[德]奥利弗・罗伊1, [波]彼得・库里基2, 陈钰 译
1.拜罗伊特大学 哲学系, 拜罗伊特 95447
2.卢布林天主教大学 哲学系, 卢布林 20950

[作者简介] 1.奥利弗・罗伊(https://orcid.org/0000-0002-9085-5701),男,德国拜罗伊特大学哲学系教授,博士生导师,哲学博士,主要从事道义逻辑、认知逻辑、博弈论、集体意向性等方面的研究; 2.彼得・库里基(https://orcid.org/0000-0001-5413-3886),男,波兰卢布林天主教大学哲学系教授,博士生导师,哲学博士,主要从事道义逻辑、量子信息处理等方面的研究。陈钰(https://orcid.org/0000-0002-0783-0969),男,清华大学哲学系博士研究生,主要从事动态认知逻辑、动态语义学、结构证明论等方面的研究。

Legal Permissibility and Legal Competences in Hierarchical Systems with Strong Permission
Olivier Roy1, Piotr Kulicki2, translated by Chen Yu
1.Department of Philosophy, University of Bayreuth, Bayreuth 95447, Germany
2.Faculty of Philosophy, John Paul II Catholic University of Lublin, Lublin 20950, Poland
Abstract

G.H.von Wright distinguished two kinds of permission, naming them weak and strong. A weak permission is simply an absence of prohibition. A strong permission, in contrast, is an explicit statement independent of any obligation or prohibition. Unlike some legal theorists claiming that there are no strong permissions in the actual legal texts and legal practice, we believe that they can be found there in the form of rights and freedom. We try to study how strong permissions issued in hierarchical legal systems can change the legal status of the elements of the systems.
We focus on the situations in which a conflict appears between authorities of different levels, e.g. in European legal system some lower level state regulations are not in accordance with the higher level EU law. What we want to discuss is the way in which such inconsistency can be modeled in terms of legal abilities and legal permissibility. The two notions imply two different interpretations of such complex legal situations. From the practical point of view, according to one of the interpretations, the lower level law is not binding whatsoever, and in the other one it is, but as the lower law was introduced against the more fundamental law, it should be changed, and the lower level authority (a member state) may be punished.
We are interested in logical modeling of both the ability and the permissibility approach in hierarchical legal systems. A logical model can be useful mainly because it makes the assumptions of each approach explicit and precise, and may show hidden consequences of those assumptions. Thus, it can serve as a tool for legal theorists and designers of legal systems to help them understand the possible options of interpretation of the results of strong permissions issued in hierarchical systems. Moreover, logical models can be used when artificial intelligence tools are designed to analyze legal systems of the kind or to act within them.
The logical model that we employ consists of two relatively modular parts: one to represent static duties and permissions, and the other one to represent the agent's potential to change these deontic relations. In the static part we need to express Hohfeldian notions of claims and freedoms that are the standard conceptual foundation in this field of research. In particular, directed obligations and permissions are important for us. These are obligations and permissions of one agent towards another. We will use here a family of obligation operators of the formOij to represent the fact that the agent i has an obligation concerning the agent j. The semantics of the operator is based on a preference-action model where preference relation plays a crucial role. In short, we understand that is obligatory when it holds in all most preferred states.
The dynamic part covers the Hohfeldian notions of power and immunity. We represent power and immunity by using tools developed in dynamic epistemic logic. We transfer them into the field of deontic logic and define how a new deontic situation is determined by a previous situation and deontic actions performed in it. Formally a deontic action model is used here along with an operator that transforms a preference-action model and a deontic action model into a new preference-action model.
This formal machinery allows us to model strong permission issued by a higher level authority understood both in the spirit of legal abilities and legal permissibility. The modeling of power and immunity in terms of deontic action model makes some interesting predictions which would be worth testing against existing legal systems. The two approaches, referring to legal abilities and legal permissibility can be related to two fundamental legal principles ″ lex superior derogat legi inferiori″ and ″ lex posterior derogat legi priori″ respectively. More generally, the present modeling of powers predicts a high level of path-dependency in the exercise of powers because deontic actions are not commutative. Of course the prediction is not that this will always be the case. The question is which one is more in line with the actual regulations of EU and its member states, which would require further empirical research, and would also make the comparison with other legal systems particularly interesting. Our intuition is that the distinction between influencing legal ability and legal permissibility of a lower level authority is strongly connected with the issue of direct applicability of the higher level regulations, specifically permissions.
Finally, the model that we have presented here raises general questions regarding logical relationships between changes in legal ability and legal permissibility. When do they ″go hand in hand″? Can any change in legal permissibility and ability be captured in this framework? Answering these questions may have important consequences for policy making and design of legal procedures. The model we have studied will help us to understand the consequences of issuing permissions and changing legal competences in specific cases. We, however, leave these questions for the further research, which may be stimulated by this study.

