Title: Reforming Sentencing and Corrections for Just Punishment and
Public Safety.
Series: Sentencing and Corrections: Issues for the 21st Century (Research
In Brief)
Author: Michael E. Smith and Walter J. Dickey
Published: National Institute of Justice, September 1999
Subject(s): Alternatives to incarceration, community-based corrections,
sentencing, criminal justice system, and restorative justice
27 pages
63,000 bytes

-------------------------------

Figures, charts, forms, and tables are not included in this ASCII plain-text
file. To view this document in its entirety, download the Adobe Acrobat
graphic file available from this Web site or order a print copy from NCJRS
at 800-851-3420 (877-712-9279 for TTY users).

-------------------------------

U.S. Department of Justice
Office of Justice Programs
National Institute of Justice

Sentencing and Corrections: Issues for the 21st Century

September 1999
Papers From the Executive Sessions on Sentencing and Corrections 
No. 4

Reforming Sentencing and Corrections for Just Punishment and Public
Safety

by Michael E. Smith and Walter J. Dickey 

Sentencing courts and corrections agencies are not communicating about
what matters. When imposing sentence, a judge rarely states clearly the
purpose of the sentence or the process by which corrections is expected to
achieve it. When correctional agencies are left guessing, they revert to
routine administration of the generic penal measures (prison, probation,
and parole) and let offenders under supervision in the community decide
who will earn revocation.

If courts and correction are to work in harmony (which our collective
interest in justice and public safety requires they do), more than
incremental investments in generic penal measures are needed. Major
restructuring is called for--restructuring of penal and corrections law, and
restructuring of correctional strategies and penal measures.[1]

Imaginative sentencing judges and innovative community corrections
professionals have the practical knowledge necessary to begin that
restructuring. The transformation will not be easy, as it requires a fresh
look at what public safety is and how correctional agencies can contribute
to it. It also requires disciplined fact finding and reasoning by sentencing
courts--an application of the rule of law familiar in every area of the law
but this one.[2]

The nature of public safety revisited

In the press and in political discourse, "public safety" usually means more
arrests, more illicit drugs seized, more sentences to incarceration, and
fewer reported crimes. These definitions have achieved currency because
publicly accountable officials do know how to arrest and imprison
offenders, have built a substantial capacity to do so, and do know how to
count crime complaints. 

But experience has taught us that rising numbers of arrests and prisoners
are often indicators of the absence of public safety. Similarly, a falling
crime rate can be cause for alarm when it means only that citizens despair
of reporting crimes or when fear of crime so restricts their activities that
they are afraid to leave home. It makes no sense to find public safety
where the crime rate in an area has fallen to zero, but where teenagers
roam the parks in a vain search for robbery victims who are home behind
double-locked doors. In other words, threats to public safety cannot be
measured by numbers of arrests and prisoners or by aggregate crime data.
They are local in nature, arising in specific places--in that park, on this
street corner, in this house--and they often exist at certain times and not
others. 

These features make aggregate crime and justice statistics virtually useless
as measures of public safety and introduce a level of complexity to which
publicly accountable government officials have an understandable
aversion. They can with some confidence promise more arrests, prison
cells, and prison time for convicted offenders. They cannot with the same
confidence claim to understand what makes a certain street corner
dangerous or, more to the point, what might be done to make it safe. The
agencies of government know relatively little about how to make
neighborhoods, parks, bus shelters, street corners, and bedrooms safe. Nor
is there a budgetary or bureaucratic reward for doing so, in part because
such things are not measured. For those who administer conventional
correctional agencies and for almost everyone else concerned (except
those at risk of crime victimization), defining the public safety problem as
"too many offenders, not enough cells" works better. 

What public safety is not

Public safety must be more than an increase in the number of imprisoned
offenders. It cannot be the same as a lower crime rate--an expression of
aggregate data reflecting the volume of complaints about a host of
different crimes committed in divergent communities where the facts and
circumstances affecting public safety change and move in different
directions all the time. 

But if public safety were defined, as it should be, by the degree to which
people and property are free from the threat of harm in particular places
and at particular times, publicly accountable officials would face the
daunting prospect of creating the conditions of safety in the many places
and at the many times in which they do not exist. We suggest that public
safety is defined as it is today because those officials as well as the body
politic lack confidence in government's capacity to produce the real thing. 

What public safety is 

It turns out to be too easy to say what public safety is not. What is needed
is a rudimentary understanding of what it is: a condition, specific to
places, in which people and property are not at risk of attack or theft and
are not perceived to be at risk. Such places would likely share the
following characteristics: 

o A set of generally agreed-upon rules of behavior. 

o A shared appreciation that rule-breaking will be punished. 

o A further appreciation that playing by the rules will be rewarded. 

Viewed this way, creating and maintaining public safety requires teaching
the lessons of responsibility and accountability and reinforcing them in
raising children, supervising adolescents, and producing law-abiding
young adults. These are tasks for parents, neighbors, schools, churches,
athletic teams, community service groups, the labor market, and--on what
needs to be relatively rare occasions--a local police, probation, or parole
officer. This is not work that can safely be left to sentencing judges and
correctional agencies. It will require imposing penal measures on
convicted offenders but it cannot be achieved by that means.

The community policing example

This conception of public safety has animated community policing.
Instead of riding around in a patrol car waiting for the chance to arrest a
bad guy, a police officer does more to create safety by finding ways to
bring other adults into a neighborhood's public spaces, particularly at
times when no officer is there. If a conventionally deployed police officer
is the only active representative of adult authority on the street, there is no
adult authority most of the time. Police officers help to displace citizens'
fear of crime with actions that prevent it, helping them organize to restore
and maintain a sense of mutual obligation in their neighborhoods and
seeking their help in devising the most effective patrol deployment plans.

From this concept of government's role in generating public safety flow
powerful analogies for the roles of probation and parole agents. They
could be active in neighborhoods where public safety is threatened--
locating, invoking, and supporting informal mechanisms of social
control--to contain the risks posed by people who are under correctional
supervision (and by those who are not). This seems ambitious only
because we have grown accustomed to passive supervision--waiting to see
which offenders will violate probation or parole--which is nicely
analogous to random motor patrol by the police.

