Thursday, August 31, 2006

What If a Commission Won't Work in Your State? Part II

In the first part of this post, I addressed concerns of folks who wondered what they could do to better rationalize their state sentencing policy if the commission/guideline concept clearly was just not going to work. (Let me make clear here—I am not saying states shouldn’t do commissions/guidelines, just what they might do if that’s not going to work for them.) In far more words than necessary, I spelled out the importance of sound and timely sentencing data (terms and outcomes) and emphasized development of what academics are calling “sentencing information systems” to inform and enforce transparency in and accountability to selected policy goals. After explaining some ways this could go, I pointed out the continued need for a mechanism to draw policymakers into the process beyond mere hope for champions or the actual White Knight who occasionally does come along.

In this post I’ll put out an idea for a possible mechanism and invite you to take shots at it, on the condition that you improve it or forward something better. What I propose would be a Sentencing Information Authority (SIA). The SIA would oversee the state’s sentencing policy and its sentencing data collection, analysis, and reporting, each of which would be devolved into two separate bodies to make recommendations to the SIA. The policy stuff would receive advisement from a traditional sentencing commission, but one now devoid of policymaking responsibility. Rec’s on the data and research stuff would come from a data and information committee composed of data analysts/IT specialists from state agencies like corrections, state courts, office of administration, legislative staff, and academics trained in corrections and public policy. It could coordinate among agencies and ensure up-to-date hardware and software usage. The split would allow the data function to survive and be preserved should the advisory sentencing commission still prove too nonproductive and/or dysfunctional to be allowed to live.

Who would make up the SIA? There could be two versions, depending on state politics and cooperation. The best would be a nine-person body, with 4 legislators (the usual “2 from each party, 2 from each house”), 4 executive appointees (from possibilities of DAs, defense bar, corrections, academe, victims’ groups, etc.), and the state court administrator to represent the judiciary (I’d say the state chief justice, but this is how it would end up anyway). The SIA would have a small staff to coordinate its two subgroups and to handle public outreach and liaison. The staff could also perform the analysis on the data from the data group and/or contract with outside researchers.

If, however, the politics are such that the legislature and governor cannot abide each other, dooming a body like this, the governor is not option-less. S/he could create a smaller SIA within the executive branch, with membership from the administration, maybe including the chairs of the commission and data subgroup. The state court director could still be invited to participate. This SIA would lack the bipartisan, bi-branch clout and interest of the first, but, depending on the governor and (as mentioned in the first post) the direct ties to his/her budget rec’s, it would still have to be listened to a lot more than most commissions are.

My experience with all three commissions I’ve worked for (still tied for the record!!!) is that legislators picked for commissions become easily frustrated with the meanderings and shenanigans of the other people on the commission. This leads to indifference and, worse, nonparticipation from the very people most needed to give a commission political teeth. Executive branch folks may be a bit more tolerant and stay a few meetings longer, but getting the DOC chief to these meetings is tough, particularly after a few said shenanigans. Thus, policymakers flee, and, by definition, all that’s left are non-policymakers staring at each other. For a supposed policymaking body. (Legal scholar Rachel Barkow has some interesting work on the importance of politicized “non-politicized” commissions.)

An SIA, however, gives policymakers direct control of the commission, the even more important data and information system, and the goals and direction of state sentencing policy. The policymakers still may be indifferent, especially if sentencing/corrections aren’t hot on political agendas at a particular time, but they can have hope of movement in the face of recalcitrant commissioners impressed with their own voices. (Those commissions are ignored by policymakers, who go their own way, anyway, so there’s no loss of “insulation” here.)

(ASIDE: I realize that this is antithetical to the traditional reasons for insulating commissions, but, really, what commission actually controls sentencing policy anywhere in this country? Why don’t they? Because they don’t have enough political influence unless it’s convenient for the policymakers, so why not just cut to the chase? Too often we’ve made simply getting to guidelines and a commission the end itself, not following up on the rational, enlightened policy that’s supposed to result and that, it turns out, doesn’t automatically flow. Even the model guidelines states have seen their work warped by time, by initiatives (WA, OR), or by mandatory legislation (MN, which now sports one of the highest incarceration rate growths in the country). Incorporating “politics” into the process in an institutionalized form from the start may, may, actually serve to improve commission “say” or at least achievement of the sentencing policy goals we want. Now back to your regularly scheduled programming.)

The SIA can set the state’s policy goals and direct commission and data subgroups specifically toward them. They can even select and remove obstacles to goal achievement from both groups. Most importantly, the SIA would be a higher profile group than any sentencing commission I know of and would be subject to greater attention and accountability by the media, advocacy groups, policy students, and the attentive public (both of them).

Too often, our focus has been on the very few commissions with demonstrated influence and not enough on the stuck ones and how to get them unstuck. SIAs would address that problem. A high-profile SIA could knock heads and take names when commissioners stall in pursuit of their own or constituency agendas, when the commissions plateau at the “all we do is report average sentences” stage, when agencies are less than forthcoming in providing data and information, when serious sentencing policy has to be made or when unserious sentencing policy is about to be. SIAs wouldn’t necessarily kill the idea or practice of sentencing commissions. A well-functioning commission would actually have the institutionalized champions it always needs (I’m not saying SIA members would be institutionalized). And the data function so important to informing effective policy would be protected and ongoing even if the commission itself is disregarded or dismembered (figuratively).

Okay, there it is. I’ve put possible solutions to the heartfelt lament of my shy, demure correspondent about her state’s failure to make the sentencing commission concept work. Her situation is not uncommon, just usually overlooked as we press the case for commissions. But those like her, who have tried commissions and watched them fail (perhaps more than once) or considered them and were not overwhelmed by their actual successes, need other options. The states with successful commissions had things come together, political climate, extraordinary leadership, plain luck, that other states haven’t. Telling my correspondent to keep trying that commission/guidelines thing over and over and over isn’t the definition of helpful. So let’s try to send her other rescue lines.

Have at what I’ve said. Holes in any of this? Of course. I didn’t talk about risk assessment modeling at all. And blood pressures went up galore when I started talking about how commissions protected from political power should be overridden in favor of bodies with it. I may post anonymously just to say how impossible this all is. But is it? You have better, wise guy? Let’s hear it. Don’t just shoot at me, though. Tell me, and her, what might get more done. Maybe we can even coax Ms. Shy-and-Demure out of hiding.

Take your shots. We’re ready.

Pigs Through a Python

Back in the late 90s when I was working in DC, at several research conferences I attended, I would hear street-level researchers talk about how the crime wave of the late 80s-early 90s had turned a bunch of children and adolescents off the "coolness" of that culture. Kids were tired of seeing siblings, cousins, uncles, parents, not all of them criminals, killed, maimed, scared witless, or hauled off to prison. Girls, the most powerful force against crime known to man (literally), were turning away more from guys whose long-term prospects included coffins and metal bars. The predictions were that these young people would turn away from crime as they got older and crime rates would head down even more.

Now most of that was the dreaded, disreputable "anecdotal evidence" that is the curse of death for people who have never been in a tornado or bombing and need a representative sample to give them evidence to get out of the way of the next one. But it rang true to me, and subsequent history fits as well. It made me think of crime wave starts and stops as cultural more than structural, that is, more dependent on attitudes than on social restrictions (like a crim just system). It also made me think in terms of the old "pigs through a python" analogy. You know, that appetizing image of the big snake ingesting pig after pig, looking like a declining sine wave (cosine?). Which meant that you could have a spike in crime, like we did with young people in drugs and guns in the 80s and 90s, which would go down as little brothers and sisters swearing off that stuff came of age. The problem would be that, after a period of quiet, the old lessons would be lost, and the new group of young people, again inspired by drugs and guns, would make a comeback about . . . now.

So the recent stories on the recent upsurge of violent crime, like this one in USA Today, on the PERF conference and the continuing UCR increase the first 6 months this year, haven't surprised me. Especially the growth in youth offenses again. These people don't have a clue about prison or "do the crime, do the time." The lessons have been lost.

But if the "pigs" analogy is even halfway right, it calls into question our "solutions" to the last bump through the belly. If the earlier decline was more the result of cultural dissatisfaction and rejection than more cells (and most research agrees that only a quarter, give or take, of the decline was due to incarceration increases, with low incarceration states having as much or more decline than high ones, indicating a national cultural wave), then responding now as we did then will probably not be as fruitful as trying to hasten the cultural feedback of distaste. This isn't an "anti-prison" statement, just trying to analyze realistically. It means that we will need containment strategies that certainly can and should include prison but much more and better. It also means that any real abatement will come due to the communities and cultures themselves down the road in a few years.

What will this produce for corrections and sentencing? Well, we do have history to guide us this time. More news and fear of crime, more political reaction, tougher sentencing, more incarceration. How much "more" and "tougher" is up for grabs, but the current zeal for Jessica's Law and other child/sex offender mandatories gives us a clue not just of punishment orientation but also potential sentencing commission (in)ability to play prescribed roles. (In WI, the state legislature never even looked at the state's commission when discussing and approving its Jessica's Law.) And for those into alternative punishments and cost-effectiveness advocacy, the future, therefore, will probably not be any kinder than it was when the last pig went through. Those things have been the first to go when crime increases meet finite budgets. I wish I could be more hopeful for those of you in this camp, but that's also known as analyzing realistically.

