Clinical decision support system

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Clinical_decision_support_system 

A clinical decision support system (CDSS) is a health information technology, provides clinicians, staff, patients, or other individuals with knowledge and person-specific information, to help health and health care. CDSS encompasses a variety of tools to enhance decision-making in the clinical workflow. These tools include computerized alerts and reminders to care providers and patients, clinical guidelines, condition-specific order sets;´, focused patient data reports and summaries, documentation templates, diagnostic support, and contextually relevant reference information, among other tools. Robert Hayward of the Centre has proposed a working definition for Health Evidence: "Clinical decision support systems link health observations with health knowledge to influence health choices by clinicians for improved health care". CDSSs constitute a major topic in artificial intelligence in medicine.

Characteristics

A clinical decision support system is an active knowledge system that uses variables of patient data to produce advice regarding health care. This implies that a CDSS is simply a decision support system that is focused on using knowledge management.

Purpose

The main purpose of modern CDSS is to assist clinicians at the point of care. This means that clinicians interact with a CDSS to help to analyse and reach a diagnosis based on patient data for different diseases.

In the early days, CDSSs were conceived to make decisions for the clinician literally. The clinician would input the information and wait for the CDSS to output the "right" choice, and the clinician would simply act on that output. However, the modern methodology of using CDSSs to assist means that the clinician interacts with the CDSS, utilising both their knowledge and the CDSS's, better to analyse the patient's data than either human or CDSS could make on their own. Typically, a CDSS makes suggestions for the clinician to look through, and the clinician is expected to pick out useful information from the presented results and discount erroneous CDSS suggestions.

The two main types of CDSS are knowledge-based and non-knowledge-based:

An example of how a clinical decision support system might be used by a clinician is a diagnosis decision support system (DDSS). DDSS requests some of the patients' data and, in response, proposes a set of appropriate diagnoses. The physician then takes the output of the DDSS and determines which diagnoses might be relevant and which are not, and, if necessary, orders further tests to narrow down the diagnosis.

Another example of a CDSS would be a case-based reasoning (CBR) system. A CBR system might use previous case data to help determine the appropriate amount of beams and the optimal beam angles for use in radiotherapy for brain cancer patients; medical physicists and oncologists would then review the recommended treatment plan to determine its viability.

Another important classification of a CDSS is based on the timing of its use. Physicians use these systems at the point of care to help them as they are dealing with a patient, with the timing of use being either pre-diagnosis, during diagnosis, or post-diagnosis. Pre-diagnosis CDSS systems help the physician prepare the diagnoses. CDSSs help review and filter the physician's preliminary diagnostic choices to improve outcomes. Post-diagnosis CDSS systems are used to mine data to derive connections between patients and their past medical history and clinical research to predict future events. As of 2012, it has been claimed that decision support will begin to replace clinicians in common tasks in the future.

Another approach, used by the National Health Service in England, is to use a DDSS to triage medical conditions out of hours by suggesting a suitable next step to the patient (e.g. call an ambulance, or see a general practitioner on the next working day). The suggestion, which may be disregarded by either the patient or the phone operative if common sense or caution suggests otherwise, is based on the known information and an implicit conclusion about what the worst-case diagnosis is likely to be; it is not always revealed to the patient because it might well be incorrect and is not based on a medically-trained person's opinion - it is only used for initial triage purposes.

Knowledge-based CDSS

Most CDSSs consist of three parts: the knowledge base, an inference engine, and a mechanism to communicate. The knowledge base contains the rules and associations of compiled data which most often take the form of IF-THEN rules. If this was a system for determining drug interactions, then a rule might be that IF drug X is taken AND drug Y is taken THEN alert the user. Using another interface, an advanced user could edit the knowledge base to keep it up to date with new drugs. The inference engine combines the rules from the knowledge base with the patient's data. The communication mechanism allows the system to show the results to the user as well as have input into the system.

An expression language such as GELLO or CQL (Clinical Quality Language) is needed for expressing knowledge artefacts in a computable manner. For example: if a patient has diabetes mellitus, and if the last haemoglobin A1c test result was less than 7%, recommend re-testing if it has been over six months, but if the last test result was greater than or equal to 7%, then recommend re-testing if it has been over three months.

The current focus of the HL7 CDS WG is to build on the Clinical Quality Language (CQL). The U.S. Centers for Medicare & Medicaid Services (CMS) has announced that it plans to use CQL for the specification of Electronic Clinical Quality Measures (eCQMs).

Non-knowledge-based CDSS

CDSSs which do not use a knowledge base use a form of artificial intelligence called machine learning, which allow computers to learn from past experiences and/or find patterns in clinical data. This eliminates the need for writing rules and expert input. However, since systems based on machine learning cannot explain the reasons for their conclusions, most clinicians do not use them directly for diagnoses, reliability and accountability reasons. Nevertheless, they can be useful as post-diagnostic systems, for suggesting patterns for clinicians to look into in more depth.

As of 2012, three types of non-knowledge-based systems are support-vector machines, artificial neural networks and genetic algorithms.

1. Artificial neural networks use nodes and weighted connections between them to analyse the patterns found in patient data to derive associations between symptoms and a diagnosis.
2. Genetic algorithms are based on simplified evolutionary processes using directed selection to achieve optimal CDSS results. The selection algorithms evaluate components of random sets of solutions to a problem. The solutions that come out on top are then recombined and mutated and run through the process again. This happens over and over until the proper solution is discovered. They are functionally similar to neural networks in that they are also "black boxes" that attempt to derive knowledge from patient data.
3. Non-knowledge-based networks often focus on a narrow list of symptoms, such as symptoms for a single disease, as opposed to the knowledge-based approach, which covers the diagnosis of many diseases.

An example of a non-knowledge-based CDSS is a web server developed using a support vector machine for the prediction of gestational diabetes in Ireland. 

Regulations

United States

With the enactment of the American Recovery and Reinvestment Act of 2009 (ARRA), there is a push for widespread adoption of health information technology through the Health Information Technology for Economic and Clinical Health Act (HITECH). Through these initiatives, more hospitals and clinics are integrating electronic medical records (EMRs) and computerized physician order entry (CPOE) within their health information processing and storage. Consequently, the Institute of Medicine (IOM) promoted the usage of health information technology, including clinical decision support systems, to advance the quality of patient care. The IOM had published a report in 1999, To Err is Human, which focused on the patient safety crisis in the United States, pointing to the incredibly high number of deaths. This statistic attracted great attention to the quality of patient care.

With the enactment of the HITECH Act included in the ARRA, encouraging the adoption of health IT, more detailed case laws for CDSS and EMRs are still being defined by the Office of National Coordinator for Health Information Technology (ONC) and approved by Department of Health and Human Services (HHS). A definition of "Meaningful use" is yet to be published.

Despite the absence of laws, the CDSS vendors would almost certainly be viewed as having a legal duty of care to both the patients who may adversely be affected due to CDSS usage and the clinicians who may use the technology for patient care. However, duties of care legal regulations are not explicitly defined yet.

With recent effective legislations related to performance shift payment incentives, CDSS are becoming more attractive.

Effectiveness

The evidence of the effectiveness of CDSS is mixed. There are certain diseases which benefit more from CDSS than other disease entities. A 2018 systematic review identified six medical conditions in which CDSS improved patient outcomes in hospital settings, including blood glucose management, blood transfusion management, physiologic deterioration prevention, pressure ulcer prevention, acute kidney injury prevention, and venous thromboembolism prophylaxis. A 2014 systematic review did not find a benefit in terms of risk of death when the CDSS was combined with the electronic health record. There may be some benefits, however, in terms of other outcomes. A 2005 systematic review had concluded that CDSSs improved practitioner performance in 64% of the studies and patient outcomes in 13% of the studies. CDSSs features associated with improved practitioner performance included automatic electronic prompts rather than requiring user activation of the system.

