MIME-Version: 1.0Content-Type: text/htmlSet-Cookie: ad_browser_id=250042768; Path=/; Expires=Fri, 01-Jan-2010 01:00:00 GMTSet-Cookie: ad_session_id=250042769%2c0%2cf5d9S7MUX1ABzUOfJM4WagRfrj2oOowQ%2c1220893665; Path=/; Max-Age=86400Set-Cookie: last_visit=1220893665; path=/; expires=Fri, 01-Jan-2010 01:00:00 GMTSet-Cookie: SERVERID=greenspun2; path=/
Cache-control: private
Connection: close
Making Way for Intelligence in Case SpaceMaking Way for Intelligence in Case SpacePhilip GreenspunDepartment of Electrical Engineering and Computer ScienceMassachusetts Institute of TechnologyCambridge, Massachusettsphilg@mit.eduMarc LauritsenHarvard Law SchoolCambridge, Massachusettslaurit@hulaw1.harvard.edupart of the MIT Journal of Computer-Aided Litigation(also presented at the Fifth International Conference on ArtificialIntelligence and Law (ACM), May 1995, College Park, Maryland)AbstractCourts should and eventually will adopt modern information systems that allowall transactions presently occurring on paper to be conducted and reviewedelectronically. The widespread deployment and interconnection of such systemswill pose dramatic opportunities and challenges for artificial intelligence andlaw. This paper describes an early effort to build a distributed system forcomputer-mediated litigation, and explores some of the issues involved in theintelligent use of such a system.1. Introduction"Paperless" litigation is still a curiosity in the United States, seen largelyin connection with major cases like those involving the Exxon Valdez oil tankerdisaster, the O.J. Simpson double murder prosecution, and the Keating savingsand loan scandal. Electronic filing is routinely available in only a fewcourts, and required only in certain complex matters by special order of ajudge. [Asay 1994; Yerton 1994]. It seems reasonable to assume, though, thatmost paper-based activities in the court system will eventually be replaced byelectronic data transactions. [McMillan 1992]. The text and images of filingsand the contents of judicial administrative databases (e.g.,dockets and calendars) will be electronically accessible. Litigants will beable to file documents and access information about their own and other casesdigitally and remotely. Internet browsers will enable the public to perusecourt dockets and pick up audio and video streams of contemporaneous orhistorical courtroom activity.It seems only slightly more speculative to assume that documents received andproduced by courts will eventually be marked up with codes that identify theircontents in terms of abstract categories relevant to the judicial process. Forinstance, paragraphs in complaints will be tagged according to their characteras factual allegations, jurisdictional statements, legal claims or defenses, orrequests for other judicial action. For each case in a particular court, adata structure will exist that contains the names and other characteristics ofall parties, claims, witnesses, and decision makers, summaries of all formallegal events, and pointers to all documents filed or generated. Conventionalquerying tools will allow one to ascertain, for example, the average award inproduct liability actions against asbestos manufacturers in which the plaintiffwas 45 years or older and resident in a southern state. For purposes of thisarticle, we call this distributed database "case space".We have developed a tool for helping litigants compose pleadings and makerelated moves in a controversy brought to court. We have also built facilitiesfor using this tool and accessing a database of disputes modelled in it overthe Internet via the World Wide Web. We will briefly summarize the workings ofthis system in Part 3 of this article. In Part 4, we take up several questionsabout how one might make intelligent mechanical use of the distributed databaseof judicial activity that would result if such a system were adopted widely.2. MotivationsLawyers and judges have been privileged relative to other professions becauseso much of the raw material of their craft judicial opinions, statutes,regulations, and secondary material has long been available online in full textform. Several graduating classes of law students in the United States have nowbeen accustomed to having free round-the-clock access to these enormous textualrepositories. Natural language tools for querying the online databases haveemerged, and research proceeds in many quarters on next-generation interfacesthat will likely combine multiple paradigms for online research. At least inmost Western countries, we are rich in legal information and tools for usingit.But this substrate of raw text, only marginally segmented and parameterized,hardly suffices for some of our informational and analytical needs. Havingaccess to the text of hundreds