Maybe I'll be proven wrong one more time, and this is just a blip. But the earlier anecdotes would predict that it's not. Anecdotes really have been known to add up to truth, and we may just have to hold on until this pig makes it through.

Wednesday, August 30, 2006

What If a Commission Won't Work in Your State? Part I

Part I of a two-part series.

A shy and demure correspondent won’t post but wants some discussion about what to do if your state needs better control of its sentencing policy for fiscal, proportionality, or other reasons but either can’t get a commission going or one(s) it’s tried sink quickly into moribund molasses (which isn’t that tasty, either). In sentencing, we do tend to focus primarily on the states with working commissions/guidelines, like the only judiciary worth studying is the federal. So let’s move from that a bit and throw some ideas out there to see if we can generate some discussion for her.

It seems to me that effective sentencing policy requires transparency and accountability. That doesn’t mean talk-radio transparency and accountability, but, for policy purposes, the ability to review clearly what is done in the aggregate (not in individual cases or judges) and to evaluate whether the results are the best possible for the selected goals. If you are trying to max out the impact of your dollars, are the sentences given in your state the best to do that? If unbiased sentences are your goal, is their distribution reasonable? If crime control is predominant, do these sentences get the job done better than in states with different sentencing patterns? Of course I’m simplifying, as most state sentencing policy will be a mix of these and other goals, but the principles of transparency and accountability will be the same.

Underlying all this will be a good data system, one that provides information to policymakers about what’s happening and to practitioners about what sentences seem to best achieve the selected goals. One of the main reasons sentencing commissions are valuable is their standing as nonpartisan repositories of data like these, analyzed and reported professionally, untainted by vested bureaucratic interests, unblown by the political winds that can warp out all value. Which means that, in my view anyway, all states should focus on creation of this data collection/analysis capacity, whether they have a commission or not. So, if you don’t or can’t have an effective commission, at least develop and nurture your sentencing data.

Where and how can that be done?

One place I recommend, maybe because I started my professional career in one, is the state budget office. In every such abode, there exists a section/unit/cubicle which has responsibility for recommending expenditures (pay raises, equipment purchases, new facilities) for courts and corrections, coincidentally the two locations with the data you need most. Neither are traditionally known for generosity with their numbers, but both must have annual budget increases each year. See where I’m going?

The problem is that most budget offices don’t see themselves with this kind of direct data collection/analysis function (you’re talking to one of the few people in human history to have done a dissertation on state budget office functioning . . . goosebumps, right?). They’d have to get over it and hire analysts with the expressed duty of maintaining data and evaluating sentencing effectiveness at stated goals. If the executive branch folks aren’t up to it, then maybe, if the legislature has a research or audit bureau, it could do it. Or cooperate with the budget office people. Whatever. The point is that those who have a say over the dollars going to the agencies which can supply needed sentencing data have a level of control even commissions don’t have. (It’s not an accident that VA’s successful “voluntary” guidelines occur in a state where the legislature picks the judges. In reality, this may make theirs the most “prescriptive” of all systems.) The danger, of course, is that executive and/or legislative offices are subject to sometimes extreme and inconsistent political winds that can compromise their integrity and their results’. (Got to work in one of those offices, way back when.) Unless that can be managed, this idea therefore has problems. Not insurmountable, but problems.

So, where else? How about someplace buffered (in part) from those winds? Contract with an existing university institute or center specializing in policy analysis, or create one if your state lacks. Tie court and correctional funding to successful provision of necessary data. (Do NOT pull the old “all state agencies shall cooperate in provision of data and information” crap you see stuck pro forma into most legislation like this. It’s toothless.) Again, the goal is to develop a body of knowledge against which to compare actual practice and to highlight who’s getting it right and who’s not.

These are not my ideas. (I can hear Steve Chanenson, Ronald Wright, Marc Miller, Judge Marcus in Portland all screaming copyright violation as I type.) I’ve just outlined the basics of a “sentencing information system,” which really does hold promise as a second round of sentencing reform, especially for states wanting to avoid the problems associated with guidelines and failed or failing commissions. The last National Association of Sentencing Commissions conference spelled out many of their thoughts, and other presenters demonstrated the software potential of these systems. It’s there, and it’s viable. It’s just waiting for some state to take the dare.

The biggest problem for SIS and commissions, however, is not the collecting, analyzing, and reporting of data to inform sentencing and sentencing policy to achieve desired goals. It’s the audience that will receive all this when completed. If data fall in a forest and no one is around to hear . . . ? What I’ve outlined above doesn’t get at Wendy Kaminer’s wonderful, realist line, “Knowledge isn’t power in criminal justice debates; knowledge is irrelevant.” So how do we draw the policymakers in, rather than just hope you run across some with interest and intelligence? What’s their role in this if there’s no commission? What if you’ve so insulated the commission from political influence there’s no point in politicians participating (hence, the data in the forest)?

Ah, that’s another post.

Around the Blogs 8-30-06

Crim Prof Blog loads us up with interesting items. This one takes us to a USA Today story on the GAO report last week questioning the effectiveness of the National Youth Anti-Drug Media Campaign, which also got dissed in 2003 by the OMB. Seems kids remembered the ads but didn't get their subsequent behavior swayed. An Office of National Drug Control Policy spokesperson responded that the GAO report was "irrelevant to us." The Crim Prof folks also direct us to an SSRN article discussing changes in the cj system that could reduce conviction of the innocent. The author IDs as problems false ID, false confessions, bad forensics, and jailhouse confessions, and throws in requiring admission of guilt before being considered for parole. And they also have one of those "don't know what to think" pieces, linking to a British article showing how gender preferences in birthing in developing nations (as in, the male:female ratio is (un)surprisingly abnormal with males) might lead to a bunch of really frustrated bachelors once adults who might need crime as a outlet release. Uh-huh. . . . Finally, Doug Berman links us to the latest NIJ report on drug courts, emphasizing "what works." As Doug notes, as our practical knowledge base grows, "research can be translated into practice."

NJ Commission Newsletter

In Ben Baryln's latest sentencing newsletter, he says, "Through the web site, I've sought to track and capture all facets of sentencing at both the state and federal levels, including the enactment of sentencing provisions, case law developments, and relevant reports issued by government entities, academia, and other interested parties. . . . The key, as I see it, is persuade the three branches of government and, of no less importance, practitioners and the general public, that New Jersey need not reeinvent the wheel with respect to sentencing reform, and that such reforms, if implemented, will indeed yield tangible benefits."

He also lets us know that we can"let your readers know that they can 'subscribe' to the newsletter and e-mail distribution list by e-mailing me at" Now we have, Ben, and good luck. It's been very good to have someone with Ben's energy and talent leap into commissioning. If NJ doesn't get to guidelines, it won't be because of Ben.

Tuesday, August 29, 2006

So You Want to Start a Sentencing Commission? Part XII (Finale)

Part XII of a series. In this conclusion of "So You Want to Start a Sentencing Commission?" we discuss why your commission will fail. That's right, "will fail." Sound fun? Thanks for staying with us for the whole thing. Send us any comments or questions that we still need to address.

Why Your Commission Will Fail

Okay, you’ve gotten this far. You’re as well prepared for starting a successful commission as most complete novices could be. And, by tapping into the commission “fraternity,” you’ll have expertise and experience at your call virtually any time. Your odds of establishing a long-term agency are actually pretty good. But not 100%. Your commission could still fail, in a year or a decade.

Why? Your state may be creative in how it does it, but some possible reasons for failure are predictable (and historical). We’ve mentioned them in passing for the most part, but let’s systematize them so you’ll have no defense for not being prepared.

* The commission failed to establish cooperative, systematic relationships and lost the resulting battle over turf and/or funding.
* It basically sat there, dithering year after year, not accomplishing much, happily classifying new offenses, issuing an innocuous report or two a year, not really having much of an impact. In other words, it failed to establish meaningful legitimacy and credibility in state policy. Come tough time or a periodic fad for “efficiency,” the commission gets reexamined, the examiners say “Wuhhh?,” and, snip, it becomes easy proof of the tough work keeping democracy accountable.
* It got too tied to a particular faction in a policy controversy, politicizing itself and losing whatever legitimacy and credibility it had built up. As noted before, the faction may even be the winner in the controversy, but it may need to (re)establish its own legitimacy and credibility. Think it will hold tightly to a compromised commission?
* It remained too neutral in a policy controversy, aggravating every side that wanted its support (see Cambodia, 1970s). Yes, it’s damned if you do, damned if you don’t. It’s also reality and reaffirms the need for politically astute chairs and members, who may nevertheless not be enough.
* It proposed or recommended sentencing policy that ran afoul of public and/or demagogic wants and needs. Sometimes that famous legitimacy/credibility will provide enough surplus for survival, but good commissions have been washed aside in the face of public panic attacks.
* It was created by an administration or party in power that is no longer there. The newcomers may lack the same interest or investment, and, ironically, if the commission has been a conspicuous success of a particularly hated former administration, it may be deliberately removed, like a splinter or a wart. Reaching out to others in advance can broaden a constituency and is strongly recommended. But don’t count on it to overcome the potential here. If the commission can get through a couple of new administrations or legislative party changes, it’s probably good to go, but don’t buy stationery in bulk.
* And then, one day, out of the blue, without word or warning, your commission may hear that a provision slipped into an authorization or budget bill on the last day passed without discussion and abolished your commission or left it without funding. Why did this happen? Your commission hacked off the wrong individual. The governor maybe, or his chief of staff, a legislative chair, maybe even a legislator on your own commission who got voted down once too often. Maybe your prison population projection shot down a favored bill, maybe the wrong judge got defended. Maybe your executive director rubbed too many people the wrong way. Maybe it was Tuesday. You may never know. All this speaks to the following: Take nothing for granted, ever. Pay attention and be prepared to respond fast. Will that save the commission? Did it save the commissions that suffered the lunacy listed above? Some it did, some it didn’t. But it does give you a better chance.


