Page 1

When most think about the “case closing” function in a criminal matter the sentence and sentencing process are the key focus.  However, if an observer only looks superficially at the requirements, like an iceberg, only a small part of the court’s data, reporting, and other needs are readily apparent.

We have seen a lot of sentencing modules in case management systems built with
database technology.  But only one that is both incredibly detailed and complex has been successful in meeting the long term needs of the court.  Others we have seen captures the core information that is of interest for statistical and key data reporting but these miss all of the detail that can be included in a criminal sentence.

But as in the “Law of the Instrument” that Abraham Maslow said in 1966, “I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.”

Before we explore some other potential solutions, we wish to remind everyone of the excellent work that was done by a national committee for the GJXDM (Global Justice XML Data Model) on Disposition Reporting information exchange in 2006 (please see both the Disposition Reporting section as well as Charge Document).  The report provides a compilation of data elements compiled from across the nation that can serve as a checklist for design.

Sentencing Data

Critically speaking, it has become apparent therefore that a relational database is probably not the best “instrument” for describing a criminal sentence that in its initial form is a document.

First, there is the fact that the sentencing document is the original source of information. Therefore, it should always be available to anyone who needs the information.  But you might say that my jail, corrections, probation, drug treatment, etc. need the data in the sentence document for their own needs?  This again is where we can use XML.
Sentencing forms are often used to support the data needs of the justice community.  If there is a form, there is the ability to assign an XML field to it.  Again the argument is what about the description parts of the form.  Certainly those to can be identified and, today we can apply either programmed XML tagging, search, or even hand-coded XML to those sections.  This is really no different process or work-wise than performing data entry.

Second, what about the complexity of relationships between the different parts of the sentence?  For example, what if some jail time depends on payment of fine and staying 500 feet away from a person for X amount of time?  So the dependencies have to be modeled and mapped.   This is easy for a human brain but difficult for an information system unless one possibly uses an alternative data structure such as a three-dimensional array.  The array can connect each of the four parts and thus if one of the aspects of the sentence is not performed, all parts are affected.   And even this four connection example is woefully inadequate to describe reality when one must connect persons in criminal cases with civil and domestic relations matters and the complexities involved therein.

So while an array approach bears consideration, there are likely other ways of doing this and we look forward to hearing of them from our tech friends in the future.

Last, this is why we are starting to use the term “consequences”.  Is staying 500 feet away from someone a punishment or something else?  As discussed above, it depends.

Risk and Hard Data

Another aspect of sentencing in the USA is the presentence investigation form.  The State of Missouri has been doing some interesting work that could be considered as part of any data sharing approach for criminal sentencing information.

The Smart Sentencing concept is a data driven feedback loop for the consequences ordered by a judge.  This is similar to drug and other problem solving courts in that it connects the sentence to the results.  So instead of relying upon personal experience of the judge or probation officers, which as we have seen is often flawed (the “Moneyball”example), it relies on hard data.

The article goes on to describe how the system provides information to the judge, but does not suggest the actual sentence.  Again, this is a data driven approach that has been proven to be successful for almost two decades (and is also being implemented in Ireland- http://www.irishsentencing.ie/ ).  Please download (PDF) and read the article for a much more complete description of the system and results of implementation.

Monitoring and Communication

And last, this article has not addressed the issues involved in the court’s monitoring of sentence performance.  The financial aspects will be discussed in a later article in this series.  But performance also relies upon effective communication.  This is the work of NIEM and GJXDM to feed the data into the court’s CMS so that it can provide the performance audit that is expected by other parts of government and the citizenry.