Keyword: dynamic deontic logic; strong permission; preference-action model; deontic action model; legal permissibility; legal ability; legal power; norms in hierarchical systems

(一) 静态法律关系

φ ::=p|¬ φ |φ φ |Doiφ |Oijφ

W是非空世界集,

・ ≤ ijW上的自反和传递关系,

・ 对每个主体iAgt, ~iW上的等价关系,

V:Prop¬ (W)是一个赋值函数。

max(≤ ij[w])={v:¬ w'∈ ≤ ij[w]使得v< ijw'}。

M, w╞p当且仅当wV(p),

M, w╞¬ φ 当且仅当M, w φ ,

M, w╞φ ψ 当且仅当M, w╞φ 并且M, w╞ψ ,

M, w╞Oijφ 当且仅当对任意w'∈ max(≤ ij[w]), 都有M, w'╞φ

(二) 动态法律关系

A是非空且有穷的道义行动集,

$≤j→kAi$A上的自反和传递关系,

Pre:AL是前提函数,

Post:A→ (PropL)是后置条件函数, 它给每个行动和每个命题变元指派L中的一个公式。对所有aA, 我们假设Post(a)与恒等函数的区别只在于Prop至多为有穷多个元素。

A中的元素可看作i能够采取的可能的法律行动, 例如出售他的房子。关系 $≤j→kAi$表示i的行动改变jk之间的(静态)法律关系的可能性, 多数情况下, 也可看作一种法律上的相对理想关系。然而, 董惠敏和罗伊(① Dong H. & Roy O., ″Dynamic Logic of Legal Competences, ″ Manuscript.)注意到这种解释并不适用于对法律权能和法律允许已做出区分的情况。但在法律关系变化的可能性方面, 我们仍使用这一解释。前提函数Pre表示对A中可执行的不同的行动a, 在初始行为偏好模型中的一个特定的世界中为真的公式是什么。例如, 只有当i是该房子的合法所有者时, 出售房子才可执行。这就对应一些为真的道义命题。最后, 后置条件函数Post表示改变一些法律事实(这些法律事实能被某些命题变元表示)的能力。技术上讲, 该函数改变了初始静态语言中的一些(但并非所有)命题变元的真值。

W'={(w, a):M, w╞Pre(a), 其中aA},

・ (w, a)≤ 'ij(w', a')当且仅当a $< i→jAi$a', 或者a $≅i→jAi$a'并且wijw',

・ (w, a)~'i(w', a')当且仅当w~iw',

V'(p)={(w, a)∈ W':M, w╞Post(a)(p)}。

M, w╞[Ai, a]φ 当且仅当如果M, w╞Pre(a), 那么M$\otimes$Ai, (w, a)╞φ

M, w╞aA< Ai, a> ¬ T(j, k, ψ )。

M, w╞aA[Ai, a]T(j, k, ψ )。

¬ Oij¬ p

 Figure Option 图1 初始行为偏好模型M

 Figure Option 图2 模型ADE

MADE进行词典式更新, 便得到图3中的模型(模型M$\otimes$ADE:德国能创设j关于pi的请求权)。这正是我们所期望的, 因为p在最理想的世界(w1, a1)上为真, 所以在更新后的模型中, ij有义务p。使用霍菲尔德的术语, 这意味着j关于pi有请求权。回到初始模型M, 公式¬ Oijp∧ ∨ i=1, 2< ADE, ai> Oijp在所有点上为真。因此, 德国在上述定义的意义上有对ij的权力。

 Figure Option 图3 模型M$\otimes$ADE

¬ Oijp∧ ∨ i=1, 2< ADE, ai> Oijp∧ < AEU, b1> (¬ Oijp∧ ∧ i=1, 2[ADE, ai]¬ Oijp)。

 Figure Option 图4 模型AEU

 Figure Option 图5 模型M$\otimes$AEU

P(a)≡ def< Ai, a> ¬ V,

¬ P(a)≡ def[Ai, a]V

 Figure Option 图6 模型A'DE

¬ Oijp∧ ∨ i=1, 2(P(ai)∧ < ADE, ai> Oijp)。

 Figure Option 图7 模型(MAEU)$\otimes$A'DE

(感谢匿名审稿人对本文提出建设性的修改意见, 同时感谢陈钰将文章的英文版翻译为中文。)