Sentencing commissions have been around long enough now to have established themselves as part of the state government and criminal justice landscape. Not having one doesn’t stop a state from having good policy and an effective process, but having one brings more light and knowledge to both. As we noted above, because of this dispensability, commissions often live on an edge of survival, but enough history exists to learn from. You should be free to create your own particular mistakes now. Good luck to you and remember—you will be part of a fraternity that helps each other. Anything helpful you got from this, please pay it forward.

Goals of Sentencing Commissions

Going through some papers from a year or so back, ran across a table I drew up after going through commission websites and examining their stated goals. Not the least bit scientific, but here's what I found for the 19 commissions I looked at. If you looked at how frequently they were mentioned as explicit goals and the order in which they were listed, you would hopefully come up with a list of goals in their priority something like this:

Public Safety/Emphasis on Violent and Career Offenders
Truth in Sentencing
Resource Availability
Alternative Sentencing
Judicial Discretion
Statement of Public Values
Public Accountability
Victim Impact
Crime Reduction
Prison Overcrowding

I wouldn't bet much on the exactitude here, but thought you might be interested. I should never start cleaning out files.

Around the Blogs 8-29-06

Quick hits today. Kent Scheidegger at Crime and Consequences alerts us to a report in the UK about incarceration rates in England and Wales being higher than the US'. You think it's the language? Maybe this is a good time to have a national language here, like Spanish or . . . no, not French [that should get some comments]. . . . The Real Cost of Prisons gives us a Neil Peirce column on the costs and benefits of our drug war, including the effort by the state bar in WA to get King County serious about treatment options. Pretty familiar stuff, but a quick reference. I liked the end the best, where they point out how marijuana use by teens has apparently gone down in states where medical use has been legalized. Says a quote, "It's not as cool when grandma uses marijuana for cancer pain." There's your next anti-drug commercial. . . . And finally, via Prevention Works, the blog of the National Crime Prevention Council, we get an update on the latest issue of the Western Criminology Review devoted solely to reentry research and book reviews. Speaking of the latter, two of them were written by extremely bright young people who worked for me when I was with the MD sentencing commission, Jill Farrell and Dave Bierie. For any future employers, they're ABD, experienced, and very pleasant to have around. Just saying. And they didn't even pay me. Yet.

Monday, August 28, 2006

Law and Social Sciences Sittin' in a Tree

Christopher Zorn continues to guest-post good stuff at Empirical Legal Studies. His latest focuses on the very real disjunction between legal scholars and social scientists regarding how they approach argument and evidence. Legal types take the adversarial approach familiar at trial and let the best case-maker, not reality or justice, win, which Zorn notes objectively is not really the way of science (at least in textbooks). Social science types, however (and I are one, huh huh), cling far too much to their aptly-named "scientism," that futile belief that positivism mixed with enough linear regression will produce the only "truth," something legal types just wouldn't understand, in the scientistic mindset. Zorn's solution in order to wring the best from both perspectives is to have academically mixed programs in all the disciplines, to cross-fertilize and open minds (and, unmentioned, open new academic positions for the bridgers).

He gets a lot right here, and it's especially applicable to my experience in sentencing. Commissions are composed predominantly of legal types, and their rules for evidence and argumentation in deciding commission policy . . . rule. Get a corrections or stat guy or an actual academic in social science justifying some recommendation because studies show that, within a 95% or 99% confidence level, the . . . what?! 5% doubt?! Even 1%, not beyond a reasonable doubt!!?? And we're believing this guy? I've also seen debates turn into cross-exams and all the tricks thereof come out in what are supposedly nonpartisan, dispassionate searches for understanding and sound decisions. "Oh, yeah, and where were you at 11:07 the night the study was done? That's not what we heard from the public defender rep over there . . . ." One very good reason why the overwhelming research on more cost-effective means of dealing with offenders has rarely caught on in policymaking is that the policymakers themselves literally do not have the mental frameworks necessary to incorporate it into their thoughts. They're not stupid. They just think as the legal types Zorn describes, not as scientists.

So any efforts to get more research findings into public results has to understand that. I part with Zorn a little bit on how to deal with it academically, but only a little. I don't think cross-listing courses, faculty, etc., is the answer in the long run. I think the effort should go full bore--Schools of Law and Policy with the primary (not secondary or tangential) purpose of linking the two perspectives into more commonly held paradigms. That's a long way off, though, if ever. His more incremental rec's are at least a good first step. Go and give him some hits before his guest days expire.

So You Want to Start a Sentencing Commission? Part XI

Part XI of a series. In Part X we started listing some predictable concerns and issues that you may face as you start and continue a sentencing commission. This part finishes that section.

Criminal history access—Structured sentencing greatly depends on accurate information about the offender’s criminal history. One of many dirty secrets about criminal justice data is that criminal history sucks. Dispositions just don’t get linked back well to arrests or charges, and out-of-state convictions are worse. Efforts have improved things greatly from the old days and enriched several consulting firms (who, when you think of it, never really have had the incentive to get the system right), but problems are still there and real. Plus, juvenile records are frequently unavailable legally. Worst for you, commissions aren’t normally considered “law enforcement” agencies legally authorized to have access to most criminal history data bases. The upshot of it all is that this is one of the most important areas for that “cooperation with other agencies” mentioned before. Figure out some good quid pro quos that you can trade off when you need data, or start on that legislative exemption for access now.

Internships and graduate students—When seeking quality employees at reasonable prices, one of the best sources is your friendly neighborhood college or university. Undergrads in criminal justice, poli sci, or public administration frequently need internships for graduation requirements, getting 3 hours credit for 120 hours of semester work as a rule, and there you are to help them out. While they can’t rewrite the criminal code for you, they can do data entry (and learn way too much about how our justice process operates at the same time), staff meetings and direct sentencing questions, and the best ones can do some writing and analysis. Grad students usually insist on being paid (although volunteers are not unheard of), but the rate is only $13-$17/hour for 10-20 hours a week (and little in the way of benefits). This may make you feel a tad exploitative, especially if they are doing statistical or other heavy-duty analysis for you, but give them a good title, access to networking with commissioners and their staffs, and their names on any publications resulting from anything they’ve worked on to add onto their resumes. Your conscience won’t be quite as hard on you. An added benefit of both grads and undergrads is that they may have access to university resources (library, intranet resources, etc.) that you don’t and can glom onto. Plus, with few exceptions, they reinvigorate a place with their interest and energy. That’s a surprisingly valuable contribution once you’ve experienced it.

Recording and reporting names of judges—To ensure unreasoned responsiveness to every panic that makes local TV news, most states require their judges in criminal cases to be elected, some frequently, most for relatively long terms after an initial shorter appointment. As a result, come election time or the latest heinous crime, political opponents and/or ADHD reporters will love to cherrypick individual cases in which sentences questionable in their view were given. Now, some judicial decisions should be second- and third-guessed, and some judges do do excellent maniac impressions. But, in our experience, this selective use of judicial data does not present an accurate or fair view of most judges or the process. (If the average sentence for child sexual abuse is only 3 or 4 years, as is frequent and a legitimate outrage, maybe it’s because it’s hard to get charges or convictions at all against “Coach” or “Father So-and-So” or “Grandpa,” not because judges or prosecutors are lunatics.)

However, start a sentencing commission and watch people suddenly want to dig through your data (actually, have you do it) to find that eye-catching, ear-numbing case. Judges not unreasonably take unkindly to this treatment, although technically the information is public and available in county files with a little work to anyone. Judges, therefore, may object to your commission maintaining their names in your data base of sentences, and the commission may have to make and defend a decision either way.

You will probably find that some judges don’t care. Those from small jurisdictions tend to be highly unsympathetic since they have to deal with a public that every week at the grocery store knows every local case that goes to court. One-judge counties are particularly easy to track in your data base. On the other hand, in large, usually urban jurisdictions with severe case pressures and local contexts that favor deals for lower sentences, the judges may opt for anonymity, especially since local TV and a major newspaper or two are likely in the city with them.

There’s no good answer here. The public has a right to know, and you’re probably bound by state law appropriately. But the judges have a right to expect buffering from your commission, too, against unfair use of your data. Pennsylvania made a conscious decision to report, Maryland one not to (although it has allowed access to its files instead and notifies the judges studied). In both cases, your commission should vocally warn users of its data against its misuse and speak out strongly when it happens.

