Charlemagne v. Board of Parole

No. 200 March 18, 2026 769

IN THE COURT OF APPEALS OF THE

STATE OF OREGON

ANTONIO ALEJANDRO CHARLEMAGNE,

Petitioner,

v.

BOARD OF PAROLE AND POST-PRISON

SUPERVISION,

Respondent.

A183595

Argued and submitted February 19, 2026.

Stacy M. Du Clos, Deputy Public Defender, argued the

cause for petitioner. Also on the briefs was Ernest G. Lannet,

Chief Defender, Criminal Appellate Section, Oregon Public

Defense Commission. Antonio A. Charlemagne filed the

supplemental briefs pro se.

Rolf C. Moan, Assistant Attorney General, argued the

cause for respondent. Also on the brief were Dan Rayfield,

Attorney General, and Benjamin Gutman, Solicitor General.

Before Tookey, Presiding Judge, Chief Lagesen, Judge,

and Kamins, Judge.

KAMINS, J.

Reversed and remanded.

Lagesen, C. J., concurring.

770 Charlemagne v. Board of Parole

KAMINS, J.

Petitioner seeks judicial review of a final order of

the Board of Parole and Post-Prison Supervision (the board)

deferring his projected parole release date for 24 months.

The board determined that petitioner suffers from a pres-

ent severe emotional disturbance (PSED) that makes him

a danger to the health or safety of the community, so as to

permit the deferral of petitioner’s parole release date under

ORS 144.125(3) (1983), amended by Or Laws 1987, ch 320

§ 53; Or Laws 1989, ch 790 § 68; Or Laws 1993, ch 334, § 1;

Or Laws 1999, ch 141, § 1; Or Laws 2009, ch 660, § 3.1 We

agree with petitioner that the board’s order lacked substan-

tial evidence and substantial reason, ORS 144.335(3); ORS

183.482(8), and thus reverse and remand.2

Petitioner was convicted in 1985 of murdering his

mother and assaulting his father in a violent attack and

was sentenced to life with the possibility of parole. When

he first arrived in prison, petitioner was given a projected

release date of April 2000. That date was pushed back, sev-

eral times, based on the board’s conclusion that petitioner

suffered from a PSED that rendered him dangerous.

In the order giving rise to petitioner’s appeal, the

board reached the same conclusion—that petitioner suffers

from a PSED that renders him dangerous—relying heavily

on petitioner’s most recent forensic mental health evalua-

tion, written by Dr. Millkey. That report evaluated petitioner

based on, among other things, an interview, Department

of Corrections records, a presentencing investigation from

1985, petitioner’s prior mental health evaluations, petition-

er’s parole plan, and specific tests related to petitioner’s per-

sonality and risk of violence.

In a section titled “DIAGNOSIS,” Millkey wrote:

1

ORS 144.125(3) (1983) provided:

“If a psychiatric or psychological diagnosis of present severe emotional

disturbance such as to constitute a danger to the health or safety of the com-

munity has been made with respect to the prisoner, the board may order the

postponement of the scheduled parole release until a specified future date.”

2

Our disposition obviates the need to address petitioner’s remaining assign-

ments of error, which concern circumstances in his parole hearing and the board’s

authority to defer his release date.

Cite as 347 Or App 769 (2026) 771

“DSM-5 DIAGNOSIS

“Antisocial Personality Traits

“Cannabis Use Disorder

“Alcohol Use Disorder.”3

Millkey did not provide any other diagnoses. In that same

section, however, Millkey explained his diagnostic reasoning:

“[Petitioner] has antisocial personality traits, although

it is not clear that he meets criteria for a personality

disorder. He has a history of failure to conform to social

norms, aggressiveness, impulsivity, and irresponsibility.

[Petitioner] has a history of behavioral difficulties in prison

resulting in placement in segregated housing, although he

has not had a disciplinary referral since 2017. However,

[petitioner’s] cognitive style suggests that he is prone to not

accepting full responsibility and attributing his behavior to

externalities (e.g., blaming his aggressive behaviors on the

actions of others).”

Millkey also opined that petitioner’s risk for future violence

can be managed in the community with a risk management

plan. After a hearing, the board deferred petitioner’s release

for 24 months. In its order deferring petitioner’s release, the

board explained that, “[b]ased on the doctor’s report and

diagnosis, coupled with all the information that the [b]oard

is considering, the [b]oard concludes that [petitioner] suf-

fers from a [PSED] that constitutes a danger to the health

or safety of the community.” The board further explained

that its decision was, in part, based on Millkey’s psycholog-

ical report which “diagnosed [petitioner] with the DSM-5

Diagnosis of Antisocial Personality Traits,” and that it was

petitioner’s diagnosis, coupled with petitioner’s history of

violence and lack of insight, remorse, or empathy, that led it

to conclude that petitioner was suffering from a PSED that

rendered him dangerous.

