Hunter, D --- "Teaching and Using Analogy in Law" [2008] LegEdDig 26; (2008) 16(2) Legal Education Digest 42

• Teaching and using analogy in law

---

Teaching and using analogy in law

D Hunter

J Assoc of Legal Writ Dir 2, 2008, pp 151–168

If law professors try to teach analogy construction at all, they usually choose one of two approaches. They might teach the ‘LSAT Model’ of analogies: ‘Arm is to Leg as Hand is to What?’ The student should answer ‘Foot,’ but even so, what has the student learned? At best, this model suggests that an analogy is some kind of relationship, but it doesn’t give much guidance beyond that. Alternatively, the professor might adopt the theories of a number of legal philosophers and teach analogies as a type of rule. In this model, the student takes a case, renders the case down into a holding, and then applies the holding to the undecided case. This model has the benefit of teaching the student that analogies actually matter in law, but it suggests that the only trick of legal analogies is to find the appropriate rule that explains a case.

One of the main problems with both these models is that, while they contain a grain of truth, they are so general as to be of no practical use in explaining to students or lawyers how to use analogies effectively. It must be the case that some analogies are better than others, but neither of these models explains why some are better than others, or how a student or a lawyer can learn to distinguish good analogies from mediocre ones.

The purpose of this article is to provide a simple model that can be used to teach and to learn how analogy actually works, and what makes one analogy superior to a competing analogy. The model is drawn from a number of theories of analogy-making in cognitive science.

The lessons from cognitive science theories of analogy can be applied to legal analogies to give students and lawyers a better understanding of this fundamental process in legal reasoning.

An analogy is a non-identical or non-literal similarity comparison between two things, with a resulting predictive or explanatory effect. This means that two items are compared, and the outcome, result or determination of one of these items is predicted or explained to be the same as the other. Typically, we compare items where we know the outcome of one, and we suggest that the outcome of the other will be the same.

The obvious example of legal analogy is case law reasoning: prior analogs, called precedents, are used to predict, explain or justify the outcome of the currently undecided case. An example will be useful here to explain the mechanism. Let us say a judge is faced with a case in which a man is suing for the loss of his luggage while he was traveling onboard an overnight ferry. The luggage was stolen from an overhead rack in the plaintiff’s compartment. The plaintiff is suing the ferry company on the theory that it was the bailee of his bags and therefore liable for their loss. Let us assume that there is no statutory pronouncement on the subject, and there are only two precedents that might be relevant to the decision. The first precedent involved a hotel proprietor who was found liable for a guest’s stolen luggage because part of the contract of hospitality involved reasonably safe storage of the guest’s belongings. The second precedent involved a railroad company, which was found not liable for the loss of the luggage of a passenger who traveled in a sleeper berth because the contract was primarily for travel and not for lodging.

Plaintiff’s counsel will suggest that the ferry is a ‘floating hotel’: it has cabins like hotel rooms, it has restaurants and other hotel-like facilities, and passengers almost inevitably spend the night onboard. Therefore, the judge should follow the hotel precedent and find for Plaintiff. Defendant’s counsel will suggest that the ferry is a ‘seagoing train’: the intention of the passengers is primarily to travel and not to stay overnight, trains have restaurants and bars, and so forth. Hence, the decision should be against Plaintiff.

The question for the judge is straightforward: Will she opt for the ‘floating hotel’ analogy or the ‘seagoing train’ analogy? In deciding this case, the judge may draw an analogy to a hotelkeeper’s duty to protect the bags of a guest, or alternatively, point to the railroad’s lack of duty to do the same for a passenger. In each analogy, the judge chooses some non-identical features of the precedent as being sufficiently similar to the current case to warrant the same outcome — perhaps the fact that both the train and ferry involve transportation or alternatively that both the hotel and ferry have sleeping quarters. Having drawn the comparison between the precedent and the undecided ferry case, the judge uses the precedent’s outcome to decide the new case. Whether the judge chooses the train or hotel precedent as analogous, the analogy has a strongly constraining, predictive effect on the outcome of the undecided ferry case.