Pursuit of grants—Like any government agency, your commission will likely have need for more funding. An obvious but very difficult source of supplementary funds jumping from the lips of everyone who has never had to do it is grants. Why is it hard? Well, the feds have never been that interested in sentencing research (it’s only the basis of everything the criminal justice system ends up doing), so you’ll have to be creative to devise the necessary link to the topics in their RFPs (request for proposal). And while some foundations are possibilities, they are hard-pressed for funds these days, too, and usually are interested more in children and families or juvenile justice than adult sentencing. That means tying your requests to topics like reentry, child abuse, or certifications for adult sentencing, which may tend to be peripheral to your commission. On top of all that, foundations often have goals identifiable with political sides, something a nonpartisan commission usually avoids, wisely. Still, this is an area in which those cooperative relationships with other agencies, like your Statistical Analysis Center or Office of Juvenile Justice, may pay off, literally, through inventive partnerships. Then all you have to do is figure how to divvy up the new funding and staff. Easy. (And, if you can tie it to “Homeland Security,” more the better.)

Performance measurement—It may sound odd to think of performance measurement for a nonpartisan body that basically sits around making recommendations about sentences. But in this heyday of state government “managing for results,” no one is immune. They might let you slide with “number of reports published” or “worksheet training sessions delivered,” but this actually gives your commission a nice chance to think through its own definition of success and what it wants to accomplish. And if you put together that mission statement, you should be able to derive objectives from which real actions are possible. These actions presumably have goals, which can then be observed for achievement.

The problem, of course, is age-old. Policymakers, and their staffs straight from an MPA program where easy performance measurement and coherent data bases are givens, will examine the measures for change in a preferred direction one year to the next. This has led to unbelievable game-playing in the four-plus decades “performance measurement” has been extolled as the savior of public management and budgeting. There has yet to be produced reason to believe any given iteration will reduce the games and produce better results (some see in this the definition of insanity, but we will be kind). Done correctly, and not abused by staffers shaking trees for budget cuts or promotions, measures can help track performance, signaling changes that need attention and the possibility of new directions. Your commission will likely have to play the game at some point. If it’s proactive, it will have more say over how it’s judged itself.

In the next part we'll tell you why your commission will fail and then we conclude. Sounds like fun, right?

Little Hoover Commission Hearing on Commissions

Our friend Ben Barlyn from the NJ Sentencing Commission sends along a message and a link:

"Hi folks:

I wanted to pass this along. The link directs you to the material presented by speakers, including Rick Kern and Judge Ross, who recently testified in support of the establishment of a permanent sentencing commission in California. I'm not through digesting all the great stuff but there appears to be a trove of pertinent material, especially the report submitted by the Stanford School of Criminal Justice. Bon appetit, and have a great weekend.


You'll also find comments from a couple of judges which are worth reading. Thanks for sending this along, Ben. Anyone else got some really good links for our readers?

CT DOC Recidivism Study

When we put up the links to state agency recidivism studies a week or two back, we inadvertently left out one of the best, this thorough report by the DOC in CT. Our apologies. Look it over if you have a chance as well as the other items there. Like we say, we'd like to be the first place you look for reports that will inform corrections/sentencing policy.

Sunday, August 27, 2006

So You Want to Start a Sentencing Commission? Part X

Part X of a series. The next two parts will address potential concerns and issues that you may have as you develop and institute your sentencing commission.

Concerns and Issues

The following is a series of “quick hits” at things that you will or should have to consider as you develop your commission’s functions. They aren’t in any particular order of importance or need. If they aren’t relevant to your commission right now, don’t worry about them (just be aware that they could pop up, though). But, as you organize and structure, it may help you to have a ready-made agenda like this from people who have been there before.

Ethics requirements—States generally have ethics commissions to handle questions of propriety about campaign contributions to and financial relationships of holders of public office with significant influence over public policy. While you may protest about your commission having significant influence over public policy, the strangely consistently humorless ethics officials believe it does. Therefore, they will require your commissioners to meet state ethics requirements. Since many if not most of your commissioners are already public officials, the requirements will affect only a few, but find out who they are and get the proper paperwork to them in time to avoid whatever fines they will otherwise have to pay for their usually thankless service.

Cooperation with other agencies—Broadly speaking, since you will find data and information from other related agencies useful and possible grants through partnerships with them, you should be willing to cooperate with them toward (legal) ends they may have. Turf wars are not unknown, however, especially as agencies react to this new commission on the block. While it won’t be true of all, be prepared for opposition from judges protecting judicial discretion and pomposity, corrections officials controlling (with varying success) their resources and bedspace, prosecutors and law enforcement suspecting liberalizing of sentences, victims groups fearing being shut out by an unhearing, mechanical process, defense attorneys and offenders groups suspecting conservatizing of sentences, criminal justice research agencies seeing a statistical center potentially competing for attention and funding, etc. In other words, from practically everyone in the process.

Since commissions have representation from virtually everyone, they make a logical place for centralization of much data and analysis (although limited in most states by staff size). That’s both a virtue and a threat. Therefore, commissions and their staffs must prioritize their activities and make them known through their commission representatives. This will signal when cooperation is to be sought and not and on what grounds. Don’t make too much of this, though. Usually cooperation is forthcoming, and noncooperation is usually passively aggressive, not something that involves bloodletting or, worse, headlines.

Listservs—One of the easiest ways to facilitate communication on a commission is to create an e-mail listserv hooking up all members. The problem is that it can be tempting to do public business this way (discuss issues, take votes, etc.). This generally violates your state’s public information laws. So, to avoid problems, use listservs primarily as a means for staff to coordinate and disseminate information and materials (agendas, articles, notices). If you have things to distribute, send them to staff to get out. And never, never, never hit “reply all.”

Worksheets—Theoretically, given a court system with good computerized records of sentences and their delivery and a corrections system with good computerized records of time served and how, a sentencing data system can be built through a merger of the relevant information from each system. Now compute in hardware and software of various ages, intentions, and compatibilities; data designed for management, not policy, purposes; and a bureaucratic disinclination to let outsiders know “more than they need to” about how each system operates. Still theoretically possible? And this is assuming that each system exists in the real world, and important records aren’t still on 3 X 5 cards somewhere. If you’re fortunate enough to have 21st century, cooperative, compatible systems, you might get along without worksheets about the sentences filled out at the time of sentencing and sent to the commission for data entry, all umpteen thousand cases a year. On the other hand, if you are not lucky, then you’ll need to either design a worksheet or pull in court sentencing documents and DOC documents for in-house coding prior to data entry.

Worksheets will need case numbers, names (all those used by the offender if possible), Social Security numbers (all those used by the offender if possible), dates of birth (all those . . . you get the idea), and other demographics. You may run into trouble on racial data (recorded by face-to-face observation by the recording agent or by self-reporting by the offender—both with major potential problems) since race and sentence are not to be formally linked. And tracking “Hispanic/Latino” apart from “black” and “white” is a giant headache, as is mixed race. Still, if your commission was formed in part to deal with racial disparity in sentencing, you have to get these data and just acknowledge the difficulties in the resulting reports.

Obviously you will need the basic information about the case(s): sentences broken down by all the particulars (including whether they are concurrent or consecutive to other sentences), all the factors required by the legislature or the commission to enhance the sentence (weapon use, vulnerable victim, amount stolen or of drug, etc.), the type of disposition process (trial by judge or jury, type of plea, type of defense counsel, etc.), a location for calculation of the applicable recommended guideline and actual sentence, with space for reasons for judicial departures. Some states also include space for information about victim knowledge about and participation in the process.

Who all gets copies will be up to your state, but obviously one will get sent to your commission. If not automated, you will need significant staff support for data entry, but the programs can be built on a small multitude of software systems. Storage is also a major consideration, including adequate security, space, and legal requirements. Three years is pretty common for public documents to be kept, but be sure to find out before you start shredding or burning. (Shredding is generally the method preferred by law and fire marshals, so find a reputable firm and get ready to pay.)

At some point in our increasingly paperless world, all worksheet information will be sent electronically directly to the commission and its data base, and more jobs will be downsized. Since commissions deal with aggregate numbers and not individual cases, this probably won’t be a bad thing. Until that giant magnetic storm goes over commission offices.

Last couple of worksheet points. Accurate completion of worksheets requires a staff “helpline,” a manual, and regular training (aka, more staff). Much can be put online, and videos and online training can be illustrative at relatively low cost. The commission should also decide how aggressively to pursue worksheet errors. The pursuit takes up valuable time, aggravates the judges, and, in the aggregate, is not likely to impact resulting data greatly (except for those 2000-year sentences). On the other hand, no pursuit may indicate a lack of concern. Clearly yet another job for the Sentencing Guidelines Committee.

Prison population projections—An important value of structured sentencing is your ability to use it to estimate future needs for prison bedspace. If you estimate well the offender intake data and fit each offender properly to his/her sentence cell in the grid and then figure an average for all offenders in that cell, and if you know well the time to be served, you will be able to predict how long incoming offenders each year will serve in prison. Added to the time to be served of your initial “stock” population already in prison and of the predicted number of offenders revoked from probation or parole, you will be able to project how many prison beds you will need each year. And how many new prisons. And how changes in sentences affecting the guidelines’ cells in the grid will change bedspace and prison needs. Policymakers with an eye to not spending more than they have to will find this useful information. Policymakers with an eye to reelection above all may see their “get tough” legislation suddenly questioned and find this information a subversive plot and the commission suddenly dispensable.

