Skip to navigation – Site map
Pravno sklepanje in argumentacija

Analogy and Balancing: The Partial Reducibility Thesis and Its Problems

David Duarte
p. 141–154
Translation(s):
Analogija i vaganje: teza o djelomičnoj svodljivosti i njezini problemi

Abstracts

With an analysis of the structure and the sequence of analogy, the paper is mainly a critique to the partial reducibility thesis: a thesis sustaining that analogy, besides a strictly analogical step, is in the remaining part reducible to balancing. Thus, the paper points out some problems raised by the partial reducibility thesis, such as the contingency of reducibility or the fact that a proper analogy is done under the cover of a balancing. The main point is, however, the claim that analogy and balancing have opposite normative conditions, being this premise the reason to a structural explanation for the unacceptability of the reducibility enterprise.

Top of page

Full text

1 Three as the Starting Point for Analogy and Balancing

  • 1 On the three basic operations, Alexy (2010: 9 and ff.), Brożek (2008: 188 and ff.), and Bustamante (...)
  • 2 The first premise seems undeniable: no legal case can be solved without the fulfilment of an antece (...)
  • 3 Brożek 2008: 188 and ff.

1The idea that the application of law includes three basic operations, subsumption, balancing and analogy, has recently become a central matter in legal theory, yielding new insights into the analysis of their connections: (i) subsumption and analogy, (ii) subsumption and balancing, and (iii) analogy and balancing.1 However, taking into account the fact that subsumption is performed in every case of the application of law, and analogy and balancing are used only under specific normative conditions, it follows that the third connection is the only one linking the basic operations that are, under this point of view, normatively contingent.2 This feature of analogy and balancing poses certain problems: (i) what are their respective specific normative conditions, (ii) is there any intersection between those conditions, and (iii) to what extent can analogy and balancing be combined or interfere with each other. It is precisely here that the partial reducibility thesis comes into play.3 Formulated as an explanation of analogy in terms of balancing, this thesis deals exactly with the problems raised by the connection between these two basic operations of the application of law.

2 Analogy by Steps: Some Basic Considerations Concerning the Sequence

  • 4 Naturally, the kind of analogy considered here is that regarding a classification and not the analo (...)
  • 5 On analogy steps, Araszkiewicz (2011: 103).

2Despite being used to represent the operation of comparison itself, analogy is, rigorously, a result: the establishment, for any purpose, of a relation of similarity.4 In order to reach the final point, some steps must be taken: (i) identification of the terms compared, (ii) list of comparison factors, (iii) evaluation of similarity or non-similarity under each factor, (iv) choice of the decisive factor, and (v) conclusion of analogy, if this is the case under the factor chosen.5 The overall operation is therefore relatively complex.

  • 6 Brewer 1996: 932. Peczenik 1996: 312.

3The operation is complex, immediately, with regard to factors of comparison. As it is known, they are naturally endless.6 Independently of what is compared, everything can be used as a factor, considering the illimitable properties of the terms and the infinite external criteria of analysis. Since the context of comparison is able to narrow the set of factors, this decreasing effect gives some manageability to the process.

- When comparing cars (c) and bicycles (b), the set of factors is endless: f1 - price, f2 - speed, f3 - metal texture, f4 - comfort, f5 - beauty, f6 - how it pleases John, and so on.
- If c and b are compared for buying purposes, the set is narrowed: some factors may become irrelevant; for instance, f3 (metal texture) or f6 (how it pleases John).

  • 7 On false analogy, among other analogy counterarguments, see Shelley (2002: 489).

4Complexity also arises from the evaluation of similarity or non-similarity that each factor confers. Beneath each one of them, a judgment has to be made in order to state or refute similarity. In a certain way, this step is the core of analogy: it is here that the terms compared are effectively confronted under the specific analysis of resemblance. Its importance cannot be overlooked. Often false analogies appear from the inconsistent judgments made at this level.7

  • 8 The symbols = and ≠ are used here for simplification purposes, just to represent similarity and non (...)

- Comparison between b and c under f1 (price) can lead to similarity (=) or non-similarity (≠).8
- Under f
2 (speed) it can also lead to ≠ or =; and so on for all the factors.