For correctional agencies to advance public safety in carrying out
sentences, and for judges to take proper account of public safety when
imposing them, restructuring the sentencing process is required, along 
with revising the deployment of corrections resources. In the first section
below, we propose a restructuring of sentencing based on rule-of-law
principles. We demonstrate that these principles require a judge to base a
sentence on findings of fact about the specific risks an individual offender
poses to public safety, the particular places where that offender is likely to
be found, and the relative plausibility of penal measures available to
reduce those risks. In subsequent sections we show how, if courts are to
impose sentences in accord with rule-of-law principles, the management of
correctional agencies--and especially community corrections--must also
change.

Sentencing under the rule of law

The plausibility of any penal measure (curfew, jail, or drug treatment, for
example) as a means to advance a penal strategy (incapacitation,
deterrence, or rehabilitation, for example) varies with an individual
offender's characteristics and circumstances. It at first appears difficult to
accommodate this fact, and the individualized sentences it seems to
require, within the current normative environment, which values equality,
proportionality, and accountability. Each of these values ought to inform,
but inevitably complicates, a court's case-by-case assessment of the
plausibility of particular penal measures.

If the value of equality requires that "like cases be treated alike," and if, as
is commonly assumed, gravity of offense and prior record are the only
facts relevant to the "likeness" of cases, then a proposal to tailor sentences
to any other characteristics of offenders and the circumstances of the
offenses would be an affront to equality. But prior record and gravity of
offense, although relevant at sentencing, are hardly sufficient to define a
category of offenders who are "alike" in culpability or in the threat they
pose to public safety. Similarly, "like treatment" ought not mean "the same
sentence." Norval Morris put it best:

The task [is] to reduce disparity in sentencing, not to achieve equality--and
by disparity in sentencing I mean the imposition of sentences randomly or
deliberately outside a range agreed, expressly or implicitly, to be justly
deserved and socially necessary punishments. . . . [As] desert is a limiting
principle of punishment, and not a defining principle, desert will allow for
the differential treatment of morally like cases.[3]

Adding to the complexity are tensions among plausibility of penal
purpose, equality of penal burden, and accountability in complying with
obligations imposed by the sentence. These tensions are most evident
when a court imposes a noncustodial sentence or when the conditions of
parole or other postconfinement supervision are set. They arise because:

o Different noncustodial penal measures serve different purposes. 

o Noncustodial penal measures serving the public safety purpose do so by
varied strategies. 

o Every sentencing condition increases the risk of nonperformance by the
offender. 

o Nonperformance of a sentence condition damages either the offenders (if
they are punished for violating the condition) or accountability (if they are
not punished), and in either case poses a risk to public safety (if the
condition was reasonably required for public safety). 

To help guide the courts in selecting penal purpose, penal strategies, and
penal measures that fit the facts as they vary from case to case, three
principles can be advanced: specificity, clarity, and parsimony.[4]

Specificity. Is the purpose public safety or just deserts? Specifying the
purpose of each sentence element is essential to its effective
implementation. For example, a 7 p.m. curfew may be a condition of
probation or parole. The enforcement of this condition and the
consequences of violating it ought to be shaped by the answer to the
question, "Was the curfew imposed because the restriction on the
offender's liberty is deserved punishment or because his unsupervised
liberty after 7 p.m. puts citizens or their property at risk?" 

Even more important is specificity about penal strategy. An example is the
sentencing of persistent petty thieves. In a court where a prison sentence of
a year, for example, is routine for such cases, the year will be imposed in
one case by a judge who believes it will dissuade other thieves, in a second
case by a judge who hopes it will dissuade this particular thief, in a third
case by a judge who is concerned only with punishing theft because theft
is bad, and in a fourth case by a judge who is content to put a dent in
thievery by incapacitating for a year someone known to have committed
theft on a regular basis. It is unlikely that the choices of these judges are
equally plausible in each petty theft case in which the 1-year sentences are
imposed.

This 1-year custodial sentence probably cannot serve both penal purposes
and all penal strategies in every case. But we believe no noncustodial
sentence could do so. That is because of the many combinations of
characteristics and circumstances thieves bring to sentencing. For
example, a generic noncustodial sentence that aims to end a thief's
thievery by incapacitation, by deterrence, and by rehabilitation too would
become unjustly burdensome and excessively costly in every case in
which it was imposed. Ensuring the rehabilitation of homeless, addicted,
persistent petty thieves would strain the purse and patience of anyone
attempting it, as would ensuring their incapacitation in the community
until the risk they will commit more thefts has passed. If deserved
punishment is a  constraint--if it is immoral to impose more punishment
than an offender deserves--the accumulating penal burdens of this
sentence, though noncustodial, would have to end long before its
strategic objective were achieved. Thus, it would be wise to specify
deserved punishment as the purpose of sentencing petty thieves who
possess these characteristics, as that purpose can be achieved within the
constraint of just deserts.

Current law does not require resolving conflicts between just desert and
public safety as the proper or plausible purpose of a sentence for persistent
petty theft or, for that matter, for any other crime. Nor does it require that
a court assess the relative plausibility of the quite different penal strategies
by which theft sentences, for example, might advance the public's interest
in reducing thievery. But resolving the conflicts of purpose and selecting
plausible strategies are not possible unless sentencing courts apply the 
principle of specificity.

Clarity. Clarity about penal purpose and strategy might seem to matter
little when confinement to prison is the only penal measure about which
we are concerned. But clarity about purpose permits reasoned resolution 
of conflicts between just desert and public safety, and lack of clarity about
penal strategy aborts inquiry into the relative plausibility of each penal
measure available to the court. This is because courts must select from an
array or continuum of penal sanctions, each of which serves some but not
all purposes. For example: 

o Community service, a form of involuntary servitude, can serve
retributive and deterrent purposes but is not well designed to incapacitate
or rehabilitate.

o Compulsory drug treatment can have punitive weight and rehabilitative
value, depending on the offender and the regimen but, depending on the
regimen, might or might not have incapacitative value.

o Electronically monitored house arrest can be used for deterrence and
retribution but is most plausibly connected to the public safety purpose by
its partially incapacitating effects.

