[A short version of this essay, about one half this length, appears as chapter 2, "Logic and Intuition" of 'The Metaphysics of Virtual Reality' (Oxford U. Press, 1993). The version here highlights the schizophrenic attitude that Blaise Pascal showed toward computers. - M.H.]

How does thinking at the computer differ from thinking with paper and pencil or thinking at the typewriter? The computer does not simply place another tool into our hands so we can symbolize our thoughts. It builds a whole environment, an information environment where our thinking breathes a new atmosphere. No sooner do we have a computer on our desk than we begin living in an information-rich world.

First, the files we create grow rapidly, making an electronic library of letters, papers, and other documents. Through online connections, we begin storing the work of colleagues and friends, notes about future projects, or leftovers from database searches. Add some serendipitous items to our storage disks - maybe the Gettysburg Address, the Constitution, or the entire King James Bible -and we soon outgrow our current disk storage capacity. We find ourselves bitten by infomania, denizens of an information environment.

Since the computer was the engine for generating and storing information, we naturally look to it also for ways of hacking our way through the dense information jungle. The steady overgrowth of information makes us seek help through computer data searching. Computer searches help us find references, phrases, or ideas instantly, plowing through massive amounts of data. A word processor or database search takes our key phrase or topic and in seconds snaps a piece of information into view. The computer seems to pull us out of the quicksand generated by computers. The computer sifts through thousands of periodicals in minutes without our ever having to know anything about silicon microchips, high-level code, or sorting algorithms. All we need for a search is access to a database and some elementary search logic you can learn in about an hour. Today most computer search logic uses elementary Boolean logic.

What is Boolean logic? Alfred Glossbrenner in How to Look it Up Online describes search logic in terms simple enough for most computer users: "AND means a record must have both terms in it. OR means it can have either term. NOT means it cannot have the specified term."1 Glossbrenner chides those who belabor the complexities of Boolean search logic and bewilder the user: "You sometimes get the impression that the authors would be drummed out of the manual-writers union if they didn't include complicated discussion of search logic laced with plenty of Venn diagrams -those intersecting, variously shaded circles you learned about in sophomore geometry. Forget it!"

But, alas, what Glossbrenner wants us to forget will soon enough slip away - as technology enfolds us in its web of assumptions.

Frequent reading and writing on computers will soon allow us little distance from the tools that trap our language. They will fit like skin. The conditions under which we work will become invisible, ever more difficult to discern, even with the most disciplined effort. Everywhere present like eyeglasses on the end of our noses, computers will not allow us to examine the angles of distortion they introduce, the loss of color, the unseen vistas they hide. Like microscopes, computer extend our vision vastly. But, unlike the microscope, the computer turns our attention from the primary world in order to represent our entire symbolic life, the whole contents of our psyche, not just the dimensions of another scale. Boolean search logic and other interface strategies will soon enough become ubiquitous enough to become second nature for literate people, something they take for granted.

What people take for granted was once something startling and unprecedented. In a felt transition like the present, we can still notice the change. We have an opportunity to consider the initial shifts in the life of the psyche. We might even consider how to compensate for the shift. We can ask, How does Boolean search logic affect our thought processes and our mental life? What side-effects of infomania hide behind those "intersecting, variously shaded circles you learned about in sophomore geometry"?

Boolean logic, displayed graphically by the circles of Venn diagrams, constitutes a central achievement of modern logic. Modern logic, which makes the computer possible, got its footing in the work of Leibniz (1646-1716) whose discoveries laid the foundations of the world we call the computer age or the age of information.2 So when we inspect Boolean logic for the side-effects of database searching, we are implicitly touching the heart of the world we inhabit. The logic of the Boolean search shapes and expresses subconscious processes affecting how we model the world we live in. Computerized text manipulation has to do with the way we structure our mental environment. Once we see how these subconscious agencies constitute our modern world, we can then grasp our possibilities - both promise and peril.

To do so, we have to turn again to those simple Venn diagrams "from sophomore geometry" and to the Boolean logic on which it is based.

George Boole (1815-1864) discovered the branch of mathematics known as symbolic logic. Boole's "Algebra of Logic" first put logical relations into formal symbols and paved the way for contemporary set theory. His Boolean logic uses groups of algebraic symbols to formulate the general connections between things and their properties. Boole's discovered how to manipulate equations to express the relationships things have to one another insofar as they belong to specific classes of things. His logic is an algebra of classes or categories, something like a map of a Chinese-box structure, which shows how the properties of items make those items internest within one another.

Take some referential terms, for instance, such as brown and cows: All objects that are brown = X; All objects that are cows = Z. One formula can then represent the logical relationship between the two terms as a product of mutual inclusion: "All brown cows" = XZ. For more complex formulas, you can add a logical NOT (-Z) as well as an AND (XZ & Z-X). You can build up terms to represent a whole series of increasingly complex relationships. You can even make symbolic formulas that represent a long deductive argument so that the logical form of the argument comes to the surface for review and criticism - making the argument into a mathematical problem.

Historically speaking, Boole's logic was the first system for algebraically calculating class membership, a system for rapidly figuring whether or not something falls into one or another class of things. As a method for organizing by the inclusiveness or exclusiveness of classes, Boolean logic became a way of determining the properties of a thing once the thing's membership or inclusion in a class has been determined.

Before Boole, logic was a study of statements about things referred to directly at hand. After Boole, logic could become a system of pure symbols. Pre-Boolean logic focused on the way direct statements or assertions connect and hold together. A set of statements that hangs together is a valid deduction. Validity is the way conclusions connect with adequate grounds or premises.

The traditional study of logic harked back to Aristotle, who first discerned patterns or "syllogisms" in the way arguments put together direct statements. Aristotle did indeed sometimes use symbols in his logic, but he used them sparingly and merely to point out a linguistic pattern. Aristotle's symbols served primarily to organize what was already given as direct statements or propositions. With Boolean logic, on the other hand, direct statements become potential instances of the relationships between abstract symbols. Direct language becomes only one possible instance of algebraic mathematics, one possible application.

Boole's logic inverted the traditional relationship between direct and symbolic languages. Boole conceived language as symbols, and he thought his symbols could absorb all logically correct language. By inverting statement and symbol, Boole's mathematical or symbolic logic seemed to swallow traditional logic, absorbing direct statements into a web of symbolic patterns. Logical argument became merely a branch of symbolic logic.

