UNITED STATES COURT OF APPEALS
UNITED STATES COURT OF APPEALS
FOR THE FIRST CIRCUIT
FOR THE FIRST CIRCUIT
No. 94-1999
RALPH C. HAMM, III,
Petitioner, Appellant,
v.
ARTHUR LATESSA, SUPERINTENDENT OF MCI, ET AL.,
Respondents, Appellees.
No. 94-2018
RALPH C. HAMM, III,
Petitioner, Appellee,
v.
ARTHUR LATESSA, SUPERINTENDENT OF MCI, ET AL.,
Respondents, Appellants.
APPEALS FROM THE UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF MASSACHUSETTS
[Hon. Walter Jay Skinner, Senior U.S. District Judge]
Before
Selya, Cyr and Stahl,
Circuit Judges.
Daniel S. Tarlow, with whom John F. Tocci and Glovsky &
Associates were on brief, for petitioner.
William J. Meade, Assistant Attorney General, with whom
Scott Harshbarger, Attorney General, was on brief, for
respondents.
December 28, 1995
SELYA, Circuit Judge. Petitioner Ralph C. Hamm, III,
SELYA, Circuit Judge.
is currently serving two concurrent, parole-eligible life
sentences in a Massachusetts state penitentiary. He faces an
additional twenty-six to forty years in prison from and after the
culmination of his life sentences. Hamm solicits a writ of
habeas corpus, naming as respondents the superintendent of the
state correctional facility where he is confined, the
Commissioner of Correction, and the Parole Board (hereinafter
collectively the respondent or the Commonwealth), and contending
that a policy implemented by the Commonwealth after his
incarceration delayed his eligibility for a parole hearing. In
his estimation, the change in policy transgressed both due
process and the ban on ex post facto laws. The district court
rejected the latter claim but granted the writ on due process
grounds and ordered, inter alia, a nunc pro tunc parole hearing.
The petitioner appeals from both the dismissal of his
ex post facto claim and from the limited grant of relief. The
Commonwealth cross-appeals from the due process ruling and the
allowance of any relief. We hold that the Commonwealth's
implementation of the challenged policy neither abridged the
petitioner's rights under the Due Process Clause nor violated the
Ex Post Facto Clause. Hence, we reverse the district court's
order and dismiss the habeas application.
I. BACKGROUND
I. BACKGROUND
We divide the introductory section of our opinion into
five segments.
2
A. The Underlying Convictions and Sentences.
A. The Underlying Convictions and Sentences.
These appeals have their genesis in events that
occurred over a quarter-century ago. In 1969, following a bench
trial, a Massachusetts court found the petitioner guilty of
charges stemming from a brutal attack and robbery that occurred
the previous year. A more complete account of the crimes,
unnecessary here, is available in Commonwealth v. Hamm, 471
N.E.2d 416, 418-19 (Mass. App. Ct. 1984) (Hamm I). The trial
court sentenced petitioner to two concurrent, parole-eligible
terms of life imprisonment for his convictions on counts of armed
robbery and assault with intent to rape, and to a series of
consecutive sentences totalling sixty-eight to eighty years on
the other counts of conviction (including mayhem and assault with
intent to murder). These consecutive sentences were to be served
"from and after" the life sentences.1 The appeals court, in an
unpublished rescript, reduced the from-and-after sentences to
twenty-six to forty years but upheld the convictions and
sentences in all other respects.
B. The Parole-Eligibility Statute.
B. The Parole-Eligibility Statute.
The Massachusetts statute governing the parole
eligibility of convicts serving terms of life imprisonment
provides (and substantially provided in 1968) that:
Every prisoner who is serving a sentence
for life in a correctional institution of the
commonwealth [with specified exceptions not
relevant here] shall be eligible for parole,
1Sacrificing originality for clarity, we refer herein to
this group of sentences as the "from-and-after sentences."
3
and the parole board shall, within sixty days
before the expiration of fifteen years of
such sentence, conduct a public hearing
before the full membership.
. . . .
After such hearing the parole board may,
by a vote of a majority of its members, grant
to such prisoner a parole permit to be at
liberty upon such terms and conditions as it
may prescribe for the unexpired term of his
sentence. If such permit is not granted, the
parole board shall, at least once in each
ensuing three year period, consider carefully
and thoroughly the merits of each such case .
. . .
Mass. Gen. L. ch. 127, 133A. Until 1977, the Commonwealth
considered inmates who were not only serving life sentences but
also facing the grim prospect of overhanging from-and-after
sentences as coming within the purview of section 133A. Based on
that interpretation of the statute, the Commonwealth granted such
inmates parole hearings (for possible parole from their life
sentences into their from-and-after sentences) once they had
served close to fifteen years. Accordingly, after the state
court sentenced Hamm, correctional officials advised him that the
parole-eligibility date referable to his life sentences would be
November 28, 1983.2
C. The 1977 Aggregation Policy.
C. The 1977 Aggregation Policy.
In 1977, the Commonwealth recast its interpretation of
section 133A. The impetus for change was the decision of the
Massachusetts Supreme Judicial Court (SJC) in Henschel v.
2The respondent fixed the parole-eligibility date in 1969
and informed the petitioner of it at that time. It should be
noted, however, that, giving credit for time served awaiting
trial and sentencing, the Commonwealth deemed the petitioner's
effective date of sentence to be November 29, 1968.
4
Commissioner of Correction, 330 N.E.2d 480 (Mass. 1975).
Henschel required the aggregation for parole-eligibility purposes
of a prisoner's consecutive county house of correction and state
institution sentences. See id. at 483-85. The SJC advanced a
thoughtful justification in support of aggregation:
To follow the defendant's [non-aggregation]
approach would require the board to make a
series of decisions granting parole from one
sentence to the next rather than a single
decision on the basis of one parole
eligibility date for all sentences. The
former procedure makes little sense since the
decision to grant parole is to be based on
whether the board believes the prisoner can
live freely outside of prison without
violating the law.
Id. at 484. The Commonwealth found this rationale to be equally
convincing in the context of making decisions to parole prisoners
serving life sentences into overhanging from-and-after sentences.
Consequently, it rethought its earlier interpretation of section
133A and revised its policies regarding parole eligibility for
certain classes of inmates, including lifers who faced impending
from-and-after sentences. Under the neoteric policy, such
inmates were not regarded as falling under section 133A and would
no longer receive parole hearings at the fifteen-year mark;
instead, the parole-ineligible portion of the prisoner's life
sentence (fifteen years) would be aggregated with the parole-
ineligible portion of his from-and-after sentences to arrive at a
"real" parole-eligibility date, that is to say, a single date at
which a favorable parole decision would result in the prisoner's
actual release from incarceration, not just his parole from one
5
sentence into another.3 While this paradigm was not compelled
by the holding in Henschel (which did not specifically address
the aggregation of life sentences with from-and-after sentences),
the respondent determined that the new arrangement more
faithfully mirrored the tenets undergirding Henschel.
In 1982 the year before Hamm would have received his
initial section 133A hearing under the former policy the
Commonwealth applied the new policy to him and recalculated his
parole-eligibility date.4 The aggregation resulted in a single,
3The document explaining the new policy, issued jointly by
the Department of Correction and the Parole Board, bore the title
"New Policies and Practices Regarding Aggregation of `From and
After' Sentences (Henschel Decision)." It states in relevant
part:
[I]t has become necessary to revise existing
procedures and policies covering aggregation
of "from and after" (i.e. consecutive)
sentences for purposes of computing parole
eligibility and good conduct deductions.
. . . .
