Jusletter IT

The current notion of transparency can be considered as an ultimate synthesis of an opposition that can be analysed under different profiles: (1) ontology, between «order» and «system», (2) epistemology, between «experience» and «knowledge», (3) philosophy of law, between «sources of law» and «legal reasoning», namely «legal domain» and «legal ontology», and (4) legal informatics, between «legal information retrieval» and «legal artificial reasoning». In this paper, I aim to assert that: (1) transparency is not an attribute of the «system» as such, (2) to support the transparency of the legal system means to adhere to a formalistic conception of law, (3) in legal informatics, transparency is a feature of the «lattice» structure of law; (4) in legal information management, transparency can be regarded as the synthesis of «legal information retrieval» and «artificial legal reasoning, and (5) in Semantic Web, transparency can be found in folksonomies, of which here is provided a definition, a short functional description, and a brief assessment of possible applications to legal documents.

In this section, I provide some preliminary remarks on transparency about the legal system. For this purpose, I outline the following issues: (1) theoretical observation on «opacity» and transparency, (2) considerations on the «opacity» of the legal system having regard to the idea of «code» in modern legal thought, and (3) definition of current meaning of transparency as a process of «codification», focusing on its main implications.

First. The «system»¹ is the typical pattern of the modern thought, which has been adopted by the scientific method in opposition to the concept of «order»², establishing – or rather replacing – its own veracity. In this sense, the «system» should be defined as «opaque» since the light of knowledge cannot pass through it and reach the constituent structures of the «being». It is for this reason that, for instance, in Hegel the Spirit converts «system» into reality, and reality into «system»³. Since the «system» is «opaque» in itself⁴, we can argue that the attribution of transparency is an attempt to overcome the theoretical difficulties of «opacity» and the raise of «scepticism» and «scientism». On first side, we find «perspectivism». According to this approach, in the contemporary age would take place the «erosion of the «reality principle»⁵, and therefore would have vanished both the trust on the existence of a universal and absolute «order» and the confidence in the ability of reason to build perfect systems. In the postmodernity paradigm⁶, indeed, men choose their temporary truth, thus creating individual worlds as projections of their own personal beliefs. In this sense, the will pretends to transcend the limits of reality and subjectivity. On the second side, we can note the recent development of the «philosophy of information»⁷. Following this view, we could stipulate a transcendental dimension that joins the «order», the system and the observer. Nature and mind would combine in a unified vision⁸, so that information would become the synthesis of reality and knowledge. In this sense, «being» and «existence» merge themselves in the «infosphere»⁹. We can say that transparency refers to a transcendental dimension in which the truth of the metaphysical «order» and the certainty of the modern «system» are not admitted, since there are only semantic fluctuations that can be represented as algebraic functions¹⁰.

Second. The opposition between «order» and «system» affects the modern legal thought. This seems clear considering how the ordo juris of the classic tradition has been replaced by modern «legal system»¹¹. We can argue that also in modern legal thought the «system» is conceived as «opaque»¹². This could be explained recalling briefly the historical experience of continental codifications, and particularly the two versions of it – the French and the German –, reconnecting each of them to one of the issues mentioned above: the French code to the definition of boundaries of the experience which has to be represented, the German code to the internal consistency of the model that has to be built¹³. On one hand, the French doctrine resumed the Roman tradition of civil law as the material of the «Code Napoleon», whose enactment was simultaneous with the elimination of all other sources of law¹⁴. On the other hand, the Pandectists in Germany used the jus commune to organize the existing legislation into a conceptual model¹⁵. Both in the sources of law and in legal reasoning we can still find a conceptual framework isolated from reality, locked in its artificial perfection, insensitive to social changes. Thus, with respect to the sources, law can be seen an entity in its own right and can be properly named as «legal domain»; having regard to the legal reasoning, conversely, law qualifies a «legal ontology» imprinted in our mind, actions and language¹⁶.

Third. The most recent studies are attempting to overcome the limitations of «opacity» combining the «theory of the sources of law» with the «theory of legal reasoning». It can be said that nowadays the «code» is not embodied in a legal text, as in the modern tradition, nor in a conceptual framework (regardless of whether it is logical or linguistic), as in contemporary thought, but should be identified in the process that continuously produces the law. In this sense, the legal system becomes both resource and result of the relentless activity of coding and recoding behavioural patterns by means of interaction with social environment. This notion of «codification» matches the meaning of transparency identified in the previous section¹⁷. In this sense, two issues emerge outwards and inwards: (1) the «system» requires opening itself to the changing influences of its context¹⁸, and (2) the «system» needs to articulate in an adaptive organization¹⁹.

