Retaliation Against People Who Speak Up

Overview

Retaliation against incarcerated people who report abuse, file grievances, or speak to outside parties is one of the most pervasive and structurally documented patterns in the U.S. carceral system, and Georgia’s prison system is not exempt. The First Amendment doctrine theoretically permits §1983 retaliation claims, but the Prison Litigation Reform Act’s exhaustion regime creates a structural trap: the same officials who allegedly retaliate also administer the grievance process that incarcerated people must complete before they can sue. Forms of retaliation range from punitive transfers and administrative segregation to falsified disciplinary tickets, withheld medical care, “lost” grievances, and physical violence — direct or coordinated. Empirical research shows post-PLRA collapse in plaintiff success rates and limited oversight against retaliation; reform models from other states (corrections ombudsmen, anonymous reporting, body‑worn cameras) demonstrate what’s possible.

Georgia’s own data and federal findings reveal that retaliation is not merely an incidental risk but is built into the sexual‑violence reporting apparatus. Across 2014–2024, Georgia prisons logged 15,542 PREA allegations; only 543 were substantiated — an aggregate rate of 3.5% ((GPS Sexual Violence & PREA Compliance in Georgia Prisons)). In 2022 alone, GDC reported 635 sexual‑abuse allegations, yet the substantiation rate remained similarly negligible, and the DOJ found that underreporting is driven by fear of retaliation or inaction ((DOJ Findings Report, October 2024; Collection: The Grievance Machine)). The U.S. Department of Justice, after an investigation that visited 17 GDC facilities, concluded in October 2024 that sexual assault is “rampant” and that Georgia’s prison system “does not reasonably protect incarcerated individuals, including LGBTI individuals,” while the PREA audit process has produced a perfect 100% compliance record for every GDC facility audited — a contradiction that underscores a system in which reporting abuse invites suppression rather than safety ((GPS Sexual Violence & PREA Compliance in Georgia Prisons)). The DOJ further found that incarcerated people “do not always report incidents because they do not expect staff to take any action in response” and that GDC failed to protect those who cooperated with the DOJ investigation from ongoing retaliation ((DOJ Findings Report, October 2024)). The sheer volume of settled claims paid by Georgia — $50,633,556 across at least 261 cases — signals a floor, not a total, for the harm that flows from a system in which the grievance machinery too often buries rather than remedies constitutional violations ((Collection: The Grievance Machine)).

Segregation is the mechanism this page has to describe. Punitive segregation is the most consequential instrument Georgia prison staff hold over a person in custody, and GDC’s Tier system is its architecture. Tier I is short-term disciplinary, protective-custody and transient segregation available at every GDC facility type — state prisons, county correctional institutions, private prisons, intensive treatment facilities, transitional centers, probation detention centers and probation boot camps; it is not confined to close-security prisons or a designated subset of sites ((GDC SOP 209.07 §II, eff. 04/30/2015)). Tier II is long-term administrative segregation with no outer time limit written into policy ((GDC SOP 209.08, eff. 04/11/2016)). Tier III is the Special Management Unit at the Georgia Diagnostic and Classification Prison, a minimum 390-day, five-phase program with a 24-month ceiling the Department reserves the right to exceed on six enumerated grounds, subject to a four-member statewide panel and the Commissioner’s personal approval ((GDC SOP 209.09 §§IV.A.1.a, IV.A.1.f, IV.A.1.h–i, eff. 04/23/2025)). Two step-down programs, Tier II STEP and Tier III STEP, sit beneath them ((GDC SOP 209.45; GDC SOP 209.55)). GDC converted 2,721 beds to Tier beds at all Level V facilities when it implemented the system in FY2014 — the only system-wide quantification of the Tier system GDC has ever published, and it is a bed count, not a population count ((GDC Annual Fiscal Report 2014)). GDC has acknowledged operating a dedicated SCRIBE module to track people in extended restrictive housing since March 2019, and a Warden’s Dashboard field for the number of days each Tier II offender has spent in restrictive housing since FY2020 — so a statewide Tier record exists inside the agency ((GDC Annual Fiscal Report 2019; GDC Annual Fiscal Report 2020)).

What this page does not claim. Segregation placement is among the methods most frequently alleged in retaliation accounts, and the Tier records are where a retaliation claim would be proved or disproved. But GPS holds no document establishing that Georgia’s Tier programs are used to retaliate against people who file grievances, contact lawyers or the press, report staff misconduct, or otherwise speak out, and no court has made a finding that Georgia uses Tier placement to retaliate for protected activity ((GPS #19312)). GPS’s case-law corpus is post-conviction and habeas-weighted and does not hold §1983 conditions litigation, so its silence is not itself evidence of absence. The sections below document the machinery, the rules and the records; they do not assert a proven pattern of retaliatory use.