Life Sentences
Life sentences on which there is no parole
eligibility . . . cannot be aggregated with
any other sentences for parole eligibility
purposes. Life sentences which do carry
parole eligibility . . . will be aggregated
with other sentences for parole eligibility
purposes . . . .
4For much of the life of this litigation, the Commonwealth
stubbornly insisted that it aggregated Hamm's sentences pursuant
to a different, long-established policy, and that its newly
contrived 1977 policy did not effect any change regarding
prisoners such as Hamm. Dissatisfied with the record on this
point, we retained appellate jurisdiction and remanded for
factfinding. The district court conducted an evidentiary hearing
and found, on the basis of the petitioner's prison records and
testimony from former and current counsel to the Parole Board,
that prior to 1977 the respondent did in fact follow a practice
of providing fifteen-year parole hearings to life prisoners
6
"real" parole-eligibility date of November 2001.5 Though this
structural change obviated the need for the petitioner to obtain
two parole permits to secure his release in 2001, he claims that
it also impermissibly deprived him of an opportunity for release
at an earlier date.
The petitioner's thesis runs along the following lines.
Massachusetts law affords prisoners serving indeterminate terms
of years various ways to reduce their sentences. These same
options, Hamm claims, are not available to prisoners who are
serving life sentences. Thus, if he had been paroled into his
(indeterminate) from-and-after sentences in 1983, he could have
availed himself of these opportunities and possibly could have
gained his freedom earlier than 2001. Under the 1977 policy,
however, he effectively remains on "life sentence status" during
the full term of his immurement and, therefore, cannot take
advantage of these early-release opportunities, which include:
(1) Establishing a "Wrap-up" Date. Once paroled into
his from-and-after sentences, the petitioner would immediately
facing from-and-after sentences. The Commonwealth now accepts
this finding and has recanted its assertion that it did not
retroactively subject the petitioner to a new policy.
5The exact manner in which the respondent arrived at this
date is inscrutable. The underlying calculation is not revealed
in the court papers and Hamm's post-1982 prison records (which
from time to time have indicated various parole-eligibility dates
ranging from 1999 to 2001) are little help. We need not probe
the point too deeply, however, inasmuch as the Commonwealth has
not disputed the petitioner's contention that his parole-
eligibility date under the 1977 aggregation policy is in November
of 2001. Like the district court, we will assume that to be the
correct date.
7
acquire, subject to divestiture for misconduct, statutory good
time under Mass. Gen. L. ch. 127, 129. This "good time" would
be based on the top end of his indeterminate sentences (forty
years) and would, the petitioner claims, amount to sixteen and
one-half years. He could earn additional good-time credits (up
to seven and one-half days per month) by participating in
educational and vocational programs.6 See Mass. Gen. L. ch.
127, 129D. Moreover, the sentencing court (both initially and
on resentencing) gave the petitioner 210 days credit on his four
from-and-after sentences for pre-sentence incarceration. Hamm
theorizes that this credit applies separately to each of his four
from-and-after sentences, yielding an aggregate credit of two
years and four months for jail time.
We assume arguendo the accuracy of the petitioner's
figures without independently verifying them.7 These potential
reductions, totalling twenty-three years and one month, would, if
garnered, enable him to leave prison without undergoing a second
parole hearing after serving just sixteen years and eleven months
6Good-time credits that a convict earns while serving a life
sentence apparently do not reduce his life sentence or his
parole-ineligible term; we are told that they are simply "banked"
and only become useful to him in the event that his life sentence
is commuted to a term of years. Upon parole into a from-and-
after sentence, the convict would lose his "banked" good-time
credits.
7The record is tenebrous as to many of Hamm's claims, and
some of them, e.g., the claim of an entitlement to an 840-day
credit for pre-sentence incarceration, strike us as
counterintuitive.
8
on his from-and-after sentences.8 Hence, if the petitioner had
been paroled into his from-and-after sentences in November of
1983, he might have established a wrap-up date in October 2000,
thus bringing about his release more than a year earlier than his
current aggregated parole-eligibility date.
(2) Early Parole. Once paroled into his from-and-
after sentences, the petitioner could also reduce the parole-
ineligibility period of these sentences, which otherwise would
remain at seventeen years and four months. First, he asserts
that he would be credited automatically with the same two years
and four months of jail time. But see note 7, supra, and
accompanying text. Second, his earned good time would
effectively count as time served toward his parole-ineligible
term. On this basis, he argues that if he had gained parole from
his life sentences in 1983 and earned section 129D credits from
then on at the maximum rate, he might have been eligible for
"real" parole as early as November of 1995.9
(3) Special Parole. The petitioner's final
8This optimistic calculation assumes, inter alia, that Hamm
would earn section 129D good time at the maximum possible rate
over the duration of his from-and-after sentences. If Hamm
serves sixteen years and eleven months of his from-and-after
sentences on his best behavior, he could theoretically accumulate
1,522.5 days of earned good time.
9The petitioner arrives at this date by taking the following
route: 208 months (Hamm's statutory parole-ineligible term) less
28 months (pre-sentence jail credits) less 36 months (maximum
possible section 129D credits during first twelve years of from-
and-after sentences) = 144 months. If Hamm had begun serving his
from-and-after sentences in November of 1983, and if his other
assumptions proved true, he had a possibility of securing a
parole hearing in November, 1995.
9
opportunity-related theory suggests that aggregation has already
deprived him of the possibility of obtaining special
consideration parole as early as 1989, after serving just one-
third of his from-and-after minimum sentences, less jail credits.
See Hamm v. Commissioner of Correction, 564 N.E.2d 1032, 1033
n.5 (Mass. App. Ct.) (Hamm II), rev. denied, 566 N.E.2d 1131
(Mass. 1991). The respondent effectively parries this thrust,
stating that Hamm may apply for this type of parole consideration
even under the 1977 aggregation policy. Finding no evidence in
the record that the petitioner has made an effort to apply for
special consideration parole, or that the Parole Board would not
consider his request, we cannot conclude that aggregation has
deprived the petitioner of this benefit. See id. Accordingly,
we do not further discuss this aspect of Hamm's claim of harm.
D. The 1988 Policy.
D. The 1988 Policy.
The Commonwealth revisited its parole-eligibility
policy anent life prisoners facing from-and-after sentences in
1988, and resumed the practice of providing them with parole
hearings at or near the fifteen-year mark. A 1990 document
prepared by the respondent, entitled "Parole Eligibility
Regulations, Policies, Procedures," explains that parole-eligible
life sentences are an exception to the general aggregation policy
"because of the statutory requirement that a parole hearing be
held after a definite period of time." The about-face did not
ameliorate the petitioner's professed plight; the respondent
declined to apply this policy retroactively because such an
10
application, it feared, might hurt prisoners approaching their
aggregated parole-eligibility dates. Thus, Hamm's parole-
eligibility date was not recalculated, and he remains
incarcerated with no parole hearing on the horizon until November
of 2001.
E. The Habeas Proceedings.
E. The Habeas Proceedings.
The petitioner initiated state habeas proceedings in
1990, arguing, among other things, that the Commonwealth's
failure to provide him with a parole hearing in 1983 deprived him
of his right to due process of law, and that the 1977 policy, as
applied to him, violated the prohibition on ex post facto laws.
A state superior court judge dismissed the petition, and the
Massachusetts Appeals Court upheld the aggregation of the
petitioner's sentences for purposes of determining parole
eligibility.10 See Hamm II, 564 N.E. 2d at 1033-34. After the
SJC denied further appellate review, the petitioner initiated
federal habeas proceedings.