Inductive inference or induction is related to analogy. Induction is, generally, the process of taking a number of specific cases or instances, classifying them into categories according to relevant attributes and outcomes, and generalising an inclusory rule from them. That is, we take a number of isolated experiences and attempt to explain them by a general rule that covers the instances examined. Within law, we can find numerous examples of this: the ability of attorneys to say from experience what particular judges’ decisions are likely to be, their ability to fit multiple precedents into a coherent framework of rules and principles, and so on.

Many lawyers fail to recognise that induction requires the generalisation of a rule from prior experience, whereas analogy is a one-to-one similarity comparison that requires no generalisation to operate effectively. The best example of this is the ejusdem generis canon of statutory interpretation. Ejusdem generis means ‘of the same kind, genus or nature’ and the canon is invoked when there is a statutory definition of the form ‘x, y, z or other.’ The canon is applied to determine whether something that is not defined should be included as ‘other.’ That is, it is used to define the scope of general words that immediately follow specific words. So, for example, in a statutory definition that indicated ‘planes, trains, automobiles, or other,’ a question might arise whether a skateboard, a solar-powered vehicle, or a snowmobile should be included within the definition of ‘other.’

The ejusdem generis rule clearly relies on inductive inference. To determine what the expression ‘other’ encompasses, courts must examine the words that make up the antecedent part of the expression in question, and then decide what the scope of those words incorporates. Courts specifically require the identification of the genus as the first step in the interpretive process. It will therefore always be necessary to create an inductive generalisation about the set of concepts that the legislature intended to be included.

Analogy is different from induction. It does not rely on any generalisation of prior experience. However, it is related to induction in that both rely on similarity comparisons of prior experience.

Analogical inference also shares a number of features with metaphorical inference. Metaphors have a less constraining effect on reasoning than do analogies, but they operate in a similar fashion. Metaphor is an expression forming a non-literal similarity comparison between two things, which has an expressive or affective content and thereby carries meaning. Unlike analogies, metaphors do not have a predictive content and do not strongly constrain the outcome of the reasoning process. Thus, in Shakespeare’s Romeo and Juliet, Romeo’s metaphor ‘Juliet is the sun’ contains an explicit similarity comparison between ‘Juliet’ and the ‘sun’: in comparing Juliet and the sun, the intent is to convey the expressive meaning that, to Romeo, she is the light of his world, she provides life and energy. Metaphors within the legal sphere operate in a similar way. To use the metaphor of the ‘corporate veil’ — to say for example that ‘Company directors are protected by the corporate veil’ — conveys the expressive meanings that the corporation is a separate entity from the directors, the directors are shielded behind some kind of protection, the directors are somehow ‘unseen’ while behind the veil, and so on. However, the metaphor does not strongly direct or predict the outcome of a case in the way that a legal analogy does.

The definitions given above are derived from cognitive science approaches to metaphor and analogy, and they are useful in both developing the connection between the two concepts and explaining their differences. Both analogy and metaphor involve a similarity relation between two objects, and the similarity relation transfers meaning from one object (the source) to another (the target). The major difference between the two is that an analogy has an explicit explanatory or predictive component which metaphors lack.

Though there are various competing models that differ in their minutiae cognitive scientists generally agree on the fundamentals. The main features are the concept of mapping, the role of source and target domains, and the operation of the parts of the process.

Both analogical and metaphorical reasoning involve a mapping of concepts from one set of ideas (the source domain) to another set of ideas (the target domain). The approach can be demonstrated by seeing how it is used to explain metaphor. Target and source are sometimes described as the ‘two halves of metaphor.’ In any given metaphor, the target is the underlying idea or principal subject. The source, which carries the metaphor, is the domain from which the salient features are drawn and then attributed to the target. So, in the simple metaphor ‘lawyers are pigs,’ ‘lawyers’ is the target and ‘pigs’ is the source. The source imports a host of features associated with ‘pigs’ without seeking to draw an identity-relation between lawyers and pigs. Analogies work in the same way: features from the source are mapped onto the target, importing a series of elements that are not present in the target.

