The Colorado Association of Law Libraries - A Chapter of the American Association of Law Libraries


From Creativity to Computer-ese, or Thinking Like a Lawyer in the Computer Age
(Originally Delivered as the 21st Annual Austin W. Scott, Jr., Lecture on October 17, 1996)

Barbara Bintliff, Library Director and Associate Professor,
University of Colorado School of Law

We all know that computers have had a significant impact on law practice, on legal education, and on the substance of law itself. Rare is the law office today without word-processing and automated billing capabilities, forms or documents on disk, and access to a computer-assisted legal research system. There are even law offices that have decided to do without paper-based records entirely, relying on computers to store all their files and documents, e-mail for most of their communication needs, and WESTLAW or LEXIS or other computer-based research system to satisfy their research needs.

Similarly, law schools of today differ significantly from those of even a few years ago. Computer labs have replaced typewriter rooms. We used to keep our CALR systems in locked rooms; only librarians had the keys and the passwords. Now LEXIS and WESTLAW training is mandatory for all students, who, together with the faculty, are given individual passwords and unlimited, free, access time, and who are encouraged to access the databases from home. Networks provide access to CD-ROM resources. Law faculty have PCs in their offices, and communicate via listservs. Classes have e-mail discussions and World Wide Web pages. Desktop publishing and word processing have changed the face of scholarly publishing.

And, of course, substantive law itself has been affected. Copyright, trademark, commercial, and privacy law will never be the same. Telecommunications law is a booming specialty. Now, other technologies-telephones, typewriters, fax machines-would seem to have had a similar impact on law and legal practice. They, too, improved communications, allowed for the rapid transmission of information, and caused changes in the substantive law. From a superficial, mechanistic, point of view, computers seem to have had little more effect than these other devices. On closer inspection, however, we will see that the computer's impact could be greater and more fundamental than almost any other development of the last hundred years.

As we move from a book-based to a computer-based profession, as we come to rely more and more on electronic means for finding and organizing the law, we are also moving to an entirely different way of thinking about legal problems. We are moving from a rule- or concept-based system to a fact-based system, in which more importance seems to be placed on matching the fact situation than on identifying and analyzing the law. The very concept of "thinking like a lawyer" is being changed by, and adapted to, the abilities of the computer.

By thinking like a lawyer, I am referring to the method of organizing and categorizing the parts of a legal problem, to allow for their discussion and possible solution, using a logical reasoning process. Thinking like a lawyer, in the traditional sense that it is taught in most law schools, encompasses finding, analyzing, and applying "the law." Its steps can be listed as follows:

1) We start with an identification of the facts of a situation. Using those facts,

2) We identify the area of law involved and legal rules that are suggested by the situation, most commonly by searching for case law.

3) We then analyze the rules, and apply the rules to our facts to predict an outcome.

4) We look for cases with similar factual situations that have applied these rules in order to see if the outcome we have predicted is reasonable.

It's not unusual for a legal problem to be broken into several discrete pieces, each being dealt with in the above manner. We may find that the process is one of refinement, as new facts suggest new rules, which may change our predicted outcomes. In the process of understanding the rules and analyzing their meaning, we begin to craft the arguments we'll use to persuade a decision-maker to agree with our interpretation of the law. The process allows us to use knowledge, intellect and creativity in the course of solving problems. It is the essence of the practice of law.

The foundations of thinking like a lawyer are laid in the first year of law school, primarily through the analysis of cases using the teaching approach called the "Socratic method." It is traditionally employed as a dialog between professor and student. The Socratic method consists of a reasoning process, moving from the specific facts of one or more cases to a general, governing concept, narrowed into a legal rule. Then, the professor begins to ask a series of questions, hypotheticals, each with slightly different facts, allowing the students to determine whether the appropriate rule has been identified, and to test the extent of its application.