10The court appeared to misconstrue the petitioner's ex post
facto argument; rather than focusing on whether the 1977 policy,
as applied, differed materially from the policy in effect in
1968, the court focused on the 1988 policy and ruled that Hamm
was not entitled to enjoy its benefits. On this point, the court
wrote:
The plaintiff's situation does not present an
ex post facto issue. The rules have not been
changed adversely to him. Rather, the
reverse has occurred: after his offenses and
sentencing, a rule has changed in a manner
that, if applied to him, would work to his
advantage or so the plaintiff seems to
think.
Hamm II, 564 N.E.2d at 1034-35.
11
The district court found that the Commonwealth had in
fact applied a change in the law to petitioner, but it concluded
that the change did not harm him and therefore posed no ex post
facto problem. On the due process claim, the court took a more
receptive stance. It interpreted section 133A as mandating that
petitioner receive a parole hearing on his life sentences after
fifteen years, and ruled that the Commonwealth's failure to
provide him a hearing in that time frame deprived him of due
process. The court ordered the state to convene such a hearing
nunc pro tunc, and to continue convening such hearings at three-
year intervals should parole be denied. See Hamm v. Latessa, No.
91-10667-WJS, slip op. at 14 (D. Mass. May 18, 1994) (Hamm III).
The court also decreed that if, despite the serial parole
hearings, the petitioner remained in custody beyond 2001, then in
such event, the 1977 policy should be applied to him as written
from that date forward. See id.
II. ANALYSIS
II. ANALYSIS
We bifurcate our analysis, examining each of the
petitioner's constitutional claims under a separate heading.
A. The Due Process Claim.
A. The Due Process Claim.
The district court found that section 133A applied to
the petitioner and afforded him a liberty interest in the
convening of a parole hearing in 1983 (as he neared the fifteen-
year mark of his life sentences). The court based this finding
on its interpretation of section 133A, emphasizing that the
statute is written in mandatory and unequivocal terms "Every
12
prisoner who is serving a sentence for life . . . shall be
eligible for parole, and the parole board shall, within sixty
days before the expiration of fifteen years of such sentence,
conduct a public hearing . . . ." (emphasis supplied) and makes
no exception on its face for life prisoners who also have from-
and-after sentences in prospect. To buttress this view, the
court noted that the aggregation policy expressed in section 133A
could not apply to the petitioner because his life sentences by
definition contain no "minimum" sentence, and therefore cannot be
aggregated with his from-and-after sentences to determine parole
eligibility. Building on this foundation, the court held that
the petitioner had an unequivocal statutory right to be
considered for parole into his from-and-after sentences once he
had served fifteen years of his life sentences, and that the
Commonwealth unconstitutionally deprived him of this liberty
interest by aggregating his life sentences with his from-and-
after sentences and by failing to grant him an initial parole
hearing in 1983.
We cannot accept the lower court's analysis. It is
settled that a statute providing for early release or other
benefits under stipulated conditions may sometimes confer upon
prison inmates a liberty interest protected by the Due Process
Clause.11 See Board of Pardons v. Allen, 482 U.S. 369, 373-81
11This is so even though, as a general rule, a convict has
"no constitutional or inherent right . . . to be conditionally
released before the expiration of a valid sentence." Greenholtz
v. Nebraska Penal Inmates, 442 U.S. 1, 7 (1979). The Greenholtz
generality like virtually all generalities admits of some
13
(1987); Greenholtz v. Nebraska Penal Inmates, 442 U.S. 1, 12
(1979); Wolff v. McDonnel, 418 U.S. 539, 556-72 (1974); see also
Kentucky Dept. of Corrections v. Thompson, 490 U.S. 454, 461-62
(1989) (restating principle and citing cases but finding no
liberty interest created by state regulations governing
visitation). At the time the district court issued its order,
constitutional doctrine suggested that a state creates such a
liberty interest "by establishing `substantive predicates' to
govern official decision-making . . . and, further, by mandating
the outcome to be reached upon a finding that the relevant
criteria have been met." Thompson, 490 U.S. at 462 (quoting
Hewitt v. Helms, 459 U.S. 460, 472 (1983), and omitting
citation). The statute at issue in Allen, for example, used
"mandatory language (`shall') to `creat[e] a presumption that
parole release will be granted' when the designated findings are
made." Allen, 482 U.S. at 377-78 (quoting Greenholtz, 442 U.S.
at 12) (alterations in original).
In recent years, the tectonic plates have shifted. In
Sandin v. Conner, 115 S. Ct. 2293 (1995), the Justices explained
that, rather than relying on the presence or absence of mandatory
language in determining whether a statute or regulation confers a
liberty interest, courts should focus their inquiry on the nature
of the interest allegedly created by the state. See id. at 2297-
3000. State-created liberty interests "will be generally limited
to freedom from restraint which, while not exceeding the sentence
exceptions.
14
in such an unexpected manner as to give rise to protection by the
Due Process Clause of its own force . . . nonetheless imposes
atypical and significant hardship on the inmate in relation to
the ordinary incidents of prison life." Id. at 2300 (citations
omitted).
While the question of whether a state law creates a
liberty interest protected by the Due Process Clause is clearly
one of federal constitutional law, the preliminary question of
parsing the state law to determine its substance is not within
the primary domain of a federal habeas court. See Estelle v.
McGuire, 502 U.S. 62, 67-68 (1991). Federal courts "are bound by
a State's interpretation of its own statute." Garner v.
Louisiana, 368 U.S. 157, 166 (1961). Particularly relevant here
is Hebert v. Louisiana, 272 U.S. 312 (1926). In Hebert, a
convicted defendant claimed that a state court's incorrect
construction of state law led to the imposition of a more onerous
sentence and thereby violated the Due Process Clause. The Court
rejected that claim, stating:
Whether state statutes shall be construed one
way or another is a state question, the final
decision of which rests with the courts of
the State. The due process of law clause in
the Fourteenth Amendment does not take up the
statutes of the several States and make them
the test of what it requires; nor does it
enable this Court to revise the decisions of
the state courts on questions of state law.
Id. at 316. The rule, then, is that a federal habeas court will
not disturb the state courts' construction or application of
state law unless it can be shown that such construction or
15
application offends the Constitution or some (applicable) federal
statute. See Bowser v. Boggs, 20 F.3d 1060, 1065 (10th Cir.
1994); Smith v. McCotter, 786 F.2d 697, 700 (5th Cir. 1986); cf.
Martinez v. California, 444 U.S. 277, 282 (1980) (explaining that
a state's interest in fashioning its own rules of state law is
paramount to any federal interest except protecting individuals
from state action that is wholly arbitrary and irrational).
Given the clearly demarcated boundaries of federal
habeas review, the proper function of the court below was not to
second-guess the state court as to what substantive guarantees
the Commonwealth's statutory and regulatory mosaic provided under
the particular circumstances, but, rather, simply to determine
whether the respondent's application of its parole scheme, deemed
lawful by the state's courts, violated the Due Process Clause.
The district court set out to accomplish the former task instead
of the latter. In so doing, it erred.
Here, a Massachusetts state court has already ruled
that section 133A, as it read both in 1968 and 1983, conveyed no
right to a fifteen-year hearing, and, moreover, that the
respondent's aggregation of Hamm's sentences was a permissible
policy notwithstanding any contrary signposts in the text of the
state statute. See Hamm II, 564 N.E.2d at 1033-34. We believe
that we are duty bound to follow this authoritative exposition of
state law and, concomitantly, to reject the district court's
16
impromptu interpretation of state law.12 See McGuire, 502 U.S.
at 67-68; Garner, 368 U.S. at 166.
We have considered and rejected the petitioner's
argument that the state appellate court's decision is "so
inconsistent with the statute's language and history that the
state court decision itself [comprises] a wholly arbitrary and
irrational action in violation of due process." Ellard v.