Absent clarity by the sentencing court about the penal purpose of a
sentence and its strategic relationship to public safety, correctional agents
who carry out the sentence must simply guess what to do. This makes it
less likely that cases seen as "alike" by the court will receive like
treatment--indeed, less likely that the penal strategy the court had in mind
will be pursued, and less likely that either purpose (public safety, deserved
punishment) will be achieved.

Parsimony. There is a major pitfall for those who construct, advocate, or
carry out noncustodial sentences. By promising to accomplish more than
would be required if the court's purposes were clear and if the court
specified the penal strategy by which it expects to advance public safety in
a particular case, they risk accomplishing nothing. The danger of this
happening is most evident when a defense attorney or sentence planner,
attempting to improve the odds that a judge will impose a noncustodial
sentence, recommends adding multiple conditions, justifying them by
reference to every permissible sentencing purpose and strategy without
regard to the plausibility of each element of the sentence or to the
offender's ability to comply. 

If the court placed on record its purpose and strategies, the reasons a
purpose urged by an advocate is not desirable or achievable, and the
reasons a proposed strategy or penal measure is implausible or excessive
in the particular case, there would be benefits beyond the case. Such
documentation is necessary because excess in selecting penal strategies
and in imposing sentence conditions sets offenders up for failure, whether 
or not public safety is advanced thereby. Conversely, parsimony in
selecting penal strategies and penal measures helps preserve accountability
while making it more likely that the court's penal purposes will be
achieved. 

What does the rule of law require at sentencing? 

It has been more than a quarter of a century since Judge Marvin Frankel's
indictment of unfettered sentencing discretion precipitated the continuing
search for ways to bring sentencing decisions within the rule of law.[5]
Almost as strong as Judge Frankel's passion for the rule of law was his
doubt that judges or legislators would resolve either the "questions of
justification and purpose" that surround imposition of penal measures or 
the questions about their efficacy that plague their execution. But the
movement toward presumptive sentencing guidelines that his book
inspired was not inevitable. Although he called for "a detailed chart or
calculus to be used by the sentencing judge in weighing the many
elements that go into the sentence," he also wrote of the need to apply to
sentencing decisions the factfinding and reasoning processes on which the
rule of law ordinarily relies when complexity and competing interests
require the exercise of discretion.

Judges and others tend to view the numbers arrayed in guideline grids as
end points of a decisionmaking process rather than as starting points. Is the
rule of law observed when two facts suffice to locate an offender in a grid
of prescribed sentences? We think not. The rule of law requires that a
court first determine what facts are relevant in light of the purpose for
which a penal sanction is being imposed and then, by inference from those
facts, reason to the type and amount of penal measure. 

Such application of the rule of law would specify how, at the trial court,
the purpose of the sentence in a particular case is to be determined, and
how the facts relevant to choosing the type and amount of sanction are to
be distinguished from facts that are irrelevant. Such application would
specify the process by which the relevant facts are to be found and
how strong the inferences drawn from them must be for a sentence based
on them to be lawful. The rule of law would set the standard of review
applicable when this factfinding or reasoning is alleged not to satisfy these
requirements. While promulgation and enforcement of rules--even the
rules embedded in guideline grids--is "law" of a kind, the "rule of law" is
something more.

What has been lost by using presumptive guidelines? Presumptive
guideline grids have sometimes altered sentencing practices and the
distribution of punishments, have reduced the appearance of sentencing
disparity (as conventionally defined), have in some jurisdictions made
prison population projections more reliable and (at least theoretically)
more sensitive to resource constraints, and have in a few jurisdictions
generated a body of useful appellate case law. Nevertheless, we think the
decisional rules embedded in most of them fall short of rule-of-law
requirements. 

Using presumptive guideline grids to structure sentencing exaggerates the
importance of current offense and prior record--facts made relevant, to the
near exclusion of others, by their placement along the two axes of the
grids. Rule-of-law sentencing would structure the factfinding by which a
sentencing court determines which circumstances  of the case are relevant,
and the reasoning by which it selects the penal strategy (e.g., deterrence,
incapacitation, rehabilitation) and the particular penal measures that most
plausibly advance public safety in light of those facts. The rule of law
requires that, for any sentence, the strategy linking penal purpose to the
penal measures imposed be at least plausible--maybe "more plausible than
the available alternatives." This requirement calls in turn for factfinding
and reasonable inferences about how well the penal strategy and penal
measures employed suit the facts of the particular case and offender.  

Perhaps the core deficiency of presumptive guidelines is that they
encourage judges to avoid this underlying complexity. In examining this
complexity, we use noncustodial penal measures for illustration, because
in most jurisdictions a sentence to prison is imposed in only about half of
all felony cases (and is, even in those cases, almost always followed by a
period of noncustodial supervision) and because any effort to bring
consistency and reason to the use of noncustodial measures immediately
brings the complexity to the surface. 

Noncustodial penal measures are of many types, can be combined in
almost infinite variation, and cannot be reduced to "duration" or another
single measure of penal burden. Extending the reach of guideline grid
systems to encompass such sentences has proved difficult and may have
reached a dead end. This is because noncustodial penal measures cannot
serve the purposes for which they are imposed unless they are tailored to
the varied nonlinear characteristics and circumstances of individual
offenders. The challenge is to determine how "tailoring" sanctions can be
governed by the rule of law and not left to the unfettered (and perhaps
biased) discretion of judges. If that challenge could be met, there would be
no reason not to rely on those principles and procedures in imposing
custodial sentences as well. 

What is gained by reasoning from penal purpose to sentence by inference
from facts and circumstances in each case? At the most general level, the
rule of law by which sentences should be imposed may be stated as
follows:

Decisions about the types and amount of penal measures to impose should
be reached, at a specified degree of confidence, through a publicly
explained and reviewable application of a legal standard, to facts found to
a specified degree of subjective certainty by fair inference from reliable
evidence--with the specifications, the legal standard, and the burden of
persuasion established by lawful authority other than the decisionmaker. 