John Venn introduced the term "symbolic logic" in a book by that title in 1881. Venn continued Boole's effort to absorb the direct statements of language under a general system of abstract algebra. With mathematics as a basis, Venn could solve certain logical difficulties which had perplexed traditional Aristotelian logicians. With mathematical precision, modern logic could present linguistic arguments within a total system, a formal organization of axioms and theorems. Systemic consistency became more important than the direct reference to things in our experience.

Now we have a first telltale sign of infomania: the priority of system. Where system precedes relevance, the way is clear for the primacy information. John Venn revealed a feature of logic that is crucial for Boolean search logic. Venn noted that his new logic uses classes or sets or terms in the same sense that we use the notion of compartments. That is, we can speak of compartments that are empty. Terms having no actually existing members are considered null sets, a kind of empty compartment. As modern logicians say, logical terms do not of themselves necessarily have existential import. Which is to say that the relationships described in modern logic are purely abstract relations uncontaminated by actual existence or by the direct reference to first-hand experience. Such logical relations are pure abstractions, fictions no longer embedded in the natural language we use to describe things we experience in the scope of our activities.

Now this seemingly small shift the meaning of logical terms has drastic implications for logic itself - and for logic as a formal study. Traditional Aristotelian logic presupposed an actual subject, ideal or real, to which logical terms or words refer. Traditional logic presupposed that logical thinking is - like spontaneous thought and speech - intimately involved with a real subject matter. Mathematical logic attained its superiority by severing its significance from the conditions in which we make direct statements. Today, logicians like Willard Quine can argue that a concrete and unique individual thing (which we refer to as such) has no more reality than "to be the value of a variable," at least when we consider things "from a logical point of view." The modern logical point of view begins with the system, not with any concrete content. It operates in a domain of pure formality and abstract detachment. The modern logical point of view proceeds from an intricate set of abstract relations having no inherent connection to the things we directly perceive and experience.

We can contrast this type of abstraction with the traditional logic that still swam in the subject matter of direct experience. Traditional logic began with direct statements insofar as the logician presupposed the necessarily existential interpretation of propositions or statements. When we state something in everyday language, we attribute something to something; we attribute the color white to the wall, the quality of mercy to a judge. That is, we speak of what is before us and speak in the context of others who also have access to what we are talking about. We normally assume the existence or at least existential relevance of that to which we address ourselves. Mathematical logic, on the contrary, like mathematics itself, does not concern itself with the actually existing world - not even the world of direct statements. In this sense, mathematical logic operates at a remove from everyday involvement with reality.

The remoteness from everyday, intuitive involvement with things characterizes infomania as it afflicts us in Boolean search logic. Boolean searches on the computer affect our relationship to language and to thought: We are placed at a new remove from subject matter. We are less directly involved in the texture of what we are exploring.

Ok, but is the Boolean interface any less alienating than other procedures? Do not the words we use to pose any question distance us from the wordless subject matter we are thinking about? Does not every logic we use place us in a more abstract relationship to whatever it is we are dealing with? Western logic tends to remove us from any specific subject matter, because Western logic (as opposed to, say, the logic of Chinese Taoism) is a formal logic, an explicit set of rules. The rules implicit in everyday thinking have been made explicit. Once made explicit, the rules function as a kind of feedback which gets recycled into our everyday thought. The feedback effect can be see even in traditional logic.

Traditional or syllogistic logic makes its effects evident, even painfully so, when we observe someone arguing with emphatic logic for a conclusion contrary to our own gut feelings. In such situations, we often feel overwhelmed by the sheer logical power of the argument itself. Logic moves like a juggernaut, apart from any concern with its subject matter. Someone with a great deal less experience than we, for example, can make us feel compelled to accept a conclusion we know instinctively to be wrong. We feel the logical coercion even though we may have much more familiarity with the matter under discussion. Listening to someone like Socrates, or even William F. Buckley, can be disconcerting. We sense the power of thought pushing powerfully through the topic - perhaps even despite the topic. Like mathematics, logic operates outside the intuitive wisdom of experience and common sense. Unlike the wisdom of experience, logic abstracts from particular facts and circumstances. You can be perfectly logical yet be completely insane. Logic by its very nature operates with abstractions.

Abstract by nature, traditional logic too operates at a distinct level of abstraction. The level of abstraction differs with the kind of logic, with the degree of formalism. Modern symbolic logic works with greater abstraction than traditional Aristotelian logic, placing us at a further remove from experience and from felt insight. Through it we stand at a further remove from the subject we are exploring.

Even traditional logic masters our direct experience by organizing it. But Boolean logic on the computer removes thought even further from whatever we may be pondering. If traditional logic aids us in working out thoughts in ways that juxtapose thoughts and argue them out in printed books, then the computer bursts the linear logic of books by making relationship between the thoughtful person and the subject matter even more tangential. Consider, then, how those Venn diagrams "from sophomore geometry" exemplify the existential remove I am pointing out. Circles can map statements connecting universal terms. The statement "All cows are brown" can be mapped with two overlapping circles, one representing cows and the other brown things. Shade in (exclude) the areas that are cows and that do not overlap the circle representing "brown things" and you have mapped the statement "All cows are brown." Add a third circle to represent spotted things and you can map "No brown cows are spotted" or "All brown cows are spotted," and so on. What have we really mapped? According to Boolean logic, no cows or brown things or spotted things need actually exist. All we have mapped is the relationship between sets or classes. The sets could be called cows or custards or quanta.

When adding particular statements to our map, like "Some spotted cows are brown," we must introduce further symbols. Statements about particular things can be mapped on the diagrams by introducing another convention, often a star, or an asterisk, or some other sign. Statements implying that particular members of a class actually exist must be specifically marked as such. This is because the classes or general terms of Boolean logic are potentially empty compartments. This logic stands at another remove from direct statements about things in which we existing beings are actually, personally involved.

The disengagement I speak of is evident too in the domain of logic pedagogy. In college, students learning modern logic must learn how to apply that logic to arguments in "ordinary language." When first learning how to symbolize statements and arguments in symbolic logic, students undergo a lengthy and painful process of converting into symbolism the statements and arguments of the English language. So removed is this logic from our direct uses of language that the textbooks commonly refer to the process of converting arguments into symbols as "translation." Students first must translate their thoughts before logically analyzing them. In order to fit into the system of modern logic, all statements in direct English must first undergo translation.