Alabama Bd. of Pardons and Paroles, 824 F.2d 937, 944 n.7 (11th
Cir. 1987) (citation and internal quotation marks omitted), cert.
denied, 485 U.S. 981 (1988). A federal court must not exercise
the raw power to strike down state laws in the name of the
Constitution with too much gusto. Section 133A is silent on the
parole eligibility of life prisoners facing from-and-after
12We reach this conclusion cognizant that what we deem a
controlling state court interpretation of state law emanated from
an intermediate appellate court. Intermediate appellate court
decisions "are trustworthy data for ascertaining state law,"
Losacco v. F.D. Rich Constr. Co., 992 F.2d 382, 384 (1st Cir.),
cert. denied, 114 S. Ct. 324 (1993), and, in the absence of other
telltales indicating that the state's highest tribunal would have
ruled otherwise, we believe it is prudent to accept the appeals
court's interpretation as authoritative. See, e.g., Fidelity
Union Trust Co. v. Field, 311 U.S. 169, 177-78 (1940) ("An
intermediate state court in declaring and applying the state law
is acting as an organ of the State and its determination, in the
absence of more convincing evidence of what the state law is,
should be followed by a federal court in deciding a state
question.").
The respondent's 1988 about-face and its return to a
non-aggregation policy for life inmates does not suggest a
different outcome. Given the language of the statute and the
discretion reposed in the Parole Board, it is entirely plausible
that both the 1977 and 1988 policies are permissible
interpretations of state law. Cf. Strickland v. Commissioner,
Me. Dept. of Human Servs., 48 F.3d 12, 17-18 (1st Cir.) (holding
that an agency's rule may receive the usual degree of deference
even when it represents a "sharp departure from a longstanding
prior interpretation"), cert. denied, 116 S. Ct. 145 (1995).
17
sentences; a Massachusetts court had never before ruled on the
issue: and the appeals court based its holding on Henschel, which
provided a defensible rationale for an aggregation policy.13
Under these circumstances, it would be unprincipled to declare by
federal fiat that the Due Process Clause broadly nullifies the
Commonwealth's power to construe and apply its laws correctly.
See Lerner v. Gill, 751 F.2d 450, 459 (1st Cir.), cert. denied,
472 U.S. 1010 (1985).
The petitioner has also asserted that his fourteen
years of state-induced reliance on a prospective 1983 parole
hearing, followed by the state's abrupt shift in policy, deprived
him of due process. In mounting this challenge, he embraces our
decisions in Lerner and in DeWitt v. Ventetuolo, 6 F.3d 32 (1st
Cir. 1993), cert. denied, 114 S. Ct. 1542 (1994). Hamm's ardor
is mislaid. Those cases addressed the power of a state court
retroactively to correct an erroneous sentence or a mistaken
interpretation of state law. See DeWitt, 6 F.3d at 34-35;
Lerner, 751 F.2d at 458-59. However, the calculation of the
petitioner's original parole-eligibility date was not
"incorrect," so he arguably possessed a greater interest in
seeing it carried out than would a prisoner who was the
beneficiary of a botched calculation. Nevertheless, the
13Henschel supports the view that, since parole decisions
are premised on whether the Parole Board believes a convict can
live outside prison without behaving in an antisocial manner, the
Parole Board should not normally be required to make a series of
decisions paroling a convict from one sentence to another. See
Henschel, 330 N.E.2d at 484.
18
Lerner/DeWitt line of cases does not invalidate the
Commonwealth's actions. We explain briefly.
Though we observed in Lerner, 751 F.2d at 458, that
"unforeseeable changes . . . made after the passage of a
substantial period of time may, in some presumably extreme
circumstances, be fundamentally unfair and hence violative of due
process even if designed to correct an illegal sentence," this
observation is inapposite here. In the first place, we do not
think that the Massachusetts Appeals Court decision was
unforeseeable; as stated above, the statute was silent on the
precise situation, there was no decisional law directly on the
point, and Henschel adumbrated the result reached in Hamm II. In
the second place, a convict must show special prejudice stemming
from a changed interpretation. See DeWitt, 6 F.3d at 35. This
requisite showing must consist of something more tangible than
merely demonstrating that "hopes were raised only to be dashed."
Lerner, 751 F.2d at 459. Hamm has not suffered prejudice even
remotely approaching that sustained by Lerner, whose parole-
ineligibility period was extended from ten years to twenty years
after he had (1) undergone two parole hearings, (2) moved into a
minimum-security facility and accepted other privileges, and (3)
transported his family to another state and caused them to invest
in a business to create an employment opportunity for him, see
id. at 453 to whom we denied relief under the Due Process
Clause, see id. at 459.
In this case, all roads lead to Rome. We hold that the
19
Commonwealth did not infract the petitioner's rights under the
Due Process Clause when it failed to provide him a parole hearing
in 1983. The administrative scheme in force at that time,
approved as lawful by a state appellate court, did not mandate
that petitioner receive a parole hearing after fifteen years. In
ruling to the contrary, the district court erred.
B. The Ex Post Facto Claim.
B. The Ex Post Facto Claim.
Article 1, 10 of the Constitution ("No State shall .
. . pass any . . . ex post facto Law") has been interpreted to
forbid the enactment of
any statute which punishes as a crime an act
previously committed, which was innocent when
done; which makes more burdensome the
punishment for a crime, after its commission,
or which deprives one charged with crime of
any defense available according to law at the
time when the act was committed . . . .
Beazell v. Ohio, 269 U.S. 167, 169-70 (1925); see also California
Dept. of Corrections v. Morales, 115 S. Ct. 1597, 1601 (1995)
(stating that "the Clause is aimed at laws that retroactively
alter the definition of crimes or increase the punishment for
criminal acts") (citation and internal quotation marks omitted);
Collins v. Youngblood, 497 U.S. 37, 42 (1990) (quoting Beazell).
Petitioner asseverates that, by depriving him of opportunities to
obtain his release earlier than November of 2001, the 1977
aggregation policy made his punishment more burdensome and is,
therefore, an impermissible ex post facto law. The Commonwealth
demurs. In its view, the 1977 policy is not a "law" subject to
the ex post facto proscription, and in all events, the resultant
20
aggregation did not increase the petitioner's punishment.
Because we agree with the Commonwealth's second contention, we
need not decide the thorny question of whether the 1977 policy
comprised a "law" subject to ex post facto analysis.14
In line with the foregoing, we assume for argument's
sake, but do not decide, that the 1977 aggregation policy
constituted a regulation possessing the full force and effect of
law, and that it is therefore subject to analysis under the Ex
Post Facto Clause. This assumption brings us to the decisive
question: Does the 1977 policy, as applied to the petitioner,
infringe the constitutional proscription against ex post facto
laws?
14We note in passing that, although the Supreme Court has
not addressed the question of whether an administrative policy or
regulation can be an ex post facto law, a number of courts have
held that binding administrative regulations, as opposed to those
that serve merely as guidelines for discretionary decisionmaking,
are laws subject to ex post facto analysis. See, e.g., Akins v.
Snow, 922 F.2d 1558, 1561 (11th Cir. 1991) (holding that Georgia
parole board's new regulation promulgated pursuant to delegated
legislative power that changed period between inmate's parole
hearings from one year to eight years was a law subject to ex
post facto analysis), cert. denied, 501 U.S. 1260 (1991);
Rodriguez v. United States Parole Comm'n, 594 F.2d 170, 174 (7th
Cir. 1979) (deeming a new regulation that eliminated a parole
hearing after one-third of a prisoner's sentence "tantamount to a
statute" for ex post facto purposes); Love v. Fitzharris, 460
F.2d 382, 385 (9th Cir. 1972) (holding that a state's
recalculation of a prisoner's parole-eligibility date under a new
interpretation of the governing statutes violated the Ex Post
Facto Clause because the state had changed its interpretation
midstream), vacated as moot, 409 U.S. 1100 (1973). There are, of
course, cases mostly involving the federal Parole Commission's
guidelines that can be read as holding the other way. See,
e.g., Kelly v. Southerland, 967 F.2d 1531, 1532-33 (11th Cir.