Clearly, these kinds of projections can make commissions vital players in effective sentencing policy and fiscal impact analyses, and, if that’s what your commission is there for, go for it. Know these few things, however. You won’t be the only game in town, and folks who don’t like your projections may wheel out their “equal and opposite Ph.D.’s” with projections that magically fit what these folks want to do. And, you better be pretty accurate. More than one or two really bad projections will destroy commission credibility, perhaps on more than projections. Plus, remember “if you build it, they will revoke”—that is, projections of bedspace needs leading to more prison-building within a given time frame may lead to increased probation and parole revocations as space becomes available, filling the beds faster than projected. And, potentially worst of all, the technical wizardry and bells and whistles that go into projections can give policymakers a false sense of confidence in results. Staff may be asked for projections ten or even twenty years into the future. Weathermen have better luck with seven day forecasts. So, while projections can be very empowering for commissions, they truly are a “be careful what you wish for” situation. Some states literally create triangulated projections using the predictions of their commissions, DOCs, state budget office, and others. Not as empowering, and controversial when someone goes off on their own, but probably safer politically and statistically.

In part XI we will finish Concerns and Issues. Don't give up on us. We only have a couple of more parts to go after that.

Friday, August 25, 2006

So You Want to Start a Sentencing Commission? Part IX

Part IX of a series. In Part VIII we talked about commission structuring, including standing committees. This part finishes that section.

Data and Research
Since most of the commissioners will likely have legal training (meaning, they ran from math classes), it is sometimes hard to populate the sentencing statistics committee with people knowledgeable enough to contribute effectively. As with the overly-knowledgeable folks on Sentencing Guidelines, this can be a mixed blessing. Deference to statistical staff is far more likely, and good staff will respond well. Poor staff won’t. However, the commission’s data reports will have enough potential impact on practitioners and others to generate feedback to highlight problems with the data.

Unfortunately, the feedback can go beyond locating inaccurate data. Good numbers will likely be challenged as well by those negatively affected by them, and commissions and this committee in particular need to be prepared. Staff should be sensitive to the political ramifications of their work and not report anything reactive (even under time pressure, such as a call from the governor’s office for stats in ten minutes) without alerting at least the chair of this committee and/or the commission chair. The committee and commission, however, must resist the inevitable temptation to vet every single piece of data reported in a regular report, a request from the public, or a demand from a policymaker. Staff should be trusted and live up to that trust, or the commission will delay and deny information necessary in sentencing deliberations. In the long-term, this will hurt the commission’s reputation with sentencing stakeholders.

This committee will not likely deal only with statistics. It will probably have to consider sharing of data about sentences, costs, offenders, etc., with other agencies and governments. It may also look at types of hardware and software for data systems, requests for data from academic researchers, public information act requests, and even subpoenas for evidence in trial, appeals, and sometimes civil proceedings against government officials. As mentioned, the average commissioner does not have this background. It is, therefore, often helpful to establish a “workgroup” of people in the state criminal justice and information technology communities to provide advice and assistance. Whether this is a formal offshoot of the committee or merely folks “on call” to staff doesn’t matter as much as having the expertise at hand.

Ultimately, the value of your sentencing commission to the state and public is not simply maintenance of formalistic sentencing grids. It’s also the collecting and reporting of the data about the sentencing that will inform later deliberation and decisions by the commission, legislature, governor. Good data will not immunize a state from rash and opportunistic political acts, especially in criminal justice, but, without good data in which people are confident, the process is flying blind. That’s not necessarily a fear for politicians who hope to be off the plane before it crashes, but it’s hell for the taxpayers and citizens who have to pay the price when it does.

Public Outreach
At some point your commission is going to have to set down policies on public information act requests (formal form or not?, within the law, how much to charge for copying, staff time, etc.?, notify a judge if s/he has been the specific target of a request or not?), on who says what when the media come calling (staff on technical questions, chair on policy?), on whether to do public polling or focus groups, on newsletters, press releases, staff presentations, on dealing with public controversies or attacks on the commission itself. Again, the formal backgrounds of commissioners in public relations are likely to be limited, but most of them got appointed for political involvement and ties and should have more experience and expertise than with statistical data.

A prudent commission will be proactive in its public outreach. Unless the commission is on the front page of the newspaper (which will never happen if it’s good news), the average citizen will never hear about it. But outreach to active community groups, churches, victims’ and offenders’ organizations, and specialized practitioner communities can alert them of the commission’s existence and purpose and put names and faces from more pleasant and informative encounters in their minds if/when hooey hits fan.

And, keep in mind that most of the data and information that a commission brings to the table is actually helpful to the parties “outreached.” Practitioners usually benefit from knowing what the practice is. Citizens and interest groups benefit from accurate information beyond anecdote and/or their own unique experiences. And Lord knows the media benefit from access to good data, whether they actually use or report it appropriately or not.

Plus, what is there to fear from open channels to sources outside your commission? Yes, sometimes people find out embarrassing things or take data out of context mistakenly or mendaciously. But data and information that people really want usually come out at some point anyway. Is it better to have little or no control or input over its use and interpretation? And the commission can, of all things, actually learn from its interaction with the public, with groups, with the folks in the field doing the heavy lifting. Too much policy is made by authorities from on high who then beknightedly drop it on an unsuspecting population who could have told them all the things that then screw up if anyone had bothered to ask them. So ask them. Yes, you’ll get a lot of talk radio-like wisdom, but you’ll also get some gems that may save you time, resources, and grief. And, cynically, you’ll also get a buffer from charges that you’re just another of those unresponsive, self-absorbed government agencies. No citizenry will ever come rushing to defend the life of a sentencing commission, but good, proactive public outreach can help prevent them rushing to tear it down.

Other Possible Committees
You’ll likely have several things pop up that do require some intensive investigation and thought delegated to a few commissioners for later recommendation but that do not last long enough to require forming a permanent committee. And you may have particular needs that do require one but that other states don’t have. However, as you plan your commission’s organization and structure, there are a few other areas that you might want to consider for standing committees.

Strategic Planning—Anyone who has been through this will groan, but careful early planning and then constant, consistent monitoring of performance will help keep your commission from straying into irrelevance or navel-gazing. The success of “problem-oriented policing” with its emphasis on Scanning, Analysis, Response, Assessment suggests a slightly different approach, “problem-oriented sentencing,” you may want this committee to develop, for example.
Policy and Legislation—Depending on how long your legislature meets and how many sentencing-related bills it considers, you might want a “quick strike team” that can support or oppose bills between commission meetings. As a rule, it’s easier to avoid controversy by ignoring bills until they pass. In practice, controversy is hard to ignore if you’re doing fiscal impact statements on those bills or if the legislature wants you to report on race of offenders’ birth mothers from other states or something similarly genius. But, given the pace and schedule of state legislatures, especially at the end of a session, getting the full commission for action is hard. A small committee for the purpose is much easier logistically, and should be populated by your most politically astute and/or influential commissioners.
Alternative Sentencing—While this can be part of Sentencing Guidelines, the intricacies and ramifications of alternative sentencing and its various means and programs may require more careful and specific consideration than the Sentencing Guidelines people can fit onto their loaded plates. If your state is seriously considering broadening its range of possible sanctions rather than just investigating occasional questions, your commission will likely benefit from a separate body here.
“Technocorrections”—Although legislatures, governors, courts, and commissions seem determined to ignore careful thought about the scope and consequences of our ongoing and unstoppable evolution of the use of surveillance, pharmaceutical, and genetic technologies for sanctions and behavior modification, the great hope of a democracy is that policymakers will consider world-shattering events such as these in something other than a piecemeal basis. Should your commission be the first, you will definitely need this committee, supported by workgroups of experts and laypeople with sense, if they can be found.
“Costing Out”—A technique becoming popular in state and local government management these days is known as “costing out” of functions and programs. The method may vary, but the idea is to establish among relevant communities a set of program and policy outcome priorities which can be triaged as necessary according to available funding. Once the priorities are known, the costs and benefits associated with what is known to work to achieve those priorities are established. The priorities are then reexamined to determine if the net of gains and losses are truly what is desired. If appropriate, the priorities are revamped to reflect what can be done with available dollars. Clearly, since one purpose of a sentencing commission and structured sentencing is to monitor costs/benefits of sentencing decisions and policies, this technique would seem relevant to a pro-active commission and fits well with the “evidence-based practice” buzz that’s currently in the air. Therefore, a committee to do regular costing out of sentencing impacts might do the taxpaying public a lot of good and rein in the opportunists in the policy community. It’s not regularly done at this time, however, so, again, your commission could be the guinea pig if you want.