The translation into empty symbols marks a chronic stage of infomania - especially now that the computer does the translating for us. Instead of the human mind twisting language to fit the system, the computer swallows all symbols and pre-digests them for digital regurgitation. The passage from intuitive content to smooth information becomes invisible and painless. A digital absorbs our language so we can massage and squirt symbols at lightning speeds, or can scan human thought with Boolean search logic. With the Boolean search, it is the computer, however, not the student, that does the translating. That way, the shift is hardly noticeable from the direct awareness of things to the detached world of logical distance. By encoding language as data, the computer has already translated the language we use into mathematized ASCII (American Standard Code for Information Interchange). The computer then allows us to operate with the certitude of Boolean formulas. The logical distance has all the allure of control and power without the pain of having to translate to and from our everyday approach to the things we experience.

Boolean searches places us at a great remove from direct context-embedded language. In order to optimize the search through vast amounts of data, key terms are used to categorize and distill the data. Key terms categorize the topics, general conclusions, and even the quality of the data. To optimize the user's time, the program will gather relevant material quickly before going down to a more detailed level of raw data. The human user saves time by skimming through key terms and by relying on summaries of the data provided by the computer. Since printouts of first-order material cost more, the user comes to rely on the quality judgments supplied by the purveyors of database information. Very often the key terms and quality judgments in the database come not from the hand of relevant experts but from general readers hired to organize the database. The result is a second-hand, low-quality filtration system. Not only do we end up with a digest of potential literature, but we even absorb the prejudices of an unqualified reader. Knowing as we do the vagaries of even expert reviewers, how can we trust the built-in guidance of a non-expert search system? At least one author has already felt shock at seeing his article keyed and rated by an outrageous database. The database by its nature has nothing to do with the qualitative feel of its data, but, unlike the Dewey Decimal System, the database appears to organize and prioritize at the level of significance.

This is our first observation. It remains a preliminary one and may seem to some extent self-evident. It does not plumb the depths of the question, How are we affected by the increasing importance of the Boolean search? Many of our textual tools are far from linguistic in nature, many of them operate "at a great remove from direct context-embedded language." If the Boolean search operates at a great remove from direct oral address, so do pen and paper, not to mention rubber erasers and linotype printing presses. Being non-linguistic tools, erasers operate at a great distance from direct context-embedded language. Does it affect us significantly to use a rubber eraser? Only in a subtle, psychological sense. Teachers point out that learning to use an eraser on your writing is the first significant step toward a self-critical attitude that distinguishes a good from a bad writer. Erasing means we are learning to evaluate our own thoughts and words. In a similarly subtle, psychological sense, Boolean search logic is a significant step in the way we, as an articulate species, approach our own thought and language. In other words, Boolean search logic is part of the psychic framework of electronic text processing. Electronic text processing comprises the symbolic element proper to the accelerated world of information.

What happens with us as we use computers to search through language, as we employ the powerful tools of the information age? Modern symbolic logic enables us, in a double sense, to manipulate language as data. It does so in the first sense insofar as this logic organizes microswitches in the central processing unit. In a second sense, the Boolean search empowers us by applying modern symbolic logic to human consciousness. This essentially spider-like, non-direct logic places us at a new remove from concretely embedded language.

What is language in a condition of abstract manipulation? When no longer embedded in the context of things we address directly, language has become information. The horizon of significance for information differs from that of book language. Digital language exists in a different element. We can contrast the information element with that of traditional books. Books do, of course sometimes contain information. But they do not carry information essentially, as books. In 'Electric Language', especially chapter 6, I described some of the vast differences between the traditional book and the new computer-mediated texts. But here, in Boolean search logic, we see both symbolic elements from a new angle. Information characterizes the psychic framework of computer texts.

Information is not the same as knowledge in books. The difference between information and book knowledge only becomes clear after the computer revolution. Information comes into its own with computerized word processing, because information is essentially electronic, and the word processor - with all its permutations of database and hypertext - is the electronic interface for language. With the computer, language is accessed at speeds below the threshold of human perception. We accommodate these high speeds by adjusting to information. Information is such that the context for its meaning has already been formed. Information subsists in an already formed, already formulated state. Information is an inherently non-creative use of language, and what partially offsets this lack of creative participation is the variety of ways we can access the information. As long as we accept it as information, we can interact with digital language, and we can be creative in this interaction. But the interaction takes place in a symbolic element that differs from books as much as books differ from stone carvings or from papyrus scrolls.

Because they are interactive, computer texts affect us as much as we affect them. We place our demands on the texts through computers, and the software we use in turn channels our range of perception and expectation. There is constant feedback. The term feedback comes from cybernetics and means that one aspect of a system provides information about the status of the system as a whole and the system in turn reacts to that information. I use the term feedback to emphasize the intimacy the human mind has with the computer, an intimacy typical of the postmodern age. >From our contemporary perspective, our ancestors' ways of manipulating symbols seem no more than primitive versions of information storage. We might say that even earlier ways of manipulating symbols also created a distance from the concrete language that is always embedded in a context. Humankind has always used artificial memory, even if only through elaborate verbal narratives like epic poems. But our computers introduce a quantitative leap in distance so that information is something altogether different. In fact, you might say that only with computers has knowledge become information. The computer makes written knowledge interact with us in ways that make information qualitatively different from any previous receptacle or storage medium. Language as information establishes the greatest pre-given distance from concretely embedded language. And this is a distance with a difference.

How then describe the impact of Boolean search logic? How measure its greater distance? One way is to describe the abstractive, cybernetic distance as it affects us individually and psychologically. We could understand remoteness as an activity happening within our own brains, say, on the analytical left side of the brain. Many people today associate the left side of the brain with reasoning and logic. American neurologists examine human activities in terms of left-brain, right-brain work. Teachers apply it to everything from artistic drawing to management techniques. Hemispheric specialization - left-brain, right-brain thinking, the bicameral model of the mind - functions nowadays as a universal metaphor. Boolean search logic might exercise that already dominant left-brain, so that computer interaction enhances a part of ourselves. In a high-tech world, the left-brain is already dominant, and calculation of all kinds is fostered. Our left-brain diligence creates computers, and then computer feedback reinforces that same part of ourselves. Can we come to terms with Boolean search logic by thinking of ourselves as organisms exercising different sides of the brain as we work in the interface?