This statement is admittedly abstract, but it is not unknown in criminal
court, where a guilty verdict is lawful only if the factfinder is confident
beyond a reasonable doubt, in light of all relevant evidence, that the facts
required by statutory definition of the crime have been proved. 

It would be paralyzing to require a court to find beyond a reasonable doubt
that the sentence imposed in a given case will advance its specified penal
purpose, but the rule of law requires this level of confidence only for
decisions about guilt. A burden of persuasion better suited to the
sentencing decision might be that the sentence imposed is "more likely to
advance the penal purpose than alternatives available to the court." But
rule-of-law principles are slighted if there is no burden of persuasion on
this question; that is, if it were lawful for a court to impose a sentence it
believes, on the basis of known evidence, is not likely to advance the penal
purpose. This is far too often the case today, most obviously when an
addicted offender is sentenced to prison for drug treatment even though
the court has good reason to doubt that a treatment slot will be available.

It would indeed be possible to ground sentencing decisions in the kind of
factual inquiry that criminal courts are designed to conduct. That could
happen if courts' sentencing power were exercised with case-by-case
attention to the plausibility of the sentence as a means to achieve just
punishment and public safety, and if care were taken to invoke only penal
measures that are actually available. This type of factual inquiry would
assign prosecution and defense a function for which they are (or should
be) well prepared. In addition, factfinding of this kind is especially
suitable to appellate or other review. On review, the sufficiency of
evidence on which a sentencing court makes its findings of fact, and the
strength of the inferences drawn from them that lead it to the sentence
imposed, can be tested against a prescribed legal standard.

How would rule-of-law sentencing work? 

To ease the task of illustrating how the rule of law could be applied to
sentencing, and why it would be worth applying, we assume a jurisdiction
in which public safety and just punishment are by statute the only
permissible purposes of sentences. On the basis of these assumptions,
rule-of-law sentencing might be structured this way:

1. When imposing sentence--

(a) the court shall specify the following: the primary purpose (just desert
or public safety), the facts relevant to its finding of desert, the penal
strategy or strategies by which it finds public safety would be advanced by
its sentence, the penal measures by which it expects the sentence to serve
that strategy, the facts upon which it finds the sentence more likely to
advance the purposes of desert and public safety than alternatives known
to it, and the facts justifying any exchange of desert for public safety;

(b) the court shall consider public safety the primary penal purpose,
although just desert should be an element of and a constraint upon every
sentence;

(c) public safety trumps desert when the court finds the magnitude of
deserved punishment would interfere with the most plausible penal
strategy for advancing public safety; but 

(d) just desert is the only permissible purpose when the court finds no
strategy plausible for advancing public safety through imposition of any
penal measures authorized for the offense. 

2. The court shall impose the sentence it finds more likely to advance the
specified penal purpose(s) than the alternatives. 

This procedure may appear burdensome if weighed against the
requirements of sentencing hearings in indeterminate sentencing systems
today, or even in the most exacting presumptive guideline systems.
However, the law in this country requires this sort of rigorous factfinding
as a matter of course in other fields, where often much less is at stake.
Moreover, although some facts will vary greatly from offender to offender,
many will not, thus limiting the factfinding burden. 

The procedure should be guided by a new style of sentencing commission,
one that reviews evidence bearing on commonly arising questions of fact,
further narrowing the original factfinding required in individual cases.
Such a commission could, for example, make findings about the
circumstances in which one penal strategy might generally be preferred
over another for particular types of cases or for a particular purpose (for
example, deterrence over incapacitation or rehabilitation over
incapacitation for public safety; prison over fine or community service
over prison for just punishment). 

Similarly, a new style of sentencing commission could guide sentencing
courts in making findings about the punishment a particular offender
deserves. Just desert is notoriously difficult to "find." It cannot properly be
specified by a legislature or commission that is unfamiliar with the
case-specific facts and circumstances bearing on an offender's culpability,
but we should be concerned that, without guidance, sentencing judges will
"find" just desert by personal and essentially unreviewable hunch. A
court's findings about the range of just punishment in a particular case
should be tied to norms and values widely shared in the body politic, not
to those of judges, legislators, or the media. The sentencing commission
envisioned here would undertake systematic inquiry of the general
population, to find the range of punishments widely thought to be
deserved in archetypal offense/offender scenarios and to discover how the
public's view is affected by knowing that particular penal measures tied to
the public safety purpose are available to the sentencing court. On the
basis of such inquiries, ranges of "not undeserved" punishments could be
provided to the courts as starting points from which to reason to sentences. 

What would rule-of-law sentencing orders look like?

The rule of law would require that sentencing orders link a limited set of
conditions im- posed on an offender to the lawful purpose or purposes
specified. Multiple-conditioned, all-purpose sentences undermine the
development of operational capacity in community correctional agencies.
Correctional agents flounder when they try to deliver everything to a court
that did not make clear what it wanted or how it expected the penal
measures imposed to produce the intended effect. Such sentences also
undermine accountability because violations of nonessential conditions
often either go unpunished or are punished without good reason. They
undermine cost effectiveness because staff who supervise an offender
burdened with such a sentence either must waste resources on
interventions and enforcement not tied to a penal purpose or not based on
a plausible strategy, or they must guess which, if any, conditions of the
sentence should be treated seriously.

Sentences to community service, drug or alcohol treatment, home
detention, and other noncustodial penal measures serve sentencing
purposes of desert or public safety (or, in some cases, both) through
plausible penal strategies. But this is true only if the sentence is imposed
in cases in which the purpose and the strategy are congruent with the penal
measures. For example, when community service sentences are imposed
on individuals whose liberty poses immediate unacceptable threats to
public safety, those sentences may indeed punish but will before long
come into disrepute for their failure to incapacitate. When home
confinement, curfews, and other surveillance techniques are imposed too
widely on offenders whose unsupervised behavior is not viewed as
particularly risky but whose offenses deserve to be punished, the
conditions are not likely to be rigorously enforced. Again, the
noncustodial penal measure is likely to fall into disrepute for its apparent
inability to accomplish the penal purpose that would justify its imposition.

