Automated Legal Processing

Question

Will it ever be possible to make legal systems quantifiable enough to process with computer algorithms? What technologies would have to be in place before this is possible? Are there any existing technologies that are already trying to accomplish this?

Out of curiosity, I downloaded the text for laws in my local municipality, and tried applying some simple NLP tricks to extract rules from sentences. I had mixed results. Some sentences were very explicit (e.g. "Cars may not be left in the park overnight"), but other sentences seemed hopelessly vague (e.g. "The council's purpose is to ensure the well-being of the community").

I apologize if this is too open-ended a topic, but I've often wondered what society would look like if legal systems were based on less ambiguous language. Lawyers, and the legal process in general, are so expensive because they have to manually process a complex set of rules codified in ambiguous legal texts. If this system could be represented in software, this huge expense could potentially be eliminated, making the legal system more accessible for everyone.

Answer 1

Doubtful. The facts and circumstances vary too much between cases. The ambiguity in contracts is often the reason why the courts have to step in later to sort them out.

However, there are many tools in processes that courts have adopted to make litigation more efficient and less expensive, such as electronic discovery. There are also many software packages on the market that help attorneys organize and present cases.

And of course, case research online via Lexis or even FindLaw has proven to be a huge time saver.

Answer 2

I'm writing a MSc thesis right now touching on the subject you're interested in. It's a controversial subject. The author(s) of this paper seems appalled by the mere idea of using formal logic and automatic reasoning and suggests the rule-based approach is hopelessly flawed. Others believe it possible (searching the web for "formal legal reasoning" is an imperfect indicator of the level of interest in the subject) and are investigating different approaches.

I would say that automatic legal reasoning is a much less complex (saying "simpler" would definitely give the wrong idea) problem in European law because it's based on regulations (constitution, laws, international treaties etc.) as opposed to all that plus the entire legal history available to look for precedents, as is characteristic of Anglo-Saxon legal systems. In the latter case, we have to come up with a good way of "understanding" previous cases to be able to identify similarities so a case can be used as a precedent.

Another aspect of the problem is what exactly is the goal. It's much, much harder, for example, to create a system which has to guarantee that the conclusions it draws about a case are of the same quality as those of a good lawyer than it is to create a system which might provide advice or guidance to a lawyer or judge by identifying parts of the law the reasoning engine found to be important for a specific case. This simpler case, basically a decision support system, is what I'm investigating right now.

In my view, the most important aspect of this area of research is that it raises significant philosophical questions about the organization of our societies. For a start, it's obvious you could easily reason automatically about simple/trivial systems, e.g. bike trip rules:

• if we're lost, attempt to match the surroundings to something on the map
• if the trip is long, bring enough food and water
• if we're going into the mountains, we have to carry warmer clothes

...and so on. Now suppose for a minute that encoding laws into formal logic and reasoning automatically is not possible. This can be either because of the way our laws were written (vague programming in a natural language) or because the nature of the emergent life of our society makes it impossible for us to formally describe it using rules. If it's just a matter of existing laws being poorly written, that's solvable. If, on the other hand, society's inner workings can't be formalized in a way which allows automatic reasoning, it sheds completely new light on the legal system, the foundation of any modern country: it becomes clear that the law is fundamentally subjective and any idea of "right" or "wrong" is subject to interpretation, impression etc., reducing the quality of the primary deliverables of law: an absolute sense of justice (the one sold to voters during every political campaign) and dependability and stability of law, which gives people the security to take certain steps (go into a business, invest...whatever) because they expect their actions to be seen in a certain light regardless of the judge or lawyer.

It's a complex, challenging and supremely interesting area of research, one that this handful of sentences only touches on.

Answer 3

This isn't really a suitable topic for SO, but because I am in a good mood:

"Cars may not be left in the park overnight"

What:

• is a "car"? Can I leave a bus?
• is meant by "left"? Abandoned, casually discarded?
• is a "park"? A green leafy place? Or somewhere you "park" things?
• is "overnight"? Specific hours? dusk to dawn?

This is why we have always had lawyers, and will have them for a long time yet.

Answer 4

You are treating the ambiguity as if it's a flaw. On the contrary; often the ambiguity is deliberately there so that the law can be interpreted based on the will of the people and past precedent, not to mention to account for unforeseen circumstances. Laws are necessarily forward-looking and often have to be worded in a way that allows them to be interpreted based on the original spirit of the legislation.

With this in mind, my answer to your original question is 'no', or at least 'not before we understand psychology far, far, far better than we already do'.

Answer 5

Applying any kind of automated reasoning system to legal cases is a difficult computational problem that many have looked at. There may be a subset of cases or legal categories where this is feasible, but a general-purpose legal reasoning system is problematic since legal reasoning involves multiple methods, not all of which can be expressed algorithmically or as formal logic, e.g., human judges use inconsistent and sometimes conflicting criteria to decide cases.

Some useful resources to look at which help illustrate the challenges involved and what research has been done include the following:

• Argumentation theory
• Argumentation methods for artificial intelligence in law (book)
• Predicting Outcomes of Case-based Legal Arguments (paper)