With the new level of abstraction, we do more left-brain thinking than ever before. Boolean search logic is an instrument of homolateral left-brain patterns. It cuts down on right-brain holistic work by turning us toward logical relationships and abstract connections. It makes us think about the overlapping of categories, of logical exclusion and inclusion. But notice that we always need some amount of holistic guesswork. Boolean searches cannot even begin without our vague hunches and surmises.

We need inklings to start with, and hunches are associated with right-brain activity. Nevertheless, the Boolean search places our hunches in the service of a skeletal logic far removed from the direct operations of language. For the most part, Boolean search logic puts the left-brain to work. But in general, can human-computer interaction be understood on the order of a physical muscle, a brain which we exercise by working out on an external device?

Split-brain analysis has become a staple, even something of a clich , easy and familiar as the old phrenology once was. Phrenology classified mental qualities by the shape of the skull and by the kind of indentations you have on your head. The theory of hemispheric specialization originated with research into brain damage and with studies on how a brain malfunction can relate to organic dysfunctions. The theory is a way of precisely locating physical skills by mapping them onto areas of the brain. It has been a useful metaphor, based as it is on a fundamental dichotomy.

But there is danger substituting a quick metaphor for an investigation that probes more deeply into ourselves, into our psyches. The theory works with a simple dualism: right-brain, left-brain. Dividing things into two sides seems unavoidable for human beings. Our organism comes with built-in symmetry. We get our orientation on the dichotomies of up-down, right-left, good-bad. The very habit of making dichotomies like these stems itself from the tendency we might call left-brain. Ironically, the widespread use of lateral brain terminology shows the deep roots of our intellectual dichotomies and of symmetrical categories. Emphasis on right-brain thinking - or left-brain thinking - can just perpetuate a set of overly simple oppositions and categories.

Our questions about the Boolean search goes deeper than a few psychological side-effects. The question has to do with our stance in the world, with the way we apprehend realities. It has to do with the postmodern world itself as we experience it. Modern thinking designs powerful external instruments to augment natural, in-born capabilities. From the time of Descartes, modern thinking began to build instruments systematically, from telescopes and microscopes to cyclotrons and radio-wave amplifiers. Over the centuries, the technology has expanded to include our thinking. Thought augmentation is not a neutral amplification of a naturally stable faculty, but it is a total reconfiguration of the psyche, of our mental-emotional complex. The theory of hemispheric specialization - left-brain, right-brain - does not go deep enough. The question about Boolean logic must go to the heart of the world we inhabit, to our subconscious processes that model the world we live in. Computerized texts have less to do with our passing individual states of mind and more to do with the whole human world we inhabit.

The split-brain, however, supplies an important clue. A hemispherically specialized brain implies a delicate balance, an unstable dualism that can cripple. For healing lateral imbalances, therapeutic psychology recommends exercises like cross-crawling to offset asymmetries developed in childhood. The need to restore balance and communication between two polarities of the psyche, two sides of ourselves, is an old story.

More than a decade ago, cultural critics saw we were afflicted by a dualism. The "two cultures debate" heated up with the writings of C.P. Snow. Two poles of culture, the scientific and humanistic, seemed at loggerheads, with no middle ground between them. Each culture spoke its own dialect, mutually unintelligible dialects, and each cherished its own priorities. The two cultures debate no longer survives as a living lament about the lost connection between objective science and the poetic humanities. Recent transformations of cultural phenomena has rendered the debate largely moot. The dualism faded as the older positivistic science became a science of less logical coherence and rigidity, so that today's statistics-based speculation would look bizarre in the eyes of the older positivism. With the advent of chaos theory and indeterminacy, even physical laws seem less apodictic. The humanities too have changed. In the fine arts, computers display all manner of graphic design, and the literary imagination is venturing forth into interactive fiction. Writers are joining programmers in engineering new software.

But cultural dualism does not vanish simply and cleanly in a few decades. Ancient problems of the one and the many, of identity and difference, go underground only to surface in new forms. When suppressed and overcome in one guise, they re-emerge in another, sometimes in startling shapes. A new unity is emerging. The computer interface is a common dwelling place for both scientist and humanist. If the computer is the calculator of the scientist, then the word processor is the calculator of the humanist. We must look, therefore, to the depths of the computer interface if we want to discover the newly emerging polarity.

The dualistic metaphor of the bicameral brain replays that older two-cultures debate around which cultural conversation turned a generation ago. Hemispheric specialization suggests that we are afflicted by dualistic polarities, that we suffer from imbalances. The split-brain theory echoes age-old polarities and life paradoxes. But now the dualism is internal to the individual. This dualism affords us an opportunity to go more deeply into our roots in modern philosophy where the technological interface first emerged.

A deep, unsettling polarity was present already at the inception of the computer. Polarity tormented the computer's inventor. Dualism haunted the mind of Pascal. Blaise Pascal (1623-1662) engineered and produced the first automated calculating machine. His Pascaline used eight rotating gears and wheels to automate operations of addition and subtraction.4 No accident, then, that PASCAL spells the name of a high-level programming language, in tribute to the inventor of automated calculation. More important, Pascal names a certain state of mind haunting the interface he invented.

PASCAL is a metaphor for the double nature of contemporary thought. It is a programming language that also names a mind transfixed by religious mysticism. Pascal doubted the pretensions of the rational interface that he - and we - built. For us, Pascal is both a high-level computer programming language and the name of a thinker who struck the first chords of existentialist philosophy. The computerized text conceals the major struggles of the human spirit that first dreamt the modern period. In asking what is happening with us as we use computers and databases, we are also asking about our own intellectual forebears. They bequeathed to us not only the outlines of computer hardware but also the general attitude we need to use computers in organizing things. In the case of Pascal, we are also bequeathed the doubt that the probes the computer interface.