of thousands of judicial opinions is wonderful,but they represent only a small percentage of the cases filed, and even inthose cases rarely contain any of the authentic pleadings or evidence. Mostare appellate opinions that provide only summaries of the proceedings "below".However central such texts are to our jurisprudence, they provide a woefullyincomplete picture of what actually goes on in the courts, and yield toautomated analysis only insofar as our computers can understand naturallanguage and apply common sense. Once businesses, consumers, courts, and lawyers are automated and networked, itbecomes natural to want access to the full grist of litigation: pleadings,motions, orders, opinions, transcripts, and documentary evidence such as thetext of contracts, engineering drawings in product liability cases, and mapsand photographs of crime scenes. The likely demand for and increasingtechnological feasibility of achieving this form of access opens up theprospect of an alternative to the full-text substrate described above. A goodsubstrate that we can build with current technology is one that does not depend on natural language understanding, common sense reasoning, orlaborious formalization of facts, can present multiple views of disputes simultaneously (plaintiff, defendant,judge, clerk, public), reduces the administrative burdens of a lawsuit, saving time for both partiesand the judge, incrementally builds a substantial database of formalized legal argument,and uses the Internet/Web infrastructure to make the database available to legaland computer science researchers and to open the courts to the public.We believe that it will be upon such a substrate that machine learning programswill learn the law, probabilistic reasoning systems will make reasonablepredictions, and new ideas for computer-aided law will grow.3. KTA: A System for Computer-Aided Litigation over the InternetKTA (Knowledge Theorist for Attorneys) is a Common Lisp program that capturesstructured legal argument from litigants and judges. KTA includes productivitytools for those directly involved in a lawsuit and a World Wide Web interfaceso that the public can browse court dockets.Briefly, our design goals were to create a unified database for allinformation in a lawsuit, protect confidential and privilegedinformation, eliminate all physical transmission and storage ofdocuments, supply a unified user interface for each class of user(e.g., judge, litigant), capture legal argument in a finer grainedform than current pleading documents, have a strong enough model ofprocedure to explain cases to lay people, and achieve isomorphism withcurrent paper practices. A. DatabaseAt the core of KTA is a distributed object database. As shown in Figure 1, theviews of the database presented to parties, the judge, the court clerk, and thepublic may be quite different. In particular, the clerk's view is the databaseof record and contains filed, immutable pleadings, motions, and documents. Aparty's private view may contain draft pleadings as well as argument andauthority that he has not yet been forced by the rules of procedure to revealto the court or his opponents. A judge's private view may contain draftopinions. The public's view via the Web obscures documents that have beenfiled under seal and contains additional computer-generated explanations ofjargon and court procedure.KTA's object database is distributed in the sense that objects are storedacross multiple networked computers. It is also distributed in the sense thatthe instance variables of a single object may be stored across multiplecomputers. For example, in the complaint that starts a lawsuit, a plaintiffneed only claim "breach of warranty." The lawyer who files the complaint mayhave already looked up case law to find authorities that give the elements ofbreach of warranty. He may have already associated facts asserted by theclient with each of those elements. He may have already associated pieces ofevidence brought to him by the client with each of those facts. Figure 2 showsan example of a claim for intentional infliction of emotional distress, wheresome material is kept on the plaintiff's machine, shielded from other partiesuntil the plaintiff chooses to expose it.B. Capturing legal argument in fine-grained form, or "If you have strongsyntax, maybe you don't need semantics"A plaintiff does not file a complaint in KTA as a monolithic block of ASCIItext. It is broken up into allegation objects and claim objects. Theplaintiff is encouraged but not required to pick a claim from a standard set sothat the system automatically picks up a canonical title such as "breach ofcontract". The claim object has a slot for arbitrary text, perhaps settingforth which of the allegations lead to this claim, and a slot