  • 9 And vice-versa. On quantitative similarity, Davis & Russell (1987: 265).

5Even if narrowed by context, a comparison may still have to be done under a plurality of factors. From this it follows that different evaluations of similarity and non-similarity may be carried out. Therefore, for the terms compared, a table with different results appears: terms are similar under some factors and non-similar under others. As would be expected, under a common list of factors, the more proximate the terms, the fewer the results of non-similarity.9

- For b and c: f1 ≠, f2 ≠, f3 =, f4 ≠, f5 ≠, f6 =.
- For b1 and b2, hypothetically: f1 =, f2 =, f3 =, f4 =, f5 ≠, f6 =.

  • 10 Or factors: all references to a decisive factor include, naturally, a set of decisive factors.
  • 11 Alexy 2010: 17. Aarnio 1987: 104.

6The complexity of the analogy lies, however, in the choice of the decisive factor.10 If certain terms are similar under a factor and non-similar under another, the conclusion of analogy is wholly dependent on the factor selected. This choice is, nonetheless, external to the comparison: the equal position of each factor with respect to the terms implies that the operation, in itself, holds no criteria for defining any kind of prevalence.11 Thus, the overall analogy operation is decided through a meta-factor: the one that decides which factor, all things considered, is to be chosen.

- For b and c: f1 ≠, f2 ≠, f3 =, f4 ≠, f5 ≠, f6 =.
- For
b and c: (f1 ≠) (f2 ≠) (f3 =) (f4 ≠) (f5 ≠) (f6 =).
- For
b and c, hypothetically, with a meta-factor (mfa) that chooses f3: b = c.

3 The Same Steps in a Legal Cases Analogy

  • 12 Nino 2011: 293. Weinreb 2005: 97 and ff. Guastini 1993: 370.

7The previous scheme is entirely applicable when the operation is used to provide a solution for a case unforeseen under any norm of the legal order. Here, where no answer to the legal question is provided, an operation of analogy is required to define whether the case at hand is similar to another which fulfils the conditions foreseen in an enacted norm. Legal cases, then, become the terms to be compared. If the conclusion is an analogy, then the prima facie inapplicable norm becomes the decision-norm of the case and the legal question is answered.12

  • 13 This example presupposes, obviously, a gap concerning the entrance of motorcycles into the park, in (...)

- Case: Motorcycle entrance into the park.
- No norm on motorcycle entrance into the park.
-
Rule1: “Cars are not allowed to enter the park.”
- “
Car entrance” (c) and “motorcycle entrance” (m) are terms of comparison.
- If
c and m are analogous, “entrance on a motorcycle” is not allowed.13

  • 14 Araszkiewicz 2011: 102. Roth 2000: 115.

8When terms of comparison are cases, as in legal reasoning by analogy, the context of comparison is given by the legal question: what matters here is to obtain an answer to the deontic status of an unforeseen action. The endless list of factors is narrowed precisely by that question. Factors with no bearing on that normative issue will be irrelevant for the comparison.14

- Cases: “Car entrance into the park” (c) and “Motorcycle entrance into the park” (m).
- Legal question: Is a motorcycle allowed to enter the park?
- Irrelevant factors: f1 - number of wheels, f2 - comfort for the driver, and so on.
- Relevant factors: f3 - pollutant, f4 - danger to pedestrians, and so on.

9Even if narrowed, there may still be a plurality of factors which may lead to a table with different results. No matter what the reliability of each evaluation is, any two cases may be similar under some factor and non-similar under another.

- For cases c and m, the factors of comparison are: f3 (pollutant), f4 (danger to pedestrians), f5 (drivers' freedom of action), f6 (damage to the park vegetation), and so on.
- For c and m: f3 ≠, f4 ≠, f5 =, f6 ≠.

10Now the results of similarity and non-similarity offered by the table of comparison present the main problem in the analogy operation: the selection of the decisive factor (step iv). As we have seen, if the decisive factor implies an evaluation of similarity, then a conclusion of analogy follows. If not, the legal question remains without a legal solution with regard to the analogy operation.