Even a brief intellectual profile shows the dualism in Pascal's mind. He was a brilliant logician and mathematician, who enjoyed a correspondence with friends like Fermat and Descartes. Pascal pioneered the scientific method, applying traditional deductive reasoning to experiments on atmospheric pressure and on the nature of the vacuum. An intellectual achiever, Pascal nevertheless thought much about the fragility and limitations of the human mind. He defines the human being as a "thinking reed," as a center of thought beleaguered by floods of emotional weakness and by the winds of changing moods.5 Today Pascal is known as one of the first philosophers to use reasoning to show the inadequacies of reason. As an explorer at the interface, Pascal doubted whether technology could answer our deepest needs.

Pascal resembled many mathematical minds today. Like many logicians and mathematicians, Pascal set himself rigorous standards of clarity and verifiability. Measured by mathematical standards, much of our thinking appears irrational, arbitrary, and without proper justification. Most of our spontaneous thought processes elude deductive patterns, and placed against the standards of mathematical logic, most thinking looks like mysticism. For Pascal, mystical insight was the single alternative to reasoning. By admitting a chasm between rationality and mystical insight, Pascal became the model not only of the contemporary mathematicians but also of the contemporary existentialist. The either/or dilemma, forking off into either science or mysticism, includes existentialists like Heidegger and Sartre as well as analytical types like Bertrand Russell and Ludwig Wittgenstein. Mathematical minds are often so dedicated to the purity of rational thought that anything else seems dark, unprovable, mysterious. You can guarantee rational purity if you stipulate meanings and postulate procedures, but leave that clarity and you feel blind venturing into the world at large. So Bertrand Russell, the mathematical-empirical rationalist, leads the first "Ban the Bomb" movement and declares the ultimate premise of his position an inexplicable conviction of rightness.

Pascal's Dualism runs through Modern Rationalism.

Pascal was of two minds. He could write sophisticated studies of the mathematics of the cycloid and exchange scientific correspondence with Christopher Wren and Christian Huyghens. At the same time, he could compose religious arguments in his 'Provincial Letters' and explore contemplative depths in his famous notebook of 'Pens es' ['Thoughts']. The notebook, for which he is most well-known, probes the dualism within himself. For instance, he writes:

"He's a good mathematician," you say. But I'll say "I have nothing to do with mathematics." Otherwise you might take me for a proposition.6

Refusing to be reduced to a logical proposition, Pascal tried repeatedly to prove that his devotion to a radical Christianity had from no rational justification. Reasoning was beside the point, argued Pascal, because faith can only be considered a matter of the heart, not of the mind. Hence his famous dictum, "The heart has its reasons of which reason knows nothing."7

In Pascal, mind and heart split, reason and emotion parted company. Pascal originated the self-standing, automated reasoning device. Pascal also names the thinker who most deeply doubted the powers of reasoning. Because of his skepticism, then, Pascal symbolizes both the origination of the interface in which we work as well as the deepest doubts we have about the interface. Pascal reminds us of the limits of rational civilization. The name Pascal guides our question about the Boolean search.

For the historical Pascal, science was objective and one's personal life could be expressed only in emotional terms. Pascal was his own best observer - especially as he observed his personal thinking, which belongs not to science but to private moral life. His 'Thoughts' proved to himself that his revealed religion could not be set down with the precision of mathematics. Nor could any faith or moral belief meet the demands of logic. Our seventeenth-century French skeptic is a moral rigorist who makes stringent demands for rational organization but cannot view his life foundations through any rational perspective. He sees human nature at war with itself:

An inner war goes on between reason and the passions. If humans had only reason without passion... If humans had only passion without reason... But having both, humans cannot exist without conflict, being unable to be at peace with the one without being at war with the other. So human beings are always inwardly divided and self-opposed.8

Truth is a harmonious, impervious, deductive fortress. Humans get to visit but must spend their lives outside it, troubled by the memories they carry home.

Pascal saw life differently from Descartes, whose dualism split animal/reason and organism/machine. Pascal too admired rationality, mathematics, and machines, but, unlike Descartes, Pascal had an interest in subverting rationality with subconscious, intuitive processes. Like a good modern Cartesian inventor, Pascal sought methods to filter out the irrational animal organism. At the same time, he acknowledged the rights of what had been suppressed in the name of reason. "The arithmetical machine produces effects which approach nearer to thought than all the actions of animals. But it does nothing which would enable us to attribute will to it, as to the animals."9

Pascal dwelt on the animal life within himself, in contrast to Descartes, who learned to direct his intelligence outward to mechanisms. For Pascal, volition issues from impulses opaque to thought, and the intuitive process follows behind reasonable systematic order like a hidden twin. "Reason's last step is to realize that there are an infinite number of things which lay beyond it. Reasoning is puny if it does not realize this."10 Pascal took this last step. He not only realized the feebleness of reason but also explored the realm of intuitive mysticism where certainty comes in unexpected, non-deductive, non-rational terms.

The duality splits cultures and personalities. Pascal describes two personal or cultural styles, two fundamental ways of being human: the mathematical attitude ('l'esprit de geometrie') and the intuitive attitude ('l'esprit de finesse').11

The mathematical attitude insists on definite, clear principles, delighting in the step-by-step pursuit of the outcome of principles. The mathematical attitude moves from the clarity of a small number of things to the implications contained in that small number. By contrast, the intuitive attitude feels or senses more than the clarity of a few points. It notices a multitude of intricately connected and delicately organized complexes of detail - "tacitly, without technical rules." The intuitive mind can hardly express how or why it proceeds as it does. Rather than going from premise or principle to conclusion, the intuitive mind proceeds by "a logic of digression," circling around a point and returning to it without ever striking it squarely and explicitly.

The two styles run parallel without ever meeting.

"Instinct and reason, marks of two natures."12 Pascal believed he was describing two human or cultural styles, but his description goes beyond the depiction of immediate mundane things. His insight suggests more than human or cultural types. The two intellectual attitudes disclose two sides of the interface he helped build, and these two types constitute an insight into the modern project to set up a interpenetration of human and machine.

The Pascalian insight suggests two aspects of the human psyche required for the marriage of human and machine. As a modern, Pascal was enthralled by the power of this dualism. By dividing the rational from the intuitive, he could fashion instruments that replicate rational procedures. Once fashioned, the instruments could amplify natural, intuitive activities, thereby enhancing human power. Pascal was dreaming of the cyborg, the implantation of rational devices into the biophysical organism. On the logical level, he was searching for a system to replace the weaknesses of human reason. Pascal shows the split in the psyche as it relies on the technological interface.