1992); Inglese v. United States Parole Comm'n, 768 F.2d 932, 936
(7th Cir. 1985). We see nothing to be gained from entering this
thicket without a compelling need to do so.
21
It is a universal truth that, for a law to offend the
Ex Post Facto Clause, it must be "more onerous than the prior
law." Dobbert v. Florida, 432 U.S. 282, 294 (1977). The
prescribed inquiry demands that we compare the new law with the
old in its totality to ascertain "if the new may be fairly
characterized as more onerous." Id. The inquiry must be carried
out in practical, as opposed to purely theoretical, terms; the ex
post facto prohibition does not foreclose every change in the law
that possesses some imaginable risk of adversely affecting an
inmate's punishment. See Morales, 115 S. Ct. at 1602. In the
last analysis, "the question of what legislative adjustments will
be held to be of sufficient moment to transgress the
constitutional prohibition must be a matter of degree." Id. at
1603.
There is no mechanical formula for identifying which
legislative changes have a sufficiently profound impact on
substantive crimes or punishments to cross the constitutional
line and which do not. Consequently, courts must determine, case
by case, whether a particular change in the governing law
"produces a sufficient risk of increasing the measure of
punishment attached to the covered crimes." Id. If so, the Ex
Post Facto Clause comes into play. See Hill v. Jackson, 64 F.3d
163, 167-170 (4th Cir. 1995).
Morales is the touchstone of modern ex post facto
jurisprudence. There, the Justices examined a state statute that
permitted parole boards to defer parole suitability hearings for
22
up to three years for double murderers and certain other
prisoners if the board specifically found that it was
unreasonable to expect that parole would be granted during the
intervening years. The Court concluded that the statute created
"only the most speculative and attenuated possibility of
producing the prohibited effect of increasing the measure of
punishment for covered crimes," and held that these "conjectural
effects" were insufficient to animate the Ex Post Facto Clause.
Morales, 115 S. Ct. at 1603 (footnote and internal citations
omitted).
Buttressing its conclusion, the Court identified
several aspects of the statute that neutralized the risk of
increasing the measure of punishment. First, the statute applied
only to "a class of prisoners for whom the likelihood of release
on parole is quite remote." Id. at 1603. Second, the statute
cabined the parole board's discretion by requiring it to make a
specific finding that a particular prisoner was unlikely to be
paroled. See id. Third, the statute only applied to those who
were denied parole eligibility the first time around, thus
restricting the affected class to those least likely to be found
suitable for parole. See id. Finally, though the statute
addressed the frequency of suitability hearings, it empowered the
board to "tailor the frequency of subsequent suitability hearings
to the particular circumstances of the individual prisoner." Id.
The Court also noted that "the possibility of immediate release
after a finding of suitability for parole is largely theoretical;
23
in many cases, the prisoner's parole release date comes at least
several years after a finding of suitability," and so "the
practical effect of a hearing postponement is not significant."
Id. (citation and internal quotation marks omitted).
At bottom, Morales is about the risk or likelihood of
harm: the Court upheld the California statute largely because it
found the risk of increased punishment posed by the new law to be
speculative and, at any rate, to fall within acceptably small
boundaries.15 Viewed against this backdrop, Hamm's case
founders. Though the parole scheme in effect in 1968 may have
afforded him the possibility of terminating his incarceration as
early as 1995 if a series of events materialized, it seems highly
unlikely that these events would have come to pass. The
likelihood of harm, therefore, is tiny. And, moreover, it is at
least equally possible that, because of the structural
differences imposed by the two policies, Hamm may well have
languished in prison longer under the old scheme than under the
new. On balance, then, it is virtually impossible to
15The dissent strongly suggests that Morales stands for the
bright-line proposition that any action which substantially
delays, or deprives a prisoner of, an initial parole hearing
works a per se violation of the Ex Post Facto Clause. The
Morales Court rejected a similar argument, emphasizing that such
an "arbitrary line has absolutely no basis in the Constitution.
If a delay in parole hearings raises ex post facto concerns, it
is because that delay effectively increases a prisoner's term of
confinement, and not because the hearing itself has independent
constitutional significance." Morales, 115 S. Ct. at 1603 n.4.
The dissent's mechanical approach not only ignores this
admonition but also overlooks Morales's central holding, namely,
that a proper ex post facto inquiry must focus on the risk that
the prisoner will be subject to an increased measure of
punishment.
24
prognosticate the risk that the petitioner may be released at a
later date because the 1977 policy rather than its predecessor,
controls in his case.
Under the original policy, unless the petitioner
managed to obtain parole in 1983 and thereafter amassed all
possible credits, he would still have to be paroled a second time
in order to be released as early as 1995, or, alternatively, to
garner every conceivable credit in order to attain a wrap-up date
in October of 2000. The record is bereft of any evidence
suggesting that Hamm would probably achieve prompt parole into
his from-and-after sentences, become a model prisoner, go on to
earn all available credits, and then be paroled out of, or
otherwise released from, his from-and-after sentences at any time
before 2001.16 In fine, this case, like Morales, involves a
situation in which the possibility of harm is entirely
speculative.17
16If past is prologue, cf. W. Shakespeare, The Tempest, act
II, sc. i (1612), all of these prospects seem extremely dubious.
Hamm's disciplinary record reveals a cavalcade of misconduct,
including episodes of inciting a prison riot, arson, assaulting a
guard, attempting an escape, conspiring to take a hostage,
organizing a work stoppage, and possessing controlled substances.
These are not the emblemata of an inmate who is likely either to
inspire a parole board to act favorably on his behalf or to
accumulate good-time credits at a rapid rate.
17There are, of course, other similarities to Morales. We
mention two of them. First, the challenged policy here like
the statute at issue in Morales, 115 S. Ct at 1603 applies only
to a limited class of prisoners (here, life inmates who face the
overhanging prospect of from-and-after sentences) for whom the
likelihood of release on parole is considerably below the norm.
Second, the availability of special parole, see supra Part
I(C)(3), offers the Parole Board the flexibility that the Morales
Court deemed important. See id. at 1604.
25
This case also possesses a further dimension that
weighs against the petitioner's position. Whereas the new law in
Morales could not conceivably have inured to a prisoner's
benefit, the new aggregation policy that the Commonwealth adopted
in 1977 might very well redound to the petitioner's advantage.
After all, the 1977 policy eliminates the need for two parole
permits and ensures that the petitioner will be eligible for
parole from all his sentences at one fell swoop. Under the old
policy, if the respondent denied the petitioner parole into his
from-and-after sentences in 1983, 1986, and 1989 not an
unlikely eventuality in light of Hamm's mottled record his
wrap-up date, even assuming the accrual of all conceivable
credits, would not occur until sometime in 2006. This is a full
five years after the date on which he could be paroled from all
his sentences under the 1977 policy.