Legal questions, whether in law school or real life, are rarely solved with a quick "yes or no" answer; solutions take time and effort. Thinking like a lawyer allows us to explore possible solutions to client's problems within a framework of legal rules and concepts, using a Socratic dialog-like thought process. We begin by identifying rules suggested by our facts at hand, then search for those rules of law in other cases to see how they have been applied. Once we understand the nature and context of the rule, we look for factually similar cases, to test the limits of the rule in our specific situation. The framework, our organization, is legal concepts refined into rules-that which we call "the law."

Thinking like a lawyer places a premium on the identification of rules, and on approaching a problem within a legal framework. We can counsel clients on anticipated outcomes if we have identified the correct rules. We can suggest courses of action, recommend strategies, and explore alternatives, all based on known and understood rules. Over the years our legal reasoning process has changed, from emphasizing an almost exclusive application of formal rules to recognizing that outside forces may influence the application of rules. But the identification and application of rules has been the main thrust of legal reasoning for over a century.

Public opinion polls show an almost unwavering confidence in the integrity of the judicial system. Unlike Congress and the legislatures, the Presidency and the governors, schools, and religious institutions, all of which have experienced ebbs and flows in public confidence, the courts have maintained a fairly even keel. They are repeatedly described as fair, impartial and stable institutions. Their moral authority remains high. With our emphasis on predicting outcomes based on existing rules, we have contributed to a sense of predictability and stability in the law. For we, as attorneys, give the courts their raw materials in our briefs and written documents, and in our oral arguments.

Thinking like a lawyer. That's what we spend three years in law school learning. It's what we use every day in our legal practice, or wherever our professional lives take us. It's certainly what law faculty spend a lifetime teaching. It's the fundamental skill of lawyering, the skill that sets an attorney apart from a paralegal, or a layperson. And it's what's changing with computers.

This change is well illustrated through an examination of the basic skill of legal research. Now, for whatever inexplicable reason, legal research isn't considered particularly glamorous. Its challenges and its necessity are not always fully appreciated by those who have time pressures and competing priorities. But it is an integral part of thinking like a lawyer, for it is the way we find out what "the law" is, and the way we begin the process of understanding its application to our situation.

The law today comes in many forms; statutes and administrative regulations probably play a larger role in the daily life of the average citizen than does case law. However, it is in the written decisions of the judicial branch that the law, whether statutory, administrative, or common law, is explained and interpreted. When a question arises about the proper application of a statute or regulation, parties must look to the courts for guidance. It is, after all, the judicial interpretation of a statute or a regulation, rather than its bare language, that becomes the precedent, and is binding under the doctrine of stare decisis. When lawyers are called upon to find and interpret the law, in whatever form, for their clients, rare is the research project that can be completed without resort to court decisions.

Furthermore, reading and discussing court opinions is the primary method by which law students are taught the legal reasoning process-the process we refer to as "thinking like a lawyer." Finding, reading, and analyzing opinions hones the legal reasoning skills of law students, and helps prepare them for a career during which they will use these skills repeatedly. Court decisions are generally marked by a discussion of the reasons by which a result is reached, explaining and analyzing competing strands of thought and various rules, using much the same process as attorneys do when considering how to proceed with a given fact pattern. Decisions provide a framework for learning the law, both in school and in practice. Because the use and understanding of court decisions is so fundamental to the practice of law, I am concentrating on them as I discuss the manner in which computers have impacted legal thinking.

For well over a hundred years, our "thinking like a lawyer" skills have been shaped by and, as some would argue, even determined by the simple device of the case digest. While there are a number of digests in existence today, and there were many others in the past, the most influential system has been that developed by John West, and the West Publishing Company-the American Digest System. It is recognized as the comprehensive digesting system, for all reported cases in America. The classification scheme used by West has organized the law, and guided our thinking, since 1896.

As early as the 1830s and 1840s, legal publishers and individuals were experimenting with ways to organize the growing body of case law in America. By 1860, the Abbott brothers of New York had devised a classification system for New York cases. Based on this system, they began to revise a digest of American case law, the United States Digest, published by Little, Brown and Company. In 1889, West Publishing Company, a growing court reports publisher in Minnesota, purchased the United States Digest.