A 2005 systematic review found... "Decision support systems significantly improved clinical practice in 68% of trials." The CDSS features associated with success included integration into the clinical workflow rather than as a separate log-in or screen., electronic rather than paper-based templates, providing decision support at the time and location of care rather than prior and providing care recommendations.

However, later systematic reviews were less optimistic about the effects of CDS, with one from 2011 stating "There is a large gap between the postulated and empirically demonstrated benefits of [CDSS and other] eHealth technologies ... their cost-effectiveness has yet to be demonstrated".

A 5-year evaluation of the effectiveness of a CDSS in implementing rational treatment of bacterial infections was published in 2014; according to the authors, it was the first long-term study of a CDSS.

Challenges to adoption

Clinical challenges

Much effort has been put forth by many medical institutions and software companies to produce viable CDSSs to support all aspects of clinical tasks. However, with the complexity of clinical workflows and the demands on staff time high, care must be taken by the institution deploying the support system to ensure that the system becomes an integral part of the clinical workflow. Some CDSSs have met with varying amounts of success, while others have suffered from common problems preventing or reducing successful adoption and acceptance.

Two sectors of the healthcare domain in which CDSSs have had a large impact are the pharmacy and billing sectors. Commonly used pharmacy and prescription-ordering systems now perform batch-based checking orders for negative drug interactions and report warnings to the ordering professional. Another sector of success for CDSS is in billing and claims filing. Since many hospitals rely on Medicare reimbursements to stay in operation, systems have been created to help examine both a proposed treatment plan and the current rules of Medicare to suggest a plan that attempts to address both the care of the patient and the financial needs of the institution.

Other CDSSs that are aimed at diagnostic tasks have found success, but are often very limited in deployment and scope. The Leeds Abdominal Pain System went operational in 1971 for the University of Leeds hospital. It was reported to have produced a correct diagnosis in 91.8% of cases, compared to the clinicians' success rate of 79.6%.

Despite the wide range of efforts by institutions to produce and use these systems, widespread adoption and acceptance have still not yet been achieved for most offerings. One large roadblock to acceptance has historically been workflow integration. A tendency to focus only on the functional decision-making core of the CDSS existed, causing a deficiency in planning how the clinician will use the product in situ. CDSSs were stand-alone applications, requiring the clinician to cease working on their current system, switch to the CDSS, input the necessary data (even if it had already been inputted into another system), and examine the results produced. The additional steps break the flow from the clinician's perspective and cost precious time.

Technical challenges and barriers to implementation

Clinical decision support systems face steep technical challenges in a number of areas. Biological systems are profoundly complicated, and a clinical decision may utilise an enormous range of potentially relevant data. For example, an electronic evidence-based medicine system may potentially consider a patient's symptoms, medical history, family history and genetics, as well as historical and geographical trends of disease occurrence, and published clinical data on therapeutic effectiveness when recommending a patient's course of treatment.

Clinically, a large deterrent to CDSS acceptance is workflow integration.

While it has been shown that clinicians require explanations of Machine Learning-Based CDSS, in order to able to understand and trust their suggestions, there is an overall distinct lack of application of explainable Artificial Intelligence in the context of CDSS, thus adding another barrier to the adoption of these systems.

Another source of contention with many medical support systems is that they produce a massive number of alerts. When systems produce a high volume of warnings (especially those that do not require escalation), besides the annoyance, clinicians may pay less attention to warnings, causing potentially critical alerts to be missed. This phenomenon is called alert fatigue. 

Maintenance

One of the core challenges facing CDSS is difficulty in incorporating the extensive quantity of clinical research being published on an ongoing basis. In a given year, tens of thousands of clinical trials are published. Currently, each one of these studies must be manually read, evaluated for scientific legitimacy, and incorporated into the CDSS in an accurate way. In 2004, it was stated that the process of gathering clinical data and medical knowledge and putting them into a form that computers can manipulate to assist in clinical decision-support is "still in its infancy".

Nevertheless, it is more feasible for a business to do this centrally, even if incompletely, than for each doctor to try to keep up with all the research being published.

In addition to being laborious, integration of new data can sometimes be difficult to quantify or incorporate into the existing decision support schema, particularly in instances where different clinical papers may appear conflicting. Properly resolving these sorts of discrepancies is often the subject of clinical papers itself (see meta-analysis), which often take months to complete.

Evaluation

In order for a CDSS to offer value, it must demonstrably improve clinical workflow or outcome. Evaluation of CDSS quantifies its value to improve a system's quality and measure its effectiveness. Because different CDSSs serve different purposes, no generic metric applies to all such systems; however, attributes such as consistency (with and with experts) often apply across a wide spectrum of systems.

The evaluation benchmark for a CDSS depends on the system's goal: for example, a diagnostic decision support system may be rated based upon the consistency and accuracy of its classification of disease (as compared to physicians or other decision support systems). An evidence-based medicine system might be rated based upon a high incidence of patient improvement or higher financial reimbursement for care providers.

Combining with electronic health records

Implementing EHRs was an inevitable challenge. This challenge is because it is a relatively uncharted area, and there are many issues and complications during the implementation phase of an EHR. This can be seen in the numerous studies that have been undertaken. However, challenges in implementing electronic health records (EHRs) have received some attention. Still, less is known about transitioning from legacy EHRs to newer systems.

EHRs are a way to capture and utilise real-time data to provide high-quality patient care, ensuring efficiency and effective use of time and resources. Incorporating EHR and CDSS together into the process of medicine has the potential to change the way medicine has been taught and practiced. It has been said that "the highest level of EHR is a CDSS".

Since "clinical decision support systems (CDSS) are computer systems designed to impact clinician decision making about individual patients at the point in time that these decisions are made", it is clear that it would be beneficial to have a fully integrated CDSS and EHR.

Even though the benefits can be seen, fully implementing a CDSS integrated with an EHR has historically required significant planning by the healthcare facility/organisation for the CDSS to be successful and effective. The success and effectiveness can be measured by the increased patient care being delivered and reduced adverse events occurring. In addition, there would be a saving of time and resources and benefits in terms of autonomy and financial benefits to the healthcare facility/organisation.

Benefits of CDSS combined with EHR

A successful CDSS/EHR integration will allow the provision of best practice, high-quality care to the patient, which is the ultimate goal of healthcare.

Errors have always occurred in healthcare, so trying to minimise them as much as possible is important to provide quality patient care. Three areas that can be addressed with the implementation of CDSS and Electronic Health Records (EHRs), are:

1. Medication prescription errors
2. Adverse drug events
3. Other medical errors

CDSSs will be most beneficial in the future when healthcare facilities are "100% electronic" in terms of real-time patient information, thus simplifying the number of modifications that have to occur to ensure that all the systems are up to date with each other.

The measurable benefits of clinical decision support systems on physician performance and patient outcomes remain the subject of ongoing research.

Barriers

Implementing electronic health records (EHR) in healthcare settings incurs challenges; none more important than maintaining efficiency and safety during rollout, but in order for the implementation process to be effective, an understanding of the EHR users' perspectives is key to the success of EHR implementation projects. In addition to this, adoption needs to be actively fostered through a bottom-up, clinical-needs-first approach. The same can be said for CDSS.

As of 2007, the main areas of concern with moving into a fully integrated EHR/CDSS system have been:

1. Privacy
2. Confidentiality
3. User-friendliness
4. Document accuracy and completeness
5. Integration
6. Uniformity
7. Acceptance
8. Alert desensitisation

as well as the key aspects of data entry that need to be addressed when implementing a CDSS to avoid potential adverse events from occurring. These aspects include whether:

• correct data is being used
• all the data has been entered into the system
• current best practice is being followed
• the data is evidence-based

A service oriented architecture has been proposed as a technical means to address some of these barriers.