This scenario prompted the district court to conclude
that "[i]n 1982, when the respondent recalculated the
petitioner's parole eligibility, it was not clear whether the
petitioner would be helped or harmed by aggregation; the balance
sheet is unclear even today." Hamm III, slip op. at 10.18
18The Morales Court pointed out that the relevant inquiry
must not focus "on whether a legislative change produces some
ambiguous sort of `disadvantage,' . . . but on whether any such
change . . . increases the penalty by which a crime is
punishable." Morales, 115 S. Ct. at 1602 n.3. Despite the fact
that the district court issued its opinion without the benefit of
Morales and framed the corresponding part of its ruling as an
inquiry into whether the 1977 policy resulted in a
"[d]isadvantage to the petitioner," it nonetheless made a
thoroughgoing examination of the uncertainty inherent in
comparing potential results under the old and new policies.
26
This statement is unarguably accurate, and the uncertainty that
it portends strips the veneer of plausibility from the
petitioner's ex post facto initiative. Morales makes it crystal
clear that such uncertainty militates against the petitioner
because any other approach would "effectively shift[] to the
State the burden of persuasion as to [the prisoner's] ex post
facto claim." Id. at 1603 n.6. A party who asserts an ex post
facto claim must show a real possibility of cognizable harm, not
a theoretical possibility bound up in gossamer strands of
speculation and surmise.
We find illuminating a recent decision of another court
that needed to construct the ex post facto balance of prospective
benefits and burdens. In United States v. McGee, 60 F.3d 1266
(7th Cir. 1995), the court addressed an amendment that became law
following the defendant's conviction but before the imposition of
sentence. The amendment eliminated a twelve-month minimum
sentence for the offense of conviction and replaced it with a
range of twenty-one to twenty-four months. The district court
imposed the maximum twenty-four-month sentence. The defendant
appealed, claiming that the amendment violated the ex post facto
prohibition because it removed the possibility of a shorter
sentence, i.e., a sentence of between twelve and twenty-one
months. The Seventh Circuit disagreed. It noted that, although
the amendment eliminated an opportunity for a milder sentence, it
also placed a ceiling on the maximum available sentence, thus
narrowing "the range of punishment to [the defendant's] benefit."
27
Id. at 1271. This potential benefit made an evaluation of the
risk that the new law might subject the prisoner to harsher
punishment too speculative to constitute an ex post facto
violation. See id.
The Seventh Circuit's rationale is persuasive here.
Due to the peculiar concatenation of circumstances especially
the profound uncertainty over how the petitioner would have fared
under the old system and the potential benefits that may accrue
to him under the new regime the potential risk of more
Draconian punishment under the 1977 policy defies reliable
measurement. As a result, we hold that the application of the
new policy to the petitioner did not insult the Ex Post Facto
Clause.
III. CONCLUSION
III. CONCLUSION
We need go no further. For the foregoing reasons, we
reverse the district court's order and dismiss the application
for a writ of habeas corpus.
It is so ordered.
It is so ordered.
Separate Opinion Follows
STAHL, Circuit Judge (concurring in part and
STAHL, Circuit Judge (concurring in part and
dissenting in part). I agree that the Commonwealth did not
dissenting in part).
violate Hamm's rights under the Due Process Clause when it
failed to provide him a parole hearing in 1983. Unlike my
28
colleagues, however, I am persuaded that, as applied to Hamm
and other similarly-situated prisoners, the 1977 aggregation
policy is an unconstitutional ex post facto law. First, I
believe that the 1977 aggregation policy, which effectively
altered the date of Hamm's initial parole hearing, is a "law"
subject to ex post facto limitations. Second, contrary to my
colleagues, I believe that the 1977 policy, as applied to
Hamm and other similarly-situated prisoners, clearly produces
a risk of increasing the measure of punishment sufficient to
violate the Ex Post Facto Clause. Hence, I respectfully
dissent from parts II.B.-III.
I.
I.
Article 1, 10 of the Constitution clearly
proscribes the authority of a state to enact any ex post
facto law. As the majority explains, it is long settled that
the Clause forbids
any statute which punishes as a crime an
act previously committed, which was
innocent when done; which makes more
burdensome the punishment for a crime,
after its commission, or which deprives
one charged with crime of any defense
available according to law at the time
when the act was committed.
Collins v. Youngblood, 497 U.S. 37, 42 (1990) (quoting
Beazell v. Ohio, 269 U.S. 167, 169-70 (1925)); see also
California Dep't of Corrections v. Morales, 115 S. Ct. 1597,
1601 (1995) ("the Clause is aimed at laws that retroactively
alter the definition of crimes or increase the punishment for
-29-
29
criminal acts") (internal quotation omitted). In general, an
ex post facto inquiry requires a two-step analysis. See
Miller v. Florida, 482 U.S. 423, 430 (1987). A court should
ask (1) whether the challenged provision is a "law" that acts
retrospectively, and (2) whether the burden the law
retrospectively imposes is of sufficient type and degree to
violate the Constitution.
In this case, Hamm argues that the 1977 aggregation
policy violated the Ex Post Facto Clause by retroactively
depriving him of opportunities to obtain his release earlier
than November 2001. In response, the Commonwealth contends
that the 1977 aggregation policy was not a "law" subject to
ex post facto limitation, and that, in any event, the
aggregation did not increase Hamm's punishment. My
colleagues agree with the second contention, and therefore
find it unnecessary to consider the first. Because, as I
explain more fully infra at 35-43, I believe that the 1977
aggregation policy engendered a sufficient risk of increasing
Hamm's punishment, I cannot avoid the first prong of the
Commonwealth's argument. Accordingly, I proceed first to
explore fully whether the 1977 aggregation policy is a "law"
subject to ex post facto proscription, and, second, to
discuss my disagreement with the majority over whether the
-30-
30
new policy produces a risk of increasing the measure of
punishment sufficient to violate the Constitution.19
A. Is the 1977 Aggregation Policy a "Law"?
I agree with the district court that the 1977
aggregation policy was a "law" for purposes of ex post facto
analysis. Although the aggregation policy was not formally
promulgated as a regulation governing the Parole Board, it
was as binding on the Parole Board, on a case-by-case basis,
as an act passed by the legislature would have been.
Moreover, the Commonwealth does not argue that, once the
policy had been promulgated, the Parole Board had any
discretion to deviate from the policy in any particular
instance.
The Supreme Court has not addressed the question of
whether an administrative policy or regulation can be an ex
post facto law. A number of circuit courts, however, have
held that binding administrative regulations, as opposed to
those that serve merely as guidelines for discretionary
decisionmaking, are laws subject to ex post facto limitation.
For example, in a case factually similar to this one, the
Ninth Circuit held that the California Department of
Corrections's recalculation of a prisoner's parole-
eligibility date under its new interpretation of the
19. I do not restate the facts or outline the prior
proceedings. For a complete discussion of these matters, see
Majority Opinion at 2-12.
-31-
31
governing statutes violated the Ex Post Facto Clause because
"the Department has changed its interpretation of the
authority itself." Love v. Fitzharris, 460 F.2d 382, 385
(9th Cir. 1972), vacated as moot, 409 U.S. 1100 (1973). The
Love court stated that:
the interpretation of the relationship
between the statutes . . . by the
administrative agency charged with their
enforcement has the force and effect of
law. . . . [N]ot only defendants, in
contemplating their pleas, but also trial
courts, in imposing sentences, are
entitled to rely on such administrative
interpretations. . . . A new
administrative interpretation which
subjects the prisoner already sentenced
to more severe punishment has the same
effect as a new statute lengthening his
present term . . . .
Id. (citations omitted). The Eleventh Circuit similarly
concluded that a regulation, promulgated pursuant to the
Georgia parole board's delegated legislative power, that
changed the period between inmate's parole hearings from one
to eight years was a "law" subject to ex post facto
limitation. Akins v. Snow, 922 F.2d 1558, 1561 (11th Cir.),
cert. denied, 501 U.S. 1260 (1991); see also Rodriguez v.