West refined the Abbott's classification system, reducing its major conceptual categories from ten to seven "grand categories" (in order of importance: persons, property, contracts, torts, crimes, remedies, and government). It divided the categories into major subdivisions (for example, persons is divided into "relating to natural persons in general," "particular classes of natural persons," "personal relations," etc.). These subdivisions were then organized into the various digest topics with which we are familiar (approximately 430; "relating to natural persons in general," the subdivision of the grand category persons, is divided into topics such as "civil rights," "dead bodies," "domicile," etc.).

Topics are then organized into subtopics, expressing various actions or concepts, and eventually refined to discrete "points of law." This is the level at which we find our case digests, the annotations or brief paragraphs explaining how a rule, "the law," is applied in a particular case. This is a very hierarchical structure with deeply layered levels. It can represent several ideas at once. The refinements of each topic work in much the same way as a Socratic dialog-narrowing and testing the limits of each concept, based on varying fact patterns. These smallest refinements of each topic, the specific points of law, have been assigned a number, referred to as a key number. There is a strict order of precedence by which cases are assigned to a particular topic; in general, cases are placed into the topic that is in the highest "grand category" possible. This system permits the digest to maintain an intellectual coherency despite its massive and rapid growth.

For many years, we learned to find case law rules by searching through the West digests. Using a combination of legal terminology and major "fact words," we located the key number-the specific point of law-suggested by the facts of our situation. Then, by sorting through brief descriptions of the many other cases included under that same point of law, we were able to identify the rule or rules which were applicable to our situation. At this point, we spent time developing arguments and predicting of outcomes. Finally, we could identify cases with similar facts under that key number, that specific point of law, which we could use to buttress our argument for the application of the law.

By scanning through the related key numbers, we found cases that expanded or narrowed our rules, giving us a better understanding of the context and nature of the rules. We found helpful and problematic cases, allowing us to test our interpretation of the rules. We could find old cases, allowing us to trace the development of the rules. We could find recent cases, affirming our own thought processes. Lawyers were able to identify and assess minor differences in the language and phraseology of opinions, and be flexible and creative in drawing parallels and distinctions. The print allowed us to see and understand the complex relationships between the words themselves. We could tell when our arguments had been used, and when we were pushing the envelope of interpretation through the use of innovative logic. Sometimes the best digest research was the result of serendipity. Something seen out of the corner of your eye suddenly inspired a thought-provoking argument. The limits of a rule were relatively clear, because the law was right there on the printed page.

The West digest system is, and has been, a tightly controlled method of organizing American case law despite its size (over 100,000 separate key numbers; digesting close to 3,000,000 cases). It is comprehensive, covering all published cases (which generally means appellate level cases). Its outlines of topics provide a syndetic structure for each area of law, allowing researchers to understand the relationship, context and hierarchy of identified rules. To use the digest, you have to think in terms that match its organization; you have to think of rules and hierarchies. The digest's organization follows the same pattern as our legal reasoning process, and has almost come to be the physical manifestation of thinking like a lawyer.

Critics of the topic and key number system are many, and they have many valid points. The system was developed by individuals in another century, and does not always reflect today's legal developments. It is considered inflexible in dealing with new topics. There are occasional errors in assigning headnotes, and sometimes a case is "lost" by being classified in the wrong place. Digests are the product of human judgment, and many consider them rife with the biases of those who compile the information, particularly given West's policy of "normalizing" certain judicial language. There is too much information from too many cases, and it is difficult to find cases with similar facts. It uses a cumbersome, multi-step process of indexes and bound volumes, pocket parts, supplements, reporter volumes and advance sheets. Several regional digests have ceased publication, forcing researchers into the extremely unwieldy and inefficient Decennials. The digests don't include all "the law," but only cases.

These charges all have truth to them, and West has shown some responsiveness to criticisms of the digests over the years. Changes in the way we perceive the legal system and its decision-making processes have changed and shaped the digest. For example, the legal realism movement, with its emphasis on social science methods and the policy basis of decisions, has resulted in the incorporation of much more policy information in the digests (in the "in general" headings, primarily). New topics have been added ("drugs and narcotics" was added in 1972), old topics have been renamed and refined ("bastards" became "illegitimate children," and then "children out-of-wedlock "). An elaborate system of training and checking headnote editors has been put in place. But the system remains intellectually and practically organized as it was at its introduction in the 1890s.