Status in Australia

As of July 2015, the planned transition to EHRs in Australia is facing difficulties. Most healthcare facilities are still running completely paper-based systems; some are in a transition phase of scanned EHRs or moving towards such a transition phase.

Victoria has attempted to implement EHR across the state with its HealthSMART program, but it has cancelled the project due to unexpectedly high costs.

South Australia (SA) however is slightly more successful than Victoria in the implementation of an EHR. This may be because all public healthcare organisations in SA are centrally run.

(However, on the other hand, the UK's National Health Service is also centrally administered, and its National Programme for IT in the 2000s, which included EHRs in its remit, was an expensive disaster.)

SA is in the process of implementing "Enterprise patient administration system (EPAS)". This system is the foundation for all public hospitals and health care sites for an EHR within SA, and it was expected that by the end of 2014, all facilities in SA will be connected to it. This would allow for successful integration of CDSS into SA and increase the benefits of the EHR. By July 2015 it was reported that only 3 out of 75 health care facilities implemented EPAS.

With the largest health system in the country and a federated rather than a centrally administered model, New South Wales is making consistent progress towards statewide implementation of EHRs. The current iteration of the state's technology, eMR2, includes CDSS features such as a sepsis pathway for identifying at-risk patients based upon data input to the electronic record. As of June 2016, 93 of 194 sites in-scope for the initial roll-out had implemented eMR2

Status in Finland

The EBMEDS Clinical Decision Support service provided by Duodecim Medical Publications Ltd is used by more than 60% of Finnish public health care doctors.

Research

Prescription errors

A study in the UK tested the Salford Medication Safety Dashboard (SMASH), a web-based CDSS application to help GPs and pharmacists find people in their electronic health records who might face safety hazards due to prescription errors. The dashboard was successfully used in identifying and helping patients with already registered unsafe prescriptions and later it helped monitoring new cases as they appeared.

Legal expert system

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Legal_expert_system 

A legal expert system is a domain-specific expert system that uses artificial intelligence to emulate the decision-making abilities of a human expert in the field of law. Legal expert systems employ a rule base or knowledge base and an inference engine to accumulate, reference and produce expert knowledge on specific subjects within the legal domain.

Purpose

It has been suggested that legal expert systems could help to manage the rapid expansion of legal information and decisions that began to intensify in the late 1960s. Many of the first legal expert systems were created in the 1970s and 1980s.

Lawyers were originally identified as primary target users of legal expert systems. Potential motivations for this work included:

• speedier delivery of legal advice;
• reduced time spent in repetitive, labour intensive legal tasks;
• development of knowledge management techniques that were not dependent on staff;
• reduced overhead and labour costs and higher profitability for law firms; and
• reduced fees for clients.

Some early development work was oriented toward the creation of automated judges.

Later work on legal expert systems has identified potential benefits to non-lawyers as a means to increase access to legal knowledge.

Legal expert systems can also support administrative processes, facilitating decision making processes, automating rule-based analyses and exchanging information directly with citizen-users.

Types

Architectural variations

Rule-based expert systems rely on a model of deductive reasoning that utilizes "if A, then B" rules. In a rule-based legal expert system, information is represented in the form of deductive rules within the knowledge base.

Case-based reasoning models, which store and manipulate examples or cases, hold the potential to emulate an analogical reasoning process thought to be well-suited for the legal domain. This model effectively draws on known experiences our outcomes for similar problems.

A neural net relies on a computer model that mimics that structure of a human brain, and operates in a very similar way to the case-based reasoning model. This expert system model is capable of recognizing and classifying patterns within the realm of legal knowledge and dealing with imprecise inputs.

Fuzzy logic models attempt to create 'fuzzy' concepts or objects that can then be converted into quantitative terms or rules that are indexed and retrieved by the system. In the legal domain, fuzzy logic can be used for rule-based and case-based reasoning models.

Theoretical variations

While some legal expert system architects have adopted a very practical approach, employing scientific modes of reasoning within a given set of rules or cases, others have opted for a broader philosophical approach inspired by jurisprudential reasoning modes emanating from established legal theoreticians.

Functional variations

Some legal expert systems aim to arrive at a particular conclusion in law, while others are designed to predict a particular outcome. An example of a predictive system is one that predicts the outcome of judicial decisions, the value of a case, or the outcome of litigation.

Reception

Many forms of legal expert systems have become widely used and accepted by both the legal community and the users of legal services.

Challenges

Domain-related problems

The inherent complexity of law as a discipline raises immediate challenges for legal expert system knowledge engineers. Legal matters often involve interrelated facts and issues, which further compound the complexity.

Factual uncertainty may also arise when there are disputed versions of factual representations that must be input into an expert system to begin the reasoning process.

Computerized problem solving

The limitations of most computerized problem solving techniques inhibit the success of many expert systems in the legal domain. Expert systems typically rely on deductive reasoning models that have difficulty according degrees of weight to certain principles of law or importance to previously decided cases that may or may not influence a decision in an immediate case or context.

Representation of legal knowledge

Expert legal knowledge can be difficult to represent or formalize within the structure of an expert system. For knowledge engineers, challenges include:

• Open texture: Law is rarely applied in an exact way to specific facts, and exact outcomes are rarely a certainty. Statutes may be interpreted according to different linguistic interpretations, reliance on precedent cases or other contextual factors including a particular judge's conception of fairness.
• The balancing of reasons: Many arguments involve considerations or reasons that are not easily represented in a logical way. For instance, many constitutional legal issues are said to balance independently well-established considerations for state interests against individual rights. Such balancing may draw on extra-legal considerations that would be difficult to represent logically in an expert system.
• Indeterminacy of legal reasoning: In the adversarial arena of law, it is common to have two strong arguments on a single point. Determining the 'right' answer may depend on a majority vote among expert judges, as in the case of an appeal.

Time and cost effectiveness

Creating a functioning expert system requires significant investments in software architecture, subject matter expertise and knowledge engineering. Faced with these challenges, many system architects restrict the domain in terms of subject matter and jurisdiction. The consequence of this approach is the creation of narrowly focused and geographically restricted legal expert systems that are difficult to justify on a cost-benefit basis.

Current applications of AI in the legal field utilize machines to review documents, particularly when a high level of completeness and confidence in the quality of document analysis is depended upon, such as in instances of litigation and where due diligence play a role. Among the numerically most quantifiable advantages of AI in the legal field are the time and money saving impact by freeing lawyers from having to spend inordinate amounts of their valuable time on routine tasks, aiding in setting free lawyers’ creative energy by reducing stress. This in turn increases the rate of case load reduction by accomplishing better results in less time, which unlocks potential additional revenue per unit of time spend on a case. The cost of setting up and maintaining AI systems in law is more than offset by the attained savings through increased efficacy; unbalanced cost can be assigned to clients.

Lack of correctness in results or decisions

Legal expert systems may lead non-expert users to incorrect or inaccurate results and decisions. This problem could be compounded by the fact that users may rely heavily on the correctness or trustworthiness of results or decisions generated by these systems.

Examples

ASHSD-II is a hybrid legal expert system that blends rule-based and case-based reasoning models in the area of matrimonial property disputes under English law.

CHIRON is a hybrid legal expert system that blends rule-based and case-based reasoning models to support tax planning activities under United States tax law and codes.

JUDGE is a rule-based legal expert system that deals with sentencing in the criminal legal domain for offences relating to murder, assault and manslaughter.

Legislate is a knowledge graph powered contract management platform which applies legal rules to generate lawyer-approved contracts.

The Latent Damage Project is a rule-based legal expert system that deals with limitation periods under the (UK) Latent Damage Act 1986 in relation to the domains of tort, contract and product liability law.