United States Parole Comm'n, 594 F.2d 170, 174 (7th Cir.
1979) (new regulation eliminating parole hearing after one-
third of sentence is "tantamount to a statute" for ex post
facto purposes).
In those cases holding that particular
administrative regulations or guidelines were not laws
-32-
32
subject to the Ex Post Facto Clause, courts have often
premised their holdings, at least in part, on the advisory
nature of the regulation or guidelines in question. See,
e.g., Kelly v. Southerland, 967 F.2d 1531, 1532-33 (11th Cir.
1992) (rescission guidelines promulgated by federal Parole
Commission did not violate Ex Post Facto Clause because they
both were subject to amendment by the Commission and merely
guided, but did not dictate, actual parole decisions); Smith
v. United States Parole Comm'n, 875 F.2d 1361, 1367 (9th Cir.
1988) (finding parole "regulation" was not an ex post facto
law and noting that "the operative factor in assessing
whether a directive constitutes a `law' for ex post facto
purposes is the discretion that the Parole Commission retains
to modify that directive or to ignore it altogether as the
circumstances may require"); Inglese v. United States Parole
Comm'n, 768 F.2d 932, 936 (7th Cir. 1985) ("The power to
exercise discretion indicates that the [parole] guidelines
are merely guides, and not law: guides may be discarded when
circumstances require; laws may not."). Moreover, these
cases involve the federal Parole Commission's guidelines,
which are "truly advisory" because the Commission possesses
the authority to disregard them in the appropriate
circumstances. Bailey v. Gardebring, 940 F.2d 1150, 1158
(8th Cir. 1991) (Lay, C.J., dissenting), cert. denied, 112 S.
Ct. 1516 (1992). The Commonwealth makes no claim that its
-33-
33
aggregation policy was merely "advisory" or that it was free
to disregard the policy in a particular case.
The Commonwealth does postulate, however, that
because it had discretion to adopt the 1977 aggregation
policy in the first place and to modify the policy
subsequently, as it did in 1988, the policy should not be
considered a law. Although a number of the federal Parole
Commission cases have relied in part on this reasoning,20
see, e.g., Smith, 875 F.2d at 1367, I would reject it here.
The argument not only exalts form over substance but its
logic suggests that even legislative acts, because they too
may be modified, should be immune to challenge under the Ex
Post Facto Clause. See Bailey, 940 F.2d at 1158 (Lay, C.J.,
dissenting). A binding policy or regulation, promulgated
pursuant to delegated legislative authority by an
administrative body that implicitly retains authority to
amend it in the future, is no different in its force and
effect than a law passed by a legislature that retains
authority to amend or revoke that law. The Commonwealth's
20. The Commonwealth claims that the Eighth Circuit adopted
this reasoning in Bailey, which held that a change in
Minnesota parole board regulations abolishing annual review
of prospective release dates and limiting the board's
discretion in changing such dates did not constitute a law
for ex post facto purposes, even though the board lacked
discretion to disregard its regulations in any given case.
However, the relevant section of Bailey, 940 F.2d at 1156,
drew the concurrence of a second panel member as to the
result only, and not its reasoning, id. at 1157 (Stuart, J.,
concurring).
-34-
34
Parole Board possessed delegated legislative authority to
promulgate the aggregation policy: "The parole board shall .
. . make rules relative to the performance of its duties."
Mass. Gen. L. ch. 27, 5(e). Furthermore, under
Massachusetts law, an agency regulation21 "promulgated
pursuant to a legislative grant of power generally [has] the
force of law." Kenney v. Commissioner of Correction, 468
N.E.2d 616, 619 (Mass. 1984). Thus, because the 1977
aggregation policy was effectively a regulation having the
full force and effect of law, I would hold that it is subject
to limitation under the Ex Post Facto Clause.
21. That the 1977 policy was not formally deemed a
"regulation" also does not seem to matter: Under
Massachusetts law, a "regulation"
includes the whole or any part of every
rule, regulation, standard or other
requirement of general application and
future effect . . . adopted by an agency
to implement or interpret the law
enforced or administered by it.
Mass. Gen. L. ch. 30A, 1(5) (emphasis added).
-35-
35
B. Does the 1977 Aggregation Policy Produce a Sufficient
Risk of Increasing the Measure of Punishment?
I now turn to the issue at the heart of my
disagreement with the majority: Whether, as applied to Hamm
and other similarly-situated prisoners, the 1977 aggregation
policy produces a risk of increased punishment sufficient to
violate the Ex Post Facto Clause? My colleagues answer this
question in the negative, basing their conclusion on two
premises. First they deem it highly unlikely that, under the
prior policy, Hamm would have won early parole from his life
sentence and acquired the necessary good-time credits to
advance the date of his ultimate parole hearing to a point
earlier than 2001. Hence, they conclude that any harm to
Hamm ensuing from the 1977 aggregation policy is highly
speculative. Second, they posit that, due to structural
differences between the two policies, a "real" benefit
accrues to Hamm under the new policy. Then, combining these
two premises, my colleagues ultimately conclude that, on
balance, the 1977 aggregation policy does not violate the Ex
Post Facto Clause. I strongly disagree.
My colleagues favorably compare the risk of
increased punishment occasioned by the 1977 aggregation
policy with the risk of increased punishment recently
examined by the Supreme Court in California Dep't of
Corrections v. Morales, 115 S. Ct. 1597 (1995). In Morales,
-36-
36
the Court considered an amendment permitting the California
state parole board to defer annual parole-suitability
hearings for up to three years for prisoners at least twice
convicted of murder. Id. at 1600. In upholding the
amendment, the Court rejected the contention that it violated
the Ex Post Facto Clause simply because the deferral of
subsequent suitability hearings deprived affected prisoners
of an opportunity to gain an earlier release from prison.
Id. at 1602 n.3. The Court explained that just because the
amendment caused the loss of some theoretical opportunity to
gain an earlier release did not mean that it necessarily
violates the Constitution. Id. Instead, the Court held
that, for ex post facto purposes, the test is whether the
loss of that opportunity actually produces a "sufficient risk
of increasing the measure of punishment for the attached
crimes." Id. at 1603.
In applying this test, the Court focused on several
factors that significantly minimized the California
amendment's risk of harm. Morales, 115 S. Ct. at 1603-05.
First, the Court noted the amendment's limited application.
Id. at 1603. The amendment had no effect on any prisoner
unless the California parole board first found that the
prisoner was both unsuitable for parole and unlikely to be
found suitable at subsequent hearings during the deferral
period. Id. at 1604. Moreover, the Court noted that the
-37-
37
amendment did not affect "the date of any prisoner's initial
parole suitability hearing: it affected the timing only of
subsequent hearings." Id. at 1605.
Next, the Court observed, inter alia, that, even
with respect to a prisoner who might have actually received a
favorable recommendation at an omitted hearing, the practical
effect of the amendment on that prisoner's ultimate release
date was only slight. Morales, 115 S. Ct. 1605. At the
deferred hearings, the parole board determined only a
prisoner's "suitability" for parole but did not set actual
parole dates. Id. The Court noted that, significantly, in
many cases, an actual parole date comes several years after a
finding of suitability. Id. Moreover, under California law,
evidence that a prisoner in fact had been "suitable" for a
year or two prior to the date of the prisoner's delayed
hearing would be relevant in setting the prisoner's actual
parole date. Id. Hence, the Court concluded that, in most
cases, any delay resulting from the amendment could be
corrected by the parole board when it set the prisoner's
ultimate release date. Id.
