A recent paper by Colleen F. Shanahan, Anna E. Carpenter and Alyx Mark makes a very important point that, as the abstract puts it:
Access to justice interventions that provide a little representation, including nonlawyer representation and various forms of limited legal services, may be valuable solutions for low- and middle-income Americans. However, a thoughtful approach to improving access to justice efforts should recognize that a little representation may have risks. In particular, one potential risk of a little representation is that while it provides assistance with a discrete legal need in a specific moment, the nature of the assistance is incompatible with challenging the law. As a result, individual litigants do not have the benefit of legal challenges in their own cases, and our legal system develops devoid of law reform that reflects the needs of low- and middle-income litigants. (Links added by this blog.)
The core idea, based on detailed analysis of hearings in the Washington DC unemployment insurance system, focuses on the actual evidence techniques used in the hearings, and the costs of lack of strategic judgements. The more detailed analysis is in this parallel paper from the same authors.
To me, this suggests a two pronged approach to improving courts that thinks about the system as a whole, rather than individual cases. In the first phase, you make heavy use of experienced counsel who can essentially “clean up” the court’s practices, including where necessary taking up the appeal to change the law or make sure that it is followed in the court. When that process has been successful, then you move in with nontraditional services, in which those providing them are trained in the skills needed to make sure that facts and law are presented to the court. You keep traditional legal representation in a small number of cases, and make sure that referral is both available, and known to be available for those having to rely on those without full legal training.
A sophisticated triage system would actually be designed to adjust recommendations based on an analysis of where the court or forum was in this reform and support system.
Similarly, a sophisticated access to justice strategy would be designed to develop such a multi-state approach, test and modify it in a few forums, and then apply it statewide.
Different stakeholders would play different roles, with advocacy organizations providing representation, and the courts providing parallel staff training and perhaps support systems for the new forms of assistance.
Let’s think strategically here.
P.S. I should note that it was reading these papers that also got me going on the analysis and suggestions in my earlier post here about the relationship between triage and court simplification redesign.
Richard Zorza's Access to Justice Blog 
Richard,
Thank you for focusing on the important complexities of this “legal triage” concept. As we discuss in our article, my co-authors Prof. Carpenter, Prof. Mark and I believe the concept of triage is a helpful and underexplored area of research and reform. Your two posts align closely with the three key questions that guide my co-authors and my ongoing research in this area:
1. What are the problems we are trying to solve, both for an individual litigant and for the system as a whole?
2. What expertise is necessary to solve these problems?
3. What expertise do different forms of assistance (from expert lawyers to self-help materials) lend to these problems?
Your description of a multi-actor, multi-phase, calibrated set of interventions is exactly the type of system we consider in our article where different actors use appropriate expertise to work together to bring “case-focused” and “system-focused” challenges. Though our published work has focused on lawyers and nonlawyer/lay representatives, a current project considers law clinic students as actors in such a system. We are also working on a project considering the role of judges in such a system. Of course, others are engaged in understanding different aspects of these crucial problems and we believe (and hope) our numbers are growing.
Prof. Greiner is quite right that we know very little about legal triage. We don’t know “what a case needs” or “what works” in legal triage. My co-authors and I think the situation is even direr than that. We have woefully undeveloped theories of triage in legal services. We don’t have a comprehensive understanding of how expertise is being allocated to problems on the ground right now. Specifically, we don’t even have a reasonable map of (1) the range and nature of services we provide now, (2) how providers currently choose to connect those services to specific problems (whether those problems are case-focused or system-focused), and (3) the consequences that individuals and the system experience now. The randomized controlled trials of the type Prof. Greiner and his colleagues are undertaking are imperative for considering the causal interaction of these factors. But first we need to understand what all these factors are and to ground our research in theories tied to our aspirations for our justice system, which necessarily includes understanding what litigants are experiencing today. As academics, we think of these as research questions. As lawyers, advocates, and citizens, we think of these as a societal imperative.
Hi, Richard, I think this is the most important post you’ve ever written. Level-of-service or type-of-service triage is perhaps the most consequential decision that legal aid providers make. And remarkably little thinking goes into it. Most of the thinking is toward the idea that triage involves getting each case “what it needs,” with little thought about how one defines “need,” how one operationalizes that definition, and what one does when one discovers (surprise!) that resources are insufficient to provide the level or type of service that cases “need.”
Implicit in your discussion above are some extremely deep concepts, concepts that are both difficult and necessary to swallow:
1) We might triage a case/client not based on what is best for that case/client but rather based on what is best for the overall goal of providing A2J for a broad service population.
2) That means we might use a case/client as a means to a broader end.
3) We have absolutely no clue, none at all, of how to take any of the steps you outline because we lack ALL of the critical information one would need to take any of these steps.
Note: As a start to attempting to gather the information, the A2J Lab is pursuing a triage study, see http://a2jlab.org/current-projects/signature-studies/violence/.