A rule-of-law sentencing order would reflect the principle of parsimony--
parsimony in the number of penal strategies and measures. It would also
reflect consideration of public safety risks that require management while
the sentence is carried out. When a noncustodial penal measure is imposed
or a prison sentence reaches its postconfinement phase, the offender's
conditioned liberty may or may not put victim, family, or community at
risk of harm. An essential task for sentencing courts (and for correctional
staff responsible for implementing noncustodial sentences and the
noncustodial phases of prison sentences) is to specify the risks the
sentence is intended to contain. They should ask how serious the potential
harm is, how likely the offender is to cause harm in the absence of penal
intervention, and how perfectly the correctional intervention must contain
the harm.

The answers can be found only by inference properly drawn from the
individual characteristics and circumstances of an offender the
circumstances surrounding his or her conditioned liberty, the purposes of
the sentencing court, the penal strategy the court adopted, and the
correctional resources available to manage the risk. When public safety is
a dominant purpose, it is hard to imagine findings more important for the
sentencing court to make--or findings more useful to correctional agencies
charged with carrying out the sentences.

What would be required of corrections under rule-of-law sentencing? 

For sentencing to be governed by the rule of law in the way proposed
here, judges must be offered a richer array of penal measures--
noncustodial ones in particular--that are plausible for advancing public
safety, and judges must be confident that corrections will execute them
properly. For this to happen will require a shift in the mission of
correctional agencies and new approaches to deploying and managing
their resources. 

Enlarging the corrections mission 

Corrections has long focused on individuals--using authority and force in
the routine business of punishing and treating them. Thus, we are
unaccustomed to thinking about other ways public safety might be served
by the considerable and increasing resources of corrections departments.
But once public safety is understood as a condition of particular places at
particular times, rather than as an exercise of correctional authority over
some number of individuals, it is difficult to imagine an agency of the
executive branch achieving it by confinement or "correction" of sentenced
offenders or by "sending messages" to the not-yet convicted. 

Collaboration with naturally occurring guardians. What might be more
plausible ways to use the authority and resources of correctional agencies?
Routine activity analysis and crime pattern theory explain crime as the
result of a confluence of factors: a motivated offender, a vulnerable target,
and a place and time when guardians are absent or ineffective.
"Guardians" in this usage are people who have a protective relationship to 
a potential victim, people who are responsible for or are willing to assume
responsibility for a targeted place, and people who have an intimate or
supervising relationship with the potential offender. Such guardians are
abundant in safe places. But they can also be found in dangerous places.
The most effective police, probation, and parole officers know how to find
them and enlist their help.

This is the line of thought that has led to the more complex and more
intensely local engagements that characterize problem-oriented
community policing. For more than a decade, police have been
learning to collaborate with "naturally occurring" guardians to solve crime
and disorder problems. Probation and parole officers--"official guardians"-
-need also to find and invoke the authority of naturally occurring
guardians of the offenders under their supervision, of the people made
vulnerable by proximity to them, and of places where the resulting risks
arise. 

The theoretical framework for restructuring corrections, set out here,
emphasizes the power of naturally occurring forces in the community to
create and maintain public safety and invites corrections to form
relationships with them. The framework emerged from the authors' work
on the 1996 Wisconsin Governor's Task Force on Sentencing and
Corrections.[6] (See "The Need To Restructure Corrections--Evidence
From a Milwaukee Neighborhood.") 

Of course, correctional resources are for the most part tethered to
individual offenders by the sentences imposed, and correctional agencies
have no license to trade their statutory, offender-centered responsibilities
for place-based ones. But they can learn from the experiences of police
departments that have moved beyond reactive patrol, and beyond trying to
arrest more bad guys each year, toward partnership with community to
produce and maintain public safety. Correctional agencies can also learn
from scattered, low-visibility attempts to adapt some of the
problem-solving techniques of effective community police officers to
neighborhood supervision of offenders.[7]

Principles for managing corrections for public safety. From observation of
various police and corrections operations whose design reflects analysis of
the public safety threats arising in particular places, we draw a half dozen
principles to guide correctional agencies embracing a public safety
mission: 

o The nature and degree of supervision of an offender should be directly
related to the risk of harm he or she poses if unsupervised. In other words,
correctional measures should be tailored to the gravity of the potential
harm and to the likelihood of its occurring without those measures. Thus,
for example, corrections should not devote the same kind or quantity of
resources to a petty thief (even though the likelihood of another theft is
high) as devoted to the opportunistic burglar (whose chance of committing
another offense may be lower but whose potential for harm is greater). 

o The graver the potential harm and the more likely it is to occur if nothing
is done to reduce the risk, the more active must be the supervision when
the offender is not in prison. The nature of supervision should change as
the risk to public safety the offender presents changes over time.

o Alacrity, flexibility, and parsimony should characterize responses to
changes in the risks to public safety posed by offenders under supervision.
For corrections staff in prisons as well as in the community, both legal
authority and resources need to be configured to permit them to tailor
correctional measures to changes in offenders' circumstances. Proper
management of staff in exercising such discretionary power must also be
active and not rely on procedural formalism.

o "Active supervision" requires that offenders who pose a risk to public
safety not be allowed anonymity--not in the community and not in prison.
Anonymity enables them to conceal themselves from naturally occurring
agents of social control as well as from correctional agents. Active
supervision is the broad engagement of correctional staff with offenders in
whatever setting they are found, as well as with police; other members of
offenders' communities; and offenders' families, neighbors, employers,
friends, and enemies. This type of supervision necessitates reorganization
and a new management style for correctional agencies, enabling them to
replicate in large communities, where anonymity is more the norm, the
active supervision and neighborhood orientation more often found in small
communities. 

o Active supervision also requires creative program development in
correctional agencies. Here the aim should be for every offender who
poses a risk warranting supervision in the community to have a stable
housing situation; to be in the labor market; and to be bound to a
supportive network of family and neighbors, when possible, or to others
recruited to that role by correctional agents in the community. These
requirements represent major challenges to conventional practice because
most correctional agencies have not developed the capacity, for example,
to create jobs and job placement services where they are lacking. 

o Finally, while correctional agencies committed to the public safety
purpose of sentencing need to be more active in supervising offenders and
will have to seek more intensive engagement with the communities where
the offenders are found, it is important that these agencies not adopt
strategies and programs that conflict with the naturally occurring forces of
social control that can be found in every neighborhood, even the most
crime-ridden. Active supervision requires familiarity with the operation of
those forces in the lives of offenders and others who live and work where
correctional agents have public safety responsibilities.