- For c and m: (f3 ≠) (f4 ≠) (f5 =) (f6 ≠).
- For
c and m: if f5 then “Motorcycles are not allowed to enter the park”.
- For
c and m: if f4 then no analogical answer.

11As is generally the case, the decisive point in legal reasoning by analogy also rests upon the definition of the meta-factor according to which the prevailing factor will be selected. However, this meta-factor must be sustained by the legal order: if in an analogy operation the conclusion supports something that works as a “new norm”, no other option can be upheld. The problem is, of course, how.

4 The Partial Reducibility Thesis

  • 15 Brożek 2008: 193.
  • 16 Brożek 2008: 194.

12In the context of the basic operations in the application of law, the partial reducibility thesis says that analogy can be partially explained through balancing.15 The thesis sustains only the partial autonomy of the analogy operation: besides a strictly analogical step, the remaining part is reducible to balancing. Thus, for the legal question at hand, the solution is drawn by weighing up the principles in conflict, as in any balancing.16

  • 17 Which is, as stated, an unproblematic level of similarity (Brożek 2008: 191).

13The partial reducibility thesis depends, however, on a distinction between two levels of similarity: similarity1 and similarity2. Similarity1 is equivalent to what was referred to here as the context of comparison and stands for the proximity between the cases brought by the legal question at hand. This question defines as similar1 all cases whose solution is an answer to the same legal problem, setting the boundaries for an initial stage of similarity.17 The cases selected here are then subject to further evaluation.

- Legal question: Are motorcycles allowed into the park?
- Rule1: “Cars (c) are not allowed into the park.”
- Rule2: “Bicycles (b) are allowed into the park.”
- No norm on motorcycle (m) entrance into the park. -
- Rule3: “Speed limit for bicycles (b) in the park is 10 km/h.”
- Cases with cars (c) and bicycles (b) are similar1 to the cases with motorcycles (m).
- Speed limit case b is dissimilar to the case m at this level of similarity1.

  • 18 Brożek 2008: 191.

14Similarity2 is the decisive level of similarity since it creates the norm for the case establishing the legal solution. Here, the cases that were selected at the level of similarity1 have to be compared and the decision on the “relevant similarity” must be taken. As similar1 cases are linked to different solutions, the answer to the legal question is given by the case that is evaluated as similar2. For this reason, similarity2 stands for the conclusive and deeper choice of similarity.18

- No norm on motorcycle (m) entrance into the park.
- Car case (c): “Cars are not allowed into the park.”
- Bicycles case (b): “Bicycles are allowed into the park.”
- If m = c then m ~P park.
- If m = b then m P park.

  • 19 Brożek 2008: 194.

15The core of the partial reducibility thesis lies, however, in the way similarity2 is defined and solved. As the cases selected under similarity1 are linked to a different solution, the main point is to discover the principles backing each one of those solutions. Since these principles by definition point in different directions, a common scenario of principles in conflict is created.19

  • 20 In its original presentation, the principle mentioned here is “people are entitled to rest actively (...)

- Car case (c) Rule1 (“Cars are not allowed into the park.”) is backed by principle P1.
-
Principle P1: “The environment should be protected by law.”
- Bicycles case
(b) Rule2 (“Bicycles are allowed into the park.”) is backed by principle P2.
-
Principle P2: “Everyone has freedom of action.”20
- With the question if motorcycles are allowed into the park, principles P
1 and P2 conflict.

  • 21 On the “collision rule” (law of competing principles), for instance, Alexy 2002: 54, and Pino 2010: (...)
  • 22 Brożek 2008: 195.

16In the case of conflicting principles, the solution is obtained through balancing, namely the use of the “weight formula”: here, principles are weighed and one of them will prevail over the other. Accordingly, the prevailing principle yields the solution of the case and a “collision rule” is created with the case facts and the consequence withdrawn from the winning principle.21 With this approach, the problem of similarity2 is transformed into the problem of the weighing of principles, which encompasses, as the partial reducibility thesis sustains, a significant advantage: instead of asking which terms are similar2, only simple balancing must be carried out.22

- Principle P1 (“The environment should be protected by law.”) → m ~P park.
- Principle P2 (“Everyone has freedom of action.”) → m P park.
- If with the “weight formula” P1 > P2m ~P park.