In this section, I claim that the legal informatics can be properly defined as the study to represent the legal system in terms of transparency. In support of this statement, I provide the following arguments: (1) few preliminary remarks on transparency in legal informatics, (2) a brief insight into legal information management, and (3) a focus on the resulting meaning of the «lattice» model commonly used in recent studies.

First. At the level of legal informatics, transparency is the goal of maximum efficiency and effectiveness of information’s processes. It denotes the perfect situation in which there are no barriers to the exchanges since structures are constantly adapting to their changing environment, and functions feedback to each other. Applying this notion, we can imagine a continuous exchange of information among legal system, society and individuals. However, we should point out that if institutions change their pattern according to the needs of the social context, and rules continuously aggregate and shape social relations, then ethical values ​​are nothing more than symbols that we, human beings, need to be driven to respect social rules²⁰.

Second. At this point, we could broadly define the legal information management as the study of automatic elaboration of the law. The two major research directions are the legal information retrieval and legal artificial reasoning: the first approach, which can be defined as inferential or «bottom-up», from legal texts aims to build legal concepts and formulate legal rules²¹; the second, which is called deductive or «top-down», intends to classify legal documents using criteria derived from a pre-established conceptual representation²². What is interesting here is that these perspectives feature the same problems mentioned above. Regarding the «legal domain», the issue is the «openness» of the system, namely the identification of the limits of the represented experience and the definition of criteria for data mining. Concerning the «legal ontology», the difficulty is in the construction of knowledge, namely the need to adapt the existing categories to information collected further. The central point is that the most recent studies attempt to combine these two methodologies, on one hand by improving the selection and analysis of data, on the other hand by increasing the change in taxonomies²³. This synthesis precisely confirms the importance of transparency in the «legal informatics»: the artificial representation of the experience and knowledge – «legal domain» and «legal ontology» – appears as a whole information process.

Third. After observing that the researches concerning legal informatics achieve their natural completion in the paradigm of transparency, it is useful to dwell on how it is expressed. In this regard, it is interesting to note that the most recent studies use the conceptual tool of the «lattice» structure. It is not simply a suggestive trope or a widespread topos, but it has a precise meaning. The decentralized network, indeed, is the conceptual pattern that best describes also two huge phenomena: (1) the structure of social relations²⁴, (2) the backbone of computer networks²⁵. There is no denying that today «many-many communication» – which is the hallmark of the Internet²⁶ – is the most efficient technique to share resources on a global level. Therefore, we can say that transparency needs a distributed «lattice» structure to perform the processes of resources» sharing.

In this section, I argue that the folksonomies are today the most advanced tool to achieve transparency on the Internet and that they can be fruitfully used in legal information management. To support this assumption I wish to: (1) provide some preliminary notions to define folksonomies; (2) describe their main features, (3) explain broadly their meaning relying on what reported in the previous paragraphs, and (4) evaluate their potential applications to legal information management.

First. As we know, in the last twenty years the Web²⁷ has evolved in the Semantic Web²⁸. Today users may publish content and spread it worldwide, even if they do not have technical skills. The fast growth of unstructured data has required tools for cataloguing information quickly and easily. The most effective solutions are tagging tools, which consist in provide the URI (Uniform Resource Identifier) with metadata that describe the resources according to users preferences. By tagging we can: (1) describe the contents of an object, (2) label the item freely, without having to follow a pre-set taxonomy, (3) use any lexical expression, even belonging to natural language, (4) allocate many tags to an object or assign the same tag to different objects, and (5) share or recommend our choices and preferences. Semantic patterns resulting from the use of tags are commonly defined as «folksonomies»²⁹ and consist of sets of associations among three elements: (1) the users (people who actually place the tagging), (2) the tags themselves, and (3) the resources being tagged³⁰.