Allocating corrections resources 

More than clarity is required. The nature of public safety also demands
that any correctional agency hoping to promote it will need to build new
capacity, from the bottom up, and will need to construct new systems to
manage that capacity and the flow of information it requires. That is
because information from the local level will be required for monitoring
and revising deployment plans at individual agent, regional, and State
levels--and this information should be expected to change rapidly.

The experience of police agencies that are grappling with the demands of
problem-oriented approaches makes it evident that victim vulnerability
and public safety assets and deficits vary greatly from place to place. This
necessitates planning and overseeing a State's responses to variations in its
corrections investments--variations more complex, changeable, and subtle
than "caseload" or "offender needs." The process ought to be grounded in
information, gathered systematically and periodically, about the particular
public safety threats and the naturally occurring community capacities to
contain them found, neighborhood by neighborhood, throughout a State. 

We know of no correctional agency that has such capacity today. But
rule-of-law sentencing would call for judges to be provided information of
this kind, and it is needed by corrections to support locally oriented
allocation of resources (for example, prison cells, treatment slots,
transitional housing, various types of community supervision officers, job
training, and drug treatment programs).

Probing for specification of public safety problems. In pursuing this issue
for the Wisconsin Governor's Task Force on Sentencing and Corrections,
we wanted to find out whether different communities would nominate
different public safety problems as warranting attention. It was clear to us,
and to the judges and other officials with whom we worked, that a
corrections department ought not conduct a public opinion survey of how
many prison cells are needed or what sentence should be imposed on a
particular offender about to come before the court. It was just as clear,
however, that without systematic inquiry, the State's policymaking
apparatus was in no position to specify the nature and patterns of public
safety threats in different parts of the State, and that without such inquiry
policymakers would remain ignorant about what, in the view of those
closest to the threats, might be done to reduce them.

In the time available to the task force, it was possible only to conduct a
survey to test the proposition that perceptions of public safety threats vary
from one location to another throughout the State. Variation was indeed
what the pollsters found. Survey respondents' fears were specific to the
places where they lived, the places where they worked, the places they
visited, and the places they avoided for safety's sake.

Categories of risks and "best practices" for managing them. To find out
what might be done to manage place-specific public safety risks in
Wisconsin, we first set out to categorize them in a way that would lend
itself to finding best practices for dealing with them. However, no
routinely gathered data were available to categorize the kinds and degrees
of risk to public safety posed by the 32,000 felony offenders under
probation and parole supervision in Wisconsin in 1996. Nor were such
data available for the 11,000 felony offenders then in prison or the 6,000
expected to be released to parole over the following 5 years. In the absence
of these data, we relied on focus group sessions conducted among
experienced probation and parole staff to define behavioral categories of
risk and to estimate the number of offenders in each category, in locations
throughout the State. From this information, a typology of public safety
risks was developed (see the exhibit). 

For each of the 24 categories of risk in the typology, the focus groups
sketched what would be required by "best practices" if public safety were
the correctional objective. Their work permitted the task force to estimate,
by location, how many prison cells and how much local confinement
capacity would be needed; how many community supervision agents
would be needed; and what drug treatment slots, jobs, and housing would
have to be made available.

The focus groups revealed differences by locality in the distribution of
offenders' criminal propensities and of the circumstances in which they
posed risks to public safety. They also revealed differences in community
assets and correctional assets locally available to contain those risks. The
result was a task force plan reflecting what best practices would require in
allocating State corrections resources for public safety.

We then developed a computer application in which information of the
kind provided by the focus groups was combined with data routinely
gathered on related topics. This was done so that the task force and
Wisconsin's Department of Corrections could see how a "bottom-up"
public safety strategy might be constructed and implemented and how
variations on that strategy would affect the demand for and deployment of
correctional resources.

Legal authority for restructuring corrections. Flexibility in corrections'
response to change in individual offenders' circumstances--necessary if
public safety is the purpose--is incompatible with the core idea of
determinate sentencing: that the specifics of a sentence should be
determined by the court at the time sentence is imposed. Of course, more
than half the States are still properly characterized as having indeterminate
sentencing systems, and the flexibility public safety requires can be
achieved in at least some determinate systems. In any event, sentences
with public safety purposes, whether determinate or indeterminate as to
duration, need to be indeterminate as to content.

It was with an eye to achieving the necessary flexibility that the Wisconsin
Governor's Task Force asked how State law would have to be amended to
support "best practice" in corrections management of public safety risks. It
recommended abolishing felony probation and creating a new form of
confinement, one merging features of prisons with varying degrees of
liberty and obligations of the offender in the community. This new penal
measure, "Community Confinement and Control" (CCC), was designed to
ensure substantially more supervision and control of behavior, and also
substantially more treatment, employment, and other socialization
programs than conventional probation and even "intensive" probation ever
had--but for shorter periods. And because the task force recognized that
parole supervision often approaches unsupervised liberty, it felt that no
felony offender sentenced to prison should be granted parole without
successfully completing some period in CCC status. It followed that
offenders sentenced to CCC would be paroled from it, just as those
sentenced to prison would be paroled from CCC after serving at least a
brief time in that status on their way out.

There was remarkably little objection, from any quarter, to the proposed
abolition of felony probation and to the use of CCC when purposeful
supervision in the community is required. The desire to do away with
felony probation stemmed from a nearly universal lack of confidence in it
and from a sense that offenders convicted of felonies in Wisconsin either
present, at the start of their sentence, risks too serious for conventional
probation or require very little supervision at all (because all that is
intended is that they fulfill a particular condition). For the latter class of
cases, the task force recommended a new sentence, "conditional
supervision," which would require almost no supervisory resources unless
the condition(s) were not met. 

Where does this all lead? 