Section 1 — The legal frame

Retaliation against incarcerated people who report abuse, file grievances, or speak to outside parties is a First Amendment injury. The doctrine is straightforward on paper and grueling in practice. To prevail under 42 U.S.C. § 1983 — the statutory vehicle for suing state and local officials, including state prison staff — an incarcerated plaintiff must establish three elements: (1) the underlying speech or petition was constitutionally protected; (2) the official’s conduct was an adverse action that “would likely deter a person of ordinary firmness from the exercise of First Amendment rights”; and (3) a causal link between the protected activity and the adverse action ((CourtListener, https://www.courtlistener.com/opinion/77110/danny-m-bennett-v-dennis-lee-hendrix/)). When the defendant is a federal officer, the same theory survives — if at all — through a Bivens action, but the Supreme Court has narrowed Bivens aggressively over the last two decades, and prison‑retaliation claims rarely fit its surviving categories.

The Eleventh Circuit, which controls federal litigation arising out of Georgia, Florida, and Alabama, formally adopted the “ordinary firmness” objective test in Bennett v. Hendrix, 423 F.3d 1247 (11th Cir. 2005), framing it as an issue of first impression and aligning the circuit with the Sixth, Second, and Fourth Circuits ((CourtListener, https://www.courtlistener.com/opinion/77110/danny-m-bennett-v-dennis-lee-hendrix/)). Six years later, in O’Bryant v. Finch, 637 F.3d 1207 (11th Cir. 2011), the court applied that test specifically to prisoner‑on‑officer retaliation but added a doctrinal trapdoor: if a disciplinary report is even partially valid, it can break the causal chain, making retaliation claims extremely difficult to sustain even if the officer’s action was motivated by protected speech.

Yet the most formidable barrier is not the First Amendment merits test but the exhaustion requirement imposed by the Prison Litigation Reform Act. Under 42 U.S.C. § 1997e(a), “[n]o action shall be brought with respect to prison conditions under section 1983 … by a prisoner confined in any jail, prison, or other correctional facility until such administrative remedies as are available are exhausted.” The Supreme Court in Woodford v. Ngo, 548 U.S. 81 (2006), held that the PLRA demands “proper exhaustion” — compliance with all procedural deadlines and steps the prison prescribes. Jones v. Bock, 549 U.S. 199 (2007), makes that prison’s own rules the boundary of what counts as proper exhaustion, effectively granting corrections agencies the power to define the exhaustion standard. Ross v. Blake, 578 U.S. 632 (2016), created a narrow safety valve: a prisoner need exhaust only “available” remedies, and a remedy is unavailable when it operates as a dead end, is “so opaque that it becomes, practically speaking, incapable of use,” or “when prison administrators thwart inmates from taking advantage of a grievance process through machination, misrepresentation, or intimidation.” The Supreme Court recently held in Perttu v. Richards, 605 U.S. 460 (2025), that parties are entitled to a jury trial on PLRA exhaustion when the issue is intertwined with the merits of a claim triable to a jury under the Seventh Amendment.

The Eleventh Circuit’s Georgia‑rooted exhaustion jurisprudence shows both the potential and the fragility of the unavailability doctrine. Turner v. Burnside, 541 F.3d 1077 (11th Cir. 2008), a Georgia case, established a two‑step framework and held that “a prison official’s serious threats of substantial retaliation against an inmate for lodging or pursuing in good faith a grievance makes the administrative remedy ‘unavailable.’” Yet Johnson v. Meadows, 418 F.3d 1152 (11th Cir. 2005), affirmed dismissal of a GDC prisoner’s claim because an untimely grievance means “the inmate … has spurned the administrative process.” The result is a minefield: a prisoner who fears retaliation enough to delay filing may be barred for untimeliness, while one who files promptly risks the retaliation the PLRA framework purports to prevent.

Subsequent Eleventh Circuit decisions illuminate the procedural gauntlet. Whatley v. Warden, Ware State Prison, 802 F.3d 1205 (11th Cir. 2015) (Whatley I), reversed a dismissal because the district court improperly enforced a procedural rule without applying the Turner two‑step; Whatley v. Smith, 898 F.3d 1072 (11th Cir. 2018) (Whatley II), held that a prison waives its exhaustion defense when the warden reaches the merits of an appeal. Geter v. Baldwin State Prison, 974 F.3d 1348 (11th Cir. 2020), remanded for a prisoner with severe mental and physical disabilities because the grievance process was unavailable when the prisoner could not understand or navigate it. McIlwain v. Burnside (11th Cir. 2020) ordered an evidentiary hearing after five grievances were allegedly ignored. More recently, Brinson v. Gillison (S.D. Ga. Mar. 22, 2024), dismissed a GDC prisoner’s action after Ross and Geter, because the prisoner failed to argue the process was unavailable. The district‑court casualties are legion: Varner v. Shepard, 852 F. App’x 411 (11th Cir. 2021), dismissed a claim with prejudice for no timely grievance; Major v. Toole, No. 5:15‑cv‑483 (M.D. Ga. Feb. 27, 2017), dismissed an excessive‑force claim because the timely grievance mentioned only confiscation of property, not force; Nolley v. Warden, Macon State Prison (11th Cir. 2020), affirmed dismissal for lack of any grievance evidence.