Petitioner requested administrative review of the

board’s decision. In his request, petitioner argued that

substantial evidence did not support the PSED finding,

because Millkey only diagnosed petitioner with “traits” and

3

On appeal, the board acknowledges that it did not consider the diagnoses of

Cannabis Use Disorder or Alcohol Use Disorder as bases for a PSED.

772 Charlemagne v. Board of Parole

not a personality “disorder.” The board responded that the

PSED determination is a legal determination for the board

to make, and that, although a psychiatric or psychological

diagnosis is a prerequisite to its consideration, the diagno-

sis alone does not dictate the eventual result. See Weidner

v. Armenakis, 154 Or App 12, 17, 959 P2d 623 (1998), with-

drawn by order, reaffirmed in Merrill v. Johnson, 155 Or App

295, 964 P2d 284, rev den, 328 Or 40 (1998) (so stating).4

The board affirmed its decision, and this petition for judicial

review followed.

We review the board’s order for legal error, Morrison

v. Board of Parole, 277 Or App 861, 863, 374 P3d 948,

rev den, 360 Or 465 (2016), and substantial evidence, ORS

144.335(1), (3); ORS 183.482(8), including substantial rea-

son, Jenkins v. Board of Parole, 356 Or 186, 195, 335 P3d

828 (2014). “Substantial evidence exists to support a finding

of fact when the record, viewed as a whole, would permit

a reasonable person to make that finding.” ORS 183.482(8)

(c). In reviewing for substantial evidence, we defer to the

board’s reasonable inferences and do not reweigh the evi-

dence. Mendacino v. Board of Parole, 287 Or App 822, 834,

404 P3d 1048 (2017), rev den, 362 Or 508 (2018). Substantial

reason, on the other hand, requires that the board “articu-

late a ‘rational connection between the facts and the legal

conclusions it draws from them.’ ” Jenkins, 356 Or at 195

(quoting Ross v. Springfield School Dist. No. 19, 294 Or 357,

370, 657 P2d 188 (1982)).

On appeal, petitioner reiterates his argument

that the board’s PSED conclusion lacked substantial evi-

dence and substantial reason because it was missing a

necessary fact: a psychiatric or psychological diagnosis

of a disorder. We agree. Although the PSED standard is

a legal determination, Weidner, 154 Or App at 17 n 2, the

diagnosis required to support it is a factual one. See State

v. Meighan, 324 Or App 136, 150, 525 P3d 78, rev den, 371

Or 175 (2023) (noting that, under certain circumstances,

diagnosis involves a “complex factual determination”). A

4

Although Merrill is the precedential case, as has been the practice of the

appellate courts, we refer to and cite Weidner, because that is the case in which

we first announced our reasoning. Gordon v. Board of Parole, 266 Or App 405,

408 n 4, 338 P3d 185 (2014), rev den, 357 Or 299 (2015)

Cite as 347 Or App 769 (2026) 773

psychiatric or psychological diagnosis of a disorder is a “pre-

requisite” from which the board draws its conclusion of a

PSED. Christenson v. Thompson, 176 Or App 54, 59, 31 P3d

449 (2001). “Diagnosis” is the “ ‘art or act of identifying a

disease from its signs and symptoms.’ ” Id. (quoting Webster’s

Third New Int’l Dictionary 622 (unabridged ed 1996) in

interpreting ORS 144.125(3)). “The standard and commonly

recognized psychiatric and psychological diagnoses are set

forth in the American Psychiatric Association’s Diagnostic

and Statistical Manual of Mental Disorders (DSM) * * *.”

Weidner, 154 Or App at 17.