Split-Up is a rule-based legal expert system that assists in the division of marital assets according to the (Australia) Family Law Act (1975).

SHYSTER is a case-based legal expert system that can also function as a hybrid through its ability to link with rule-based models. It was designed to accommodate multiple legal domains, including aspects of Australian copyright law, contract law, personal property and administrative law.

TAXMAN is a rule-based system that could perform a basic form of legal reasoning by classifying cases under a particular category of statutory rules in the area of law concerning corporate reorganization.

Controversies

There may be a lack of consensus over what distinguishes a legal expert system from a knowledge-based system (also called an intelligent knowledge-based system). While legal expert systems are held to function at the level of a human legal expert, knowledge-based systems may depend on the ongoing assistance of a human expert. True legal expert systems typically focus on a narrow domain of expertise as opposed to a wider and less specific domain as in the case of most knowledge-based systems.

Legal expert systems represent potentially disruptive technologies for the traditional, bespoke delivery of legal services. Accordingly, established legal practitioners may consider them a threat to historical business practices.

Arguments have been made that a failure to take into consideration various theoretical approaches to legal decision making will produce expert systems that fail to reflect the true nature of decision making. Meanwhile, some legal expert system architects contend that because many lawyers have proficient legal reasoning skills without a sound base in legal theory, the same should hold true for legal expert systems.

Because legal expert systems apply precision and scientific rigor to the act of legal decision-making, they may be seen as a challenge to the more disorganized and less precise dynamics of traditional jurisprudential modes of legal reasoning. Some commentators also contend that the true nature of legal practice does not necessarily depend on analyses of legal rules or principles; decisions are based instead on an expectation of what a human adjudicator would decide for a given case.

Recent developments

Since 2013, there have been significant developments in legal expert systems. Professor Tanina Rostain of Georgetown Law Center teaches a course in designing legal expert systems. Open-source platforms like Docassemble and companies such as Neota Logic and Checkbox have begun to offer artificial intelligence and machine learning-based legal expert systems.

Legal informatics

From Wikipedia, the free encyclopedia

https://en.wikipedia.org/wiki/Legal_informatics

Legal informatics is an area within information science.

The American Library Association defines informatics as "the study of the structure and properties of information, as well as the application of technology to the organization, storage, retrieval, and dissemination of information." Legal informatics therefore, pertains to the application of informatics within the context of the legal environment and as such involves law-related organizations (e.g., law offices, courts, and law schools) and users of information and information technologies within these organizations.

Policy issues

Policy issues in legal informatics arise from the use of informational technologies in the implementation of law, such as the use of subpoenas for information found in email, search queries, and social networks. Policy approaches to legal informatics issues vary throughout the world. For example, European countries tend to require destruction or anonymization of data so that it cannot be used for discovery.

Technology

Cloud computing

The widespread introduction of cloud computing provides several benefits in delivering legal services. Legal service providers can use the Software as a Service model to earn a profit by charging customers a per-use or subscription fee. This model has several benefits over traditional bespoke services.

• Software as a service is much more scalable. Traditional bespoke models require an attorney to spend more of a limited resource (their time) on each additional client. Using Software as a Service, a legal service provider can put in effort once to develop the product and then use a much less limited resource (cloud computing power) to provide service to each additional customer.
• Software as a service can be used to complement traditional bespoke services by handling routine tasks, leaving an attorney free to concentrate on bespoke work.
• Software as a service can be delivered more conveniently because it does not require the legal service provider to be available at the same time as the customer.

Software as a service also complicates the attorney-client relationship in a way that may have implications for attorney–client privilege. The traditional delivery model makes it easy to create delineations of when attorney-client privilege attaches and when it does not. But in more complex models of legal service delivery other actors or automated processes may moderate the relationship between a client and their attorney making it difficult to tell which communications should be legally privileged.

Artificial intelligence

See also: Artificial intelligence

Artificial intelligence is employed in online dispute resolution platforms that use optimization algorithms and blind-bidding. Artificial intelligence is also frequently employed in modeling the legal ontology, "an explicit, formal, and general specification of a conceptualization of properties of and relations between objects in a given domain".

Artificial intelligence and law (AI and law) is a subfield of artificial intelligence (AI) mainly concerned with applications of AI to legal informatics problems and original research on those problems. It is also concerned to contribute in the other direction: to export tools and techniques developed in the context of legal problems to AI in general. For example, theories of legal decision making, especially models of argumentation, have contributed to knowledge representation and reasoning; models of social organization based on norms have contributed to multi-agent systems; reasoning with legal cases has contributed to case-based reasoning; and the need to store and retrieve large amounts of textual data has resulted in contributions to conceptual information retrieval and intelligent databases.

History

Although Loevinger, Allen and Mehl anticipated several of the ideas that would become important in AI and Law, the first serious proposal for applying AI techniques to law is usually taken to be Buchanan and Headrick. Early work from this period includes Thorne McCarty's influential TAXMAN project in the US and Ronald Stamper's LEGOL project in the UK. Landmarks in the early 1980s include Carole Hafner's work on conceptual retrieval, Anne Gardner's work on contract law, Rissland's work on legal hypotheticals and the work at Imperial College London on the representation of legislation by means of executable logic programs.

Early meetings of scholars included a one-off meeting at Swansea, the series of conferences organized by IDG in Florence and the workshops organised by Charles Walter at the University of Houston in 1984 and 1985. In 1987 a biennial conference, the International Conference on AI and Law (ICAIL), was instituted. This conference began to be seen as the main venue for publishing and the developing ideas within AI and Law, and it led to the foundation of the International Association for Artificial Intelligence and Law (IAAIL), to organize and convene subsequent ICAILs. This, in turn, led to the foundation of the Artificial Intelligence and Law Journal, first published in 1992. In Europe, the annual JURIX conferences (organised by the Jurix Foundation for Legal Knowledge Based Systems), began in 1988. Initially intended to bring together the Dutch-speaking (i.e. Dutch and Flemish) researchers, JURIX quickly developed into an international, primarily European, conference and since 2002 has regularly been held outside the Dutch speaking countries. Since 2007 the JURISIN workshops have been held in Japan under the auspices of the Japanese Society for Artificial Intelligence.

The interoperable legal documents standard Akoma Ntoso allows machine-driven processes to operate on the syntactic and semantic components of digital parliamentary, judicial and legislative documents, thus facilitating the development of high-quality information resources and forming a basis for AI tools. Its goal is to substantially enhance the performance, accountability, quality and openness of parliamentary and legislative operations based on best practices and guidance through machine-assisted drafting and machine-assisted (legal) analysis. Embedded in the environment of the semantic web, it forms the basis for a heterogenous yet interoperable ecosystem, with which these tools can operate and communicate, as well as for future applications and use cases based on digital law or rule representation.

Scope

Today, AI and law embrace a wide range of topics, including:

• Formal models of legal reasoning
• Computational models of argumentation and decision-making
• Computational models of evidential reasoning
• Legal reasoning in multi-agent systems
• Executable models of legislation
• Automatic legal text classification and summarization
• Automated information extraction from legal databases and texts
• Machine learning and data mining for e-discovery and other legal applications
• Conceptual or model-based legal information retrieval
• Lawbots to automate minor and repetitive legal tasks
• Risk assessment, pricing and timeline predictions of litigation using machine learning and artificial intelligence.

Formal models of legal reasoning

Formal models of legal texts and legal reasoning have been used in AI and Law to clarify issues, to give a more precise understanding and to provide a basis for implementations. A variety of formalisms have been used, including propositional and predicate calculi; deontic, temporal and non-monotonic logics; and state transition diagrams. Prakken and Sartor give a detailed and authoritative review of the use of logic and argumentation in AI and Law, together with a comprehensive set of references.

