "We do not, it seems, have a very clear and commonly agreed upon set of notions about data -- either what they are, how they should be fed and cared for, or their relation to the design of programming languages and operating systems. This paper sketches a theory of data which may serve to clarify these questions. It is based on a number of old ideas and may, as a result, seem obvious. Be that as it may, some of these old ideas are not common currency in our field, either separately or in combination; it is hoped that rehashing them in a somewhat new form may prove to be at least suggestive" [Mealy]. That opening paragraph of a now classic paper, some ten years old, is still distressingly apt today.

There is a wonderful irony at work here. I may be trying to overcome misconceptions which people outside the computer business don't have in the first place. Many readers will find little new in what I say about the nature of our perceptions of reality. Such readers may well react with "So what's new?" To them, my point is that the computing community has largely lost sight of such truisms. Their relevance to the computing disciplines needs to be re-established.

People in the data processing community have gotten used to viewing things in a highly simplistic way, dictated by the kind of tools they have at their disposal. And this may suggest another wonderful irony. People are awed by the sophistication and complexity of computers, and tend to assume that such things are beyond their comprehension. But that view is entirely backwards! The thing that makes computers so hard to deal with is not their complexity, but their utter simplicity. The first thing that ought to be explained to the general public is that a computer possesses incredibly little ordinary intelligence. The real mystique behind computers is how anybody can manage to get such elaborate behavior out of such a limited set of basic capabilities. The art of computer programming is somewhat like the art of getting an imbecile to play bridge or to fill out his tax returns by himself. It can be done, provided you know how to exploit the imbecile's limited talents, and are willing to have enormous patience with his inability to make the most trivial common sense decisions on his own. Imagine, for example, that he only understood grammatically perfect sentences, and couldn't make the slightest allowance for colloquialisms, or for the normal way people restart sentences in mid-speech, or for the trivial typographical errors which we correct so automatically that we don't even see them. The first step toward understanding computers is an appreciation of their simplicity, not their complexity.

Another thought, though: I may be going off in the wrong direction by focussing so much concern on computers and computer thinking. Many of the concerns about the semantics of data seem relevant to any record keeping facility, whether computerized or not. I wonder why the problems appear to be aggravated in the environment of a computerized data base. Is it sheer magnitude? Perhaps there is just a larger mass of people than before who need to achieve a common understanding of what the data means. Or is it the lost human element? Maybe all those conversations with secretaries and clerks, about where things are and what they mean, are more essential to the system than we've realized. Or is there some other explanation?

The flow of the book generally alternates between two domains, the real world and computers. Chapter 1 is in the world of real information, exploring some enigmas in our concepts of "entities". Chapter 2 briefly visits the realm of computers, dealing with some general characteristics of formally structured information systems. This gives us a general idea of the impact the two domains have on each other. Chapters 3 through 6 then address other aspects of real information. Chapters 7 through 11, dealing with data processing models, bring us back to the computer. We top it all off with a smattering of philosophical observations in Chapter 12.

This has been an approximate characterization -- one view -- of what the rest of the book contains. Please read on to discover what you might think the book is about.

* * * *

I want to thank the people who took the time to comment on (and often contribute to) earlier versions of this material, including Marilyn Bohl, Ted Codd, Chris Date, Bob Engles, Bob Griffith, Roger Holliday, Lucy Lee, Len Levy, Bill McGee, Paula Newman, and Rich Seidner. George Kent, of the Political Science Dept. at the University of Hawaii, provided a valuable perspective from a vantage point outside of the computing profession. Karen Takle Quinn, our head librarian, was immensely helpful in tracking down many references. I thank Willem Dijkhuis of North Holland for his substantial encouragement in the publication of this book.

And very special thanks go to my wife, Barbara, who helped make the book more readable, and who coped and sacrificed more than anyone else for this book.