As mentioned, Millkey included the words “Antisocial

Personality Traits” under the heading of “DSM-5 Diagnosis,”

but also explained that he was unable to diagnose petitioner

with a personality disorder, because petitioner did not meet

the criteria for a personality disorder. Under the DSM-5,

personality traits are different from disorders. See DSM-5

at 647 (defining “personality traits” as “enduring patterns

of perceiving, relating to, and thinking about the environ-

ment and oneself that are exhibited in a wide range of social

and personal contexts”). Personality traits constitute disor-

ders “only when [the traits] are inflexible and maladaptive

and cause significant functional impairment or subjective

distress.” Id. (emphasis added). In other words, while per-

sonality traits may be a necessary component of a disorder,

they do not, standing alone, constitute a disorder. Millkey’s

report thus lacked the necessary prerequisite for the board’s

conclusion. See also Newcomb v. Thompson, 176 Or App 167,

168, 33 P3d 319 (2001) (psychological report that diagnosed

the plaintiff merely with “some elements” of a personality

disorder insufficient to support a PSED conclusion). Or, put

differently, the board legally erred by interpreting our case

law to mean that a psychological diagnosis of “Antisocial

Personality Traits” was sufficient to support a PSED con-

clusion. The PSED finding was thus not supported by sub-

stantial evidence and substantial reason.

The board, however, raises two substantively new

arguments on appeal as to why we should affirm its order.

First, the board argues that the record entitled it to find

that Millkey actually diagnosed petitioner with a different

774 Charlemagne v. Board of Parole

disorder, Unspecified Personality Disorder, notwithstand-

ing the fact that Millkey did not include that diagnosis in

his report. That disorder, as defined by the DSM-5, applies

when there are symptoms present of a personality disorder,

but a person’s characteristics nonetheless do not meet “the

full criteria” for a particular personality disorder and the

clinician chooses not to specify the reason that the criteria

are not met. Second, the board argues that case law requir-

ing a DSM diagnosis of a disorder by a psychiatrist or psy-

chologist as a prerequisite for a PSED conclusion is plainly

wrong and should be overruled.

We begin with the board’s argument that the board

was entitled to read Millkey’s “DSM-5 Diagnosis” as “the

equivalent” of an Unspecified Personality Disorder diagno-

sis. That argument is unpersuasive, as it fails to grapple

with our scope of review. On judicial review, we look to the

order itself for substantial evidence and substantial reason.

See Washington v. Board of Parole, 294 Or App 497, 499, 432

P3d 372 (2018) (“The issue in this case, then, is: Does the

board’s order contain a logical explanation of the connection

between adequately supported facts and the board’s conclu-

sion that [the] petitioner has a psychological condition that

is present, severe, and makes him a danger to the health or

safety of the community?” (Emphasis added.)). Nowhere in

the board’s order did the board explain that it was finding

that petitioner had a diagnosis of Unspecified Personality

Disorder, or that it was concluding that petitioner suffered

from a PSED that rendered him dangerous due to a diag-

nosis of Unspecified Personality Disorder. Rather, the only

psychiatric diagnosis identified by the board to support the

legal finding of a PSED was “Antisocial Personality Traits,”

not Unspecified Personality Disorder. And, the board

expressly relied on Millkey’s report—which, in turn, explic-

itly provided that it could not diagnose petitioner with any

personality disorder. The board’s argument thus fails.

The board’s argument requesting that we overrule

our case law requiring a psychological or psychiatric diagno-

sis as a prerequisite is similarly unavailing. In order to over-

rule our precedents, we must be convinced that those cases

are “plainly wrong,” which is “a rigorous standard, satisfied

Cite as 347 Or App 769 (2026) 775

only in exceptional circumstances.” State v. Civil, 283 Or App

395, 417, 388 P3d 1185 (2017). The board contends that our

previous decision in Weidner did not undertake a proper stat-

utory interpretation analysis, and had we then looked to the

text, context, and legislative history of ORS 144.125(3), we

would have found no requirement of a psychological or psy-

chiatric diagnosis as a prerequisite. The board also contends

that the language in Weidner was dicta and subsequent

cases that rely on it for a holding are plainly wrong.

In Weidner, the issue was whether ORS 144.125(3)

contained a requirement that a psychiatrist or psychologist

diagnose a petitioner with a “severe emotional disorder”

before the board could defer release on the basis of a PSED.

155 Or App at 19. We concluded that the statute contained no

such requirement. Id. at 19-20. We reasoned that “the text of

ORS 144.125(3) (1991),5 particularly when read in the context

of other related statutes” contemplated that the board could

“consider all information,” and not just the psychiatric or psy-

chological diagnosis, “in reaching a conclusion about whether

a prisoner suffers from a [PSED].” Id. at 17-18 (emphasis

added). We noted, however, that “[a]lthough a psychiatric or

psychological diagnosis is a prerequisite to the [b]oard’s con-

sideration of whether the statutory criteria have been met,

that diagnosis alone does not dictate the result.” Id. at 19-20.

