What Is Jurimetrics;( scientific method to tackle legal problems)

Jurimetrics deals with the use of the scientific method to tackle legal problems, so as to strengthen legal certainty and predictability of the decision, together with its verifiability. The matter is growing: legislation and jurisprudence begin to make use of it.

  1. Historical origins

The historical origin of the term “jurimetry” is usually traced back to the publication of an American jurist Lee Loevinger ( Jurimetrics. The next step forward , in MinnLRev , 1949, 455), who thought of connecting a so-called exact science such as mathematics, capable to ensure certainty, with the so-called inaccurate law, with a lower degree of certainty.

The basic idea was to make the law measurable: that is, we wanted to make the sentence as a result of an operation that was always verifiable and not characterized by unpredictability.

In this way, it would always have been possible to verify its correctness, scrutinizing each passage: as for a complex mathematical operation, it is possible to go back to all the individual passages, so it should have been for a judicial measure; the sentence, in practice, was seen as the result of a series of measurable and verifiable operations.

George Boole, with the work The Mathematical Analysis of Logic, can also be mentioned in this context: he worked on formal logic built through mathematical / algebraic operations .

On the other hand, certainty has always been perceived as necessary in a system of law; legal certainty means that the clarity, knowability and uniqueness of the rules and the coherent and prompt application of the same by the authorities in charge, and especially by the judges, offer associates reasonable certainty about the legal consequences deriving from conduct or from a situation

In the past, Max Weber had already talked about the predictability and calculability of the law, while more recently the same issue has been tackled not only by Frosini, Losano, Borruso, Bellomo, but also by Irti, according to whom ” the rationality of calculation embraces even the right. The entrepreneur counts on the functioning of the administrative systems and on the application of the law. Therefore, he counts on the subsumptive judgment, in the sense that, opening a dispute, the concrete event will be brought back within the normative case: that case, which he knows and which allows him to calculate the juridical future ”

 

  1. Definition

Jurimetry can be defined as the science that studies the application of mathematical methods for solving legal problems; in particular, she has dealt with the “measurement” of judicial decisions and the construction of models for their predictability (Baade, HW, The Methodology of Legal Inquiry , in Baade, HW, Jurimetrics , New York-London, 1963, 8 ff. ).

It responds to the need to make the law more predictable, as a natural consequence of its certainty.

The same predictability, on the other hand, is a value because it makes citizens more free, equal and aware; this is especially true in criminal law where the need for guarantee is greater. The Constitutional Court (sentence of 29.5.2019, n. 134) recently affirmed that “the principle of legality, predictability and accessibility of the punishable conduct and of the sanction having a punitive-afflictive character, whatever the nomenattributed to it by the legal system, it cannot, by now, be considered a heritage derived not only from constitutional principles, but also from those of European conventional and supranational law, on the basis of which it is illegitimate to in the condition of “knowing”, in all its typified dimensions, the illegality of the omissive or commissive conduct actually carried out » .

Jurimetry does not aim at replacing human beings with something else, but at strengthening human activity itself through the aid of scientific subjects, among which mathematics stands out for its importance.

 

  1. Banking matters

In this aspiration for greater certainty to protect rights and the principle of equality, there are some reflections by operators who are convinced that some mathematical trigger in legal reasoning can be useful to obtain more calibrated decisions, at least in the presence of specific circumstances (Morelli, C. Jurimetry, increasingly mathematical in the world of justice, in Altalex.com , 2019).

Banking is the one that, from the beginning, has been more permeable to jurimetry, mainly due to a massive presence of calculations; by way of example, just think of art. 1283 cc that deals with the cd. anatocism, which involves the rejection of any exponential law (Manca, P., Banking jurimetry. Technical bases , Florence, 2019), or art. 1283 of the civil code on the calculation of the interest rate.

In this case there is no doubt that there is an inseparable interpenetration between law and mathematics, endorsed by the same letter of the law, where it says: “the rate of legal interests is determined in an amount equal to 0.05 percent year “, pursuant to art. 1284 cc

 

  1. Criminal law

Among the subjects that physiologically appear compatible, if not even permeated, of the cd. scientific method there is criminal law.

This, in its physiology, having to reach a degree of certainty beyond any reasonable doubt, pursuant to art. 533 cpp, can only rely on science; otherwise, it would easily lend itself to doubts.

The same jurisprudence makes use of the scientific method, above all to ascertain the fact, on some occasions, such as – merely by way of example – to determine the age of the victim (with the so-called Greulich – Pyle method, indicated in Cass. Pen. , 6.12.2019, no. 49685).

It must be said that no scientific method – due to its intrinsic fallibility – can prove the truth of a scientific law, with the consequence, inevitably, that even the expert opinion (as well as partisan advice) cannot be considered the bearer of an absolute truth ( thus Criminal Court, SU, 28.1.2019, n. 14426).

The validity of the scientific test passes through the verification of the reliability of the expert (skills, generic and specific professional experience), the reliability of the scientific method used and its correct application to the procedural case, all operations that also allow to distinguish the irrelevant or false opinions of the expert (so-called junk science) with motivated opinions on the basis of laws and methods scientifically tested and accredited by the scientific community (Criminal Cassation, 5.8.2019, n.35601).