for a textprayer, e.g., asking the court for $10,000 in damages. KTA makes no attempt tointerpret either of these text strings. To facilitate automated processing,the claim object also contains a slot called PRAYER-DOLLAR-AMOUNT that theplaintiff is encouraged to fill in with the number, e.g., 10,000. This enablesan expected value calculation to be performed on a lawsuit without needing towade into text prayers and pull out the numbers with natural language smarts.The claim contains slots called AUTHORITY and RESPONSES. If the user pickedthe claim from an on-line horn book or practice guide, then the authority forthat claim is preserved. This doesn't go to the court, but may prove useful ayear later when a new associate takes up the case by providing hypertext accessto, for example, the statute that the filing attorney thought relevant.RESPONSES is a list of responses to the claim by opposing parties and thecourt. When a defendant files an answer, it is in the form of response objectslinked back to their counterparts in the complaint, plus new allegations andaffirmative defenses. The defendant is encouraged but not required tocategorize his response as an "admission", "partial admission", "partialdenial", "denial" or "insufficient information". He can choose the "other"category, but can very likely save time by choosing one of the predefinedcategories because KTA automatically brings in appropriate boilerplate.The overriding theme here is that structure is captured when possible, in amanner that is painless for the participants, but KTA never limits what a partycan ask of a court. Anything that can be pleaded on unstructured paper can bepleaded in KTA.When a judge wants to review the status of a case, KTA produces an easilyunderstood hypertext outline. Allegations or claims that have been admitted orpreviously decided are shown in special colors so that the judge canconcentrate on parts of the case that remain in dispute. (Figure 3 illustrateswhat a judge may see mid-way through a lawsuit.)KTA's attempt to capture legal argument does not stop with the parties. Thejudge is encouraged to rule in a structured manner, allegation by allegationand claim by claim. This enables one to ask a computer to "show me all thebreach of contract claims that Judge Smith denied in the last two years." C. World Wide Web interfaceAlthough it is theoretically possible to represent KTA's rich set of linkedobjects with SGML, it is certainly impractical to do so with HTML, a simpleSGML document type definition that is the lingua franca of the World Wide Web.[Berners-Lee 1992]. KTA makes no use of SGML or HTML in its internal database,but is capable of walking through its network of Common Lisp Object Systemobjects and reformatting all the information into HTML files ready forinstallation on a Web server. Some of the data is even improved in theprocess. For example, KTA maintains a small dictionary of legal terms (asubset of Black's Law Dictionary) and walks through strings of English text asit writes them into HTML files. When KTA finds a word that is in thedictionary, it links that word to its definition, something that might be veryuseful to a lay person checking out a local court's docket.With the KTA Web interface, court documents that are supposed to be publicbecome truly public, available to anyone in the world with an Internetconnection, 24 hours a day. D. Software StatusMuch more work remains to be done on KTA to make it a practical system forcourts and attorneys. High priority items would be the following: mating a reliable disk-based object database to the current memory-basedobject database, and implementing classes for modelling complex legal argument such as that foundin summary judgment motions.KTA is free software that will be available to anyone on the Internet under thesame conditions as tools such as GNU Emacs. It is currently limited to runningon Apple Macintosh computers due to its use of the Macintosh Common Lisp windowsystem and user interface extensions. There are a few files ofmachine-dependent code, some of which are fairly complex (e.g., fordrag-and-drop editing).4. Discussion A. Do we need "AI"?Most of what arguably needs to be done in the judicial information processingarena and has been prototyped with KTA can be achieved with thoroughlyconventional computer science techniques. There is a vast range ofimprovements that could be made by applying well understood andnon-experimental technologies to our court system. The difficulties there seemoverwhelmingly sociological, political, and organizational ones.To be sure, there will be enormous engineering challenges in building,maintaining, and querying massive distributed databases of judicialinformation. Standards will have to be developed. Policy will need to