CHAPTER 12: PHILOSOPHY

Reality and Tools

I have tried to describe information as it "really is" (at least, as it appears to me), and have kept tripping over fuzzy and overlapping concepts. This is precisely why system designers and engineers and mechanics often lose patience with academic approaches. They recognize, often implicitly, that the complexity and amorphousness of reality is unmanageable. There is an important difference between truth and utility. We want things which are useful -- at least in this business; otherwise we'd be philosophers and artists.

Perhaps it is inevitable that tools and theories never quite match. There are some opposite qualities inherent in them.

Theories tend to distinguish phenomena. A theory tends to be analytical, carefully identifying all the distinct elements and functions involved. Unifying explanations are abstracted, relationships and interactions are described, but the distinctness of the elements tends to be preserved.

Good tools, on the other hand, intermingle various phenomena. They get a job done (even better, they can do a variety of jobs). Their operation tends to intermix fragments of various theoretical phenomena; they embody a multitude of elementary functions simultaneously. That's what it usually takes to get a real job done. The end result is useful, and necessary, and profitable.

Theories tend toward completeness. A theory is defective if it does not account for all aspects of a phenomenon or function.

Tools tend to be incomplete in this respect. They incorporate those elements of a function which are useful and profitable; why bother with the rest? The justification for a tool is economic: the cost of its production and maintenance vs. the value of its problem solving functions. This has nothing to do with completeness. (In 1975, a government official asked to have his job abolished, because nobody actually needed the services of his office. His job did have a well defined function, in theory. "Completeness" would have dictated that his job be retained.)

Useful tools have well defined parts, and predictable behavior. They lend themselves to solving problems we consider important, by any means we can contrive. We often solve a problem using a tool that wasn't designed for it. Tools are available to be used, don't cost too much, don't work too slowly, don't break too often, don't need too much maintenance, don't need too much training in their use, don't become obsolete too fast or too often, are profitable to the toolmaker, and preferably come with some guarantee, from a reliable toolmaker. Tools don't share many of the characteristics of theories. Completeness and generality only matter to the extent that a few tools can economically solve many of the problems we care about.

Thus the truth of things may be this: useful things get done by tools which are an amalgam of fragments of theories. Those are the kinds of tools whose production and maintenance expense can be justified. Theories are helpful to gain understanding, which may lead to the better design of better tools. This understanding is not essential; an un-analytic instinct for building good tools is just as useful, and often gets results faster.

It may be a mistake to require a tool to fit the mold of any theory. If this be so, then we'd better be aware of when we are discussing theory and when we are discussing tools.

Data models are tools. They do not contain in themselves the "true" structure of information. What really goes on when we present a data model, e.g., hierarchies, to a user? Does he say "Aha! Of course my information is hierarchically structured; I see how the model fits my data"? Of course not. He has to learn how to use it. We generally presume that this learning is required only because of the complexity of the tool. Difficulties are initially perceived as a failure to fully understand the theory; there is an expectation that perseverance will lead to a marvelous insight into how the theory fits the problem. In fact, much of his "learning" is really a struggle to contrive some way of fitting his problem to the tool: changing the way he thinks about his information, experimenting with different ways of representing it, and perhaps even abandoning some parts of his intended application because the tool won't handle it. Much of this "learning" process is really a conditioning of his perceptions, so that he learns to accept as fact those assumptions needed to make the theory work, and to ignore or reject as trivial those cases where the theory fails.

Tools are generally orthogonal to the problems they solve, in that a given tool can be applied to a variety of problems, and a given problem can be solved in different ways with different tools. Versatility is in fact a very desirable property in a tool. It is useful then also to understand separately the characteristics of a tool and the nature of the problems to which it can be applied.

Points of View

A conceptual model, by its very nature, needs to be durable -- at least in form, if not content. Its content should be adjusted to reflect changes in the enterprise and its information needs -- only. The form of the conceptual model -- the constructs and terms in which it is expressed -- should be as impervious as possible to changes in the supporting computer technology. We can postulate that the man-machine interface will continue to evolve toward man; data processing technology will move toward handling information in ways that are natural to the people who use it. It follows then that a durable conceptual model should be based on constructs as close as possible to the human way of perceiving information.