In short, the Court recognized that the
amendment's built-in limitations, severely restricting both
its application and potential effect, effectively minimized
any risk of increased punishment caused by the elimination of
subsequent suitability hearings. Furthermore, the Court
-38-
38
carefully limited the breadth of its holding, expressly
disavowing any opinion "as to the constitutionality of any of
a number of statutes that might alter the timing of parole
hearings under circumstances different from those present
here." Morales, 115 S. Ct. at 1603 n.6.
On close analysis, I believe the effect of the 1977
aggregation policy challenged here differs significantly from
the risk of increased harm produced by the Morales amendment.
First, in contrast to the Morales amendment, the adoption of
the 1977 aggregation policy potentially affects all
Massachusetts prisoners previously eligible for parole from a
life sentence into consecutive from-and-after sentences. No
provision in the policy limits the class of affected
prisoners to only those adjudged by the Commonwealth's Parole
Board (or some similar body) to be unlikely to win early
parole or to earn significant good-time credits. Moreover,
where the Morales amendment affected only subsequent
hearings, the 1977 aggregation policy essentially delays an
affected prisoner's initial parole hearing.
Second, also in contrast to Morales, the impact of
the 1977 aggregation policy on those it affects is
substantial. For example, under the prior policy, Hamm could
have terminated his incarceration as early as 1995, through
the acquisition of earned and statutory good-time credits and
-39-
39
the application of his jail credits.22 The 1977
aggregation policy extinguished that possibility; Hamm's term
of incarceration cannot end under the 1977 policy before his
first-available parole hearing in 2001. Thus, in contrast to
the amendment in Morales, which will have little, if any,
real impact on an affected prisoner's actual time in prison,
the 1977 aggregation policy could potentially increase Hamm's
prison term by up to six years. See Weaver v. Graham, 450
U.S. 24, 26-27, 31-34 (1981) (holding that new state statute
reducing amount of good time that could be earned
prospectively by current inmates violates Ex Post Facto
Clause because it removed existing opportunity for shortened
prison time).
My colleagues gloss over these clear distinctions
by positing that, whatever the risk might have been at the
outset, given the brutal nature of Hamm's crime and his poor
record as a prisoner, it is highly unlikely that Hamm could
have availed himself of the opportunity to obtain an earlier
release. Such analysis, however, is more akin to a harmless
error inquiry focusing on the particulars of Hamm's case than
to a proper ex post facto inquiry into whether the new law
22. As does the majority, I assume the accuracy of Hamm's
claim of entitlement to 840 days of jail credit. See
Majority at 8. I note, however, that the claim is not
critical to my analysis. Even without the 840 days, the 1977
aggregation policy deprives Hamm of the opportunity to
advance his initial ultimate parole date by over three and
half years.
-40-
40
posed a sufficient risk of increasing the measure of
punishment. Moreover, while it is clear that Hamm bears the
ultimate burden of establishing that the new law changes the
measure of punishment, Morales, 115 at 1603 n.6, this does
not mean that he must necessarily show "that he would have
been sentenced to a lesser term under the measure or range of
punishments in place under the previous statutory scheme."
Id. (citing Lindsey v. Washington, 301 U.S. 397, 401 (1937));
see also id. at 1607 (Stevens, Souter, J.J., dissenting);
Miller v. Florida, 482 U.S. 423, 432 (1987) (reaffirming
Lindsey). Indeed, the proper "inquiry looks to the
challenged provision, and not to any special circumstances
that may mitigate its effect on the particular individual."
Weaver, 450 U.S. at 33.
In any event, the fact of the matter is that the
1977 aggregation policy completely deprived Hamm of his once-
existing opportunity to gain a release from prison as much as
six years earlier than he can now. Moreover, notwithstanding
my colleagues' post hoc evaluation of Hamm's chances, because
Hamm never received a parole hearing, no findings exist to
inform us whether or not the Commonwealth would have found
Hamm to be a likely candidate for early parole from his life
sentence. Indeed, without such findings or even knowledge
concerning the standards and policies that guide the
Commonwealth's Parole Board in making such recommendations,
-41-
41
this court can only speculate as to whether the 1977
aggregation policy posed a sufficient risk to Hamm.23
In Morales, the Court reasoned that the delay in
parole suitability hearings caused by the challenged
amendment did not produce a sufficient risk of punishment
because, in major part, the amendment affected only a
carefully limited class of prisoners, and the impact of any
delay on an affected prisoner's actual time in prison was
negligible. Implicit in the Court's holding, however, is the
recognition that delay in a parole hearing produces some
possibility of an increase in punishment. Where, as here,
the delay is not predicated on a finding that the prisoner is
an unlikely candidate for parole, and the delay may
significantly increase the prisoner's sentence, I believe,
even in Hamm's case, such delay produces a "sufficient risk
of increasing the measure of punishment." Morales, 115 at
1603.
As noted, my colleagues also base their conclusion
on the premise that the 1977 aggregation policy arguably
provides a "real" benefit to Hamm and other affected
prisoners. I believe, however, that it is this putative
"benefit" that is too "speculative" to merit significant
23. The fact that the record lacks the opinion, much less
the findings, of the Commonwealth's Parole Board on Hamm's
suitability for early parole clearly underscores the
inappropriateness of my colleagues' "harmless error" style
review.
-42-
42
weight in the ex post facto inquiry. Any fair analysis
reveals that the supposed benefit arising from the 1977
aggregation policy assumes several rather contradictory
predicates. For example, to find that Hamm would benefit
from the 1977 policy, I would need to assume both (1) that,
under the prior policy, the Commonwealth's Parole Board would
have refused to grant Hamm parole from his life sentence at
least three times (in 1983, 1986, and 1989), or that, if the
board did grant him such initial parole, he would have
subsequently failed to accrue good-time credits, and (2)
that, under the new policy, the Parole Board would then grant
him "real" parole into society at large in 2001
(notwithstanding that the board would not even have granted
Hamm parole from his life sentence into his lengthy from-and-
after sentences on at least three prior occasions). In other
words, the Parole Board would have to deny Hamm's request for
parole from one lengthy sentence into another at least three
times, but then, a short time later, be willing essentially
to grant Hamm a complete release from prison. The inherent
contradiction in such assumptions discloses the difficulty of
quantifying such a "benefit," or even determining whether one
genuinely exists. Thus, I believe that any benefit
engendered by the 1977 aggregation policy is much too
-43-
43
speculative to serve as an effective counterweight to its
real risk of harm.24
II.
II.
In sum, I believe that the 1977 aggregation policy
is a "law" subject to ex post facto limitation, and that the
policy produces a risk of increasing the measure of
punishment sufficient to violate the Ex Post Facto Clause of
the Constitution. Accordingly, I dissent from parts II.B.-
III.
24. My colleagues find further support in the Seventh
Circuit's recent decision in United States v. McGee, 60 F.3d
1266, 1271 (7th Cir. 1995). In McGee, the Seventh Circuit
rejected an ex post facto challenge to a sentencing provision
that substituted a mandatory range of 21 to 24 months in
place of an open-ended 12 month minimum sentence for the
offense of conviction. Id. I find the analogy inapt because
in McGee the district court had actually sentenced the
defendant to the maximum 24 months under the new sentencing
range prior to the Seventh Circuit's review. Thus, the
Seventh Court could fairly quantify both the benefit and the
harm produced by the new sentencing range. Significantly,
the fact that the district court had sentenced the defendant
to the maximum possible under the new sentencing scheme
clearly suggested that, if anything, it would have given the
defendant a higher, not lesser, sentence under the old
scheme. In our case, however, Hamm received no analogous
treatment. He received no parole hearing. Thus, instead of
fairly quantifying the risk as the McGee court did, we can
only speculate as to the effect the 1977 aggregation policy
will ultimately have on his sentence.
-44-
44