LEGAL KNOWLEDGE AND AGILITY IN PUBLIC ADMINISTRATION

• Author: Alexander Boer,Tom Van Engers
• DOI: http://doi.org/10.1002/isaf.1339
• Published date: 01 April 2013
• Date: 01 April 2013

LEGAL KNOWLEDGE AND AGILITY IN PUBLIC

ADMINISTRATION

ALEXANDER BOER*AND TOM VAN ENGERS

Leibniz Center for Law, University of Amsterdam, Amsterdam

SUMMARY

To address agility in public administration, we have developed a knowledge acquisition infrastructure for legal

knowledge, based on an implementation-oriented conceptualization of the legal system. Our objective is to reframe

legal knowledge as a knowledge source in a design-oriented task ontology, building on insights from the

CommonKADS methodology for intelligent system design. The main result is to make case law and legal expert

knowledge of critical incidents in organizations, two diagnostic knowledge sources underutilized in modern man-

agement and engineering, more accessible as a resource for design of agile organizational structures and intelligent

systems. Copyright © 2013 John Wiley & Sons, Ltd.

Keywords: knowledge management; knowledge acquisition; law

1. INTRODUCTION

Traditionally, legal theory insists on a division between legal theory per se, addressing the sphere of

law, and a sociology of power, addressing the sphere of power. By making this distinction, legal theory

has made the problem of organizing system-level implementation of the law into social structures –for

instance, business process compliance –somebody else’s problem.

Legal theory focuses its attention on the resulting frictions, and its effects in case law. Because legal

knowledge engineers tend to follow dominant positivist legal theory, intelligent systems in the field of

law are usually either based on the concept of nor mative reasoning by agents following formalized legal

rules, or the concept of legal argumentation about individual cases. Both represent operational perspec-

tives on legal knowledge: the first is addressed as a kind of planning problem, the second as a kind of

assessment problem. Both are used indirectly to validate organizational str uctures, and to evaluate

policy proposals. The formalization of normative rules prior to implementation of services, in support

of business processes, has become common practice in the executive suborganizations of the state we

work with. Critical incidents, however, happen after an implementation has been chosen, and legal

knowledge about these incidents influences the legal requirements engineering process in an indirect

way, mediated by legal experts in the organization. Mainly in this second area, organizational agility

is lacking.

The work reported in this article is part of a larger project that also included development of a refer-

ence service-oriented architecture for agility in public administration, reported in Gong (2012). The

reference infrastructure mainly deals with the problem of deploying new rules without interruption of

* Correspondence to: Alexander Boer, Leibniz Center for Law, University of Amsterdam, Amsterdam. E-mail: A.W.F.

Boer@uva.nl

Copyright © 2013 John Wiley & Sons, Ltd.

INTELLIGENT SYSTEMS IN ACCOUNTING, FINANCE AND MANAGEMENT

Intell. Sys. Acc. Fin. Mgmt. 20, 67–88 (2013)

Published online 20 May 2013 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/isaf.1339

running operations. Agility, however, is not an infrastructural concept. It is a matter of awareness of the

organizational environment and the ability to foresee problems and react appropriately swiftly.

In this articlewe present a knowledge acquisition infrastructure for legal knowledge, based onan appli-

cation of complexadaptive system and multi-agent system concepts tothe legal system. The agility prob-

lem called for a solution that makes the complex dynamics of implementation of law manageable, rather

than dealing withit as a matter of uncertainty and organizational robustness. Our approachwas to reframe

legal knowledge as a knowledge source in a design-oriented task ontology, building on insights from the

CommonKADS methodology for intelligent system design in Breuker (1994) on problem solving. This

has led us to considerthe problem of model-based diagnosis,of real and hypotheticalproblems, in the legal

system. Themain result of our work on diagnosisis to make case law and legal expertknowledge of critical

legal incidents in organizations, two knowledge sources underutilized in modern management, better ac-

cessibleas a knowledge resource for designof organizational structuresand supporting intelligentsystems.

We first sketch the problem setting, agility in public administration, in Section 2. Then we present the

concepts we developed with the Tax and Customs Administration and Immigration Service of the

Netherlands. The central concept is a design methodology that includes the legal diagnosis problem

type as a knowledge source for the design of social structures implementing the law.

In Section 4, we extend the model-based diagnosis problem definition, normally only applied to well-

understood physical domains, to include the domains of law and social structure. We introduce agents

as systems and agent roles as components implementing functions, based on a similarity between func-

tions and social roles. Moreover, we distinguish between case-specific remediations of operational

problems, system-level repairs of implementation problems and policy-level domain theory revision.

Finally, we show that the narrative and argumentative structure of case decisions are intimately re-

lated to the diagnosis problem definition, and the ingredients of a diagnosis problem space are apparent

in legal knowledge. As an example knowledge domain we use real-estate tax evasion, because the

Dutch tax administration has shown a particular interest in this domain. It has invested resources in a

large-scale survey of tax evasion patterns in this area, in parallel with our methodological project,

and shares expert knowledge with us to test the methodology. The tax evasion subject will certainly

appeal to an audience interested in finance and accounting, and increasingly attracts attention of the

public at large. An example case is developed in Section 4.3, before we conclude this article with an

overview of results and conclusions.

2. PROBLEM SETTING

Law is a social construct, a system of systems. It provides society with a mechanism to steer behaviour,

by promoting desired behaviour and discouraging undesired behaviour. To remain effective, law must

be regularly adapted to, and by, a changing regulated society. This makes the law a dynamic system, as

most complex systems are. A comprehensive theory of law should, therefore, account for the mecha-

nisms behind observed dynamics.

The executive suborganizations of the state, like the Dutch Tax and Customs Administration and

Immigration Service, traditionally plays a key role behind the scenes as drivers and observers of

change. Playing this part has become increasingly complex over the last decades, and the state, as a

consequence, has difficulty maintaining grip.

Legal pluralism is one of the problems. Besides the nation state, others have increasingly accumu-

lated power to create legal norms. The European Union, and international bodies, received institutions

with the power to create laws that sometimes contain direct norms, and more often indirect norms

68 A. BOER AND T. VAN ENGERS

Copyright © 2013 John Wiley & Sons, Ltd. Intell. Sys. Acc. Fin. Mgmt. 20,67–88 (2013)

DOI: 10.1002/isaf