There's a catch right there: the implicit assumption that there is just one "technology" by which all people perceive information, and hence which is most natural and easy for everybody to use. There probably isn't. Human brains undoubtedly function in a variety of ways. We know that some people do their thinking primarily in terms of visual images; others hear ideas being discussed in their heads; still others may have a different mode of intuiting concepts, neither visual nor aural. Analogously, some people may structure information in their heads in tabular form, others work best with analytic subdivisions leading to hierarchies, and others naturally follow paths in a network of relationships.

This may well be the root of the debates over which data model is best, most natural, easiest to learn and use, most machine independent, etc. The camps are probably divided up according to the way their brains function -- each camp advocating the model that best approximates their own brain technology.

A View of Reality

"I do not know where we are going, but I do know this -- that wherever it is, we shall lose our way." (Sagatsa)

"If you're confused, it just proves you've been paying attention." (G. Kent)

This book projects a philosophy that life and reality are at bottom amorphous, disordered, contradictory, inconsistent, non-rational, and non-objective. Science and much of western philosophy have in the past presented us with the illusion that things are otherwise. Rational views of the universe are idealized models which only approximate reality. The approximations are useful. The models are successful often enough in predicting the behavior of things that they provide a useful foundation for science and technology. But they are ultimately only approximations of reality, and non-unique at that.

This bothers many of us. We don't want to confront the unreality of reality. It frightens, like the shifting ground in an earthquake. We are abruptly left without reference points, without foundations, with nothing to stand on but our imaginations, our ethereal self-awareness.

So we shrug it off, shake it away as nonsense, philosophy, fantasy. What good is it? Maybe if we shut our eyes the notion will go away.

What do we know about physical entities, about ourselves?

Lewis Thomas tells us that a human being is not exactly a single discrete living thing, but more a symbiotic interaction of hordes of discrete living things inhabiting and motivating our cells. We are each an enormously divisible social structure [Thomas].

Sociobiologists are telling us that the human being is not the unit of evolution and survival. It is our genes which are motivated to survive and perpetuate themselves. Individual people are merely vehicles whose survival serves that higher purpose -- sometimes! [Time Magazine, Aug. 1, 1977.]

Our precious self image is being challenged from another quarter, too. Some scientists aren't quite so sure any more that they can clearly distinguish between the categories of "man" and "animal". "People" might not be a well defined category! Recent experiments have demonstrated the capabilities of chimpanzees and gorillas to acquire language, concepts, symbols, abstractions -- traits held by some to be the only significant hallmarks of the human species. A lawyer is prepared to argue that such animals are entitled to some of the protections accorded individuals under the law -- such animals may be "legal persons". An article in the New York Times Magazine of June 12, 1977 observes: "If apes have access to language, can they not be expected to reason? And if they can reason, what distinction is there remaining between man and beast?" "Separately, and in some instances collectively, these animals have demonstrated the ability to converse with humans for as long as 30 minutes, to combine learned words in order to describe new situations or objects, to perceive difference and sameness, to understand `if-then' concepts, to describe their moods, to lie, to select and use words in syntactic order, to express desire, to anticipate future events, to seek signed communication with others of their species and, in one extraordinary sequence .... to force the truth from a lying human." "... It's a heretical question, really. I was brought up a good Catholic. Man is man and beast is beast. I don't really think that now. You can't spend four or five years with a chimp, watch it grow up, and not realize that all the going on in her head is pretty much the same as that going on in mine ..."

Which brings to mind that our vision of ourselves as uniquely intelligent creatures is also threatened from quite another quarter -- the one we've been dealing with all along here. What, in some people's view, is one of the objectives of artificial intelligence, if not to endow machines with an intelligence competitive with humans? Is science fiction really mistaken in its visions of humanoids and robots functioning like, or better than, human beings? How often have those visionaries been wrong before?