Even if that language from Weidner could be construed as

dicta, the board has not persuaded us that its reasoning is

plainly wrong. See Engwiler v. Persson / Dept. of Corrections,

354 Or 549, 559 n 6, 316 P3d 264 (2013) (“Of course, even if

the court’s prior construction were dictum that would not,

by itself, mean that it was incorrect and without any force

whatsoever. It merely [would mean] that we are not required

to follow it as precedent. The prior construction, even if dic-

tum, could have persuasive force because of the soundness of

its reasoning.” (Internal quotation marks omitted.)).

As we have explained, the board legally erred in its

interpretation of our case law, and its PSED conclusion was

not supported by substantial evidence or substantial reason.

That means that petitioner is entitled to a reinstatement

5

Although Weidner interpreted a more recent version of ORS 144.125(3), the

statutory language it interpreted is identical to the version applied to petitioner.

776 Charlemagne v. Board of Parole

of his release date, as the board cannot demonstrate that,

at the time of his hearing, petitioner suffered from a PSED

that rendered him dangerous. See Porter v. Board of Parole,

281 Or App 237, 245, 383 P3d 427 (2016), rev den, 361 Or

100 (2017) (observing that after a release date has passed,

“it does not suffice to later find a reason to have postponed [a

petitioner’s] release”); Rivas v. Persson, 256 Or App 829, 835-

36, 304 P3d 765 (2013), rev dismissed, 354 Or 841 (2014) (“[I]f

a release date was scheduled and elapsed without the board

first having found a valid reason to postpone release, but the

inmate was erroneously not released, later events cannot

furnish a basis for postponing release; the inmate is enti-

tled to immediate release.” (Emphasis added.)). Accordingly,

there is no basis on which to defer petitioner’s release date

associated with the sentence giving rise to this appeal. We

therefore reverse and remand with directions to reinstate

petitioner’s release date.

Reversed and remanded.

LAGESEN, C. J., concurring.

I concur in the majority opinion fully. I write sep-

arately to address the board’s assertion that we should

reconsider and overrule Weidner v. Armenakis, 154 Or App

12, 959 P2d 623 (1998) (en banc), withdrawn by order, reaf-

firmed in Merrill v. Johnson, 155 Or App 295, 964 P2d 284

(1998); Christenson v. Thompson, 176 Or App 54, 31 P3d 449

(2001); and Newcomb v. Thompson, 176 Or App 167, 33 P3d

319 (2001). The board asserts that we should undertake that

exercise because we did not employ the statutory construc-

tion methodology announced in PGE v. Bureau of Labor

and Industries, 317 Or 606, 859 P2d 1143 (1993), and State

v. Gaines, 346 Or 160, 206 P3d 1042 (2009). In the board’s

view, an application of that statutory construction method-

ology to the relevant version of ORS 144.125 demonstrates

that our conclusions in those cases are plainly wrong, such

that the right thing to do is overrule them.

I tend to agree with the board that the construction

of ORS 144.125 adopted by the en banc majority opinion in

Weidner—and built upon by Christenson and Newcomb—

might be wrong. The dissenting opinion’s analysis in Weidner

Cite as 347 Or App 769 (2026) 777

suggests as much. Weidner, 154 Or App at 20-23 (Warren, J.,

dissenting).1 That analysis rested on the ordinary meaning

of the plain text of ORS 144.125 and, for that reason, has

the hallmarks of a typical PGE analysis. But the Weidner

majority opinion has been the settled law of this court for

28 years, and I am unpersuaded that there is a reason to

devote our limited institutional resources to the destabiliz-

ing exercise of re-evaluating our longstanding precedent.

That is particularly so in a case like this one in which our

predecessors devoted the resources of the en banc court to

resolving the question—twice.

I recognize that the proper construction of the stat-

ute is important. The proper construction of the statute at

issue here is particularly important because it affects both

liberty and, potentially, public safety. But we are Oregon’s

intermediate appellate court, and we were not built to be

the last word on the matter, or on any important question of

law. Our role, to the extent we are called upon to announce

law in addition to correcting errors, is to adjudicate thought-

fully, fairly, and impartially the legal questions presented to

us, and then to apply the precedent we set in an even-handed

way. Once we have issued precedent providing an authorita-

tive interpretation or construction of an important statute,

like the one at issue here, ordinarily it is our responsibility

to apply it to similarly-situated cases, and it is the respon-

sibility of the Supreme Court—our state’s primary law-an-

nouncing court—to address whether our interpretation or

construction is the correct one.