5 First Problem: Random Weighing Outcome

  • 23 Of course, nothing prevents there being more than one principle on each side of the question, reque (...)

17Despite it being of undeniable interest, the partial reducibility thesis poses several problems. The first and most intuitive concerns the choice of the principles that stand behind the rules governing similar1 cases. Albeit constructed as a simplified model, the thesis seems to disregard the fact that different principles may be available to back the rules that sustain similar1 cases. This is immediately relevant for two reasons. First, because it shows that the principle choice is more complex than it seems and, mainly, because the final result depends on the principles brought to weighing.23

- Car case (c): Rule1 (“Cars are not allowed into the park”) is backed by principle P1 and or principle P3.
- Principle P1: “The environment should be protected by law.”
- Principle P3: “Physical integrity is inviolable.”
- Bicycles case (b): Rule2 (“Bicycles are allowed into the park”) is backed by principle P2.
- Principle P2: “Everyone has freedom of action.”
- Legal question is still the same: Are motorcycles (m) allowed into the park?
- If interferences of P2 in P1 and in P3 are different (how motorcycles pollute and how they can damage physical integrity), the weighing result can differ.
- Hypothetically: P1 > P2 → m ~P park, but P3 < P2m P park.

6 Second Problem: Analogy Covered by Balancing

  • 24 When one considers that balancing, in its proper sense, just decides which norm among all those app (...)

18Related to the previous point, the partial reducibility thesis also seems to disregard the fact that the principles chosen to carry out the weighing process are materially connected with the factors used to compare cases: each principle within balancing describes a factor for the cases selected during the similarity1 phase. Therefore, not only does the choice of a principle describe the selection of a factor, but, furthermore, the balancing process is merely set out as a scheme for organising the choice of the decisive factor. Hence, this leads to the claim that balancing is used here only as a tool for determining the meta-factor: among the factors selected by the principles chosen, balancing just decides which one of them prevails. Under the cover of balancing, proper analogy is performed.24

- For c, b and m: f1 (pollutant), f2 (freedom of movement), f3 (danger to pedestrians).
- Principles P1, P2 and P3 represent f1, f2, and f3.
- Principle P1: “The environment should be protected by law.”; in f1: m = c; mb.
- Principle P2: “Everyone has freedom of action.”; in f2: m = c; m = b.
- Principle P3: “Physical integrity is inviolable”; in f3: mc; m = b.
- Selection of principles P1 and P2 means that the “main factors” are f1 and f2.
- If P1 > P2, then the decisive factor is f1: m = c; mb.
- P1 → f1m = c.

7 Third Problem: Balancing Irrelevant Principles

19Taking into account that, under the partial reducibility thesis, the replacement of the similarity2 phase is achieved by balancing principles that support rules inapplicable to a case, another problem arises: the irrelevance of the principles called on by those rules. In fact, in spite of similarity1, there is nothing to ensure that those principles are not irrelevant for the unregulated case, in which case they would be unable to justify any solution. The main reason for this result comes from the inapplicability of one of those principles to the case requiring an answer, which follows from the fact that the features describing that case do not match the principle's antecedent. Under this scenario, the principles called on under similarity1 play no role here. This leads to the following claim: the partial reducibility of analogy to balancing only works if the unregulated case, in spite of similarity1, is analogous to the regulated case to the point that it triggers exactly the same principles as those backing the rules whose application by analogy is being considered.

  • 25 This is even more visible if it is used as the original principle P2: “people are entitled to rest (...)

- Car case (c): Rule1 (“Cars are not allowed into the park”) is backed by principle P1.
- Principle P1: “The environment should be protected by law.”
- Bicycles case (b): Rule2 (“Bicycles are allowed into the park”) is backed by principle P2.
- Principle P2: “Everyone has freedom of action.”
- New similarity2 case: May a memorial tank (mt) enter the park?
- No norm on memorial tank (mt) entrance into the park.
- Under similarity1, nothing has changed: the question is still the permission to enter.
- Balancing of principles P1 and P2 for mt is meaningless: at least, principle P2 is irrelevant.
- The unregulated case does not fulfil the antecedent of principle P2.25

8 Fourth Problem: Not Enough Principles for Balancing

  • 26 For instance, Pino 2010: 185 and ff., and Duarte 2010: 56 and ff.