Regardless of its shortcomings, the digest system worked well for many years. It guided our thinking and analysis of the law by providing us with a structure used across the country. Lawyers in Florida and South Dakota, Ohio and Nevada, consulted the same books, used the same organizing framework, found the same cases. The arguments crafted from these cases encouraged the best legal thinking, and gave judges the opportunity to explore the many sides of an issue and make a decision that was understandable. Many are the arguments about which came first: the digest or a predictable, stable judicial system. The fact remains, however, that they became almost inextricably intertwined.

However, as the body of case law grew, spurred by social, economic, political, demographic, and other causes, a book-based system of organization could no longer control the sheer volume of decisions. A new method for organizing and retrieving the information was needed. Enter the computer, first the large, expensive, mainframe variety, rapidly followed by the affordable, user-friendly, desktop (or laptop), personal computer.

Court decisions were a natural for organization by the earliest computers. Decisions are text-based, rarely incorporating pictures or graphics. They are relatively short, individual documents, not necessarily related to other documents (i.e. there are generally no separate appendixes, or reports, or supporting documentation). They require minimal editing, and can be easily input. They come with pre-arranged internal segments (names of the parties, citation, name of judge, date, majority opinion, dissents and concurrences, etc.), which allowed for easy, understandable access points. Their primary users-attorneys-were intelligent and highly educated, making training on a retrieval system less of a problem. It's no surprise that, by 1976, there were two companies (Mead and West) developing massive legal databases. Law was the first field to have the full-text of its primary materials placed in machine-readable form, and remains a leader in this area.

Attorneys-and law students and law professors-took to computer-assisted case retrieval like ducks to water. The computer gave us a way to get some control over the thousands of cases (65,000 to 100,000, depending upon to whom you speak) being handed down each year. We could find and manipulate information in ways we had only dreamed of previously. We could find all the cases written by Judge Williams in the last six years dealing with administrative law; we could find cases in which a certain law firm or attorney had appeared; we could find cases from an appellate court that were affirming a lower court during a certain year. These projects are almost impossible using books.

And, the best part was that searching for this information was so easy! Just plug in a couple of words, maybe add a date or a judge's name or some other specific identifier, and the computer did the rest. No muss. No fuss. No book dust. In some instances, research time was cut from several hours to several minutes. This saved the attorney time, and had the added benefit of allowing the attorney to bill less time to the client.

In addition, computers got rid of the "middle man," the West editor and all his (or her) possible biases or mistakes. In fact, most computer-assisted legal research systems do away with the entire West digest system itself; only WESTLAW and West's CD-ROMs make topics and key numbers available. We the searchers decided what words to use, what we'd be looking for, what was important. We can find cases that were decided mere hours or day ago. It is as if the only limitations in a computer search come from us. Our search results are just exactly what we've asked for.

But let's think for just a minute about the processes we use when we prepare to research a problem using a computer system. What happens to our legal reasoning processes, our "thinking like a lawyer"?

We start at the same place: the facts of our client's situation. And we use those facts to identify the area of law involved (bankruptcy, torts, etc.). We then start distilling the information we have into a computer search strategy.

And what is it that we use to devise a computer search strategy? Facts. Not rules or concepts, but words or short phrases. Computer-ese. We have the ability to search for every published case in America that involves whatever situation we can describe. You want cases about patients biting their dentists? No problem. Cases about the smells coming from pig farms? Gotcha. Computers are terrific in matching factual situations, and searches can be limited, as noted above, to provide very precise information about parties or judges or courts, for example.