In the monthly magazine published by the American Museum of Natural History, we read: "Some futurists ... view the current difference between human and artificial intelligence as one of degree, not of kind, and predict that the gap between humans and machines will be crossed about the year 2000" [Jastrow]. Data processing people are fond of saying that the category of employees is a subset of the category of people. How long before we have to expand that to include animals and robots? I wonder if that question will really sound as foolish to someone reading this, say, twenty or fifty years from now.

What does all this do to our sense of identity, to our egocentric view of people as entities? If we have to rebuild our world view so radically again (as, for example, Copernicus forced us to do once before), then how much faith can we have in the permanence of any world view?

Our notions of reality are overwhelmingly dominated by the accidental configurations of our physical senses. We are very parochial in our sense of scale. Bacteria and viruses and sub-atomic particles are not very real to most of us, nor are galaxies. We don't really know how to comprehend them. Our concept of motion is bounded by the physiology of our eyes: the continental plates don't move, but motion pictures (sequences of still pictures!) do. Most of us think of continents and islands as permanent and discrete entities -- rather than as accidents of the current water level in the oceans. Are islands and mountains such different things? Have you ever had the opportunity to observe a reservoir get filled, or emptied?

And our sense of reality is quite conditioned by the very narrow frequency range to which our eyes respond. Imagine if we couldn't see the "visible" spectrum, but could see ultra-violet, or infra-red, or x-rays -- or maybe sound waves! We might not have any notion of opaque objects; everything might be translucent or transparent. Things might appear to have entirely different shapes or boundaries. We might not have such a primary notion of things having sharp or fixed boundaries; the normal mode of things might be a state of flux, like the wind or clouds or currents in the ocean. Think of perceiving people in terms of the thermal gradients around their bodies, rather than gradients in the visible spectrum. We might have no concept of day or night. Those concepts are only so "real" and "fundamental" because we are so dependent on visible light. Clumps of heat might look like "things" to us, just as clouds do now. We might see sounds as physical things moving through the air, and we might see the wind.

Or suppose that senses other than sight dominated our world view. The universe of many animals -- their sense of what things exist, and what they are -- is based on smells. To them, the existence and nature of a thing is defined primarily by what it smells like. What it looks like is an occasional, trivial consideration (like the smell of things is to us). In a heavy fog, we suddenly live in a universe of things heard, rather than things seen.

The shark seems to have sense organs responding directly to electrical phenomena. What image of reality could it have, which we don't even know how to imagine? (And what view of reality do we have, which a blind person doesn't even know how to imagine? Can you even begin to imagine how it feels to have no comprehension at all of what the verb "see" means?)

To a greater or lesser extent, we all operate with somewhat different foundations for our perceptions of reality. Biologist Robert Trivers comments: "The conventional view that natural selection favors nervous systems which produce ever more accurate images of the world must be a very naive view of mental evolution." [Time Magazine, Aug. 1, 1977.] Among many of us, the differences are trivial. Between some of us they are enormous.

Compare your view of reality with that of a mathematical physicist, or an astronomer. (If you are one, how does it feel to be singled out as having a peculiar view?) The world view of such people includes as regular features such notions as Einsteinian time and space, particles of light, light being bent by gravity, everything accelerating away from everything else, black holes, and seeing things (stars) which may have vanished thousands or millions of years ago. How often do these crop up in your world view?

Your brain may be obliged to confess such views are real, but your intuition isn't. What shall we make of it? The earth does look flat, after all, doesn't it? And, no matter how much schooling we've had, we can't seem to stop thinking of the sun as rising and setting. Incidentally, do your children share your world view of this phenomenon?