20The obstacle to the partial reducibility thesis that underlies the previous problem can be extended to all normative situations in which the rules governing similar1 cases are not based on any principle or, probably most often, in which those rules are supported by the same principle. In such situations no balancing is possible for the simple reason that there are not enough principles in play. It is known that balancing is an operation used to solve the normative conflicts unsolvable by norms of conflicts: for that reason, balancing requires two or more norms.26 Therefore, when cases are backed by the same principle, no balancing can take place.

- Rule1: “Car circulation is restricted to three days per week.”
- Rule2: “Plates with an even first digit: Mondays, Wednesdays and Fridays.”
- Rule3: “Plates with an odd first digit: Tuesdays, Thursdays and Saturdays.”
- No rule was enacted for the few cars whose plate starts with a letter.
- Even first digit case (e): Rule2 is backed by principle P1.
- Odd first digit case (o): Rule3 is baked by principle P1.
- Principle P1: “The environment should be protected by law.”
- Legal question: When can cars whose plates start with a letter circulate?
- Balancing is unusable: principle P1 backs both rules; no normative conflict exists.

9 The Background Problem: Analogy and Balancing Do Not Match

21All of the problems with the partial reducibility thesis identified so far are, in a sense, no more than a consequence of a larger one: analogy and balancing do not match. This becomes rather clearer when we take into account the fact that each one of these basic operations demands the opposite normative circumstances: while analogy depends on the absence of an applicable norm, balancing relies on the applicability of two or more norms. Based on their reverse opportunity, this opposition points towards a mutual exclusion. A case requiring an analogy does not call for balancing and vice-versa.

- Case (m): “Motorcycle entrance into the park.”
- Normative circumstances of analogy: No norm on m.
- Normative circumstances of balancing: About m, principle P1 conflicts with principle P2.

22The probable explanation for the overlap between analogy and balancing, notwithstanding their mutual exclusion, seems to be in a reductive understanding of the role played by principles as effective regulating norms. Acceptance of principles as norms like all others, which is an inevitable consequence of their deontic character, has to imply that they govern cases in the same way that rules do. All the particular features of principles, starting with their ability to be applied in various degrees, do not interfere with the fact that they provide for legal solutions as well: if a case fulfils the antecedent of a principle, then there is a legal consequence for that case. It is obvious that, if this principle conflicts with another, the solution becomes dependent on their balancing. However, once there is a weighing outcome, a final consequence is obtained and there is no place for intromitting an analogy.

- Legal question: Are motorcycles allowed in the park?
- Principle P1: “The environment should be protected by law.”
- Principle P2: “Everyone has freedom of action.”
- Motorcycle entrance is an instance of principles P1 and P2.
- No norm on “car entrance into the park”.
- No norm on “bicycle entrance into the park”.
- Legal solution is obtained through balancing of principles P1 and P2.

  • 27 But surely without analogy, as its normative circumstances are not present. This justifies why some (...)

23The reductive understanding of the role played by principles as effective regulating norms turns out to affect, then, the proper comprehension of what a gap is. If it still represents an absence of regulation, its extension must consider that both principles and rules are in the same way providing legal consequences, even though principles, bearing expansive antecedents which cover larger amounts of reality, have considerably narrowed the space for unregulated cases. Thus, for the application of law no gap exists if the case at hand is an instance of a principle's antecedent: if it triggers a rule, the rule is applicable, but if it triggers “only” a principle, the case is legally foreseen as well and, with or without balancing, the principle has to be applied.27

- Legal question: Are motorcycles allowed in the park?
- Principle P1: “The environment should be protected by law.”
- Principle P2: “Everyone has freedom of action.”
- Motorcycle entrance is an instance of principles P1 and P2.
- Rule1: “Cars are not allowed into the park.”
- Rule2: “Bicycles are allowed into the park.”
- Legal solution is not achieved by analogy.
- Legal solution is still obtained through balancing of principles P1 and P2.
- Legal question: Are cars allowed in the park?
-
Legal solution is given by Rule1.

