
Understanding Rules with LogicMaps
Consider the following rule:
Psychiatric (mental health) service to an outpatient is a covered service only if it constitutes an active preventative, diagnostic, therapeutic, or rehabilitative service with respect to emotional or mental disorders, and only if (a) the service is furnished by a group practice organization, by a hospital, or by a community mental health center or other mental health clinic which furnishes comprehensive mental health services, or if (b) the service is furnished to a patient of a day care service with which the Board has an agreement under section 49(a)(3), or to the extent of twenty consultations during a benefit period (as defined in regulations), if the service is furnished otherwise than in accordance with clause (a) or (b).
What does this rule say? What does this rule mean? Suppose you are advising a client who has already scheduled an appointment with a prospective provider. Would the service be covered? You would like to know the specific consequences of this rule as it applies to your client’s situation. If you paste the text of the rule into LogicMaps and hit enter, you would get this:
The decision procedure for this kind of logic can only give you a definitive true or false answer if the rule states necessary and sufficient conditions for the conclusion. In most cases, a rule will enumerate only a set of sufficient conditions for its conclusion — the conclusion is true if the conditions are met. But there may be other sets of conditions, in other parts of the document, that are also sufficient for the conclusion, so you can’t determine (from the rule alone) whether the conclusion is false. More rarely, the rule will assert only necessary conditions (as this one does). These are conditions that must be met for the conclusion to be true, but they may not be sufficient for its truth by themselves. So they can only be used to definitively establish that the conclusion is false (by not being met). When you click on the light blue roundtangle to see the the LogicMap for this rule, you get this:
Necessary Conditions for:
Psychiatric service to an outpatient is a covered service
The rectangular nodes represent all of the basic propositions that can be either true or false, and the connections between them enumerate all of the logical combinations that lead to a positive or negative outcome. To determine how this rule applies to your client, you simply walk the diagram from the top, taking the true or false path out of each node as it applies to their case. Again, if this rule had been necessary and sufficient for covered services, the left (positive) side would end in a green (true) outcome. But these are only necessary conditions, so we can only definitively determine when your client is not (or will not be) covered. This is an important distinction, often overlooked. Humans naturally read rules as if they were necessary and sufficient (even legal professionals), when the conclusion is stated first, and the conditions follow. The key difference here is the two occurrences of ‘only if’ in the text, meaning that the conclusion implies the conditions rather than the other way around.
Your client clearly wants to end up on the left side of the diagram, not the right. If you click on the left terminal node, the LogicMap is re-rendered to show all of the paths leading to the positive side in green. Every node that must be true in every positive scenario will also turn green, every node that must be false in every positive scenario will turn red, and the remaining nodes, which could be either true or false in some positive scenario, will turn grey.
Necessary Conditions for:
Psychiatric service to an outpatient is a covered service.
Ending on the left side, in this case, is clearly better than ending on the right, but you must advise your client that they are not out of the woods yet. You will have to map other rules from the regulations, looking for some sufficient ones.
Conversely, if you were interested primarily in the false scenarios, you could click on the final false node in the diagram and have the false paths highlighted in red. The same node coloring convention applies but this time relative to invariant truth values in all negative scenarios, so the node colors will likely change.
Necessary Conditions for:
Psychiatric service to an outpatient is a covered service.
Why would you be interested in the false paths? Well, the rule in question may be defining prohibited behavior, so you would like to verify that none of it applies to your client. In that case, all of the false paths would represent your safe harbor options according to the rules. And it doesn’t matter if the rule states only necessary conditions because you only need to not satisfy one of them.
Thinking Abstractly
Why is the LogicMap easier to understand than the rule itself? Because as humans we find it challenging to think abstractly, in generalizations. We are much better at understanding concrete examples. Some humans are adept at thinking abstractly -- scientists, logicians, lawyers. Someone has to write the rules that apply to all of the imagined examples. But to achieve this generality, and cover all of the cases while still remaining precise, the rules often become obscure, particularly as they become larger. Understanding all of the implied consequences then becomes challenging even for the abstract thinkers. So scientists use calculators, logicians use theorem provers — and now legal professionals can use LogicMaps.
Trying to understanding the consequences of legal rules is somewhat like trying to understanding the consequences of computer programs. It's actually worse for computer programs, because unlike laws, they typically have infinite consequences. Programmers must think abstractly when they write them, but their inability to predict all of the possible consequences of the code they write means software almost always contains bugs, no matter how hard we try to avoid them. It has become such an accepted fact of life that software, unlike most consumer products, comes with a disavowal of implied warranty and fitness for a purpose. It’s just too hard to predict all of its possible consequences.
A Meeting of Minds
Fortunately for legal professionals, legal rules have finite consequences — typically, comfortably small sets of consequences. And because these consequences can be exhaustively enumerated as LogicMaps, the maps can achieve a crucial common understanding of the meanings of rules. When legislators or regulators are first proposing rules, there is a need for a common understanding with legal drafters as to intended vs actual consequences. When the enacted laws are later refined by regulatory agencies, the refiners need to compare unrefined consequences to the prospective refined ones. When either rules or regulations are later amended, there is a need to understand the net change in consequences. Without the LogicMaps, a fallible estimate of the consequences is held privately in each party’s head, so it’s not surprising that these estimates often don’t entirely match.
Once produced, there are several venues where LogicMaps can aid in the consumption of rules, even if the rules are not automated. When the rules are adjudicated — by a regulator deciding whether a violation has occurred, by an administrator deciding whether an applicant qualifies for a benefit — the adjudicator can mentally walk down the LogicMaps. When advising clients — clients whose factual scenarios are already fixed, or clients contemplating actions not yet committed — the legal advisor can mentally walk down the LogicMaps. There is even a direct to the consumer application where ordinary citizens can view the LogicMap versions of rules online to make a preliminary self-assessment.
In all of these applications, the LogicMaps express the meanings of rules in a form that any competent speaker of the language can readily understand. No one needs to learn any new skills. Perhaps of more importance is that LogicMaps enable stakeholders with different skill sets to achieve a meeting of the minds by comparing the consequences of their different rule conceptions in a common, verifiable format.