bearticulated and executed at many levels of federal and state government. Butthese challenges are comparable to those involved in other large transactionalsystems like airline and hotel reservation networks.Our view is that artificial intelligence is of little importance to the basictask of "electrifying" the courts, but will play an increasingly central rolein the legal world that results after that process has occurred. We need tobegin thinking systematically about the knowledge representations and processesneeded to perform these activities. A few main realms of effort come tomind. B. Enforcing the rulesA natural first context for deploying knowledge-based technology is that ofdeveloping clerical/administrative tools to monitor and enforce compliance withprocedural rules. While some rules can be straightforwardly modelled inprocedural code such as having to file an answer within a certain number ofdays after the service of a complaint absent court approval, others involvecomplexities (such as context dependence and defeasance) that may be besthandled with declarative knowledge representation techniques. The early workby Jeff Meldman [1978] on the use of Petri nets to model civil proceduresuggests the surprising complications one can encounter. One more recent andvery elegant approach is Thomas Gordon's normative formalization andcomputational implementation of civil pleading [1993]. C. Following the rulesAnother obvious place for AI is in advisory systems for litigants, attorneys,and judges. Such systems ought to be able to advise parties what moves arepossible or recommended in given contexts. A good model of procedure is neededbefore a system like KTA can generate good explanations for lay people.Companies and individuals will be interested in programs that monitor the netand notify users when claims are asserted or post-complaint filings are made.A firm or government agency that frequently finds itself in litigation can begreatly assisted both in terms of efficiency and quality of effort.Case-based reasoning models will likely play a role here. Strategic reasoningwill also be called for as multiple intelligent players try to "game" thesystem. D. Making the rulesCourt and legislatures contemplating proposed reforms will want tools thataccess the distributed case database as an information refinery in aid oflegislative policy judgments. ("If we shorten the statute of limitations onmail fraud claims, how is that likely to affect the number of filings?")Similarly, parties to disputes will often find themselves better off enacting"private legislation" through forms of computer-aided negotiation. Tools likeKTA can assist people in narrowing the scope of disagreement, assessingprobabilities of success, and identifying opportunities for win-winsolutions. E. Calling the shotsPerhaps the most ambitious efforts in this area will be those that attempt toexploit systems like KTA to aid judges in making decisions and writingopinions. Karl Branting [1993] has described an issue-oriented approach tojudicial document assembly that involves the structured entry of caseinformation and subsequent rule-based guidance of a judge through decisionmaking and justification. One can imagine decision-support tools that makesuch use of large databases of similar electronic case records.5. ConclusionEven an elaborately developed and painstakingly implemented electronic datasystem for court information will fail to capture the full particularity ofhuman controversies that make up the business of the judicial system. Withoutrobust natural language and common sense abilities, no computer system willyield results regularly comparable to those of competent jurists inparticipating in and making sense of most judicial activity. But a great dealof the transactions taking place in court can and should be supported by moreintelligent information technology. If the rule-governed and formulaic aspectsof judicial activity can be crystallized out of the muddle of its human contextand subjected to intelligent distribution and analysis, gains in rationalityand efficiency will be achieved that might ultimately yield better justice.ReferencesAlan Asay. Toward Paperless Utah Courts. 1994. MIT Journal of Computer-AidedLitigation, http://kta.org/mjcal.Berners-Lee, T.J., Cailliau, R., Groff, J-F, Pollermann, B., CERN 1992."World-Wide Web: The Information Universe", in "Electronic Networking:Research, Applications and Policy", Vol. 2, No. 1, pp. 52-58. Meckler Publishing, Westport.L. Karl Branting. An Issue-Oriented Approach to Judicial Document Assembly.In Proceedings of the Fourth International Conference on ArtificialIntelligence and Law, pp. 228-235. ACM, 1993.Thomas F. Gordon. The Pleadings Game. In Proceedings of the FourthInternational Conference on Artificial Intelligence and Law, pp. 10-19. ACM,1993.Jeffrey A. Meldman. A Petri-Net Representation of Civil Procedure. IDEA19(2). 