One side of the psyche is the rational order inscribed in the computer's microcircuits. This is the modern symbolic or mathematical logic discussed earlier. The rational side of the interface has many levels, such as various languages in which the computer is programmed. The languages are stipulated, artificial creations of conscious reasoning. The higher-level languages provide the windows, as it were, through which encoded textual material can be stored, manipulated, viewed, and linked.

The other side of the psyche is the indeterminate, spontaneous, creative side. It is not fixed but moves in fluid, unpredictable ways - constantly inconstant. The ancient Greeks called this side of the psyche 'pathos' and the Latins 'passio' or passion, both from the verb 'pathein', meaning "to suffer, to accept passively," and "to be affected by." This is a receptive side that is affected by what goes on at the moment. It is affection. Taken together, the logical and the affective sides of the psyche constitute the interface as a whole.

Our thought is always sandwiched between the logical and affective sides of the psyche. But now our logical thought takes place at the interface. So does much of our natural-language thinking and communicating. Even when our thoughts move to the steps of traditional logic, they do so increasingly on devices which assimilate everything to mathematical and Boolean logic. Why could this new position of the psyche impact on thought? Because the psyche is constantly in motion. The affective side of the psyche does not stay put but ranges back and forth and into the logical. Being affective, it can be affected strongly by the logical, especially by the logic assumptions of the interface. Pascal calls affection or predilection "the logic of the heart," 'la logique du coeur'.

Sometimes we assume, like naive Platonists, that our minds do - or at least ought to - move through every condition without ever being affected by any condition. We presume to be god-like judges of the world, supreme and uncontaminated by the world around. Pascal mocks such presumption:

"The mind of this supreme judge of the world is not so independent that it is impervious to things going on nearby. It doesn't take a cannon blast to stop a human thought. The creaking of a pulley or a weathervane will do. Don't be surprised if reasoning is not too sound at the moment - there's a fly buzzing around my ears."

"That's enough to incapacitate good judgment. If you want us to be able to find the truth, drive away the creature that is paralyzing our reason and disturbing the mighty intelligence that rules over cities and kingdoms. What absurd gods we are! What ridiculous heroes!"13

Inconstancy, even arbitrariness assails human thinking. Pascal shows how doubtful our faith should be in the sovereign powers of the mind. We may believe we command our own thoughts, but this belief is an illusion brought on by thought itself. We are not skippers of the stable ship of thought headed firmly on a self-determined course. Because of the weatherhelm of passion, there is a difference between the course we steer and our actual course. The origins of thoughts are obscure and the constancy of thinking uncertain. In fragments that could have been written on a word processor, Pascal observes: "Chance gives rise to thoughts, and chance removes them. No art can keep or acquire them. A thought has escaped me. I wanted to write it down. I write instead that it has escaped me."14

Memory and attention fade for all minds. No device can eliminate the essential fluctuation. Pascal calls this the wave of "nothingness," the vacuous emptiness that is always poised to overwhelm us: "In writing down my thought, it sometimes escapes me. But this makes me remember my weakness, that I constantly forget. This is as instructive to me as my forgotten thought; for I strive only to know my nothingness."\15\ Inconstancy and the void appear elsewhere in Pascal.\16\ The breath of nothingness,'le n ant', blows coolly on the neck of human thought.

In the fear of nothingness Pascal found the human need to depend on external aids to mind and memory. This nothingness gave impetus to augmenting thought with the calculating machine. This nothingness was the felt gap that originally opened up the interface. Pascal's search to overcome inconstancy sought stabilizing support in mechanical systems, in the reliable rationality of the machine. His research into thinking machines was the other side of his fear of the nothingness that threatens human attentiveness. The interface was a flight into systematic thought, further splitting intuition from reason.

Pascal keeps the nothingness in mind. Unlike cognitive theorists today, Pascal does not forget the nothingness that first impels the creation of machines to mimic mind. The theoretical attempt to model the mind on its computer creation can only arrive later, once the machine has reached a high degree of development. Only after human energies construct artificial intelligences does it seem possible to think of mind as computer, to forget the desperate, annihilating experience which gives rise to the creation of computers.

The nothingness at one's back can also be an ally of creativity. On the negative side is helplessness and ennui. On the positive side is creativity and freedom. You can fight nothingness by configuring autonomous systems. But nothingness accepted works also as a positive chaos against which the human apprehension of things gains definiteness and vivacity. The 'n ant' or nothingness of Pascal is also a rich emptiness. It is omnidirectional potential. This nothingness is an open space holding apart the duality, the in-between where mind and machine meet. Modern information systems are never so air-tight that they shut out the unsettling nothingness that breeds novel thinking. The spur for new formulations of things pierces through routine information gathering.

We now begin to locate the impact of Boolean search logic. Boolean logic offers the mind an occasion for achieving closer conformity with the power of computerized devices. The mind can identify more fully with a closed system of rational control. Boolean search logic offers a bridge between human questioning and the texts written by humankind. Boolean logic invites the illusion that human knowledge and experience become available primarily as information. Information is essentially something we can control and retrieve without regard for creative reformulation.

Boolean search logic is a kind of question. Through it we pose questions to texts as if we were the sovereign lords over the world of words. The Boolean search puts questions to written texts as data, as information. Every question posed is also a net for catching - in its own set terms - all possible answers to the question. Every Boolean search imposes a range of possible findings. In a sense, then, each search contains something of its own answer, because the range of discoveries is limited by the terms of the search. This is especially true of Boolean search logic. The more remotely a logic stands apart from the material it operates upon, the more unilaterally that logic limits interaction with the material on which it works. As I argued earlier, modern symbolic logic is especially removed from direct statements and from the inner landscape of thought. Remoteness pervades the interface. Language as information is remote from the primal uses of language, which include the shared context in which one addresses oneself to things directly.

Dealing with texts through Boolean searches enhances the power of conscious, rational direction. Rationality in this sense is not the contemplative, meditative perusal of a line of thinking, as searching through books can be. Books exercise linear thought processes and move sequentially, page-by-page, chapter-by-chapter. Of course, books do not have to be read in this sequential a fashion, but they nevertheless present themselves as linear. Even though books can be objects for serendipity and creative browsing, they present themselves as instances of sequential thinking, of natural non-Boolean logic.