Second. Scholars observed that in folksonomies an implicit agreement typically arises among users in the choice of tags, thus creating a stable and consistent core of meaning which may be suitable as a classification scheme for the resource. In other words, through the analysis of tags it is possible to build a semantic representation of the data collected. Without commenting technical details³¹, we can detect some exciting features. First, immediacy. We can observe that, looking to the future, this aspect would likely increase with the growth of the mean-device interaction through voice commands (e.g. SIRI). Second, spontaneity. Nobody forces a user to publish content and to tag items, yet thousands of people do it every single second. Of course, users share information and metadata for different personal reasons, but we can say that the main aggregating factor is a common underlying interest. We have to admit that it works. Third, language. The task of organizing information is done by inserting lexical terms in the system and through the association of their meaning with the tagged object. We can argue that tagging could be considered a kind of language-game concerning the description of reality³². Fourth, «lattice» structure. From the analysis of the relations between tags we can obtain – besides descriptions of resources and their meanings – also details about the structure of the links between users, and hence elements which might be useful for describing the social structure.

Third. I consider appropriate to face the theoretical aspects of folksonomies resuming the order of the issues previously addressed. Thus, I begin with engaging in epistemological question, I continue with those related to the philosophy of law, and I conclude with some remarks from the point of view of legal informatics. Regarding the first issue, I can argue that folksonomies express the synthesis between «perspectivism» and «philosophy of information». According to «perspectivism», the activity of tagging could confirm that the individual is naturally led to ascribe meaning to its context in a way that escapes his full control, since the truth is believed not to be exclusively rational. In this sense, users shape the resources in order to build their own reality just matching their needs, which besides can be wrong, misunderstood or provisional. It does not matter since simply the «domain» created by user is the synthesis of what previously we called «order» and «system». With regard to the «philosophy of information», we observed that tags automatically organize resources spontaneously, continuously and collectively. The fact that these processes are the result of the synergy of individual autonomous choices could lead one to consider that there are links underlying raw data, and so to believe that there should be an overall and unifying vision that could always explain the apparent chaos of our experience of life. I’d say this is quite a mystic point of view. With regard to the philosophy of law, it can be argued that folksonomies fits perfectly into the flow of information that surrounds people, things and institutions, and which is produced by the continuous process of «codification» of social relations. Within such a deterministic context, in which the events are caused – directly or indirectly – by interactions among operators, even law must adapt passively to this ever-changing context. Furthermore, I may suggest that also human existence, till in its inner dimension, becomes nothing more than a semantic process, which cannot be nothing but provisional, unceasing and fast: the individual conscience is guided by slogans, icons and keywords, rather than by deep convictions, which require an awareness that the individual is not supposed to have or improve. As for the legal informatics, we should note that labelling belongs to our common experience. In folksonomies, resources are described with metadata by humans and not by computers³³. On this basis, we can make two observations regarding: (1) the structure of the communication, (2) the effects of folksonomies from the cognitive point of view. With regard to the first aspect, it is true, indeed, that through tag’s distribution users create a set of semantic relations among the resources, but it is also true that the connections replicate the structure of relationships among the same users. Essentially, what enhances and reinforces the sharing of resources is precisely the «lattice» conceptual pattern of the connections built by folksonomies. Considering the second issue, scholars claim that the association of tags – and thus the increase of the metadata – influences the way users select the resources and browse the Web, since tags facilitate the semantic associations between resources, using the same model that allows the human mind to form a chain of connection among memories³⁴.

Fourth. With particular regard to the legal information management, attention must be focused on three features of folksonomies: (1) the possible combinations of tags is virtually infinite, (2) metadata may refer not only to the resources, but also to the way they interact with their environment, and (3) the descriptions may refer to the individual attitude towards the resources. As regards the first aspect, the practical disadvantages given by the lexical and linguistic extreme variety – thus difficult to manage by computers – can be reduced by technical measures, as seen above. On the second point, it is important to outline that the legal documents can be described in all their aspects, both extrinsic (for instance, in a legislative text: the issuing authority, serial number, date of publication, date of appeal) and intrinsic (for example, in a decision: specific rhetoric figures, arguments for appealing). As for the third profile, users can annotate their reaction about the legal document (for example, impressions, feelings or comments). In this latter issue can be appreciated the major difference compared to the «top-down» and «bottom-up» approaches – or to their combinations – in legal information management. Hence, we can identify four implications: (1) regarding the «top-down» approach, ontologies could become more resilient to the «lattice» structure of social relations, (2) on the «bottom-up» approach, there could be combined documents that do not belong strictly to the theory of the sources of law (mainly: literature, judicial sentences and administrative rulings), thus allowing to widen the «legal domain» however minimizing the risk of inserting irrelevant data, selected manually by users, (3) with regard to the feedback on the laws, it should be possible to disclose the «sentiment»³⁵ of citizens, thus providing an additional tool in the hands of the lawmakers³⁶, and (4) with respect to individuals, self-organizing legal documents could help navigate within a legal system increasingly complicated.

