Comment by Zhyl

A programming Language for Law is a bad idea. Or, certainly, one must have a very clear idea of what the goals are for such a programming language. For example this (almost throwaway) comment from another post:

>I'm wondering what could result of the combination of smart contracts on a blockchain and this legal language

Is precisely the wrong way to be thinking about this.

The expectation of Law formalised into code is that the benefit lies from the execution: that if you can provide it with the 'inputs' then the neutral computer will do the 'hard work' of working through the consequences and output a decision.

The other intended hope is that the computer can act as a neutral mediator (presumably why what I quoted above also includes the words 'blockchain' and 'smart contracts') which will come to its conclusions justly and without prejudice. Indeed, if you have an agreed upon system via both parties, then it acts like the spinning table at the beginning of the Temple of Doom - allowing two parties with competing interests to come to an accord without trusting each other.

But it is pretty clear to anyone who has worked within a legal or law enforcement context that this is not usually the bottleneck. Lawyers and Justices are usually intimately familiar with the law, and so much of a trial or a case is working through the establishment of the inputs and any of the logical crunching is normally able to be followed through fairly easily by those present.

There is absolutely a case for tools around the law that may superficially seem similar to the above, but in spirit are the complement: search tools for users to easily find related documents, markup to declare and highlight evidence formally (for human consideration, not machine), editors that can assist with writing up legal documents etc.

Something that would seem to be the same thing, for example, would be to create a data model over the existing law to describe the legal architecture as alluded to in the description of the OP: dependencies, exceptions, related articles etc. However, the purpose of doing this in the OP would be to make this executable. The actually useful thing is a lot more banal - to allow things to be surfaced for human inspection and consideration more quickly and easily. It would be to try to spot contradictions and inconsistencies in the law making stage, not in the law interpretation stage. It would be to spot the impact of making a change to one paragraph beyond just the Act that is being scrutinised.

These things are decidedly unsexy to a technical audience, and don't have that same electric feel of 'solving a problem'. Indeed, many of the biggest advances to be made in Law are to be using de rigeur tools in the tech world (such as version control) in a legal context.

A TED talk that summarises some further points:

https://www.youtube.com/watch?v=CEN4XNth61o