Over at Concurring Opinions, Dave Hoffman has a very nice post entitled Must District Judges Give Reasons?. Here's a taste:
So here's the issue: in the ordinary case, to what extent are judges required to explain themselves?
I ask as a facet of the work I'm doing on when district judges write opinions (versus orders). The large-scale empirical project I've undertaken to answer that question is still ongoing, but my preliminary findings highlight the relative scarcity of opinions as a proportion of judicial workproduct (3% of judicial actions taken in my study, around 10% of non-ministerial/scheduling actions). Previous work on this subject has suggested that judges provide long-form explanations, complete with citations to "cases . . . statutes[s] . . . treatise[s] . . . article[s] . .. note[s] . . . [and] blawg[s]" when the issues are hard. When the issues are "easy" they write orders. When the issues are really easy, and time is crunched, they simply rule. "Just because I said it." But I think that this story is wrong.
My hypothesis, which I hope the data will prove out, is that judges explain themselves to maximize certain ends: (1) avoiding reversal; (2) gaining reputation with certain members of the bar; (3) intellectual satisfaction (i.e., more opinions in con law cases than social security cases); (4) free time (i.e., less opinions as workload increases); (5) reputation as a function of race and gender;etc. That is, the universe of opinions (what we call doctrine) is likely to be significantly different from the universe of what judges do in contested matters, in ways that we can predict. This isn't necessarily bad, although I am looking to see if certain classes of litigant - those represented by AMLaw 100 firms; corporate parties - are more likely to get gold-standard justice instead of "because I said it" justice.
A very fine post--well worth reading. This is, of course, a complex question. It's quite obvious that trial judges do not have an obligation to write a written opinion justifying each and every action they take: that obligation would mean that judge's could not engage in the management function during "live events" like trials and hearings. Every hour of trial would require many hours of opinion writing! On the other hand, we do expect trial judges to offer some sort of justification when they make a major decision--the clearest case being a decision that disposes of a claim or defense: one rule of thumb might be that any decision that creates an appeal as of right is a decision for which an explanation should be provided.
Why is there such an obligation? The place to look for an answer to this question is in a theory of procedural justice. Such theories generally divide into three families: (1) accuracy theories, (2) balancing theories, and (3) participation theories. Each of these approaches offers insights relevant to the question at hand. An obligation to offer justification has obvious accuracy-enhancing effects: it forces the decision maker to engage in an internal process of deliberation about explicit reasons for an action and to consider whether the reasons to be offered are "reasonable" and whether they are likely to be sustained in the event of appeal. Balancing approaches, which consider the costs of procedural rules as well as their accuracy benefits, point us in the direction of the costs associated with requiring justifications on too many occasions and of the costs of requiring justificatory effort that is disproportionate to the benefitsto be obtained. Requiring reasons facilitates a right of meaningful participation as well: when a judge gives reasons, then the parties affected by the action can respond--offering counter reasons, objecting to their legal basis, and so forth. Moreover, the offering of reasons provides "legitimacy" for the decision.
For more, see the following:
Procedural Justice (on SSRN)