DNA and Laws are ultimately sequence of strings. Both of them work by the process through modification, interpretation, and substantiation.
The two types of text have also feedback loop that a product made by a sequence will maintain or change the sequence itself.
For DNA, you can already access base arrangements of many species on databases, for example NCBI. In the datasets, base sequences are tagged with expressed gene region, functional RNA region, regulatory sequence, and other categories.
What’s more, BLAST can retrieve similar genes based on a sequence query, which allows us to infer evolutional tree and a type of gene that the DNA code.
Returning to law, How would it be when sentences of laws are tagged and collected like DNA sequences?
To compare law and DNA, let’s think about the correspondences of both.
- Genome = Constitution; since they govern entire genes or laws.
But there is a few differences. Genome holds entire sequence, while a constitution has a tiny abstracted sets of rule.
- Genes = Laws; since both of them care for a single topic or a function roughly.
A gene will later be translated to protein to take a task as an enzyme or structural building block.
A law will be executed by administrative bodies to accomplish a task.
- Regulatory genes = Regulatory laws; among other genes or laws, this type of element can control other genes or laws.
For genes, it is probably an enhancer, promoter or a transcription factor.
For laws, it would be a regulation over political donation or election, which controls how laws are changed by political participants.
- Protein = Administration; since they are in charge of execution. Misinterpreting blueprint can be fatal for both. Wrong folding protein and wrong organization don’t work as desired.
- Metabolism = governmental organization; it illustrates how protein/government interact until actual output comes inside the cell/institution.
So this is the main topic of the post. How Can I tag on sentences of law?
It would be like the interactive sequence view on NCBI which puts tags in parallel to DNA sequence.
Then what kind of tags can be applied on law sentences?
- Ministries which is responsible for the law. It seems to be easy to get the data.
- Status of political parties when the law was established.
- History of the law. It is equivalent to history of a gene.
- function of the law. It is equivalent to function of protein.
- A Law being the base to other laws. It is equivalent to a gene modification factor.
Next, how about sequence alignment for laws?
For law, similarity search may not work with actual sentences because natural language is more complicated than gene’s ATGC.
Feature extraction by natural language processing could work for it, but alternatively, tag similarity search might fit for the purpose.
Benefits from the attempt:
- Knowledges that which part of constitution/law backs up other laws.
This shows law version of intron/exon ratio. It also shows disparity of dependency tree of the law.
- History of law. It is equivalent to editing history of wikipedia pages.
- laws that govern changing/establishing laws. It shows which laws can affect making new law or updating law. For instance, regulations of political donation and political activity.
Data and resource that can help the idea:
- Popular data format for law.
- Diversity of law system, such as China and Muslim nation.
- Database website for law.
- Models of how a law is generated by political power game.
- D3’s library to tag a text.
- Data to show relationship between laws.