  • 28 For instance, Alexy 2002: 47 and ff., and Brożek 2012: 223.
  • 29 The point here is that normative gaps exist only at the level of rules if rules created the gap or (...)
  • 30 All this presupposes that analogy, as an operation adoptable in order to create a decision-norm not (...)

24It seems that what is meant by “gap” has significantly changed pursuant to the distinction between principles and rules and, consequently, to what follows from the “optimisation requirement” character of principles.28 If one accepts that both principles and rules stipulate legal consequences, and thus that both are capable of solving cases in the very same way, the extension of “gap” now has a range that seems to cover only two normative situations: (i) the unlikely situation in which no norm is applicable to a case, neither a rule nor a principle, and (ii) the situation in which a rule, blocking the applicability of principles, has a consequence that has not been specified for a category that, among others, also belongs to its sphere.29 If these normative situations define what can enter into the extension of “gap”, it follows therefrom that the basic operation of analogy is confined to them.30

- Case1: Circulation of cars whose plates start with a letter.
- Gap situation1: If case1 is not covered by any rule or any principle (gap1).
- Gap situation2: If existing norms have unspecified consequences for case1, for instance:
- Rule1: “Car circulation is restricted to three days per week.”
- Rule2: “Plates with an even first digit: Mondays, Wednesdays and Fridays.”
- Rule3: “Plates with an odd first digit: Tuesdays, Thursdays and Saturdays.”
- No rule was enacted for the few cars whose plates start with a letter (gap2).

10 Final Remark

25Analogy is goal-oriented: no choice among comparison factors, particularly at the level of similarity2, can be carried out except in view of some purpose. Without some such purpose, the meta-factor deciding which factor prevails is indefinable and, for this reason, the choice among factors lacks justification and becomes a strictly arbitrary option. Since principles provide the ends adopted by a legal order both on a larger scale and more perceptibly, and since these ends are usually introduced in analogy in order to solve the meta-factor problem, the goal-oriented character of analogy has found its main source of operability in these kind of norms. Here, however, principles play a double role: as “end-providers” and, while giving direct solutions for cases, as “gap-decreasers”. This duplicity has to be treated carefully: whenever principles govern a case, they immediately put analogy aside and thus remove the possibility of them constituting the criteria with regard to what is similar to what.

Acknowledgments. I thank Hendrik Kaptein, Luís Duarte d’Almeida, Pedro Múrias, Pedro Moniz Lopes and the two anonymous referees for their comments on an earlier version of this paper.

Top of page

Bibliography

Aulis AARNIO, 1987: The Rational as Reasonable. Dordrecht: Kluwer.
Robert ALEXY, 2010: Two or Three? On the Nature of Legal Principles. Ed. Martin Borowski. Stuttgart: Franz Steiner Verlag. 9–18.
——, 2002: Theory of Constitutional Rights. Oxford: Oxford University Press.
Michal ARASZKIEWICZ, 2011: Analogy, Similarity and Factors. Proceedings of the 13th International Conference on Artificial Intelligence and Law. New York: ACM. 101–105.

Scott BREWER, 1996: Exemplary Reasoning: Semantics, Pragmatics, and the Rational Force of Legal Arguments by Analogy. Harvard Law Review 109 (1996) 5. 923–1029.
Bartosz BROŻEK, 2008: Analogy in Legal Discourse.
Archiv für Rechts- und Sozialphilosophie 94 (2008) 2. 188–201.
——, 2012: Legal Rules and Principles: a Theory Revisited. i-Lex, Rivista di Scienze Giuridiche, Scienze Cognitive ed Intelligenza Artificiale (2012) 17. 205–226.
Luka BURAZIN, 2014: Antinomies Between Implicit Legal Principles: a Solution to the Total-Partial Antinomy.
Courts, Interpretation, The Rule of Law. Eds. Miodrag Jovanović and Kenneth E. Himma. The Hague: Eleven International Publishing. 167–180.
Thomas BUSTAMANTE, 2012: Finding Analogies Between Cases. On Robert Alexy´s Third Basic Operation in the Application of Law.
On the Philosophy of Precedent. Ed. Carlos Bernal Pulido. Stuttgart: Franz Steiner Verlag. 59–71.