1978.Jim McMillan. Judicial EDI: The Need for National Standards. The CourtManager. Winter 1992, pp. 17-22.Stewart Yerton. "CLAD" for Litigation. The American Lawyer. October, 1994,pp. 111-112.Figure 1 Figure 1: KTA is a system for structuring and storing legal discourse.Each box above represents the portion of a distributed database thatis stored on a lawsuit participant's computer. The court clerk'scomputer holds the primary record of the case, i.e., all theinformation that has been filed, most of which is publicly accessible(in this case, there is only one letter, an exhibit for a motion todismiss, that is sealed and protected via encryption). Note that a ¹Pi in front of a "document" indicates that it was filed by the plaintiff,a Delta by the defendant, a J by the judge.The plaintiff's local database contains only a draft motion forsummary judgment, currently hidden from all other parties in thisaction.A dynamic trial/settlement model on the defendant's machine reflectshis concern about the ultimate cost of this case. This model isupdated when events in the litigation change probabilities. Forexample, if an motion to exclude evidence succeeds, that may make itdifficult for the plaintiff to prove a critical element of a claim.The probability of loss on that claim will be reduced, thus reducingthe expected cost of the case. Because over 90% of cases settlebefore trial, this kind of information can be very valuable in acomplex case.The judge in this case has retained her draft denial of thedefendant's motion to dismiss because she may want to use the samepresumptions of fact in hearing a summary judgment motion. KTAassists judges by outlining cases and could quickly show which claimscannot succeed given certain assumptions about the facts.Figure 2 Figure 2: The plaintiff's view of his breach of his intentionalinfliction of emotional distress claim is very different from thepublic's at this early stage in litigation. American civil procedurerequires a plaintiff only to state a claim by name and hence that isall that shows up in the public view: "intentional infliction ofemotional distress." Note that in the Plaintiff's private view, hisattorney has sketched in some notes about what the elements of thisclaim are according to a legal authority and what pieces of evidenceexist to support those elements.KTA's distributed object database keeps private information on theplaintiff's private machine until it becomes public on the plaintiff'sinitiative. Thus, the letter of 3/12/93 that supports the element ofan outrageous defendant act, is available as a hypertext link (notethe underlining) but will be supplied from the local disk. TheFigueiredo case (in which a psychiatrist had an affair with apatient's wife) is also available as a link, to be supplied eitherfrom a local cache, the World Wide Web, or a commercial service suchas Lexis.Note that in the public view of the case, no authorities have beencited by either side to support or attack the fit of the facts of thiscase to the claim of intentional infliction of emotional distress.That is presumably because no motions for summary judgment have beenfiled. In any case, the defendant apparently denied the claim, thefull text of the denial would be available with a mouse click.Figure 3 Figure 3: What a Massachusetts judge might see mid-way through alawsuit. Greenspun is the plaintiff. He alleges that Smyly Dodge stolehis car stereo while servicing his car. The ovals marked "¹F" arefacts that Greenspun asserts support all the elements of a ConsumerProtection Act claim. While it might be valuable to present oneside's case to a judge in just this schematic form, it is even betterto see the other side's counterarguments superimposed in a differentcolor (alas impossible in this proceedings).Big A's indicate that Smyly admits that Greenspun is a consumer andthat they are a business. The big D's and supporting defendant facts("ÆF") are part of Smyly's argument but another critical part is showby the big X's. What is being disputed with the topmost X is not thatthe demand letter contained a 2nd paragraph, but whether or not thetext of that paragraph stated the injury suffered sufficientlyprecisely to satisfy the statute. Becuase of this X, KTA has markedthe element "¹ gave notice" under dispute by surrounding it with a redrectangle (in practice, this is done on-screen by displaying the textin red).(Note: this figure is loosely based on a real lawsuit, Greenspunv. Smyly Autos, but does not accurately represent the legal argumentsactually made. The original documents are available on the Web athttp://smyly.com/smyly)_uacct = "UA-315149-1";urchinTracker(); |
|