An important role of formal models is to remove ambiguity. In fact, legislation abounds with ambiguity: Because it is written in natural language there are no brackets and so the scope of connectives such as "and" and "or" can be unclear. "Unless" is also capable of several interpretations, and legal draftsman never write "if and only if", although this is often what they intend by "if". In perhaps the earliest use of logic to model law in AI and Law, Layman Allen advocated the use of propositional logic to resolve such syntactic ambiguities in a series of papers.

In the late 1970s and throughout the 1980s a significant strand of work on AI and Law involved the production of executable models of legislation, originating with Thorne McCarty's TAXMAN and Ronald Stamper's LEGOL. TAXMAN was used to model the majority and minority arguments in a US Tax law case (Eisner v Macomber), and was implemented in the micro-PLANNER programming language. LEGOL was used to provide a formal model of the rules and regulations that govern an organization, and was implemented in a condition-action rule language of the kind used for expert systems.

The TAXMAN and LEGOL languages were executable, rule-based languages, which did not have an explicit logical interpretation. However, the formalisation of a large portion of the British Nationality Act by Sergot et al. showed that the natural language of legal documents bears a close resemblance to the Horn clause subset of first order predicate calculus. Moreover, it identified the need to extend the use of Horn clauses by including negative conditions, to represent rules and exceptions. The resulting extended Horn clauses are executable as logic programs.

Later work on larger applications, such as that on Supplementary Benefits, showed that logic programs need further extensions, to deal with such complications as multiple cross references, counterfactuals, deeming provisions, amendments, and highly technical concepts (such as contribution conditions). The use of hierarchical representations was suggested to address the problem of cross reference; and so-called isomorphic representations were suggested to address the problems of verification and frequent amendment. As the 1990s developed this strand of work became partially absorbed into the development of formalisations of domain conceptualisations, (so-called ontologies), which became popular in AI following the work of Gruber. Early examples in AI and Law include Valente's functional ontology and the frame based ontologies of Visser and van Kralingen. Legal ontologies have since become the subject of regular workshops at AI and Law conferences and there are many examples ranging from generic top-level and core ontologies to very specific models of particular pieces of legislation.

Since law comprises sets of norms, it is unsurprising that deontic logics have been tried as the formal basis for models of legislation. These, however, have not been widely adopted as the basis for expert systems, perhaps because expert systems are supposed to enforce the norms, whereas deontic logic becomes of real interest only when we need to consider violations of the norms. In law directed obligations, whereby an obligation is owed to another named individual are of particular interest, since violations of such obligations are often the basis of legal proceedings. There is also some interesting work combining deontic and action logics to explore normative positions.

In the context of multi-agent systems, norms have been modelled using state transition diagrams. Often, especially in the context of electronic institutions, the norms so described are regimented (i.e., cannot be violated), but in other systems violations are also handled, giving a more faithful reflection of real norms. For a good example of this approach see Modgil et al.

Law often concerns issues about time, both relating to the content, such as time periods and deadlines, and those relating to the law itself, such as commencement. Some attempts have been made to model these temporal logics using both computational formalisms such as the Event Calculus and temporal logics such as defeasible temporal logic.

In any consideration of the use of logic to model law it needs to be borne in mind that law is inherently non-monotonic, as is shown by the rights of appeal enshrined in all legal systems, and the way in which interpretations of the law change over time. Moreover, in the drafting of law exceptions abound, and, in the application of law, precedents are overturned as well as followed. In logic programming approaches, negation as failure is often used to handle non-monotonicity, but specific non-monotonic logics such as defeasible logic have also been used. Following the development of abstract argumentation, however, these concerns are increasingly being addressed through argumentation in monotonic logic rather than through the use of non-monotonic logics.

Quantitative legal prediction

Both academic and proprietary quantitative legal prediction models exist. One of the earliest examples of a working quantitative legal prediction model occurred in the form of the Supreme Court forecasting project. The Supreme Court forecasting model attempted to predict the results of all the cases on the 2002 term of the Supreme Court. The model predicted 75% of cases correctly compared to experts who only predicted 59.1% of cases. Another example of an academic quantitative legal prediction models is a 2012 model that predicted the result of Federal Securities class action lawsuits. Some academics and legal technology startups are attempting to create algorithmic models to predict case outcomes. Part of this overall effort involves improved case assessment for litigation funding.

In order to better evaluate the quality of case outcome prediction systems, a proposal has been made to create a standardised dataset that would allow comparisons between systems.

Legal practice

Within the practice issues conceptual area, progress continues to be made on both litigation and transaction focused technologies. In particular, technology including predictive coding has the potential to effect substantial efficiency gains in law practice. Though predictive coding has largely been applied in the litigation space, it is beginning to make inroads in transaction practice, where it is being used to improve document review in mergers and acquisitions. Other advances, including XML coding in transaction contracts, and increasingly advanced document preparation systems demonstrate the importance of legal informatics in the transactional law space.

Current applications of AI in the legal field utilize machines to review documents, particularly when a high level of completeness and confidence in the quality of document analysis is depended upon, such as in instances of litigation and where due diligence play a role. Predictive coding leverages small samples to cross-reference similar items, weed out less relevant documents so attorneys can focus on the truly important key documents, produces statistically validated results, equal to or surpassing the accuracy and, prominently, the rate of human review.

Delivery of services

Advances in technology and legal informatics have led to new models for the delivery of legal services. Legal services have traditionally been a "bespoke" product created by a professional attorney on an individual basis for each client. However, to work more efficiently, parts of these services will move sequentially from (1) bespoke to (2) standardized, (3) systematized, (4) packaged, and (5) commoditized. Moving from one stage to the next will require embracing different technologies and knowledge systems.

The spread of the Internet and development of legal technology and informatics are extending legal services to individuals and small-medium companies.

Corporate legal departments

Corporate legal departments may use legal informatics for such purposes as to manage patent portfolios, and for preparation, customization and management of documents.

Defeasible reasoning

From Wikipedia, the free encyclopedia

In philosophical logic, defeasible reasoning is a kind of reasoning that is rationally compelling, though not deductively valid. It usually occurs when a rule is given, but there may be specific exceptions to the rule, or subclasses that are subject to a different rule. Defeasibility is found in literatures that are concerned with argument and the process of argument, or heuristic reasoning.

Defeasible reasoning is a particular kind of non-demonstrative reasoning, where the reasoning does not produce a full, complete, or final demonstration of a claim, i.e., where fallibility and corrigibility of a conclusion are acknowledged. In other words, defeasible reasoning produces a contingent statement or claim. Defeasible reasoning is also a kind of ampliative reasoning because its conclusions reach beyond the pure meanings of the premises.

Defeasible reasoning finds its fullest expression in jurisprudence, ethics and moral philosophy, epistemology, pragmatics and conversational conventions in linguistics, constructivist decision theories, and in knowledge representation and planning in artificial intelligence. It is also closely identified with prima facie (presumptive) reasoning (i.e., reasoning on the "face" of evidence), and ceteris paribus (default) reasoning (i.e., reasoning, all things "being equal").

According to at least some schools of philosophy, all reasoning is at most defeasible, and there is no such thing as absolutely certain deductive reasoning, since it is impossible to be absolutely certain of all the facts (and know with certainty that nothing is unknown). Thus all deductive reasoning is in reality contingent and defeasible.

Other kinds of non-demonstrative reasoning

Other kinds of non-demonstrative reasoning are probabilistic reasoning, inductive reasoning, statistical reasoning, abductive reasoning, and paraconsistent reasoning.