Clearly, organization and structure are more involved that this brief detailing can portray. Your commission will, of course, have contexts that will shape the roles of your chair and other commissioners, when and where it meets and acts, the discretion given to staff, what kinds of committees are necessary and how effective they are. As stated about the mission statement, the really important thing is to have a clear and consensual vision of what you want the commission to be and to have done a year, five years, ten years down the road and then to organize around that vision. Will the organization and operating procedures change? Sure, as the circumstances force and the vision shifts. But as long as you don’t paralyze yourself with too much deliberation or fracture over individual agendas (the two biggest threats to commission action), your commission’s vision should fairly quickly guide you into patterns that will help you accomplish your basic functions and more.

Which leads us to a more detailed look at some specific concerns you’re probably going to have to deal with that we have yet to really cover.

Part X turns to specific issues and concerns you may have.

News of the Day 8-25-06

A couple of stories on second chances, not sure whether they're optimistic or not. Robert Ehrlich, MD's Republican governor, has stepped out of the "too scared to think straight" mold of most governors today, including his predecessor. Ehrlich has increased the number of pardons and clemencies he's issued in his term there. One of the smartest things he's done here, which should be modeled by other governors, is to bring the state's major victims' group into the process. Of course, the state's major victims' group, the MD Crime Victims Resource Center, should be the national model of its kind as well. [DISCLAIMER: when in MD, I served on the Center's board, and I have since reviewed grant proposals for them. I also count its talented, insomniac director, Russell Butler, as a friend.] The offspring of a truly amazing woman, Roberta Roper, who lost her daughter in one of the worst crimes I've ever heard, the Center is a hallmark of rational consideration of criminal justice policy which recognizes that victims aren't always best served by the blind vengeance of many "advocates" and preying demogoguery. Ehrlich consults with them, limits the surprises to victims and their families, and applies basic common sense. How refreshing. His wife being a former public defender also apparently brings a perspective that assists his reasoning. Being a Republican, Ehrlich might be doing a "Nixon goes to China" thing here that spurs others. It just seems a little awkward to praise someone for doing his job the way it's supposed to be done, but I'll leave it at that. . . . Then comes the Dallas Morning News article on a job fair for former felons that only drew 10 potential employers who mainly handed out job apps to thousands of attendees. One former inmate, one with a college degree making less than $6/hr., sums it all up, "American society needs to know that the crime rate will continue to climb . . . if people can't find gainful employment." (h/t Crim Prof Blog). . . One last note. What sentencing guideline ranges will likely apply to these anti-cohabitation laws we have being considered? Will they qualify for community sentences and, if so, what on earth could they be?

Around the Blogs 8-25-06

One of the best reasons to follow blogs is the chance to learn things that you hadn't heard before. Several examples today. Doug Berman (as usual) alerts us to an ABA publication that I was unaware of, Criminal Justice, which this issue features a sentencing symposium to which he directs us. . . . Christopher Zorn at Empirical Legal Studies praises the National Science Foundation and its Law and Social Sciences funding program, encouraging us to make use of it and dispensing of the misconceptions about it. Be sure to follow through the comments. . . . At Prawfsblawg, Michael O'Hear uses a WI case to outline several key themes in the common criticisms of our cj system: defendant rights run amok, sacrifice of the truth-seeking function, plea bargaining as lawlessness, and sentencing as lawlessness. You obviously need to read the post to get the details. . . . Finally, also from Prawfsblawg, we're led to this Reuters article on Chinese halfway houses for computer addicts, including bloggers who just can't seem to stop . . . can't seem to stop . . . can't seem to stop . . . .

Thursday, August 24, 2006

So You Want to Structure Sentencing? Part II

Part I discussed the decision to pursue sentencing guidelines. Here we take up the issue of “To Grid or Not to Grid.” (You might notice a little overlap here as Mike and I are not joined at the hip, but have similar ideas about a lot of this -- see posts below)

Tom asked whether the grid approach to guidelines is preferable to some other method. That is an interesting question, and worthy of a deep thinker. Unfortunately for you, we’re all you’ve got here, and we expect to fall somewhat short of “deep thinker” status.

Sentencing Grids are a matrix usually consisting of a vertical axis, offense severity (and possibly other characteristics of the current offense), and a horizontal axis, prior criminal record (the main offender characteristic of interest, the one that most studies find dominates the judicial sentencing decision regarding the characteristics of the individual defendant).

Sentencing grids are the common choice for a guideline format in the US, since Minnesota’s system was developed in the late 70’s. The most detailed and complex, I think it’s fair to say, is the U.S. federal sentencing guidelines (see training example here and here), a 256-cell matrix with many other factors to consider. Most other grid-based schemes are simpler (The District of Columbia has two grids – here and here - and a total of 60 cells - Why we chose a grid is another story.)

However, a sentencing grid is not the only scheme. Wisconsin (example here) and Ohio (here) have gone a different direction that still considers offense severity and prior criminal record, but has a more narrative-based approach. These seem like very thoughtful schemes to me. In many ways these systems require more of judges in terms of articulating their reasons and purposes, and this has always been an issue in grid systems (e.g., reasons for departure from the grid), and I assume it is an issue for non-grid systems. Delaware's guideline system is not really a grid, but is focused on a five-level continuum of sanctions that is also quite interesting to me, and seems to embrace the principles of graduated sanctions (“stepping up” and “stepping down”) that are prominent in drug courts, for example.

Virginia (where I was research director in the ‘90’s) is not a grid but a series of factors that are scored and the final score determines both the in/out decision and the sentence length (example here). This is a sophisticated approach that is highly quantitative (Again emphasizing the need for good data – See Part I of this series). Again, it mostly focuses on the same sorts of factors one sees in grids (mostly aspects of offense severity/harm and prior criminal record) and my own opinion is that it is sort of a multi-dimensional grid, somewhat more complex but still a variation on a grid (They might disagree with that characterization – but there you go).

In Part I, we already mentioned that there are some non-grid initiatives in other countries (here, here, here, etc.).

Grids are sometimes dismissed as “cookie cutter justice.” This characterization, in my opinion, is generally unfair (Take the bait - an opportunity of reader reaction!). Even a simple grid can address many of the most important purposes of sentencing – just deserts, culpability of the offender, harm to the victim, public safety through incapacitation or specific deterrence (Prior record - found on all grids – is the best predictor of future risk to public safety in any study with which I am familiar), etc. They may even be directed at other goals such as restoration of the victim and the community, although I’m not sure any grids have gone far down this path to date.

Finally, sentencing guidelines systems are complex and context specific. There is no “right way” but there are better and worse ways and local context, politics, considerations regarding judicial discretion, whether or not there will be appellate review, etc. etc. etc. Grids can make for bad policy, but that doesn’t mean too much as a guide to other jurisdictions, as far as I am concerned. The quality of the grid usually reflects the dedication of the actors that put it together, their ability to strike reasonable compromise, etc. Now getting a diverse group of citizens, legislators and public safety professionals to agree on a particular sentencing scheme – Whew, that’s the tough part!

National Institute of Corrections Blog

We've been remiss in failing to alert you, if you didn't know already, to the excellent work being done on the weblog by the National Institute of Corrections. We're pretty focused on the relationship between corrections and sentencing here, but you get that and more at their site, so go give them a look when you get a chance.

So You Want to Start a Sentencing Commission? Part VIII

Part VIII of a series. The previous parts dealt with how to create a good commission. These next two parts discuss the structure of a good commission.

Commission Organization and Structure

At the top of an organization chart for any commission is the constitutional body in which it is housed, the legislature, executive, or courts. Even “independent” commissions have to be placed somewhere in the executive budget and are ultimately beholden to their funders. Individual commissioners frequently overlook this in their agendas, but commissions as a whole do so at their own risk. That said, being a composite body with all three branches represented, a commission must also assert its independence, as noted earlier, to be of value even to those funders. A frequently delicate line, but an unavoidable one.

More concretely, the commission chair is clearly at the top of the commission’s internal chart. Chairs are rarely given extensive power in commission enabling legislation, although some take it by Louis XIV precedent. In Maryland, they were legally given exclusive power over staff; in Wisconsin, they were not. Depending on how frequently a commission meets (or, more specifically, requires action), the chair may serve as an interim decision-maker on matters that can’t wait until the next meeting. This can sometimes, of course, be contentious, but few organizations have ever found a better way around the problem.

In light of this paucity of thought to organizational matters in enabling legislation, some commissions adopt by-laws to outline things such as vice-chairs, permanent (standing) and ad hoc committees, the authority to call special meetings, disciplining absent members, etc. Other commissions like the ambiguity. By-laws help staff with their formal delineation of procedure, but staff sometimes like the ambiguity as well. By-laws probably aren’t as important in a commission’s early days when the members are finding workable patterns for action through trial and error fitting their situations and contexts. Once these informal patterns are established, those commissions may feel that formalizing them will impede rather than enhance performance. The problem comes when, perhaps through a change of administration and some membership, the commission undergoes substantial transformation. Established by-laws, based on effective practice, will prevent these “new” commissions from having to reinvent several wheels.

The number of meetings that the commission should hold annually depends on several factors. In its beginning, commissioners may want to meet more regularly, perhaps biweekly or monthly. This can raise havoc with commissioners’ schedules, of course, but it allows the commission a chance to focus, establish routines and priorities, and direct staff as to immediate wants and needs. After a few months, as staff begins assigned projects and committee work has started (and commissioners sicken of each other), the number of meetings will likely be cut back. Some states require a certain number of meetings per year (four in Maryland, including an annual meeting for direct public input, for example) while others leave it to the commissioners. No one to our knowledge has gone to jail for failure to have the mandated number of meetings, however, and it is not clear what the sentencing guidelines for that offense would be.