What we don't necessarily have are the rules of a particular area of law, or the logic or analysis in a group of conceptually related cases. We are left to figure out rules from our list of cases with similar facts, and as we all know, a single set of facts can suggest a wide range of legal issues with their corresponding rules. Even something as seemingly straightforward as a child being accused of committing a crime raises issues such as capacity, jurisdiction, sentencing and treatment, parental rights and responsibilities, and rights to counsel. When we use computers as our primary research tool, we neither start with, nor reliably retrieve, a coherent statement of applicable rules. We don't have a framework to which to refer, as we do with a digest. It is difficult to structure and refine our arguments in a vacuum.

Thinking like a lawyer in the age of computers has compacted that methodical, four-step, traditional process into a very different procedure. Computer-assisted legal research actually starts from the opposite direction of traditional, book-based research, by looking for factually similar cases first. This leads to a thought process that puts its first and strongest emphasis on the facts; the legal rules become secondary. It's not uncommon to omit completely any attempt at the identification or inclusion of rules when we begin a computer search. (This omission has been noted repeatedly by legal software trainers and developers, and by those who teach use of the computer-based systems.)

Therefore, to use the computer, we have to think like a computer. The more we think in a highly simplified, individual word, fashion, the more effective our computer searches will be. But, the more we think in this computer-ese, the farther we move from our traditional "thinking like a lawyer."

Probably your first reaction to this is "why not search for rules too? Why focus only on the facts?" It's the obvious, and seemingly simple, solution to the problem. It is not, unfortunately, a very workable solution, given the state of computer-based research technology. The fact/word orientation is dictated by the organization of the information in the computer-based resources.

Information in legal databases is organized by words. Almost every word in each case is placed in a massive, alphabetized list, and its location in each case is noted; this is called the concordance. When we do a CALR search, the computer essentially compares the words in our request to the concordance, and notes the documents that have the word combinations we have requested. This is what we receive as a citation list. We can specify word combinations and adjacencies, we can ask for words to be in one order or another, but in the end each word is searched for individually. There is no discernible framework in the computer's organization of information. There is no overriding organization of concepts and rules.

Searching for rules, searching for concepts, is something that computers are notoriously poor at doing. Even finding cases discussing a basic concept like "burden of proof" becomes incredibly complex if you are researching on a computer. In addition to specifying the phrase or adjacency of the words "burden" and "proof," how do you account for variations such as "plaintiff has the burden of proving" or "plaintiff must prove" or "plaintiff is required to show" or any of the numerous other ways in which that common concept can be expressed? You cannot anticipate the many ways a judge may choose to describe a proceeding or explain the reasoning behind a decision.

This reality is no surprise to those who designed our computer systems. As early as the 1950's, as computer-based information retrieval systems were being contemplated and designed, a variety of retrieval methods were devised and tested. Linguists, scientists, and engineers worked together to study ways in which language was used and could be programmed for. They studied computer programs based on free-text retrieval with a concordance, controlled vocabularies, concept-based systems, and retrieval based on subject-indexing of documents, among other arrangements.

It wasn't long before concept-based retrieval was rejected because it gave the worst search results; it was too difficult to model for. Controlled vocabularies and subject-indexing were too restrictive; they limited the search results artificially, and required extensive analysis of the documents prior to their being placed in the databases in order to assign headings and subjects. "Uniterm" retrieval--the free-text retrieval of individual words--was determined to be the method that gave the best and most accurate retrieval of information. This is the method we use today in most of our computer-based legal research systems. We combine our words with the Boolean "ands," "ors," and "but nots," and specify adjacencies with the +5 or w/100 connectors, using computer-ese to communicate with the computer. The majority of our legal research systems are designed to work this way.

The two largest computer-based legal research systems, LEXIS and WESTLAW, have begun to develop concept-based systems, and have introduced "natural language" search interfaces as a step in this direction. We now have Freestyle and WIN, respectively. Natural language moves towards a conceptual search system, with a list of over 10,000 of the most commonly used legal phrases indexed in addition to words. But natural language requires a complex search interface, which substitutes a series of mechanical judgments for our decision-making process. The computer program "identifies" the "concepts," which are basically nouns or legal phrases, in your search request, and matches them against its inventory of words and legal phrases. The program identifies other documents with the same concepts, and ranks its findings by statistical relevance-primarily by the number of times the concept occurs and how close to the beginning of the document it first occurs. This program is superimposed over the concordance arrangement.