 [1] von Wright G. H. , Norm and Action: A Logical Inquiry, London: Routledge & Kegan Paul, 1963. [本文引用:] [2] Kelsen H. , Pure Theory of Law, London: The Lawbook Exchange, 2005. [本文引用:] [3] von Wright G. H. , An Essay in Deontic Logic and the General Theory of Action, Amsterdam: North-Holland Publishing Company, 1968. [本文引用:] [4] Kanger S. , ″New Foundations for Ethical Theory, ″ in Hilpinen R. (ed. ), Deontic Logic: Introductory and Systematic Readings, Berlin: Springer, 1970, pp. 36-58. [本文引用:] [5] Ross A. , Directives and Norms, London: Routledge & Kegan Paul, 1968. [本文引用:] [6] Moore R. , ″Legal Permission, ″ ARSP: Archiv für Rechts- und Sozialphilosophie/Archives for Philosophy of Law and Social Philosophy, Vol. 59, No. 3 (1973), pp. 327-346. [本文引用:] [7] Raz J. , ″Permissions and Supererogation, ″ American Philosophical Quarterly, Vol. 12, No. 2 (1975), pp. 161-168. [本文引用:] [8] Hansson S. O. , ″The Varieties of Permission, ″ in Gabbay D. , Horthy J. & Parent X. et al. (eds. ), Hand book of Deontic Logic and Normative Systems, Berlin: Springer, 2013, pp. 195-240. [本文引用:] [9] Opałek K. & Woleński J. , ″Normative Systems, Permission and Deontic Logic, ″ Ratio Juris, Vol. 4, No. 3 (1991), pp. 334-348. [本文引用:] [10] Dong H. &Roy O. , ″Dynamic Logic of Power and Immunity, ″ in Baltag A. , Seligman J. & Yamada T. (eds. ), International Workshop on Logic, Rationality and Interaction, Berlin: Springer, 2017, pp. 123-136. [本文引用:] [11] Makinson D. , ″On the Formal Representation of Rights Relations, ″ Journal of Philosophical Logic, Vol. 15, No. 4 (1986), pp. 403-425. [本文引用:] [12] van Ditmarsch H. , van der Hoek W. & Kooi B. , Dynamic Epistemic Logic, Dordrecht: Springer Science & Business Media, 2007. [本文引用:] [13] van Benthem J. , Logical Dynamics of Information and Interaction, Cambridge: Cambridge University Press, 2011. [本文引用:] [14] Hohfeld W. N. , ″Some Fundamental Legal Conceptions as Applied in Judicial Reasoning, ″ The Yale Law Journal, Vol. 23, No. 1 (1913), pp. 16-59. [本文引用:] [15] Kanger S. & Kanger H. , ″Rights and Parliamentarism, ″ Theoria, Vol. 32, No. 2 (1966), pp. 85-115. [本文引用:] [16] Sartor G. , Legal Reasoning, Berlin: Springer, 2005. [本文引用:] [17] Herrestad H. & Krogh C. , ″Obligations Directed from Bearers to Counterparties, ″ in McCarty L. T. (ed. ), Proceedings of the Fifth International Conference on Artificial Intelligence and Law, New York: ACM Press, 1995, pp. 210-218. [本文引用:] [18] Markovich R. , ″Understand ing Hohfeld and Formalizing Legal Rights: The Hohfeldian Conceptions and Their Conditional Consequences, ″ Studia Logica, Vol. 108, No. 1 (2020), pp. 129-158. [本文引用:] [19] Boutilier C. , ″Conditional Logics of Normality: A Modal Approach, ″ Artificial Intelligence, Vol. 68, No. 1 (1994), pp. 87-154. [本文引用:] [20] van Benthem J. , van Otterloo S. & Roy O. , ″Preference Logic, Conditionals and Solution Concepts in Games, ″ in Lagerlund H. , Lindström S. & Sliwinski R. (eds. ), Modality Matters: Twenty-Five Essays in Honour of Krister Segerberg, Uppsala: Department of Philosophy, Uppsala University, 2006, pp. 61-67. [本文引用:] [21] Lindahl L. , Position and Change: A Study in Law and Logic, Dordrecht: Springer Science & Business Media, 1977. [本文引用:] [22] van Benthem J. , Grossi D. & Liu F. , ″Priority Structures in Deontic Logic, ″ Theoria, Vol. 80, No. 2 (2013), pp. 116-152. [本文引用:] [23] Meyer J. C. , ″A Different Approach to Deontic Logic: Deontic Logic Viewed as a Variant of Dynamic Logic, ″ Notre Dame Journal of Formal Logic, Vol. 29, No. 1 (1988), pp. 109-136. [本文引用:] [24] Bulygin E. , Essays in Legal Philosophy, Oxford: Oxford University Press, 2015. [本文引用:] [25] Andréka H. , Ryan M. & Schobbens P. Y. , ″Operators and Laws for Combining Preference Relations, ″ s Journal of Logic and Computation, Vol. 12, No. 1 (2002), pp. 13-53. [本文引用:]