Commissions are strongly advised to establish permanent standing committees and to get them to work soon. If run properly, committees take the breadth of functions assigned to the commission and parcel them out for intensive consideration and later recommendation for action by the full commission. If the commission refuses to accept committee recommendations, however, frequently because of “my way” commissioners not on the committee, then much time and effort will be wasted recapitulating in the full body what the committee did. I strongly recommend usual deference to the committee work except in extraordinary circumstances that I really can’t think of right now. What the committees should be, of course, may vary from commission to commission, but, since many functions are the same (oversight of guidelines, collection and reporting of sentencing data, dealing with the public, media, and policymakers), many of the committees are similar. Here are a few you will likely need to consider (perhaps with different names).

Sentencing Guidelines
If you have a structured sentencing system with sentencing matrices or grids, offenses need to be classified into similar types (usually by nature of offense—violent, property, drug, sex, traffic, whatever--and by length and/or amount of statutory penalty), offenders need to be categorized (first-time, habitual, legislated penalty enhancers, etc.), and other necessary data and information need to be determined (victim participation in trial, type of counsel and disposition, demographics, reasons for guideline departures, and so on). If you are starting from scratch, this can mean complete examination and classification of your state’s criminal code, formatting your statistical data system, and constant revision as feedback and new legislated offenses and penalties each year become known. Commissions can bog down forever in minutia over these issues. It’s better to have a standing committee to get the grunt work done first.

It will be tempting to assign this work to legal practitioners (judge, prosecutor, defender), but you should consider a few others to avoid the “forest-tree” problem. And, the members should come from different parts of the state, if possible, to avoid domination of views from practitioners in one part, especially if that “part” conspicuously sentences differently. Further, the committee should be held to tough, tight timetables, or the arcane subjects could force even the smaller group into paralysis by analysis and seduction by one’s own voice. The conversations in these committees can quickly turn into “I had a case once . . ." that will take hours to be finished. A whole afternoon wasted. (The same thing can happen when the full commission debates the committee’s recommendations, so the chair will have to ride herd closely.) Keep in mind, these committees do yeoman’s work for commissions, especially at the start, and should be carefully constructed to get maximum work done in the time available.

Other committees and considerations coming next.

The Conversation Starts

As you may remember (or be tired of), we've been asking for comments and contributions to our conversation here and, thanks to Tom (last name too famous), we've begun to put together a thread. We won't always do this, but here's where we are in it now so you might be tempted to join in. Or at least to send us something yourself. We'll promise to respect your privacy as requested, just like Tom (last name too famous).

Tom--Thanks for your excellent Blog. I would appreciate your thoughts aboutwhether the grid approach to guidelines is preferable to some other method.I am familiar with the use of grids, but what are the others? Do they havesome theoretical underpinning? It seems to me that the grid approach mixesapples and oranges.

Me--Tom--Thanks for the kind words. OH is the usual example of a state without grids, but it just had parts struck down. They had written rules that judges would have to follow or explain why, such as "all first-time offenders must receive the lowest prison sentence in the legislated determinate sentence" (very rough paraphrase). They got into trouble for Blakely reasons, as I recall, and I believe you'll find similar approaches in CA, Alaska (never can remember the abreviation), AZ, probably others. They are interesting and worth considering. I've gotten less enamored of grids as I've dealt with commissions and after being a state-certified court mediator. I really don't have anything better. I would like to see all states, guidelines or not, put more effort into determining what sentences work with whom when and why. I think commissions have only gone part way in their original hopes, and I'd like for the conversations here to maybe work us more in that direction. For now,I'm interested in why you see apples and oranges in the grids. Care to elaborate? Thanks again for writing in.

Mike Connelly

Thanks for your question - re. whether the grid approach to guidelines is preferable to some other method. That is a rich subject and I suspect Mike and/or I may want to develop a longer post on this question, but I thought I'd mention a few things off the top of my head.

Grids, the common choice in the US since Minnesota's system was developed in the late 70's, is not the only system. Wisconsin (Mike's former address) and Ohio (Dave Driscoll is the director there) have gone a different direction that still considers offense severity and prior criminal record, but has a more narrative-based approach (Mike will correct me if I have misstated this). In many ways these systems require more of judges in terms of articulating their reasons, and this has always been an issue in grid systems as well as non-grid systems (e.g., reasons for departure from the grid). Virginia (where I was research director and worked for Rick Kern, executive director) is not a grid but a series of factors that are scored and the final score determines both the in/out decision and the sentence length. Again, it mostly focuses on the same sorts of factors (mostly aspects of offense severity and prior criminal record) and my own opinion is that it is sort of a multi-dimensional grid, somewhat more complex but still a variation on a grid (Rick might disagree with that characterization - but there you go).
Grids are sometimes dismissed as "cookie cutter justice." This characterization, in my opinion, is generally unfair but unfortunately may be apt for the US sentencing guidelines which are so complex and restrictive for judges. But even a simple grid can address many of the most important purposes of sentencing - just deserts, culpability of the offender, harm to the victim, public safety through incapacitation or specific deterrence (Prior record - found on all grids - being the best predictor of future risk to public safety in any study I know), etc. They may even be directed at other goals such as restoration of the victim and the community, although I'm not sure any grids have been so directed to date.
There are many good reviews of comparative systems. I will just mention a few. My DC Commission's report in 2002, Chapter 1 (see link below) includes a review of US (special focus on 4 states, three with grids) and even discusses briefly on page 5 other countries w/ non-grid systems. We didn't say a lot about our decision to opt for a grid (Frankly it had more to do with finding a system with which I could achieve consensus among my members - Whew, was that tough!). Other resources are Richard Frase's article in Columbia Law Review, v. 105:4 (May 2005) and Tonry and Frase (2001) Sentencing and Sanctions in Western Countries.
Finally, Mike and I wrote an article in the Federal Sentencing Reporter in April 2005 on Advisory guidelines v. presumptive guidelines, but one of our general points applies to the subject of grid or no grid. That is, sentencing guidelines systems are complex and context specific. There is no "right way" but there are better and worse ways and local context, politics, degree of judicial discretion granted, whether or not there will be appellate review, etc. etc. etc. Grids can make for bad policy, but just because you can find a bad grid or two doesn't mean too much, as far as I am concerned.
I hope these thoughts are useful - as I say we should probably work this into a post.

Take care,

Tom--Thanks to both of you for your responses. Now I have some homework to do. I certainly subscribe to you¹re your interest in ³determining what sentenceswork with whom, when and why.²Mike, you asked that I elaborate on ³why I see apples and oranges in grids.²I think a sentencing system should unpack the several provocations uponwhich sentences are based and then work out responses to each of those provocations. This is my understanding of how people naturally solve problems, but grid systems work in just the opposite way, making them seem counter-intuitive. Grids are used to pack all of the State¹s correctionalobjectives, strategies and tactics into a single unitary all-in-one response, thereby mixing apples and oranges..Please let me elaborate just a little. Enforcing the State¹s warning not to commit a crime and holding people accountable for committing a criminal offense are determinate objectives. They do not change once undertaken.Controlling a person¹s risk of committing another crime and reducing thatr isk are indeterminate objectives. Risk often changes over time. It follows that a workable sentencing system must be both determinate and indeterminate. This calls for tandem sentencing. Sentences that bundle all of these objectives into a single unitary response are unreasonable on their face and unworkable it seems to me.I believe that sentencing systems should emulate the way decision-makers naturally solve problems and that calls for help from the cognitive sciences. Grid based unitary guidelines are synthetic substitutes for what should be natural guidelines for each provocation response.

Me--Tom--You may be interested in the NIJ study that we started while I was in WI and should be completed this next summer. The WI guidelines worksheets ask for judges to indicate the factors they consider most important at sentencing of 11 top felonies. We did some research on them and posted it in "snapshots" on the commission website: You might look at those if you're interested (under "publications," I think) and let me know if you have any questions. Would you have any problem with me calling you "Tom" and posting our exchange (including Kim's)? It might encourage others to get in touch. I'd understand, of course, if you preferred not. In any case, it's good to hear from you. Please keep us in your loop. Thanks again.


See how easy this is???

Thanks to Grits

We need to think Grits for Breakfast for their notice of our blog the other day. They have one of the toughest jobs in blogdom, following TX corrections and sentencing, and they always have interesting stuff up. We hope any of their readers who are visiting here come to think the same of us. Thanks again.

So You Want to Structure Sentencing? Part I

The previous seven part series discussed ingredients of a good commission. Structuring the sentencing decision, a separate consideration, is outlined here.

Mike’s previous series “So You Want to Start a Sentencing Commission?” emphasizes the many functions of sentencing commissions in addition to, or even “instead of,” the construction and maintenance of sentencing guidelines. Nonetheless, judging from the “flood” of comments from our reader(s), we need to spend a little more time discussing the decision to structure sentences.