The differences between these kinds of reasoning correspond to differences about the conditional that each kind of reasoning uses, and on what premise (or on what authority) the conditional is adopted:

• Deductive (from meaning postulate or axiom): if p then q (equivalent to q or not-p in classical logic, not necessarily in other logics)
• Defeasible (from authority): if p then (defeasibly) q
• Probabilistic (from combinatorics and indifference): if p then (probably) q
• Statistical (from data and presumption): the frequency of qs among ps is high (or inference from a model fit to data); hence, (in the right context) if p then (probably) q
• Inductive (theory formation; from data, coherence, simplicity, and confirmation): (inducibly) "if p then q"; hence, if p then (deducibly-but-revisably) q
• Abductive (from data and theory): p and q are correlated, and q is sufficient for p; hence, if p then (abducibly) q as cause

History

Though Aristotle differentiated the forms of reasoning that are valid for logic and philosophy from the more general ones that are used in everyday life (see dialectics and rhetoric), 20th century philosophers mainly concentrated on deductive reasoning. At the end of the 19th century, logic texts would typically survey both demonstrative and non-demonstrative reasoning, often giving more space to the latter. However, after the blossoming of mathematical logic at the hands of Bertrand Russell, Alfred North Whitehead and Willard Van Orman Quine, latter-20th century logic texts paid little attention to the non-deductive modes of inference.

There are several notable exceptions. John Maynard Keynes wrote his dissertation on non-demonstrative reasoning, and influenced the thinking of Ludwig Wittgenstein on this subject. Wittgenstein, in turn, had many admirers, including the positivist legal scholar H. L. A. Hart and the speech act linguist John L. Austin, Stephen Toulmin and Chaïm Perelman in rhetoric, the moral theorists W. D. Ross and C. L. Stevenson, and the vagueness epistemologist/ontologist Friedrich Waismann.

The etymology of defeasible usually refers to Middle English law of contracts, where a condition of defeasance is a clause that can invalidate or annul a contract or deed. Though defeat, dominate, defer, defy, deprecate and derogate are often used in the same contexts as defease, the verbs annul and invalidate (and nullify, overturn, rescind, vacate, repeal, void, cancel, countermand, preempt, etc.) are more properly correlated with the concept of defeasibility than those words beginning with the letter d. Many dictionaries do contain the verb, to defease with past participle, defeased.

Philosophers in moral theory and rhetoric had taken defeasibility largely for granted when American epistemologists rediscovered Wittgenstein's thinking on the subject: John Ladd, Roderick Chisholm, Roderick Firth, Ernest Sosa, Robert Nozick, and John L. Pollock all began writing with new conviction about how appearance as red was only a defeasible reason for believing something to be red. More importantly Wittgenstein's orientation toward language-games (and away from semantics) emboldened these epistemologists to manage rather than to expurgate prima facie logical inconsistency.

At the same time (in the mid-1960s), two more students of Hart and Austin at Oxford, Brian Barry and David Gauthier, were applying defeasible reasoning to political argument and practical reasoning (of action), respectively. Joel Feinberg and Joseph Raz were beginning to produce equally mature works in ethics and jurisprudence informed by defeasibility.

By far the most significant works on defeasibility by the mid-1970s were in epistemology, where John Pollock's 1974 Knowledge and Justification popularized his terminology of undercutting and rebutting (which mirrored the analysis of Toulmin). Pollock's work was significant precisely because it brought defeasibility so close to philosophical logicians. The failure of logicians to dismiss defeasibility in epistemology (as Cambridge's logicians had done to Hart decades earlier) landed defeasible reasoning in the philosophical mainstream.

Defeasibility had always been closely related to argument, rhetoric, and law, except in epistemology, where the chains of reasons, and the origin of reasons, were not often discussed. Nicholas Rescher's Dialectics is an example of how difficult it was for philosophers to contemplate more complex systems of defeasible reasoning. This was in part because proponents of informal logic became the keepers of argument and rhetoric while insisting that formalism was anathema to argument.

About this time, researchers in artificial intelligence became interested in non-monotonic reasoning and its semantics. With philosophers such as Pollock and Donald Nute (e.g., defeasible logic), dozens of computer scientists and logicians produced complex systems of defeasible reasoning between 1980 and 2000. No single system of defeasible reasoning would emerge in the same way that Quine's system of logic became a de facto standard. Nevertheless, the 100-year headstart on non-demonstrative logical calculi, due to George Boole, Charles Sanders Peirce, and Gottlob Frege was being closed: both demonstrative and non-demonstrative reasoning now have formal calculi.

There are related (and slightly competing) systems of reasoning that are newer than systems of defeasible reasoning, e.g., belief revision and dynamic logic. The dialogue logics of Charles Hamblin and Jim Mackenzie, and their colleagues, can also be tied closely to defeasible reasoning. Belief revision is a non-constructive specification of the desiderata with which, or constraints according to which, epistemic change takes place. Dynamic logic is related mainly because, like paraconsistent logic, the reordering of premises can change the set of justified conclusions. Dialogue logics introduce an adversary, but are like belief revision theories in their adherence to deductively consistent states of belief.

Political and judicial use

Many political philosophers have been fond of the word indefeasible when referring to rights, e.g., that were inalienable, divine, or indubitable. For example, in the 1776 Virginia Declaration of Rights, "community hath an indubitable, inalienable, and indefeasible right to reform, alter or abolish government..." (also attributed to James Madison); and John Adams, "The people have a right, an indisputable, unalienable, indefeasible, divine right to that most dreaded and envied kind of knowledge – I mean of the character and conduct of their rulers." Also, Lord Aberdeen: "indefeasible right inherent in the British Crown" and Gouverneur Morris: "the Basis of our own Constitution is the indefeasible Right of the People." Scholarship about Abraham Lincoln often cites these passages in the justification of secession. Philosophers who use the word defeasible have historically had different world views from those who use the word indefeasible (and this distinction has often been mirrored by Oxford and Cambridge zeitgeist); hence it is rare to find authors who use both words.

In judicial opinions, the use of defeasible is commonplace. There is however disagreement among legal logicians whether defeasible reasoning is central, e.g., in the consideration of open texture, precedent, exceptions, and rationales, or whether it applies only to explicit defeasance clauses. H.L.A. Hart in The Concept of Law gives two famous examples of defeasibility: "No vehicles in the park" (except during parades); and "Offer, acceptance, and memorandum produce a contract" (except when the contract is illegal, the parties are minors, inebriated, or incapacitated, etc.).

Specificity

One of the main disputes among those who produce systems of defeasible reasoning is the status of a rule of specificity. In its simplest form, it is the same rule as subclass inheritance preempting class inheritance:

 (R1) if r then (defeasibly) q                  e.g., if bird, then can fly
 (R2) if p then (defeasibly) not-q              e.g., if penguin, then cannot fly
 (O1) if p then (deductively) r                 e.g., if penguin, then bird
 (M1) arguably, p                               e.g., arguably, penguin
 (M2) R2 is a more specific reason than R1      e.g., R2 is better than R1
 (M3) therefore, arguably, not-q                e.g., therefore, arguably, cannot fly

Approximately half of the systems of defeasible reasoning discussed today adopt a rule of specificity, while half expect that such preference rules be written explicitly by whoever provides the defeasible reasons. For example, Rescher's dialectical system uses specificity, as do early systems of multiple inheritance (e.g., David Touretzky) and the early argument systems of Donald Nute and of Guillermo Simari and Ronald Loui. Defeasible reasoning accounts of precedent (stare decisis and case-based reasoning) also make use of specificity (e.g., Joseph Raz and the work of Kevin D. Ashley and Edwina Rissland). Meanwhile, the argument systems of Henry Prakken and Giovanni Sartor, of Bart Verheij and Jaap Hage, and the system of Phan Minh Dung do not adopt such a rule.