With Boolean search logic, keywords move us through texts. We move with incredible speed through the mass of ideas conceived by our species. But through the abstract nature of our questions, our psyche no longer opens to the delicate whisperings of adjacent thoughts, to preceding and subsequent passages. In fact, the abstractness of computer access shows up in the increasing use of abstracts or the online pr cis. Rapid searches need predigested abstracts or condensations of texts. The abstract is a textual version of the Boolean search. Instead of our search terms governing all possible answers, texts have been reduced to the most compact of terms. The pathways of thought, not to mention the logic of thought, have been paved over. A meandering search through the wild country is replaced by an organized arrangement of freeways connecting predetermined points, offering fixed exit ramps. In many cases the abstracts are abstractions made by someone other than the original author. Or, maybe we should say, the notion of authorial integrity has been eroded by the linked network of computer writing, the malleability of text, as well as the new modes of access.

The Boolean search often means that texts are treated as data. When you search a database, you browse through recent material, often no more than the last ten years. Cutting off the past streamlines the search. But reflection with sheared off historical roots is a narrow reflection which no longer exposes false starts, abandoned pathways, unheard of avenues. An exclusive focus on the recent past curtails our mental musings, and active but narrower awareness sacrifices the intuitive mind.

Boolean search logic affects our mental vision. Eyes are passive receptors of the world we experience. Only when we strain to see do we fail to be surprised by perceptions. Constant straining induces a sensory myopia in which we need to strain in order to see better - what we 'wish' to see. We lose the depth of our peripheral vision when we see willfully. Likewise with the mind's eye. A relaxed and easy thought enjoys intuitive turns, and thinking at its best muses over human symbols. Boolean search logic cloaks the peripheral vision of the mind's eye. It is like the artificial lenses that help us persist in our preconceptions.

Boolean logic can unconsciously entrench us in our straining ways. We may see more and see it more sharply. But the clarity will not have the rich depth of natural vision. The world of thought we see will be flattened by an abstract remoteness. The mind's eye will, through its straining, see a thin, flattened world with less light and brightness.

Enhanced control, like all power, exacts a price. The cost is a tendency to lose touch with creative thought processes. I mean processes like the contemplative musings once fostered by the era of books and reading. If computers expand in their power to aid our searching minds, we will increasingly relegate books to special leisure time. The serendipitous search through books is less and less necessary for knowledge and learning. Scholars who are traditionally book people now use computers. With books, no programmed structure filters access to texts, only the linear thought patterns set by the author within a limited, self-contained volume. A linear thought pattern is not the same as a program. Programs are windows of access, they interpret the digital code so we have access to texts. The power of computer searching puts us squarely in the electronic element. The linear order of the book accedes to the will of the reader. The linear structure of printed books comes to the reader in a format that fosters new contexts of thought. A reader can explore a shelf of volumes like a long corridor that invites separate journeys to different times and places. Unexpected finds in a paper library lead to creative browsing, not just more information.

Of course, the computerized reader can devise a "search strategy" and browse among related items. In Glossbrenner's 'How to Look it Up Online' I cited earlier, we learn that step three of the computer searching is:

Meditate. Seriously. You may not be a Ninja warrior preparing for battle, but it's not a bad analogy. If you ride in like a cowboy with six-guns blazing, firing off search terms as they come into your head, you'll stir up a lot of dust, expend a lot of ammunition, and be presented with a hefty bill but very little relevant information when you're done.... 'Think' about the topic beforehand. Let your mind run free and flow into the subject. What do you know and what can you extrapolate about the subject?17

But this kind of meditation only sharpens an already determined will to find something definite. Meditation comes from the Latin 'meditari', which means "to be in the midst of, to hover in between." The search meditation Glossbrenner prescribes - and wisely so - is more a zeroing in on a target than a floating awareness open to new creative insights. The meditation of the computer search is more like an automotive trip over well-routed highway systems than a meandering hike of discovery. Browsing among traditional books, the old-fashioned book searcher welcomes the surprise of new terrain, new contexts where the angle of thought shifts. The programmed reader knows in advance there are certain definite exits, on-ramps, and predesignated stops. The only task for meditation is to insure adequate advance preparation. Computer search meditations should not be confused with the contemplative reflection that rides gently above its own immediate aims.

Browsing among traditional books often evokes daydreams and unsuspected connections. Analogies and pertinent serendipity happens among the stacks of physically accessible materials. Though not as efficient as the Boolean search, library browsing enriches the psyche in unpredictable ways. Looking for something in a book library frequently leads to discoveries that overturn the questions we originally came to ask. Libraries are an unsystematic, unfiltered collections of human voices and thoughts. Libraries are repositories not so much of information as of the intuitions of countless authors. The books in libraries are physical reminders of the individual voices of the authors, authors who speak to us in ways that shock and disturb, in ways that break through assumptions and preconceptions, in ways that calm and deepen. The word museum derives from the Greek word for the Muses, goddesses of dream, spontaneous creativity, and leisure geniality. Libraries may be the last museums of the stored language.

"Not to muse is to think," wrote a Chinese sage. He goes on to warn: It is a pity that our competitive Age of Institutions has placed such a premium on the technique of thinking that, as a cultural group, we are rapidly losing the art of musing. When was the last time that you sat before the fireplace on a crisp winter night or lay on the grass on a balmy spring day and enjoyed some delightful and thoroughly permeating musing?18

Thinking guided by the computer will grow narrow and sharp. But sharpness is not all. Taoist philosophers would have us keep in mind a more natural state, a more relaxed mental openness in which things can emerge unplanned and unexpected. Rather than consciously willing, we can enter a state of No-Mind - no particular or consciously narrowed aim governs our attention. The musing state of mind operates on a plane that is more sensitive and more complex than consciously controlled thought. It is not wild in the sense of undisciplined or wanton. Musing is "not thinking" because precise thinking is merely a subset of the creative mentality.