At the time we decided Weidner, the Supreme Court

appears to have viewed its role and responsibility the same

way. The Court allowed review, ultimately vacated our deci-

sion without addressing the merits, and remanded with

directions to dismiss because the case had become moot.

Weidner v. Armenakis, 327 Or 317, 966 P2d 220 (1998). Our

en banc court almost immediately readopted the reasoning

1

I note that if we were to overrule Weidner and apply the construction of the

statute advanced by the dissenting opinion in that case (a construction that the

board does not advocate), the result in this case would be the same. Under the

analysis in the Weidner dissenting opinion, the fact that there is no psychiatric

or psychological diagnosis of a present severe emotional disturbance such as to

constitute a danger to the health or safety of the community would preclude the

board from deferring petitioner’s release date. Id. at 21 (Warren, J., dissenting).

778 Charlemagne v. Board of Parole

of the vacated decision, dividing in exactly the same way. See

Merrill, 155 Or App at 296. That time, though, the Supreme

Court declined to take on the question. Merrill v. Johnson,

328 Or 40, 977 P2d 1170 (1998). The Court did so, notwith-

standing the fact that the matter satisfied the lion’s share

of the Court’s criteria for allowing discretionary review

under ORAP 9.07 both then and now, making it a textbook

case for discretionary review by a law-announcing court.

ORAP 9.07 (adopted effective 1994); ORAP 9.07 (effective

January 1, 2025). The case presented an important recur-

ring question of law involving the interpretation of a statute

that would apply to a lot of people—one of first impression

for the Supreme Court—that had been decided (possibly

incorrectly) in a precedential opinion by a divided en banc

Court of Appeals. ORAP 9.07(1)(b), (2), (3), (4), (5), (7), (8),

(11), (12), (13), (14). The Supreme Court’s exercise of discre-

tion not to consider the issue means our court has had the

last word on the meaning of this important statute for close

to three decades.

The Supreme Court’s discretionary decision not to

step in to address an important question of statutory inter-

pretation within a reasonable timeframe has costs to the

orderly administration of justice. In this case, it means that

some people may have been erroneously deprived of liberty

or, if the board is correct in its reading, that dangerous peo-

ple may have been paroled prematurely.

Beyond that, when a state’s law-announcing court

does not timely review an intermediate appellate court’s

interpretation of a statute that will govern many cases, it

allows for practices to develop and entrench around possibly

incorrect interpretations of statutes that, as a matter of sys-

tem design, were not intended to be final. That, in turn, has

the potential to give rise to long-term practices that may run

counter to the rule of law. See, e.g., State v. Hubbell, 314 Or

App 844, 848, 500 P3d 728 (2021), aff’d, 371 Or 340 (2023)

(concluding that, as a result of Court of Appeals’ mistaken

interpretation of statute, attempt crimes had wrongfully

been treated as completed offenses); State v Herried, Gray

& Clagget, 3 Or App 462, 464, 474 P2d 358 (1970) (constru-

ing phrase “memorandum decision” in ORS 19.435 (formerly

Cite as 347 Or App 769 (2026) 779

ORS 19.180) to permit the issuance of decisions unsupported

by articulated reasons, stating that doing so was “a practice

which is becoming increasingly widespread among appel-

late courts of other jurisdictions”); Aliza Milner, Written

Opinions in State Intermediate Appellate Courts: Current

Landscapes and the AI Horizon, 38 Geo J Legal Ethics 273,

282-83 (2025) (noting that 33 states affirmatively require

explanatory writings in appeals, and identifying Oregon as

one of only nine states that allow for intermediate appellate

courts “to decide at least certain appeals without explana-

tory writing”). And, one way or another, the longer a state’s

law-announcing court waits to address an important ques-

tion of state law, the more likely it is that people, as a practi-

cal matter, will have come to rely on the intermediate court’s

decision as settling the law. That means the law-announc-

ing court’s delay risks creating disruption that could have

been avoided by prompter institutional engagement.

Other states have mechanisms in place to safeguard

against this. New Jersey, for example, provides for Supreme

Court review as a matter of right when there is a dissent

in the intermediate appellate court. NJ RAR 2:2-1(a)(2).

In Idaho, where appeals are filed in the Supreme Court

and then transferred to the Court of Appeals, the Supreme

Court is ordinarily expected to retain cases “in which there

are significant issues involving clarification or development

of the law, or which present a question of first impression.”

Idaho Appellate Rule 108(b). Oregon would benefit from dis-

cussing something similar, so as to better ensure the timely

ultimate resolution of important legal questions by our

state’s law-announcing court.