Nature of defeasibility

There is a distinct difference between those who theorize about defeasible reasoning as if it were a system of confirmational revision (with affinities to belief revision), and those who theorize about defeasibility as if it were the result of further (non-empirical) investigation. There are at least three kinds of further non-empirical investigation: progress in a lexical/syntactic process, progress in a computational process, and progress in an adversary or legal proceeding.

Defeasibility as corrigibility

Here, a person learns something new that annuls a prior inference. In this case, defeasible reasoning provides a constructive mechanism for belief revision, like a truth maintenance system as envisioned by Jon Doyle.

Defeasibility as shorthand for preconditions

Here, the author of a set of rules or legislative code is writing rules with exceptions. Sometimes a set of defeasible rules can be rewritten, with more cogency, with explicit (local) pre-conditions instead of (non-local) competing rules. Many non-monotonic systems with fixed-point or preferential semantics fit this view. However, sometimes the rules govern a process of argument (the last view on this list), so that they cannot be re-compiled into a set of deductive rules lest they lose their force in situations with incomplete knowledge or incomplete derivation of preconditions.

Defeasibility as an anytime algorithm

Here, it is assumed that calculating arguments takes time, and at any given time, based on a subset of the potentially constructible arguments, a conclusion is defeasibly justified. Isaac Levi has protested against this kind of defeasibility, but it is well-suited to the heuristic projects of, for example, Herbert A. Simon. On this view, the best move so far in a chess-playing program's analysis at a particular depth is a defeasibly justified conclusion. This interpretation works with either the prior or the next semantical view.

Defeasibility as a means of controlling an investigative or social process

Here, justification is the result of the right kind of procedure (e.g., a fair and efficient hearing), and defeasible reasoning provides impetus for pro and con responses to each other. Defeasibility has to do with the alternation of verdict as locutions are made and cases presented, not the changing of a mind with respect to new (empirical) discovery. Under this view, defeasible reasoning and defeasible argumentation refer to the same phenomenon.

Fallibilism

From Wikipedia, the free encyclopedia

Charles Sanders Peirce around 1900. Peirce is said to have initiated fallibilism.

Originally, fallibilism (from Medieval Latin: fallibilis, "liable to err") is the philosophical principle that propositions can be accepted even though they cannot be conclusively proven or justified, or that neither knowledge nor belief is certain. The term was coined in the late nineteenth century by the American philosopher Charles Sanders Peirce, as a response to foundationalism. Theorists, following Austrian-British philosopher Karl Popper, may also refer to fallibilism as the notion that knowledge might turn out to be false. Furthermore, fallibilism is said to imply corrigibilism, the principle that propositions are open to revision. Fallibilism is often juxtaposed with infallibilism.

Infinite regress and infinite progress

According to philosopher Scott F. Aikin, fallibilism cannot properly function in the absence of infinite regress. The term, usually attributed to Pyrrhonist philosopher Agrippa, is argued to be the inevitable outcome of all human inquiry, since every proposition requires justification. Infinite regress, also represented within the regress argument, is closely related to the problem of the criterion and is a constituent of the Münchhausen trilemma. Illustrious examples regarding infinite regress are the cosmological argument, turtles all the way down, and the simulation hypothesis. Many philosophers struggle with the metaphysical implications that come along with infinite regress. For this reason, philosophers have gotten creative in their quest to circumvent it.

Somewhere along the seventeenth century, English philosopher Thomas Hobbes set forth the concept of "infinite progress". With this term, Hobbes had captured the human proclivity to strive for perfection. Philosophers like Gottfried Wilhelm Leibniz, Christian Wolff, and Immanuel Kant, would elaborate further on the concept. Kant even went on to speculate that immortal species should hypothetically be able to develop their capacities to perfection. This sentiment is still alive today. Infinite progress has been associated with concepts like science, religion, technology, economic growth, consumerism, and economic materialism. All these concepts thrive on the belief that they can carry on endlessly. Infinite progress has become the panacea to turn the vicious circles of infinite regress into virtuous circles. However, vicious circles have not yet been eliminated from the world; hyperinflation, the poverty trap, and debt accumulation for instance still occur.

Already in 350 B.C.E, Greek philosopher Aristotle made a distinction between potential and actual infinities. Based on his discourse, it can be said that actual infinities do not exist, because they are paradoxical. Aristotle deemed it impossible for humans to keep on adding members to finite sets indefinitely. It eventually led him to refute some of Zeno's paradoxes. Other relevant examples of potential infinities include Galileo's paradox and the paradox of Hilbert's hotel. The notion that infinite regress and infinite progress only manifest themselves potentially pertains to fallibilism. According to philosophy professor Elizabeth F. Cooke, fallibilism embraces uncertainty, and infinite regress and infinite progress are not unfortunate limitations on human cognition, but rather necessary antecedents for knowledge acquisition. They allow us to live functional and meaningful lives.

Critical rationalism

Main article: Critical rationalism

 

The founder of critical rationalism: Karl Popper

In the mid-twentieth century, several important philosophers began to critique the foundations of logical positivism. In his work The Logic of Scientific Discovery (1934), Karl Popper, the founder of critical rationalism, tried to solve the problem of induction by arguing for falsifiability as a means to devalue the verifiability criterion. He adamantly proclaimed that scientific truths are not inductively inferred from experience and conclusively verified by experimentation, but rather deduced from statements and justified by means of deliberation and intersubjective consensus within a particular scientific community. Popper also tried to resolve the problem of demarcation by asserting that all knowledge is fallible, except for knowledge that was acquired by means of falsification. Hence, Popperian falsifications are temporarily infallible, until they have been retracted by an adequate research community. Although critical rationalists dismiss the fact that all claims are fallible, they do belief that all claims are provisional. Counterintuitively, these provisional statements can become conclusive once logical contradictions have been turned into methodological refutations. The claim that all assertions are provisional and thus open to revision in light of new evidence is widely taken for granted in the natural sciences.

Popper insisted that verification and falsification are logically asymmetrical. However, according to the Duhem-Quine thesis, statements can neither be conclusively verified nor falsified in isolation from auxiliary assumptions (also called a bundle of hypotheses). As a consequence, statements are held to be underdetermined. Underdetermination explains how evidence available to us may be insufficient to justify our beliefs. The Duhem-Quine thesis should therefore erode our belief in logical falsifiability as well as in methodological falsification. The thesis can be contrasted with a more recent view posited by philosophy professor Albert Casullo, which holds that statements can be overdetermined. Overdetermination explains how evidence might be considered sufficient for justifying beliefs in absence of auxiliary assumptions. Philosopher Ray S. Percival holds that the Popperian asymmetry is an illusion, because in the action of falsifying an argument, scientists will inevitably verify its negation. Thus, verification and falsification are perfectly symmetrical. It seems, in the philosophy of logic, that neither syllogisms nor polysyllogisms will save underdetermination and overdetermination from the perils of infinite regress.

Furthermore, Popper defended his critical rationalism as a normative and methodological theory, that explains how objective, and thus mind-independent, knowledge ought to work. Hungarian philosopher Imre Lakatos built upon the theory by rephrasing the problem of demarcation as the problem of normative appraisal. Lakatos' and Popper's aims were alike, that is finding rules that could justify falsifications. However, Lakatos pointed out that critical rationalism only shows how theories can be falsified, but it omits how our belief in critical rationalism can itself be justified. The belief would require an inductively verified principle. When Lakatos urged Popper to admit that the falsification principle cannot be justified without embracing induction, Popper did not succumb. Lakatos' critical attitude towards rationalism has become emblematic for his so called critical fallibilism. While critical fallibilism strictly opposes dogmatism, critical rationalism is said to require a limited amount of dogmatism. Though, even Lakatos himself had been a critical rationalist in the past, when he took it upon himself to argue against the inductivist illusion that axioms can be justified by the truth of their consequences. In summary, despite Lakatos and Popper picking one stance over the other, both have oscillated between holding a critical attitude towards rationalism as well as fallibilism.