Todd DAVIES & Stuart RUSSELL, 1987: A Logical Approach to Reasoning by Analogy. Proceedings of the Tenth International Joint Conference on Artificial Intelligence. Ed. John P. McDermott. Burlington: Morgan Kaufmann Publishers. 264–270.
David DUARTE, 2010: Normative Conditions of Balancing: Drawing Up The Boundaries of Normative Conflicts That Lead to Balances.
Legal Reasoning: the Methods of Balancing. Ed. Jan Sieckmann. Stuttgart: Franz Steiner Verlag. 5162.

Riccardo GUASTINI, 1993: Le fonti del diritto e l´interpretazione. Milano: Giuffrè.

Hendrik KAPTEIN, 2005: Legal Progress Through Pragma-Dialectics? Prospects Beyond Analogy and A Contrario Argument. Argumentation 19 (2005) 4. 497–507.

Fabrizio MACAGNO & Douglas WALTON, 2009: Argument from Analogy in Law, the Classical Tradition, and Recent Theories. Philosophy & Rhetoric 42 (2009) 2. 154182.
Juan Ruiz MANERO, 2005: Algunas Concepciones del Derecho y sus Lagunas. Lagunas en el Derecho. Madrid: Marcial Pons. 103126.

Carlos Santiago NINO, 2003: Introducción al Análisis del Derecho. 11ª ed. Barcelona: Ariel.

Aleksander PECZENIK, 1996: Jumps and Logic in the Law. Artificial Intelligence and Law 4 (1996) 3-4. 297–329.
Giorgio PINO, 2010.
Diritti e Intepretazione. Bologna: Il Mulino.

Jorge Luis RODRÍGUEZ, 2000: Axiological Gaps and Normative Relevance. Archiv für Rechts- und Sozialphilosophie 86 (2000). 151–167.
Bram ROTH, 2000: New Reasoning Patterns in Analogical Legal Case-Based Reasoning: An Informal Investigation.
Legal Knowledge and Information Systems. Ed. Joost Breuker. Amsterdam: IOS Press. 113–122.

Cameron SHELLEY, 2002: Analogy Counterarguments and the Acceptability of Analogical Hypotheses. British Journal of Philosophy of Science 53 (2002) 4. 477–496.
Jan SIECKMANN, 2010: Balancing, Optimisation, and Alexy´s “Weight Formula”. Legal Reasoning: the Methods of Balancing. Ed. Jan Sieckmann. Stuttgart: Franz Steiner Verlag. 101–118.

Bart VERHEIJ & Jaap HAGE, 1994: Reasoning by Analogy: A Formal Reconstruction. Legal Knowledge Based Systems. The Relations with Legal Theory. Ed. Henry Prakken. Lelystad: Koninklijke Vermande. 65–78.

Lloyd WEINREB, 2005: Legal Reason. The Use of Analogy in Legal Argument. New York: Cambridge University Press.

Top of page

Notes

1 On the three basic operations, Alexy (2010: 9 and ff.), Brożek (2008: 188 and ff.), and Bustamante (2012: 59 and ff.).

2 The first premise seems undeniable: no legal case can be solved without the fulfilment of an antecedent (of a norm or a decision-norm). The second one will be explained later.

3 Brożek 2008: 188 and ff.

4 Naturally, the kind of analogy considered here is that regarding a classification and not the analogy supporting a prediction (Macagno & Walton 2009: 171).

5 On analogy steps, Araszkiewicz (2011: 103).

6 Brewer 1996: 932. Peczenik 1996: 312.

7 On false analogy, among other analogy counterarguments, see Shelley (2002: 489).

8 The symbols = and ≠ are used here for simplification purposes, just to represent similarity and non-similarity.

9 And vice-versa. On quantitative similarity, Davis & Russell (1987: 265).

10 Or factors: all references to a decisive factor include, naturally, a set of decisive factors.

11 Alexy 2010: 17. Aarnio 1987: 104.

12 Nino 2011: 293. Weinreb 2005: 97 and ff. Guastini 1993: 370.

13 This example presupposes, obviously, a gap concerning the entrance of motorcycles into the park, inexistent if a general principle regarding freedom of action is taken into account. However, the example is valid if a norm blocking the permission given by that principle is added: for instance, “entrance into the park depends on a specific rule regarding the kind of vehicle in question”. With this norm, and with a rule for cars and no rule for motorcycles, there is a gap regarding the latter kind of vehicle.