Transparency suffers from fundamental difficulties: (1) the «order» is not a «system» and the «system» is not an «order», so a «system» cannot be, as such, «transparent», (2) experience cannot be detached from «being» and knowledge is, after all, a mere representation of the experience, (3) there is no collective intelligence³⁷ in the proper sense, and there is no supposed «wisdom of crowds»³⁸, nor could they be the result of the automatic computation, and (4) to conceive the law as «transparent» means to break the link between the juridical experience and the meaning of human existence, and prevent it from giving a substance to legal values​​.

More than sixty years ago, the legal informatics was developed as the «next step forward» of jurisprudence³⁹. Today, perhaps we might ask ourselves whether folksonomies might be a «next step» in the legal informatics. Of course, it is necessary to overcome the theoretical difficulties concerning the notion of transparency, especially if – as was shown in the previous sections – the legal informatics can be considered as a study of the transparency of the legal system. Yet, folksonomies can be placed within this paradigm, and may be particularly useful in the field of legal information management. Perhaps a more thorough study, possibly deepening also the topology of social connections and the cognitive sciences, could help to overcome at least some of the difficulties mentioned above and maybe reconnect – without replacing – the «system» to «order».

American Civil Liberties Union v. Reno, decision of the Court for the Eastern District of Pennsylvania, 11th June 1996, in 929 F. Supp. 824, 830–849 (ED Pa. 1996), (1996).

American Civil Liberties Union v. Reno, decision of the Supreme Court of the United State No. 96–511, 19th March 1997–26th June 1997, in 521 U. S. 844 (1997), 850, (1997).

Article 7, Loi 30 Ventôse XII (21st March, 1804).

Baran, Paul, On Distributed Communications Networks. RAND Corporation papers, P-2626, RAND (1962).

Bateson, Gregory, Mind and nature: a necessary unity. Dutton, New York (1979).

Berners-Lee, Tim/Bray, Tim/Connolly, Dan/Cotton, Paul/Fielding, Roy/Jeckle, Mario/Lilley, Chris/Mendelsohn, Noah/Orchard, David/Walsh, Norman, Architecture of the World Wide Web. W3C (2004).

Berners-Lee, Tim/Cailliau, Robert/Luotonen, Ari/Nielsen, Henrik Frystyk/Secret, Arthur, The World-Wide Web. In: Communications of the ACM, Heft 37, S. 76–82 (1994).

Berners-Lee, Tim/Hendler, James/Lassila, Ora, The Semantic Web. In: Scientific American, Heft 284, S. 28–37 (2001).

Boer, Alexander, Legal Theory, Sources of Law and the Semantic Web Frontiers in artificial intelligence and applications, 195, IOS Press, Amsterdam (2009).

Cress, Ulrike/Held, Christoph/Kimmerle, Joachim, The collective knowledge of social tags: Direct and indirect influences on navigation, learning, and information processing. In: Computers & Education, Heft 60, S. 59–73 (2013).

de Groot, Huig, De jure belli ac pacis. Apud Nicolaum Buon, Paris (1625).

Dotsika, Fefie, Uniting formal and informal descriptive power. Reconciling ontologies with folksonomies. In: International Journal of Information Management, Heft 29, S. 407–415 (2009).

Engelbart, Douglas C., Augmenting Human Intellect: a conceptual framework. Stanford Research Institute (1962).

Fabro, Cornelio, The Transcendentality of Ens-Esse and the Ground of Metaphysics. In: International Philosophical Quarterly, Heft 6, S. 389–427 (1966).

Fernández-Barrera, Meritxell/Sartor, Giovanni, Classifications and the Law: Doctrinal Classifications vs. Computational Ontologies. Working Papers of the Law Department of the EUI, European University Institute (2010).

Floridi, Luciano, The philosophy of information. Oxford University Press, Oxford (2013).

Gödel, Kurt, Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I. In: Monatshefte für Mathematik, Heft 38, S. 173–198 (1931).

Halpin, Harry, Social semantics. The search for meaning on the Web. Semantic Web and Beyond, 13, Springer, New York (2013).