With the clarification, however, that it is not up to the Supreme Court of Cassation to formulate generally principles regarding the greater adequacy of one scientific method in place of another, as its review is limited to the logical holding of the motivation, pursuant to art. 606, co. 1, lett. e, cpp (Criminal Court, 30.10.2019, n. 44289).

 

  1. Legislation that refers to algorithms

The inclusion of mathematical models in law, and in particular in legislation, is an increasingly widespread practice: law and mathematics are no longer totally irreconcilable “subjects”, but increasingly closer to the search for a synthesis, such as to make law more certain and predictable also in its practical application.

In this direction, some regulatory interventions militate:

-art. 1, co. 492, lett. and, l. 24.12.2012, n. 228 (published in the Official Gazette of 29.12.2012, n. 302) mentions a computer algorithm, preaching that “High frequency trading activities are considered to be those generated by a computer algorithm that automatically determines the decisions relating to sending , to the modification or cancellation of orders and related parameters, where the sending, modification or cancellation of orders on financial instruments of the same type are carried out with a minimum interval lower than the value established by the decree of the Minister of the Economy and of the finances referred to in paragraph 500 “;

 art. 1, co. lett. hh, Legislative Decree 3.8.2017, n. 129 (published in the Official Gazette of 25 August 2017, n. 198) deals with algorithmic trading, clarifying that this means “the trading of financial instruments in which a computerized algorithm automatically determines the individual parameters of the orders, such as for example the initiation of the order, the relative timing, the price, the quantity or the methods of managing the order after sending, with minimal or no human intervention, with the exception of systems used only to transmit orders to one or more trading venues, to process orders that do not entail the determination of trading parameters, to confirm orders or to settle transactions “;

– also recently, the operating procedures and the amount of the relocation allowance under the so-called citizenship income (art.9 dl 28.1.2019, n.4, conv., with mod., from l.28.3.2019, n.26) have been identified through an algorithm.

 

  1. Position of the doctrine

A large part of the doctrine sees the order in an objective sense, so as to become fully permeable to the scientific method; this because:

– the juridical order, understood as a set of dispositions of laws, is incompatible with contradiction and, therefore, objective; the scientific method starts from the premise that the legal system has the function of rationalizing behavior and human relations (Bellomo, F., The scientific method , in Law and Science , 1-2, 2015);

– the legal norm would have an objective and not a subjective meaning (Falzea, A., Introduction to legal sciences: the concept of law , Milan, 2008, 235);

– a computerized application of the law would even be possible, not the ascertainment of the facts, but only the uniform interpretation of the law which must be applied to those facts (established in the traditional ways by the trial judge); some laws would already be algorithms (Borruso, R, L’informatica del Legge, Milan, 2004, 316); it would also be possible to model the interpretation of the law (Viola, L. Interpretation of the law with mathematical models , Milan, 2018, 30); after all, the provisions subject to interpretation can be articulated with a sequence of logical algorithm (Cass., 8.8.2008, n. 21478);

-in order to ensure objectivity and, consequently, predictability, come to the rescue of the rules of interpretation in order to find the meaning of the law that is inferendus and not efferendus , ex art. 12 of the so-called Preleggi, or drawn from the inside and not imposed from the outside: a calculable right – i.e. based on normative cases, subsumption judgment and hermeneutical methodology – is able to offer the precision of its method, not justice, but positivity of choices consistency between rule and decision; any right, which meets the aforementioned characteristics, would be susceptible to preventive calculation (Irti, N., Un right incalculable , in Riv. dir. civ ., 2015, 1001).

 

  1. The jurisprudence

Jurisprudence has not infrequently been called upon to deal with mathematical tools applied to law; in detail, in recent years, she has dealt with the so-called algorithms.

The Council of State, with the sentence of 4.2.2020, n. 881 (in The New Civil Procedure , 2020, 1), stated that, in general terms, “three principles emerge from supranational law, to be taken into due consideration in the examination and use of IT tools:

– knowability;

– non-exclusivity of the algorithmic decision;

– algorithmic non-discrimination ” .

The use of these tools involves  “in reality a series of choices and assumptions that are anything but neutral: the adoption of predictive models and criteria on the basis of which data are collected, selected, systematized, ordered and put together, their interpretation and the consequent formulation of judgments are all operations that are the result of precise choices and values, conscious or unconscious; from this it follows that these tools are called upon to make a series of choices, which largely depend on the criteria used and the reference data used, on which it has often appeared difficult to obtain the necessary transparency ”  (thus Cons. St ., no. 881/2020, cit.).

Also in order to “apply the general and traditional rules in terms of imputability and liability, it is necessary to guarantee the traceability of the final algorithmic decision to the authority and competent body based on the law conferring power” (thus Cons. St., n . 881/2020, cit.).