"Consider how the world appears to any man, however wise and experienced in human life, who has never heard one word of what science has discovered about the Cosmos. To him the earth is flat; the sun and moon are shining objects of small size that pop up daily above an eastern rim, move through the upper air, and sink below a western edge; obviously they spend the night somewhere underground. The sky is an inverted bowl made of some blue material. The stars, tiny and rather near objects, seem as if they might be alive, for they `come out' from the sky at evening like rabbits or rattlesnakes from their burrows, and slip back again at dawn. `Solar system' has no meaning to him, and the concept of a `law of gravitation' is quite unintelligible -- nay, even nonsensical. For him bodies do not fall because of a law of gravitation, but rather `because there is nothing to hold them up' -- i.e., because he cannot imagine their doing anything else. He cannot conceive space without an `up' and `down' or even without an `east' and `west' in it. For him the blood does not circulate; nor does the heart pump blood; he thinks it is a place where love, kindness, and thoughts are kept. Cooling is not a removal of heat but an addition of `cold'; leaves are not green from the chemical substance chlorophyll in them, but from the `greenness' in them. It will be impossible to reason him out of these beliefs. He will assert them as plain, hard-headed common sense; which means that they satisfy him because they are completely adequate as a system of communication between him and his fellow men. That is, they are adequate linguistically to his social needs, and will remain so until an additional group of needs is felt and is worked out in language" [Whorf].

So far I've dealt with variations in perceived reality which I can at least describe. They are close enough to my world view (and yours, I hope) that I can describe the differences in terms of familiar concepts. But I must acknowledge the existence of world views so alien to mine that I can't even grasp the central concepts. These are exemplified by some of the Eastern philosophies, various theologies, mystical cults. The Hopi Indians have a world view of time and causality which can hardly even be expressed in our vocabulary of concepts. "I find it gratuitous to assume that a Hopi who knows only the Hopi language and the cultural ideas of his own society has the same notions, often supposed to be intuitions, of time and space that we have, and that are generally assumed to be universal. In particular, he has no general notion or intuition of time as a smooth flowing continuum in which everything in the universe proceeds at an equal rate, out of a future, through a present, into a past." "The Hopi language and culture conceals a metaphysics, such as our so-called naive view of space and time does, or as the relativity theory does; yet it is a different metaphysics from either. In order to describe the structure of the universe according to the Hopi, it is necessary to attempt -- insofar as it is possible -- to make explicit this metaphysics, properly describable only in the Hopi language, by means of an approximation expressed in our own language, somewhat inadequately it is true ...." [Whorf].

Do you and I have the "real" notion of time? What shall we make of contemporary physics, which wants us to believe that time passes at different rates for objects travelling at different speeds? The astronaut who has been travelling a year close to the speed of light has been gone from us for ten years? Or is it vice versa? .

Language has an enormous influence on our perception of reality. Not only does it affect how and what we think about, but also how we perceive things in the first place. Rather than serving merely as a passive vehicle for containing our thoughts, language has an active influence on the shape of our thoughts. "...language produces an organization of experience... language first of all is a classification and arrangement of the stream of sensory experience which results in a certain world order..." [Whorf].

Whorf quoting Edward Sapir: "Human beings do not live in the objective world alone, nor alone in the world of social activity as ordinarily understood, but are very much at the mercy of the particular language which has become the medium of expression for their society. It is quite an illusion to imagine that one adjusts to reality without the use of language and that language is merely an incidental means of solving specific problems of communication or reflection. The fact of the matter is that the `real world' is to a large extent unconsciously built up on the language habits of the group.... We see and hear and otherwise experience very largely as we do because the language habits of our community predispose certain choices of interpretation."

"Hopi has one noun that covers every thing or being that flies, with the exception of birds, which class is denoted by another noun.... The Hopi actually call insect, airplane, and aviator all by the same word, and feel no difficulty about it.... This class seems to us too large and inclusive, but so would our class `snow' to an Eskimo. We have the same word for falling snow, snow on the ground, snow packed hard like ice, slushy snow, wind-driven flying snow -- whatever the situation may be. To an Eskimo, this all-inclusive word would be almost unthinkable; he would say that falling snow, slushy snow, and so on, are sensuously and operationally different, different things to contend with; he uses different words for them and for other kinds of snow. The Aztecs go even farther than we in the opposite direction, with `cold', `ice', and `snow' all represented by the same basic word with different terminations; `ice' is the noun form; `cold', the adjectival form; and for `snow', `ice mist'."