Like other computer searches, sometimes the results of natural language searches are extraordinary and sometimes they are worthless; usually they are somewhere in between. In any event, your ability to think in computer-ese and the underlying logic of the computer program determines the outcome of your research. This isn't the bias-free, untouched-by-human-hands results we expect of a computer, for many decisions are made for you by the computer program. Furthermore, many programmers are convinced that a better search, even for conceptual information, can be crafted using the Boolean techniques. In the words of Russ Armstrong of Geronimo Development Corporation, a Virginia company producing legal materials on CD-ROMs, natural language compared to Boolean is like an automatic transmission versus a stick shift. "You don't need to know anything about transmissions to drive an automatic, but all the race cars have stick shifts."1

A search for a unique word or term on the computer can lead to a highly successful research session. We know that computers can give us information in numerous formats. But searching for many types of information on a computer-based system can be tricky. Procedural issues, for example, are extremely difficult to research, with their necessary use of words like "trial," "motion," and "action," which appear in thousands of cases. Using a legal term such as "comparative negligence" as a search request will give us far too many cases to handle. So we avoid searches that will return too many cases, and focus on searching for facts and terms.

The bottom line, however, is that the facts don't require this kind of research. It's the legal issues, the rules, that do. The rules must be identified and understood before they can be applied to the facts at hand. Clifford Stoll has written that, "computers emphasize the answer over the reasoning process. . . . Meaning doesn't come from data alone. Creative problem solving depends on context, interrelationships, and experience. The surrounding matrix may be more important than the individual lumps of information."2 Because CALR doesn't research issues, and only researches facts, one can easily miss the broader statements of policy and principles that are included as a matter of course in the digests. But relying exclusively on computer results limits your overall knowledge to only the information retrieval request you made, and hinders your ability to see the broader picture because of the uniqueness of the search terms you must choose. The lack of structure in computer research can make your reasoning process lumpy indeed.

It is as if we completely forget our training in traditional legal reasoning when we begin using computer-ese. Once we have our list of fact-based case citations, we begin the task of harmonizing the rules they suggest, in effect devising a framework of facts. We never seem to consider locating the larger, existing framework of rules. The time pressures of practice or law school may influence our behavior, and the training talks of the CALR vendors, who assure us that all we need is one good search, may have clouded our thinking. We may not even have access to a digest. But I believe that our failure to seek out the existing, larger structure, as embodied in the digests, is primarily because of a sense of over-confidence in the computer's results.

The differences in the way that we perceive computer search results, especially as compared to book search results, is startling. Most of us, in the process of carrying out our book-based legal research projects, have experienced the nagging feeling that there's something missing. We feel that, if we just had a few more minutes, we'd find that the perfect case. We turn back to the table of contents, or index, or topic outline, in hopes of a little guidance for our thoughts. No matter how much time we spend, we continue to feel uneasy about the completeness of our research.

On the other hand, we don't seem to experience these same feelings of doubt when conducting computer-based legal research. Certainly, the differences between locating rules and matching words account for some of this disparity. However, we seem to reach a point of satisfaction with our computer results fairly quickly, once we get a good, short, citation list of documents that meets our search request. We are quicker to end our computer-based research projects because we seem to trust the computer more than we trust ourselves. This situation has been noted, and studied, in many arenas (not just law) with interesting conclusions.

Computer use patterns of lawyers, doctors, scientific researchers, graduate students, and undergraduate students show remarkable consistency in their findings. Several extensive studies have clearly documented a false sense of security on the part of computer researchers. One study commented that users felt that "because the source is 'technological,' they are finding everything or, at the very least, finding the best materials."3 Other studies have shown a high confidence level on the part of CD-ROM and online database users, attributed to the fact that the resources can be used with little or no formal instruction because of their user-friendly search interfaces. Yet another study reported that 90% of CD-ROM users would use the CD-ROM service again, because they thought they got more information than from a book-based research effort.