Libraries today are ceasing to be places for musing. They are increasingly information centers, research areas. At the Los Angeles County Public Library, the world's largest circulating library, requests for information have now grown to exceed the requests for books.19 At least one university has in 1989 christened the first library without books, a building for perusing electronic texts. Books still prevail as the primary element of thought, but they are increasingly sources of information. Large sales figures and the volume of book acquisitions do not necessarily indicate that the psychic framework of the book endures as such. Today, many books gain respect as soon as they achieve status as non-books by becoming material for films, cassettes, or other media.

The surrounding electronic element will continue to affect books. Searching through books was always more romance than busyness, more rumination than information. Information is by nature time-bound. Supported in technological systems, information always needs revision and updating. When books become mere sources of information, they lose the atmosphere of contemplative leisure and timeless enjoyment. Old books then seem irrelevant as they no longer pertain to current needs. One of the new breed of information publishers epitomizes this attitude in a pithy warning: "Any book more than two years old is of questionable value. Books more than four or five years old are a menace. OUT OF DATE = DANGEROUS."20 As book libraries become museums of alphabetic life, we should recall the original meaning of museums.

Museums are places for play, for playing with the Muses that attract us, for dreams, intuitions, and enthusiasms. Information may gear us solidly into the world of computerized productivity. But the open space of books remains the preserve of psyche.

Pascal reminds us that he will have nothing to do with mathematics - lest we take him for a proposition. We must be equally careful not to take Pascal for a computer language.

Notes:

1. 'How to Look it Up Online' by Alfred Glossbrenner (New York: St. Martin's Press, 1987), p. 109.

2. Chapter 3 of my 'Electric Language: A Philosophical Study of Word Processing' (Yale University Press, 1989, c1987) treats this point. See also Heidegger's book on Leibniz, 'The Metaphysical Foundations of Logic'.

3. Boole's main contributions are found in his 'The Mathematical Analysis of Logic, Being an Essay Towards a Calculus of Deductive Reasoning' (1847) and 'An Investigation of the Laws of Thought on which are founded the Mathematical Theories of Logic and Probability' (1854).

4. Pascal's was a two-dimensional machine, using cogged wheels rotated in proportion to numerical values that were added or subtracted. Multiplication or division had to be accomplished by tedious repetition of these same operations. The calculating machine invented later by Gottfried Wilhelm Leibniz (1646-1716) was an improvement in three dimensions, using wheels at right angles that could be displaced by a special stepping mechanism to perform rapid multiplication or division. For photographs of these two machines, see 'Digital Deli', p. 9, (New York: Workman Publishing, 1984), edited by Steve Ditlea. A diagram of Pascal's calculating machine appears in Ren Moreau's 'The Computer Comes of Age' (Cambridge, Mass.: MIT Press, 1984), p. 137.

5. 'Roseau pensant' or "thinking reed" is fragment 348 in the Brunschvicg edition of the 'Pens es', (Paris: Librairie Garnier Freres), and in the English translation of the Brunschvicg edition by William Trotter (New York: Modern Library-Random House, 1941), p. 116. For other fragments, I first give the numeration of the Brunschvicg edition and the page number from the Trotter Modern Library translation. I made the translations from the French after cross-checking with the Krailsheimer translation published by Penguin Books.

6. "C'est un bon math maticien,' dira-t-on. - Mais je n'ai que faire de math matiques; il me prendrait pour une proposition." Fragment 36; p. 14 in the English translation.

7. "Le coeur a des raisons que la raison ne connait pas," fragment 277. Pascal defended a group of ascetic Christians who inhabited the ancient Cistercian educational foundation at Port-Royale des Champs. These austere Jansenists, Pascal's educators, organized their version of Christianity around the treatise 'Augustinus' written by Bishop Cornelius Jansenius, Bishop of Ypres (1585-1638). The treatise was published two years after the Bishop's death and it gave the impetus for the long Jansenist controversy with the Jesuits.

8. Fragment 412; p. 130.

9. Fragment 340; p. 115.

10. Fragment 267; p. 93. In French: "La derni re d marche de la raison est de reconnaitre qu'il y a une infinit de choses qui la surpassent; elle n'est que faible, si elle ne va jusqu'a connaitre cela."

11. Fragment 1; p. 3.

12. Fragment 344.

13. Fragment 366; p. 120. How wonderful the French: "L'esprit de ce souverain juge du monde n'est pas si ind pendant, qu'il ne soit sujet a 'tre troubl par le premier tintamarre qui se faitautour de lui. Il ne faut pas le bruit d'un canon pour emp'cher ses pens es: il ne faut que le bruit d'une girouette ou d'une poulie. Ne vous tonnez pas s'il ne raisonne pas bien pr sent; une mouche bourdonne a ses oreille; c'en est assez pour le rendre incapable de bon conseil. Si vous voulez qu'il puisse trouver la verit , chassez cet animal qui tient sa raison en chec et rouble cette puissante intelligence qui gouverne les villes et les royaumes. Le plaisant dieu que voila! O ridicolosissime eroe!"

14. Fragments 369,370; p. 121. "La m moire est n cessaire pour toutes les op rations de la raison."

15. Fragments 372 and 372: "En Ecrivant ma pensée, elle m'échappe quelquefois; mais cela me fait souvenir de ma faiblesse, que j'oublie toute heure; ce qui m'instruit autant que ma pensée oubliée, car je ne tends qu'a connaitre mon néant."

16. For example, 125 ff; p. 46 ff.

17. Page 116 in How to Look it Up Online cited above.

18. R.G.H. Siu in Ch'i: A Neo-Taoist Approach to Life (Cambridge, Mass.: MIT Press, 1974), p. 25.

19. For the first time in the 75-year history of the Los Angeles country library, during the fiscal year ending June 30, 1987, the total asking of questions for information exceeded the number of books checked out. In the previous fiscal year, 11,189,314 questions wre asked, and 11,795,130 books were checked out. But now the figures have reversed-12,320,000 questions compared with 12,217,000 checkouts of books. See the Los Angeles Times' for October 15, 1987, Part V, page 1, "The New Library is Answering the Call." In 1984-85, the Los Angeles city central library had a total of 894,000 books and other materials checked out but there were 5.9 million questions answerd.

20. Found on the first pages of 'Legal Care for your Software' by Daniel Remer and Stephen Elias, (Berkeley, California: Nolo Press, 1987). The book is a legal guide for computer software writers and publishers, but it underscores the shift to information publishing.