14 Araszkiewicz 2011: 102. Roth 2000: 115.

15 Brożek 2008: 193.

16 Brożek 2008: 194.

17 Which is, as stated, an unproblematic level of similarity (Brożek 2008: 191).

18 Brożek 2008: 191.

19 Brożek 2008: 194.

20 In its original presentation, the principle mentioned here is “people are entitled to rest actively” (Brożek 2008: 194). However, as is legally more accurate, the principle used in the text is that adopted, in the same context, by Alexy (2010: 16). The change does not affect the scheme in any way.

21 On the “collision rule” (law of competing principles), for instance, Alexy 2002: 54, and Pino 2010: 190 and ff.

22 Brożek 2008: 195.

23 Of course, nothing prevents there being more than one principle on each side of the question, requesting an extended “weight formula” (Alexy 2002: 409; Sieckmann 2010: 110). However, the point is that, through the use of balancing, analogy becomes dependent on a dubious choice among principles (disturbing the consistency of the analogy outcome).

24 When one considers that balancing, in its proper sense, just decides which norm among all those applicable is to be applied to a case, it becomes clear that, when used for selecting one of the distinct inapplicable norms for a case, balancing merely defines the proximity and distance between the case and each one of the inapplicable norms.

25 This is even more visible if it is used as the original principle P2: “people are entitled to rest actively” (Brożek 2008: 194). It seems undeniable that in any circumstance the entrance of the memorial tank can be an instance of resting actively.

26 For instance, Pino 2010: 185 and ff., and Duarte 2010: 56 and ff.

27 But surely without analogy, as its normative circumstances are not present. This justifies why some forms of alleged analogy cannot be considered as such (for instance, Verheij & Hage 1994: 65; also in the case of a reductionist approach, Kaptein 2005: 502). Naturally, with regard to the concept of the gap, only normative gaps are relevant here (on the distinction between normative and axiological gaps, Nino 2003: 281, and Rodríguez 2000: 152).

28 For instance, Alexy 2002: 47 and ff., and Brożek 2012: 223.

29 The point here is that normative gaps exist only at the level of rules if rules created the gap or at the level of principles if any is applicable (Ruiz Manero 2005: 123). The following example, in the case of gap2, shows what is meant by this. On the other­ hand, it is important to note that the common distinction between explicit and implicit principles (for instance, Burazin 2014: 171) is of no consequence for this matter. If a principle is unwritten, one of two scenarios has to be real: it is or is not a norm of a set (legal order). If the second alternative is the case, there is no principle at all and nothing follows from it. But, if it is an effective norm of a set, which is naturally the case with a customary source, then it behaves as all principles do and is no different from written principles. Being written or not is irrelevant to defining a legal consequence, provided that it is (accepted as) a norm belonging to a set.

30 All this presupposes that analogy, as an operation adoptable in order to create a decision-norm not enacted by the normative authority, depends on a norm allowing its use under a specific condition and the fact that this condition is the existence of a gap.

Top of page

References

Bibliographical reference

David Duarte, « Analogy and Balancing: The Partial Reducibility Thesis and Its Problems », Revus, 25 | 2015, 141–154.

Electronic reference

David Duarte, « Analogy and Balancing: The Partial Reducibility Thesis and Its Problems », Revus [Online], 25 | 2015, Online since 08 June 2015, connection on 23 October 2017. URL : http://revus.revues.org/3244 ; DOI : 10.4000/revus.3244

Top of page

About the author

David Duarte

Associate Professor of Law at the University of Lisbon, Portugal

   David Duarte
Faculdade de Direito de Lisboa
Alameda da Universidade, Cidade Universitária
1649-014 Lisboa
Portugal

E-mail: davidduarte@fd.ulisboa.pt

Top of page

Copyright

All rights reserved

Top of page
  • Revues.org