Fallibilism has also been employed by philosopher Willard V. O. Quine to attack, among other things, the distinction between analytic and synthetic statements. British philosopher Susan Haack, following Quine, has argued that the nature of fallibilism is often misunderstood, because people tend to confuse fallible propositions with fallible agents. She claims that logic is revisable, which means that analyticity does not exist and necessity (or a priority) does not extend to logical truths. She hereby opposes the conviction that propositions in logic are infallible, while agents can be fallible. Critical rationalist Hans Albert argues that it is impossible to prove any truth with certainty, not only in logic, but also in mathematics.

Mathematical fallibilism

Imre Lakatos, in the 1960s, known for his contributions to mathematical fallibilism

In Proofs and Refutations: The Logic of Mathematical Discovery (1976), philosopher Imre Lakatos implemented mathematical proofs into what he called Popperian "critical fallibilism". Lakatos's mathematical fallibilism is the general view that all mathematical theorems are falsifiable. Mathematical fallibilism deviates from traditional views held by philosophers like Hegel, Peirce, and Popper. Although Peirce introduced fallibilism, he seems to preclude the possibility of us being mistaken in our mathematical beliefs. Mathematical fallibilism appears to uphold that even though a mathematical conjecture cannot be proven true, we may consider some to be good approximations or estimations of the truth. This so called verisimilitude may provide us with consistency amidst an inherent incompleteness in mathematics. Mathematical fallibilism differs from quasi-empiricism, to the extent that the latter does not incorporate inductivism, a feature considered to be of vital importance to the foundations of set theory.

In the philosophy of mathematics, the central tenet of fallibilism is undecidability (which bears resemblance to the notion of isostheneia; the antithesis of appearance and judgement). Two distinct types of the word "undecidable" are currently being applied. The first one relates to the continuum hypothesis; the hypothesis that a statement can neither be proved nor be refuted in a specified deductive system. The continuum hypothesis was proposed by mathematician Georg Cantor in 1873. This type of undecidability is used in the context of the independence of the continuum hypothesis, namely because this statement is said to be independent from the axioms in Zermelo–Fraenkel set theory combined with the axiom of choice (also called ZFC). Both the hypothesis and its negation are consistent with these axioms. Many noteworthy discoveries have preceded the establishment of the continuum hypothesis.

In 1877, Cantor introduced the diagonal argument to prove that the cardinality of two finite sets is equal, by putting them into a one-to-one correspondence. Diagonalization reappeared in Cantors theorem, in 1891, to show that the power set of any countable set must have strictly higher cardinality. The existence of the power set was postulated in the axiom of power set; a vital part of Zermelo–Fraenkel set theory. Moreover, in 1899, Cantor's paradox was discovered. It postulates that there is no set of all cardinalities. Two years later, polymath Bertrand Russell would invalidate the existence of the universal set by pointing towards Russell's paradox, which implies that no set can contain itself as an element (or member). The universal set can be confuted by utilizing either the axiom schema of separation or the axiom of regularity. In contrast to the universal set, a power set does not contain itself. It was only after 1940 that mathematician Kurt Gödel showed, by applying inter alia the diagonal lemma, that the continuum hypothesis cannot be refuted, and after 1963, that fellow mathematician Paul Cohen revealed, through the method of forcing, that the continuum hypothesis cannot be proved either. In spite of the undecidability, both Gödel and Cohen suspected the continuum hypothesis to be false. This sense of suspicion, in conjunction with a firm belief in the consistency of ZFC, is in line with mathematical fallibilism. Mathematical fallibilists suppose that new axioms, for example the axiom of projective determinacy, might improve ZFC, but that these axioms will not allow for dependence of the continuum hypothesis.

The second type of undecidability is used in relation to computability theory (or recursion theory) and applies not solely to statements but specifically to decision problems; mathematical questions of decidability. An undecidable problem is a type of computational problem in which there are countably infinite sets of questions, each requiring an effective method to determine whether an output is either "yes or no" (or whether a statement is either "true or false"), but where there cannot be any computer program or Turing machine that will always provide the correct answer. Any program would occasionally give a wrong answer or run forever without giving any answer. Famous examples of undecidable problems are the halting problem and the Entscheidungsproblem. Conventionally, an undecidable problem is derived from a recursive set, formulated in undecidable language, and measured by the Turing degree. Practically all undecidable problems are unsolved, but not all unsolved problems are undecidable. Undecidability, with respect to computer science and mathematical logic, is also called unsolvability or non-computability. In the end, both types of undecidability can help to build a case for fallibilism, by providing these fundamental thought experiments.

Philosophical skepticism

Main article: Philosophical skepticism

Fallibilism improves upon the ideas associated with philosophical skepticism. According to philosophy professor Richard Feldman, nearly all versions of ancient and modern skepticism depend on the mistaken assumption that justification, and thus knowledge, requires conclusive evidence or certainty. An exception can be made for mitigated skepticism. In philosophical parlance, mitigated skepticism is an attitude which supports doubt in knowledge. This attitude is conserved in philosophical endeavors like scientific skepticism (or rational skepticism) and David Hume's inductive skepticism (or inductive fallibilism). Scientific skepticism questions the veracity of claims lacking empirical evidence, while inductive skepticism avers that inductive inference in forming predictions and generalizations cannot be conclusively justified or proven. Mitigated skepticism is also evident in the philosophical journey of Karl Popper. Furthermore, Popper demonstrates the value of fallibilism in his book The Open Society and Its Enemies (1945) by echoing the third maxim inscribed in the forecourt of the Temple of Apollo at Delphi: "surety brings ruin".

But the fallibility of our knowledge — or the thesis that all knowledge is guesswork, though some consists of guesses which have been most severely tested — must not be cited in support of scepticism or relativism. From the fact that we can err, and that a criterion of truth which might save us from error does not exist, it does not follow that the choice between theories is arbitrary, or non-rational: that we cannot learn, or get nearer to the truth: that our knowledge cannot grow.

— Karl Popper

Fallibilism differs slightly from academic skepticism (also called global skepticism, absolute skepticism, universal skepticism, radical skepticism, or epistemological nihilism) in the sense that fallibilists assume that no beliefs are certain (not even when established a priori), while proponents of academic skepticism advocate that no beliefs exist. In order to defend their position, these skeptics will either engage in epochē, a suspension of judgement, or they will resort to acatalepsy, a rejection of all knowledge. The concept of epoché is often accredited to Pyrrhonian skepticism, while the concept of acatalepsy can be traced back to multiple branches of skepticism. Acatalepsy is also closely related to the Socratic paradox. Nonetheless, epoché and acatalepsy are respectively self-contradictory and self-refuting, namely because both concepts rely (be it logically or methodologically) on its existence to serve as a justification. Lastly, local skepticism is the view that people cannot obtain knowledge of a particular area or subject (e.g. morality, religion, or metaphysics).

Criticism

Nearly all philosophers today are fallibilists in some sense of the term. Few would claim that knowledge requires absolute certainty, or deny that scientific claims are revisable, though in the 21st century some philosophers have argued for some version of infallibilist knowledge. Historically, many Western philosophers from Plato to Saint Augustine to René Descartes have argued that some human beliefs are infallibly known. Plausible candidates for infallible beliefs include logical truths ("Either Jones is a Democrat or Jones is not a Democrat"), immediate appearances ("It seems that I see a patch of blue"), and incorrigible beliefs (i.e., beliefs that are true in virtue of being believed, such as Descartes' "I think, therefore I am"). Many others, however, have taken even these types of beliefs to be fallible.