Hegel, Georg Wilhelm Friedrich, Grundlinien der Philosophie des Rechts oder Naturrecht und Staatswissenschaft im Grundrisse. Nicolaische Buchhandlung, Berlin (1821).

Hobbes, Thomas, Leviathan or The Matter, Forme and Power of a Common Wealth Ecclesiasticall and Civil. Printed for Andrew Crooke, London (1651).

Hotho, Andreas/Jäschke, Robert/Schmitz, Christoph/Stumme, Gerd, Information retrieval in folksonomies: Search and ranking. In: Sure Y, Domingue J (Hrsg.), The semantic web: research and applications, 4011, Springer, Berlin Heidelberg, S. 411–426 (2006).

Kant, Immanuel, Kritik der reinen Vernunft. 2. Auflage, J.F. Hartknock, Riga (1787).

Law, Edith/ von Ahn, Luis, Human Computation. Brachmann RJ, Cohen WW, Dietterich T, Synthesis Lectures on Artificial Intelligence and Machine Learning, 13, Morgan & Claypool Publishers, San Rafael (2011).

Lévy, Pierre, L’intelligence collective: pour une anthropologie du cyberspace. La Découverte, Paris (1994).

Loevinger, Lee, Jurimetrics: The Next Step Forward. In: Minnesota Law Review, Heft 33, S. 455–493 (1949).

Luhmann, Niklas, Das Recht der Gesellschaft. Suhrkamp, Frankfurt am Main (1993).

Luhmann, Niklas, Soziale Systeme: Grundriss einer allgemeinen Theorie. Suhrkamp, Frankfurt am Main (1984).

Luhmann, Niklas, Vertrauen: ein Mechanismus der Reduktion sozialer Komplexitaet. Soziologische Gegenwartsfragen, N. F., 28, F. Enke, Stuttgart (1968).

Lyotard, Jean-François, La condition postmoderne: rapport sur le savoir. Éditions de Minuit, Paris (1979).

Milgram, Stanley/Mann, Leon/Harter, Susan, The Lost-Letter Technique: A Tool of Social Research. In: The Public Opinion Quarterly, Heft 29, S. 437–438 (1965).

Palmirani, Monica/Ognibene, Tommaso/Cervone, Luca, Legal rules, text, and ontologies over time. Proc. Of The RuleML2012@ECAI Challenge, at the 6th International Symposium on Rules, Montpellier, (August 27–29, 2012), CEUR-WS, S. 61–78 (2012).

Ross, Alf, Tû-tû. In: Harvard Law Review, Heft 70, S. 812–825 (1957).

Surowiecki, James, The Wisdom of crowds: Why the many are smarter than the few and how collective wisdom shapes business, economies, societies and nations. Random House, New York (2004).

Teilhard de Chardin, Pierre, L’avenir de l’homme. Oeuvres de Pierre Teilhard de Chardin, 5, Éditions du Seuil, Paris (1959).

Ulpianus, Eneus Domitius, Liber Primum Institutionum, D. 1, 1, 1..

Unnamalai, K., Sentiment Analysis of Products Using Web. In: Procedia Engineering, Heft 38, S. 2257–2262 (2012).

Vander Wal, Thomas, Explaining and showing broad and narrow folksonomies. http://www.vanderwal.net/random/entrysel.php?blog=1635 angerufen 3. Januar 2014 (2005).

Vander Wal, Thomas, You down with folksonomy. http://www.vanderwal.net/random/entrysel.php?blog=1529 angerufen 3. Januar 2014 (2004).

Vattimo, Gianni, The Transparent Society. Johns Hopkins University Press, Baltimore (1992).

von Ahn, Luis/Maurer, Benjamin/McMillen, Colin/Abraham, David/Blum, Manuel, reCAPTCHA: Human-Based Character Recognition via Web Security Measures. In: Science, Heft 321, S. 1465–1468 (2008).

Wittgenstein, Ludwig, Philosophical investigations. Blackwell, Oxford (1953).

Zagrebelsky, Gustavo, Il diritto mite: legge, diritti, giustizia. Einaudi Contemporanea, 14, Einaudi, Torino (1992).

---

 

Federico Costantini

Università degli Studi di Udine, Dipartimento di Scienze giuridiche

Via Treppo 18, 33100 Udine, IT

federico.costantini@uniud.it

 

---