“Structured Sentencing”
What is commonly called Structured Sentencing (sentencing the world over has some structure, right?) is a relatively new development, and still not the norm in the 50 US states and the District of Columbia. The indeterminate sentencing model -- broad authorized sentencing ranges with great judicial discretion, substantial parole discretion, and case-by-case decision making – was nearly universal in the US in 1975 and is still common. Judge Marvin Frankel in Criminal Sentencing: Law Without Order argued for a system of sentencing based on the rule of law, including some form of “detailed chart or calculus” to reduce or channel judicial discretion in individual cases. So what we will call Structured Sentencing seeks to channel judicial discretion in individual cases by providing a recommended disposition for each offense (e.g. prison, probation, or intermediate sanction) and a proposed sentence length or range, at least where a prison term is the recommended sentence.

Not all structured sentencing systems are sentencing guidelines (usually thought of as judicial limits fashioned with input from a broadly-representative sentencing commission). Some state legislatures refined statutory limits on sentences rather than adopting sentencing guidelines. In these jurisdictions, legislatures narrowed the sentencing range within which a judge can sentence an offender by 1) setting a new maximum sentence, 2) setting a new sentencing range, including both a minimum sentence and a maximum sentence, or 3) setting a specific presumptive sentence with departure principles. These statutory limits serve to narrow judicial discretion in individual cases, and are mandatory for judges to follow. As recently as 2002, at least five states had this form of structured sentencing without instituting guidelines: Arizona, California, Illinois, Maine, and New Mexico. (Anybody want to help update this? Are you out there?) However, these systems run the risk of being rather clumsy, and not channeling discretion in important ways.

“Sentencing Guidelines”
But sentencing guidelines are the most common form of structured sentencing in the US. Many states and the federal system use presumptive or mandatory guidelines, which limit or take away the judicial discretion to impose a sentence outside of the guidelines. Other states and the District of Columbia opted for voluntary or advisory guidelines, which guide the judicial selection of a sentence, but do not require the judge to sentence within the recommended range. Guideline systems may also be descriptive (and deliberately mirror past sentencing practices) or prescriptive (and rely more on normative principles to set sentence ranges without rigid adherence to historical practice). In a 2002 report, the DC Sentencing Commission identified twenty-two states and the federal government that have a sentencing guidelines system in effect (18 states) or under study (4 states): Alabama, Alaska, Arkansas, Delaware, Iowa, Kansas, Louisiana, Maryland, Massachusetts, Michigan, Minnesota, Missouri, Nevada, North Carolina, Ohio, Oklahoma, Oregon, Pennsylvania, South Carolina, Utah, Virginia, and Washington. (Again, a little help with an update, please) Similarly, in a fine 2005 review of state guideline systems, Prof. Richard Frase identified 21 past and present guideline systems. See M. Tonry and R. Frase, Sentencing and Sanctions in Western Countries, for a more global perspective.

“Where do I start?”
Whether descriptive or prescriptive, advisory or presumptive, most structured sentencing systems are well advised to acquire (by whatever means necessary - usually involves begging) data on current practice. This falls under the category “You can’t know where you’re going, if you don’t know where you’ve been.” There is no substitute for accurate or even semi-accurate data on prevailing sentencing practice with respect to the in-out decision (Is a prison sentence the norm for similar cases?) and the sentence length decision (If prison, for how long?). Probably nothing will jeopardize a fledgling guideline system faster than na├»ve sentence recommendations that unconsciously and radically alter the mix of probationers and prisoners while ignoring prevailing practice (“the going rate”) – I am not talking here about prescriptive guidelines that attempt to consciously and with near-surgical precision alter prevailing practice for selected crimes, such as armed crimes of violence.

Some folks have tried to make-do with hypothetical exercises in place of actual data on past practice, in place of actual data. Bad idea! D. Kahneman and A. Tversky became famous (Nobel-prize famous) for their work on decision making under uncertainty, among other things. They found that even experts make intuitive predictions and judgments with severely flawed strategies if they don’t base their strategies on existing data and probabilities. For example, people tend to ignore the typical case, and base their judgments on their recall of the atypical case, which stand out in their mind precisely because it is out-of-the-ordinary(and thus a poor general guide)….

In addition to a baseline providing by good data on prevailing practice, the purposes of sentencing must be discussed, and a mission statement developed. I will not revisit this issue, so ably covered by colleague in “So You Want to Start a Sentencing Commission? Part V.”

Next time: To Grid or Not to Grid.

Wednesday, August 23, 2006

So You Want to Start a Sentencing Commission? Part VII

Part VII of a series. Earlier parts discussed ingredients of a good commission. This part finishes those considerations.

Other Requirements for a Good Commission
A commission performing to its max will also have consistent, influential, and serious sponsors among the policymakers, adequate funding for its mission, and meaningful cooperation with the other agencies involved in sentencing (the usual suspects—courts, corrections, prosecutors, etc.—but also juvenile justice, schools, treatment providers, etc., as necessary). Often, a commission starts as an initiative of a policy entrepreneur, either a judge fed up with disparity or nonuniformity, a legislator pushing for stricter sentencing or less racial disparity, or someone in the governor’s office staking out a policy turf. If these folks end up with the clout to get a commission going, they are clearly the kind of sponsors who can support and defend a commission politically and fiscally. To the extent that they are serious about making good policy as much as or more than making their names, they can help the commission establish its legitimacy and credibility in the process.

Commissions, however, cannot let themselves be too dependent on these key benefactors, or they may find themselves without help when the benefactor moves on, finds another interest, retires, visits the Great Beyond, etc. But lack of an influential sponsor will almost always commit a commission to influence purgatory in the policy arena, at least until some sentencing-related news story hits the front page. If the legislators appointed to a commission seem pro forma appointees or lesser lights or if the governor doesn’t put upper echelon folks on it and require their actual attendance (instead of flunk . . ., er, proxies), you can tell from the beginning that the commission is more for show than go. Politically influential judges taking an active role is probably the best sign, if you have such creatures in your state, since they can pull an often skeptical judiciary along as well.

Adequate funding is the mantra of all government entities, good or bad, but it is particularly important for commissions. Most fall in that netherland of “under $1,000,000” which makes them seem insignificant and potentially irrelevant but nevertheless an easy chunk of change if budget cuts are necessary. And, if created at a time of fiscal shortage, the temptation to underfund from the beginning is usually overwhelming.

There are ways for commissions to economize, some of which we will elaborate more on later. However, for those who can’t wait, web sites can offset printing and mailing costs, making documents, forms, reports, etc., available online for user downloading (at their own expense). Partnerships and cost-sharing with other agencies can also defray costs. Hiring talented but low-cost grad students as part-time analysts can get two people for the price of one FTE (or at least three for the price of two), and getting undergrad interns for grunt work is even cheaper, especially if you can tap into programs in which courses and credit are the students’ reward in lieu of actual pay. (Again, be sure to offer report credit for their resumes as compensation and as a moral salve for their exploi . . . , er, use.)

The problem, again, as mentioned above, is that, to the extent you use means such as these successfully to defray costs in the short-run as you accomplished the top of your triaged goals, funders tend to believe that you can accomplish ALL goals into the extended future with the SAME level of funding. So don’t be shy about pointing out what ISN’T getting done or the difficulties of promising the same high levels of production in the future (if grad students and interns come to believe they’re being misused, for example—which is why you don’t misuse them in the first place). A staff of six should run between $400,000-$500,000, depending on the market, prorated for more or fewer authorized employees but recognizing that less staff will be more expensive because you’re hiring your top salaries nevertheless.

Finally, establishing productive relationships with the other players in the process is vital. Since you’ll likely have most or all of them represented on the commission anyway, those commissioners will be your initial and key links. They should always be important entry points to their constituents, although you’ll undoubtedly deal with their associates as the commission does its normal work. Your staff will need to establish personal contact with the agencies/departments as well as with the necessary professional organizations, such as the state prosecutors’ association, victims’ groups, correctional organizations, etc. Perhaps they can get staff on the agenda of workshops, conferences, training sessions, etc. It is important that the commission have faces and dispositions recognizable (preferably in a good way) to practitioners on whom it will depend for ideas and support.

Commissions may need important statewide data to supplement their own from the courts and/or DOC. Prosecutors, law enforcement, and victims’ groups may also supply useful stats and info. The state justice grant administrative office, while usually not on or linked to the commission, is an important possible data source as well as potential grant provider. So is your state Statistical Analysis Center, a criminal justice data repository and disseminator funded by the U.S. Bureau of Justice Statistics. As a source of grads and undergrads, universities are potentially important partners, as well as having faculty possibly interested in research using the commission’s data.

It may be that, like families, dysfunctional sentencing commissions are dysfunctional in their own special ways, but good ones all seem alike. It borders on trite to say that good commissions have involved and respected commissioners balancing the desires of their constituents with the needs for commission consensus on the public interest; experienced, politically skilled but nonpartisan chairs and talented executive directors and staff; political support; adequate funding; and cooperative (if not necessarily happy) relationships with others in the sandbox (see above). Recognize, though, that being good doesn’t guarantee long-term survival in the face of all the silliness that hits public policy. But that’s what they look like. If you can’t provide all or most of these from the start, then you should think twice about creating a commission. Again, no commission is better than a bad one (please interpret that sentence correctly).

In part VIII we'll turn to the basic commission structure.