We are more ready to perceive things as entities when our language happens to have nouns for them. For what reason does our language happen to have the noun "schedule" for the connection between, say, a train and a time, but no such familiar noun for the connection between a person and his salary?

The way we bundle relationships is similarly affected. If we think of the relationships "has color" and "has weight", we might be inclined to lump them into a single "has" relationship, with several kinds of entities in the second domain. But if we happen to employ the word "weighs", then that makes it easier to think of the second relationship as being distinct in its own right. By what accident of linguistic evolution do we fail to have a similar verb for the color phenomenon? ("Appears" might be a close approximation.)

Other examples: "has salary" vs. "earns", "has height" vs. what?

The accidents of vocabulary: we are most prepared to identify as entities or relationships those things for which our vocabulary happens to contain a word. The presence of such a word focusses our thinking onto what then appears as a singular phenomenon. The absence of such a word renders the thought diffuse, non-specific, non-singular.

This is all very unsatisfying. It is consistent with this philosophy of reality (perhaps even necessary, rather than just consistent), that I cannot see it applied consistently. I must accept paradoxes embedded right in the process of embracing such views. I am not, after all, such an alien creature. I see the world in much the same terms as you do. I have a name, and an employer, and a social security number, and a salary, and a birth date, etc. etc. There is a reasonably accurate description of me and my environment in several files. I have a wife, and children, and a car, all of which I believe to be very real. In short, I can share with you a very traditional view of reality; most of the useful activities of my daily life are predicated on such familiar foundations.

Well then, what's going on? What are these contradictions all about?

I'm really not sure, but perhaps I can try to frame an answer in terms of purpose and scope. I am convinced, at bottom, that no two people have a perception of reality which is identical in every detail. In fact, a given person has different views at different times -- either trivially, because detailed facts change, or in a larger sense, such as the duality of my own views.

But there is considerable overlap in all of these views. Given the right set of people, the differences in their views may become negligible. Reducing the number of people involved greatly enhances this likelihood. This is what I mean by "scope": the number of people whose views have to be reconciled.

In addition, there is a question of purpose. Views can be reconciled with different degrees of success to serve different purposes. By reconciliation I mean a state in which the parties involved have negligible differences in that portion of their world views which is relevant to the purpose at hand. If an involved party holds multiple viewpoints, he may agree to use a particular one to serve the purpose at hand. Or he may be persuaded to modify his view, to serve that purpose.

If the purpose is to arrive at an absolute definition of truth and beauty, the chances of reconciliation are nil. But for the purposes of survival and the conduct of our daily lives (relatively narrow purposes), chances of reconciliation are necessarily high. I can buy food from the grocer, and ask a policeman to chase a burglar, without sharing these people's views of truth and beauty. It is an inevitable outcome of natural selection that those of us who have survived share, within a sufficiently localized community, a common view of certain basic staples of life. This is fundamental to any kind of social interaction.

If the purpose is to maintain the inventory records for a warehouse, the chances of reconciliation are again high. (How high? High enough to make the system workably acceptable to certain decision makers in management.) If the purpose is to consistently maintain the personnel, production, planning, sales, and customer data for a multi-national corporation, the chances of reconciliation are somewhat less: the purposes are broader, and there are more people's views involved.

So, at bottom, we come to this duality. In an absolute sense, there is no singular objective reality. But we can share a common enough view of it for most of our working purposes, so that reality does appear to be objective and stable.

But the chances of achieving such a shared view become poorer when we try to encompass broader purposes, and to involve more people. This is precisely why the question is becoming more relevant today: the thrust of technology is to foster interaction among greater numbers of people, and to integrate processes into monoliths serving wider and wider purposes. It is in this environment that discrepancies in fundamental assumptions will become increasingly exposed.