The mechanism through which the robotic decision is made concrete (ie the algorithm) must be “knowable”, according to a strengthened declination of the principle of transparency, which also implies that of the full knowledge of a rule expressed in a language different from the legal one. This knowledge of the algorithm must be guaranteed in all aspects: from its authors to the procedure used for its elaboration, to the decision mechanism, including the priorities assigned in the evaluation and decision-making procedure and the data selected as relevant. This is in order to be able to verify that the results of the robotic procedure comply with the requirements and purposes established by law or by the administration itself upstream of this procedure and so that the methods and rules under which it is clear – and consequently can be questioned – are has been set. In other words, the “multidisciplinary characterization” of the algorithm (construction that certainly does not require only legal skills, but technical, IT, statistical, administrative) does not exempt from the need that the “technical formula”, which in fact represents the algorithm, it is accompanied by explanations that translate it into the underlying “legal rule” and that make it legible and understandable, both for citizens and for the judge. Forum it ., 2019, III, 606).

Administrative jurisprudence has scrutinized the topic of the algorithm applied to law also in the procedural declination: “regardless of the merits of the dispute – and, therefore, from the possibility both of qualifying the algorithm as an administrative document susceptible of access, and of granting prevalence to administrative transparency requirements with respect to those of confidentiality of the counter-interested party -, it emerges, as a preliminary ruling, the need to qualify the creator of an algorithm subject to request for access as a counter-interested party in the exhibition, being able to see compromised the right to keep secret the technical rule in which one’s creation is substantiated “(Cons. St., 2.1.2020, n. 30, in La Nuova Civile Procedure , 2020,1).

for this purpose, by dominating the same IT procedures set up as a serving function and to which an instrumental and merely auxiliary role must therefore be reserved today within the administrative procedure and never dominant or subrogatory of human activity; obstructing the deleterious Orwellian perspective of dismissing the reins of the investigative function and abdicating the provisional one, the garrison constituted by the bulwark of the constitutional values ​​carved in the articles 3, 24, 97 of the Constitution as well as art. 6 of the European Convention on Human Rights ” obstructing the deleterious Orwellian perspective of dismissing the reins of the investigative function and abdicating the provisional one, the garrison constituted by the bulwark of the constitutional values ​​carved in the articles 3, 24, 97 of the Constitution as well as art. 6 of the European Convention on Human Rights ” obstructing the deleterious Orwellian perspective of dismissing the reins of the investigative function and abdicating the provisional one, the garrison constituted by the bulwark of the constitutional values ​​carved in the articles 3, 24, 97 of the Constitution as well as art. 6 of the European Convention on Human Rights “ (TAR Lazio, Rome, 13.9.2019, n. 10964).

In terms of protection in practice, against any malfunctions of the algorithm used, in the event of an appeal, the concrete malfunction of the algorithm that governed the disputed operations must be attached and proved (Court of Milan, 14.2.2019, in La Nuova Procedure Civil , 2019, 3): id est , a generic dispute is not enough, but precision and demonstration are also needed in the concrete sense that, if the algorithm had worked correctly, the interested party would have obtained the hoped-for “advantage”.

 

  1. Jurimetry, Computer Science, Artificial Intelligence

As has been explained, jurimetry deals with the study concerning the measurement of law, understood as calculability and predictability (also in the sense of the so-called predictive justice); the approach to law is essentially mathematical; it can ignore the use of the computer and the aid of the so-called artificial intelligence.

Otherwise, information technology deals with the representation, organization and automatic processing of information; has a mathematical basis; it can disregard artificial intelligence.

Artificial intelligence consists in the study of mental faculties through computational models that make possible actions such as learning, reasoning and action, in order to design intelligent agents (de Kleer, J., An assumption-based TMS, in Artificial Intelligence , 1986, 28, 2, 127); these agents must be able to act humanly and to this end must be able to communicate with humans through processes defined by the Turing test:

  1. natural language processing(NLP), for linguistic communication;
  2. representation of knowledge to store it;
  3. automatic reasoning to use stored knowledge to answer questions and formulate new conclusions;
  4. dynamic learning ( machine learning) to adapt to new circumstances or to extrapolate new paths.

To pass the Turing test, the computer needs a computer vision to frame objects and robotic skills to manipulate and move them (Moro Visconti, R., Artificial intelligence: business models and evaluation profiles, in Riv. Dir . ind. , 2018, 421).

Artificial intelligence has been added to Boolean logic (Fiammella, B., Artificial intelligence and ethics, between progress and evolution , in Altalex.com , 2018); it is a very recent phenomenon that raises new questions of legal mediation, especially in terms of the development of tools suitable for translating the rising technoetic needs into “techno-law” (Ruffolo, U., Artificial intelligence, machine learning and algorithm responsibility , in Giur. It ., 2019, 7, 1657).

Artificial intelligence, if applied to law, takes into account jurimetry and information technology, but has its own autonomy (sometimes even decision-making).

 

  1. CONCLUSIONS

Jurimetry is assuming a growing role in the legal landscape, above all thanks to its ability to concretize the principle of equality pursuant to art. 3 of the Constitution: the more jurimetry is used, the more judicial decisions appear verifiable and impartial.

 

 

Leave a Comment