Using the State of Wisconsin as an example, we have sketched, in some
detail, how public safety would be advanced by fundamental amendments
in the mission and methods of a correctional agency. We went into detail
to demonstrate that it is possible to specify the redeployment of authority
and resources required for such a change. If we are correct about the public
safety benefits of redeploying correctional resources this way, there is
little reason not to get on with it. Rule-of-law sentencing would surely
ease such a transformation but is not essential. On the other hand,
rule-of-law sentencing would be a sham if a robust array of penal
measures were not put at the disposal of the courts, or if sentencing judges
were not provided accurate information about what penal measures are
actually available and what effects they can reasonably be expected to
have. In practice, however, adoption of rule-of-law sentencing by the
courts would create irresistible demand for corrections to respond with
place-sensitive redeployment of resources. In the same way, guideline
grids would likely give way to rule-of-law sentencing if correctional
practice were reformed along the lines described here.

Notes

1. For a discussion of the current debate among community corrections
practitioners about their mission and methods, see Dickey, Walter J., and
Michael E. Smith, Dangerous Opportunity: Five Futures for Community
Corrections, Washington, DC: U.S. Department of Justice, Office of
Justice Programs, 1999; and Joan Petersilia, ed., Community Corrections:
Probation, Parole, and Intermediate Sanctions, New York: Oxford
University Press, 1998.

2. This paper is one of four in the first "round" of publications from the
Executive Sessions on Sentencing and Corrections. Together the four offer
a framework for understanding the issues raised in the sessions. The other
three papers are Fragmentation of Sentencing and Corrections in America,
by Michael Tonry (NCJ 175721); Reconsidering Indeterminate and
Structured Sentencing, by Michael Tonry (NCJ 175722); and
Incorporating Restorative and Community Justice Into American
Sentencing and Corrections, by Leena Kurki (NCJ 175723). All: Research
in Brief--Sentencing & Corrections: Issues for the 21st Century,
Washington, DC: U.S. Department of Justice, National Institute of
Justice/Corrections Program Office, September 1999.

3. Morris, Norval, Madness and the Criminal Law, Chicago: University of
Chicago Press, 1982: 196. 

4. Smith, Michael E., "Let Specificity, Clarity and Parsimony of Purpose
Be Our Guide," Law and Policy 20 (4) (October 1998).

5. Frankel, Marvin E., Criminal Sentences: Law Without Order, New
York: Hill & Wang, 1972.

6. Walter Dickey was the chair and Michael Smith the research director of
the Wisconsin Governor's Task Force on Sentencing and Corrections.

7. Smith, Michael E., and Walter J. Dickey, "What If Corrections Were
Serious About Public Safety?" Corrections Management Quarterly 2 (3)
(Summer 1998): 12-30.

-------------------------------

Directors' Message

It is by now a commonplace that the number of people under criminal
justice supervision in this country has reached a record high. As a result,
the sentencing policies driving that number, and the field of corrections,
where the consequences are felt, have acquired an unprecedented salience.
It is a salience defined more by issues of magnitude, complexity, and
expense than by any consensus about future directions.

Are sentencing policies, as implemented through correctional programs
and practices, achieving their intended purposes? As expressed in the
movement to eliminate indeterminate sentencing and limit judicial
discretion, on the one hand, and to radically restructure our retributive
system of justice, on the other, the purposes seem contradictory, rooted in
conflicting values. The lack of consensus on where sentencing and
corrections should be headed is thus no surprise. 

Because sentencing and corrections policies have such major
consequences--for the allocation of government resources and, more
fundamentally and profoundly, for the quality of justice in this country and
the safety of its citizens--the National Institute of Justice and the
Corrections Program Office (CPO) of the Office of Justice Programs felt it
opportune to explore them in depth. Through a series of Executive
Sessions on Sentencing and Corrections, begun in 1998 and continuing
through the year 2000, practitioners and scholars foremost in their field,
representing a broad cross-section of points of view, are being brought
together to find out if there is a better way to think about the purposes,
functions, and interdependence of sentencing and corrections policies. 

We are fortunate in having secured the assistance of Michael Tonry,
Sonosky Professor of Law and Public Policy at the University of
Minnesota Law School, as project director.

One product of the sessions is this series of papers, commissioned by NIJ
and the CPO as the basis for the discussions. Drawing on the research and
experience of the session participants, the papers are intended to distill
their judgments about the strengths and weaknesses of current practices
and about the most promising ideas for future developments. 

The sessions were modeled on the executive sessions on policing held in
the 1980s and 1990s under the sponsorship of NIJ and Harvard's Kennedy
School of Government. Those sessions played a role in conceptualizing
community policing and spreading it. Whether the current sessions and the
papers based on them will be instrumental in developing a new paradigm
for sentencing and corrections, or even whether they will generate
broad-based support for a particular model or strategy for change, remains
to be seen. It is our hope that in the current environment of openness to
new ideas, the session papers will provoke comment, promote further
discussion and, taken together, will constitute a basic resource document
on sentencing and corrections policy issues that will prove useful to State
and local policymakers.

Jeremy Travis
Director
National Institute of Justice
U.S. Department of Justice 

Larry Meachum
Director
Corrections Program Office
U.S. Department of Justice

-------------------------------

The Need To Restructure Corrections--Evidence From a Milwaukee
Neighborhood

A new framework for managing correctional resources emerged from the
Wisconsin Governor's Task Force on Sentencing and Corrections,
following hearings convened in 1996 in a Milwaukee neighborhood where
public safety was in serious disrepair.* Community representatives were
clear that crimes in their community should not go unpunished, although
they were also clear that sentences were undeservedly harsh in drug
possession cases, and they bemoaned the lack of attention paid, in prison
and upon release, to the obvious mental health and substance abuse
problems of members of their community sent there for correction. But
most of all, they did not believe that routine operation of the criminal
justice system would affect the public safety problems they felt most
acutely.

The problem at 9th and Concordia

A police officer detailed the conditions, as he knew them, on one street
corner. At 9th and Concordia, he reported, 94 drug arrests were made
within a 3-month period earlier in the year. These arrests, he pointed out,
were easy to prosecute to conviction. But despite the 2-year prison terms
routinely handed down by the sentencing judges, the drug market
continued to thrive at the intersection, posing risks to the safety of all who
lived nearby or had to pass through on their way to work or school.