One of the first studies of analogical mapping between domains was Gick and Holyoak’s reinterpretation of Duncker’s early radiation problem. In Duncker’s experiment, subjects were asked to solve the problem of curing a cancerous tumor by radiation therapy. The constraints on any solution were: ‘High energy radiation would damage the patient’s tissues. Low energy rays would not kill the tumor.’

The solution lay in directing a number of low energy rays from different positions on the body, but which intersected at the tumor site and created a point of high energy. In this way the two constraints were satisfied. Few subjects solved the problem when presented in this way.

Gick and Holyoak adapted this problem to investigate analogical reasoning. Their new experiment included a number of stories that the subjects read prior to tackling the radiation problem. One story involved an army of men which sought to overthrow a tyrannical ruler who was hiding in a fortress. Unfortunately for the rebels, the roads leading to the fortress were all mined, such that if a large number of men walked over any one road a mine would explode. The solution was to break the men into small groups and send them down different roads to converge on the fortress at the same time. The analogy between the fortress and the tumor was now obvious, and the number of subjects solving the problem rose dramatically, especially when the subjects were prompted that the solution to the radiation problem might lie in one of the previous stories. This demonstrated the power of analogical mapping from the source (tyrant story) to the target (cancer story), where the structural elements were the same and where the outcome of the source story was the same as the desired one for the target.

Though research has shown that analogical reasoning involves some kind of mapping between domains, this is only the start of any model of analogy. First, the question arises as to what is mapped from source to target. The mapping seems to involve some structural elements rather than merely surface features. Thus, in the radiation problem, the important concepts seem to be the structural concepts of attack (by ray or soldiers), the concept of splitting the attack (ray or soldiers), and the concept of the attacked item being malignant (cancer or tyrant). The surface features can be (and indeed must be) ignored: one story involved the body of a patient and the other involved a fictitious country, one story involved soldiers and the other radiation rays; and so forth.

Notwithstanding that structure seems to be the key to mapping here, it cannot be the entire solution to the question of how analogy operates. Research into the retrieval of analogs indicates that humans find retrieval of structural analogs very difficult, and find analogs based on surface features much easier to recall. In studies on the radiation problem, subjects more often recalled an analog about a doctor using rays for some non-therapeutic reason than they recalled the tyrant analog, even though the doctor story was not helpful in solving the problem. Structural isomorphism, or the need to see structural similarities between the source and the target is an important constraint upon analogical inference, but other constraints exist, as I shall shortly explain.

The second question is why certain features are mapped. In another famous example, students learned more readily about the mechanics of the atom by using the analogy of the solar system. In this analogy, the student mapped the known concept of the sun being at the heart of the solar system, onto the unknown world of the atom. Thus, the atom’s nucleus became the ‘sun,’ and the electrons became the circling ‘planets.’ The known concept of attraction between sun and planets was used as an analog for the attraction between nucleus and electrons.

Keith Holyoak and Paul Thagard have presented a ‘multi-constraint’ model of analogy. They suggest that various constraints influence the generation and perception of analogies.

The surface-level constraint suggests that an analogy is guided by the direct similarity in the surface-level elements in the source and target domains. Thus, in the example of the atom-solar system analogy, there are directly similar features present in both domains. In each domain, something lies in the center of the system, and other things revolve around this central body.

How does this apply to law, and in particular, to the ferry example? Surface-level similarities are obvious between the ferry case and the train: both involve travelers moving between two places. Alternatively, the similarities between the ferry and the hotel cases are obvious: both cases involve locked rooms (whether hotel room or ferry stateroom) from which the bags were taken, both have restaurants, bars; and so on.

Holyoak and Thagard note that similarity is not a fixed concept, especially surface-level similarity. The perception of similarity differs depending on the background information which is provided, the ‘context effect.’