Whereas we never seem to quite get comfortable with our book research, we have suspended our sense of disbelief in the completeness of our research when it comes to computers. Perhaps it is because we don't understand how processors and buses and modems and those wires and chips work. Perhaps it is because they have made our lives so much easier, giving us all the resources we think we need in one place, and even letting us Shepardize, read law reviews, check encyclopedias and looseleaf services, and consult other resources without moving from our desk. There is a confidence that nothing more needs to be consulted.

And yet, when you rely solely on computer results you may not even discover the applicable legal rules. Unless you identify the correct rules, you cannot develop an accurate prediction of a case's outcome, or craft a reasonable argument for your theory. Urging a change in the law's application, pushing the envelope, is difficult when you haven't even found the envelope. As a profession, we stand to lose the opportunity for creativity, and our thinking and reasoning may lose its vitality, when our focus moves from rules to facts.

If we don't proceed with a legal reasoning process that starts with known rules, we run the risk of losing the predictability, and with it the stability, of our judicial system. The decisions of the courts are based on the information we give the judge, as well as his or her own research, which is as often as not computer-based. If the judge relies on poorly conceived rules and arguments that neither state the law of the jurisdiction nor conform or relate to other relevant cases, judicial decisions could come to reflect the inability to discern the underlying precedent from computerized results. We could find ourselves losing the coherency of the rules-based structure we have.

Neil Postman, author of Amusing Ourselves to Death and Technopoly, postulates that the uncertainty in society caused by too much information has caused citizens to turn to the courts for stability and coherence.4 It is ironic that the same problem, too much information, is causing the legal profession to change the very process that has contributed so much to stability and coherence. We need the storage power of computers, and their ability to retrieve specific and unique information is invaluable. But are we ready to so thoroughly, and most likely permanently, rearrange our thought processes by a wholesale adoption of computerized research, given the potential consequences?

Certainly there are other factors that will affect a change in our thinking. Judges themselves appear to be encouraging a move towards fact-based research by a seeming unwillingness to develop the overarching rules of a judicial doctrine. Instead, many seem to be confining their application of the rules to the narrow facts of a case, resulting in an ad hoc decisionmaking that may over-emphasize the importance of factually similar cases.

But in the computer age, our approach to thinking, our methods of research, our written output and therefore the input judges receive, is becoming less rules-oriented, less structured, and more dependent on the chance that the fact-matching of a computer search will also return the right rules. Once we begin developing arguments based on fact rather than law, we move outside the time-proven framework that has served our legal system, and our profession, so well. And, once outside that framework, we lose the guidelines against which we can gauge our creativity and innovation. Without the framework, our arguments lose their context, and ultimately their strength and persuasiveness. We find ourselves relying on sheer tonnage of citations instead of the weight of established authority.

Computer scientists are, even as I speak, working on developing concept-based computer retrieval systems. The day will come when we can ask a computer to find cases that analyze the various ways in which the federal constitutional protection against double jeopardy varies from state protections, depending upon the stage of the proceedings against the defendant, and the computer's results will reflect our needs. But that day is not yet here. The law is a complex and difficult field; devising ways to organize and find it is complex and difficult too. We need to use our existing tools carefully, understanding their intrinsic organization, and applying them to our legal problems in the ways they work best. Let us consider the limits of our resources as we start our legal problem solving processes. Let us think like a lawyer.



1. Russ Armstrong (Jan. 8, 1996), CD-ROM v. Law Books, LAW-LIB Discussion List. Available e-mail:

2. Clifford Stoll, Silicon Snake Oil: Second Thoughts on the Information Highway 135 (1995).

3. F.W. Lancaster et al., Searching Databases on CD-ROM: Comparison of the Results of End-User Searching with Results from Two Modes of Searching by Skilled Intermediaries, 33 RQ 370, 382 (1994).

4. Neil Postman, Technopoly 74 (1992).

Printed with Permission of the Author


Back to the CoALL home page This page last updated 09/23/00
Copyright 2000, Colorado Association of Law Libraries