That the removal to prison of almost 100 felons did not increase public
safety at that street corner revealed the disconnect between public safety
and corrections. It surfaced the need for further inquiry into the nature of
the problem at that location and into the deployment of corrections
resources devoted to it. Informal inquiries after the hearing revealed that
those arrested for drug offenses at the corner returned, typically after
serving about 12 months of the 2-year sentences, to "supervision" of the
most passive kind by parole agents to whom they were assigned randomly,
who did not coordinate their supervision plans, and who were apparently
ignorant of their respective clients' connections to the same location.

Deployment for public safety

To the task force, a correctional agency aiming to advance public safety at
9th and Concordia would have to know the ecology of the threats to safety
peculiar to that place and would have to deploy personnel and services in
response, rather than focusing on one returning offender at a time. The
task force noticed that if only one drug felon were arrested each week, and
if only 80 percent were convicted and drew the 2-year prison sentence
(with parole after 12 months), Wisconsin would have in effect allocated 40
prison beds and 8 correctional officers to the problem at that location. If,
instead, the Department of Corrections had hired and deployed eight
probation or parole officers to 9th and Concordia, it could have had one
there 24 hours a day, 7 days a week and, within the same budget
allocation, could have had another pair working on projects to advance
public safety in the area. 

In other words, the task force realized that corrections personnel could be
deployed to the place where public safety is at risk--to enforce sentence
conditions of unpaid labor (for example, with community groups who
were restoring abandoned houses nearby), to ensure active rather than
passive supervision of the offenders there, to block their slide back into the
anonymity that fosters crime, and to help them find and keep jobs.
Meanwhile, the $3,500 per inmate that Wisconsin spends annually on the
nutritional, health, and other basic needs of each offender in prison could
have been spent on making habitable the afflicted neighborhood from
which they came--and to which, if imprisoned, they are sure to return.

* Wisconsin Governor's Task Force on Sentencing and Corrections, Final
Report, Madison, WI: University of Wisconsin Law School, 1996;
reprinted in Overcrowded Times 7 (6) (1996): 5-17.

-------------------------------

Michael E. Smith is a professor of law at the University of Wisconsin,
Madison. Walter J. Dickey is Evjue-Bascom Professor of Law at the
University of Wisconsin, Madison. Both are members of the Executive
Sessions on Sentencing and Corrections. 

This study was supported by cooperative agreement 97-MUMU-K006
between the National Institute of Justice and the University of Minnesota.

Findings and conclusions of the research reported here are those of the
authors and do not necessarily reflect the official position or policies of the
U.S. Department of Justice. 

The National Institute of Justice is a component of the Office of Justice
Programs, which also includes the Bureau of Justice Assistance, the
Bureau of Justice Statistics, the Office of Juvenile Justice and Delinquency
Prevention, and the Office for Victims of Crime.

This and other NIJ publications can be found at and downloaded from the
NIJ Web site (http://www.ojp.usdoj.gov/nij).

NCJ 175724

-------------------------------

The Executive Sessions on Sentencing and Corrections 

Convened the following distinguished panel of leaders in the fields:

Ronald Angelone
Director
Department of Corrections
Commonwealth of Virginia

Neal Bryant
Senator
Oregon State Senate

Harold Clarke
Director
Department of Correctional Services
State of Nevada

Cheryl Crawford
Deputy Director, Program 
Development Division
National Institute of Justice
U.S. Department of Justice

Barbara Damchik-Dykes
Project Coordinator
Executive Sessions on Sentencing and Corrections

Walter Dickey
Evjue-Bascom Professor of Law
University of Wisconsin 

Ronald Earle
District Attorney
Austin, Texas

Tony Fabelo
Director
Texas Criminal Justice Policy Council

Richard S. Gebelein
Superior Court Judge
Wilmington, Delaware

John Gorczyk
Commissioner
Department of Corrections
State of Vermont

Kathleen Hawk Sawyer
Director
Federal Bureau of Prisons
U.S. Department of Justice

Sally T. Hillsman
Deputy Director
National Institute of Justice
U.S. Department of Justice

Martin Horn
Secretary
Department of Corrections
Commonwealth of Pennsylvania

Susan M. Hunter
Chief, Prisons Division
National Institute of Corrections
U.S. Department of Justice 

Michael Jacobson
Professor of Law and Police Science
John Jay College of Criminal Justice
City University of New York

Leena Kurki
Research Associate 
Law School
University of Minnesota
Project Associate
Executive Sessions on Sentencing and Corrections 

John Larivee
Chief Executive Officer
Crime and Justice Foundation

Joe Lehman
Secretary
Department of Corrections
State of Washington

Dennis Maloney
Director
Deschutes County (Oregon) 
Department of Community Justice

Larry Meachum
Director
Corrections Program Office
Office of Justice Programs
U.S. Department of Justice

Mark H. Moore
Guggenheim Professor of Criminal Justice Policy and Management
John F. Kennedy School of Government
Harvard University

Norval Morris
Emeritus Professor of Law and Criminology
University of Chicago

Joan Petersilia
Professor of Criminology, Law and Society
School of Social Ecology
University of California, Irvine

Kay Pranis
Restorative Justice Planner
Department of Corrections
State of Minnesota

Michael Quinlan
Former Director
Federal Bureau of Prisons 
U.S. Department of Justice 

Chase Riveland
Principal
Riveland Associates

Thomas W. Ross
Superior Court Judge, 18th Judicial District
Chair, North Carolina Sentencing and Policy Advisory Commission

Dora Schriro
Director
Department of Corrections
State of Missouri

Michael Smith
Professor of Law
University of Wisconsin 

Michael Sullivan
Secretary
Department of Corrections
State of Wisconsin

Morris Thigpen
Director
National Institute of Corrections
U.S. Department of Justice

Michael Tonry
Sonosky Professor of Law and Public Policy
University of Minnesota 
Project Director
Executive Sessions on Sentencing and Corrections

Jeremy Travis
Director
National Institute of Justice
U.S. Department of Justice

Reginald A. Wilkinson
Director
Department of Rehabilitation and Correction
State of Ohio