Following on from this, I suggest that the process of advocacy before a judge is, in part, an attempt to introduce this context effect. Recall the ferry example. Counsel for Plaintiff, seeking to advance the ‘floating hotel’ theory, with its attendant liability outcome, will presumably highlight the factual similarities between the ferry and a hotel and may introduce other examples that make the ‘seagoing train’ theory less palatable. Thus, a good lawyer might introduce another example where a metropolitan trolley-car company was held not liable for theft committed on its trolleys. Thus, the context effect experiment becomes: ‘Out of (hotel, train, trolley car) which is a ferry more like?’ The answer is, presumably, the hotel. Defendant’s counsel might counter by introducing another example, where the owners of a permanently berthed cruise-liner were found liable for water damage to a guest’s luggage. The context effect experiment then becomes: ‘Out of (hotel, berthed liner, train) which is a ferry more like?’ The best trial lawyers are able to influence judicial assessments of similarity by a skillful manipulation of the context effect.

Holyoak and Thagard’s second constraint is structure. This constraint involves the pressure to identify consistent structural parallels between the two domains.

Structural mapping involves a mapping of relations, not propositions.

The adequacy of the analogy is dependent on our ability to fit the target within the relational structure of the source. However, the constraint is not absolute, for it is possible to project a number of mappings onto the target: either of the ‘floating hotel’ or ‘seagoing train’ analogies is structurally as coherent as the other. This seems to accord with the feeling that in the ferry case both analogies are good ones, and the outcome could go either way.

This gives rise to the question of why we focus on the structural aspects we do, and not on an infinity of other structural features.

One answer to this problem is that the other constraints — surface and purpose level — operate in conjunction with the structure constraint. But within the structure constraint itself, there is also another mechanism that affects the choice of analogs. Dedre Gentner’s model of the structural constraint suggests that relational mapping at the structural level relies on the principle of ‘systematicity.’ This is the idea that people prefer to map systems of predicates that contain higher-order relations with inferential import rather than to map isolated predicates. That is, we pick clustered groups of relations which are able to explain why the system works as it does, rather than pick isolated predicates which may be similar but which are singletons and do not help to explain the system.

This principle of systematicity is important in legal analogies. This principle encodes the human predilection for choosing structural mappings which are higher-order and which are explanatory of the entire system. We should therefore expect to see certain analogies preferred if they operate at a higher, causative level than the alternatives.

Holyoak and Thagard’s third constraint on the analogy is the purpose for considering the analogy at all.

Within law, the purpose constraint clearly applies. Take the ferry case as an example. The purpose of the attorney for Plaintiff is to make the most plausible case for Plaintiff. Hence an analogy that operates against Plaintiff’s interest — such as the train analogy — is clearly less desirable and less compelling than the alternative hotel analogy. The opposite is true for Defendant’s counsel. Thus, the purpose for which lawyers wish to use an analogy will influence their perception of it; and as a consequence the purpose constraint will determine the analogy is a particularly good one or bad one for any given party.

The purpose constraint also applies to judicial adjudication. Since the introduction of American Legal Realism and Critical Legal Studies it has become a commonplace to suggest that laws do not completely determine legal outcomes and that judges have considerable leeway in their decision making.

Given the cognitive science account of analogy, and the evidence presented above that the multiple constraint model applies in law, what lessons might we apply to the teaching of analogy in law schools and the use of analogy by practicing lawyers?

First, I suggest that this model provides a more detailed picture of what is happening when lawyers construct analogies and of what judges are doing when they adopt a particular precedent. Students and lawyers can better critique cases, as well as individual analogies, if they understand that there are multiple levels at which an analogy can be constructed and analysed.

Second, the model provides a framework to generate interesting and more compelling analogies by students and lawyers. Beginning law students and less experienced lawyers tend to construct analogies that are based on surface level similarities. As they become more practiced, both start to recognise that the most powerful analogies operate across multiple levels and they learn to rely on Gentner’s systematicity principle, most likely without ever realising that such a concept exists and has been documented elsewhere. Recognising as early as possible in their careers that they need to construct these sorts of layered analogies can only speed their development towards becoming expert users of analogical reasoning.

Finally, numerous findings of cognitive science can be used to generate more persuasive analogies (and indeed more persuasive arguments in general). One example of this, discussed above, is Tversky’s context effect. By careful manipulation of hypothetical situations, a lawyer can make a good analogy appear even more compelling.