The PLRA adds further structural penalties that chill access to courts. The three‑strikes provision, 28 U.S.C. § 1915(g), bars in forma pauperis status after three dismissals for frivolousness, maliciousness, or failure to state a claim — a rule that can be triggered even by a procedurally defective exhaustion if the court deems the claim meritless. Attorney’s fees under 42 U.S.C. § 1997e(d) are capped at 150% of the appointed‑counsel rate, and up to 25% of any monetary judgment goes to satisfy fees — a dynamic that deters experienced civil‑rights counsel from taking prison cases. Prisoners must pay the full $405 civil filing fee (28 U.S.C. § 1915(b)), a prohibitive sum for a prison‑wage population. The physical‑injury requirement of § 1997e(e) bars recovery for mental or emotional injury without a prior showing of physical injury, further limiting meaningful relief. Crucially, exhaustion is a matter in abatement — a precondition, not a jurisdictional bar (Bryant v. Rich, 530 F.3d 1368 (11th Cir. 2008)) — and an exhaustion dismissal does not adjudicate whether the beating, the denied medical care, or the unsafe conditions occurred. Yet these procedural defaults are the end of the road for the overwhelming majority of incarcerated Georgians who file suit; nationally, 95.6% of inmate civil‑rights filings are pro se, and most exhaustion dismissals occur in unpublished magistrate‑judge reports that never appear in searchable databases, systematically undercounting the legal dead‑ends created by the PLRA ((Collection: The Grievance Machine, Data Gap note)).

Indefinite segregation and the due-process frame

Where segregation is the adverse action, the constitutional question is not only First Amendment causation but Fourteenth Amendment liberty. Wilkinson v. Austin, 545 U.S. 209 (2005), held that Ohio’s supermax conditions — near‑total solitary confinement, indefinite duration, and parole disqualification — imposed an “atypical and significant hardship” giving rise to a Fourteenth Amendment liberty interest requiring informal, non‑adversary procedural protections. The Court did not reach the Eighth Amendment. Applied to Georgia, the factor Wilkinson identified as central — indefinite duration — is present in Tier II, where SOP 209.08 sets no outer limit, and absent from Tier III, where SOP 209.09 sets a 24‑month ceiling with enumerated exceptions and a quarterly panel. On Wilkinson’s own reasoning the due‑process argument is therefore stronger for Tier II than for the program the litigation actually reformed. This is an argument from a holding, not a holding: no court has applied Wilkinson to Georgia’s Tier II program ((GPS #19327)).

The durations matter to that frame. Tier I is capped at 30 days “to include investigation and disciplinary isolation time,” with the cap expressly disapplied for both voluntary and involuntary protective custody — so a person placed in protective custody by staff may be held with no policy time limit at all ((GDC SOP 209.07 §VI.H.2)). Tier II contains no maximum duration; a person who “cannot return to general population because they pose a serious threat to the safety and security of the institutional operation” remains in Phase 3 indefinitely, subject only to a 90‑day review that may confirm retention ((GDC SOP 209.08 §§IV.A.3, IV.F)). Tier III sets a stated minimum of 390 days — 60 in E‑Wing, 60 in F‑Wing, 90 in D‑Wing, 90 in C‑Wing, 90 in B‑Wing — a person who progresses successfully through every phase still serves the floor ((GDC SOP 209.09 §IV.A.1.a)). The 24‑month ceiling is stated and then immediately qualified: the Department “reserves the right to retain offenders … for a longer period of time” on six criteria, the last of which is that the person poses “such an exceptional, credible, and articulable risk … that no facility other than the Tier III Program facility is sufficient to contain the risk” ((GDC SOP 209.09 §IV.A.1.f)). Countervailing safeguards on that override are real: quarterly review on a named form against eight enumerated criteria including the 60‑ or 90‑day mental‑health evaluation and total duration of SMU confinement; a four‑member statewide panel of the Director of Field Operations, the Statewide Mental Health Director, the Statewide Medical Director and a member of the Office of Legal Services; and personal approval by the Commissioner or Assistant Commissioner for Facilities ((GDC SOP 209.09 §§IV.A.1.h–j)). None of that exists anywhere in Tier II ((GPS #19330)).

Conditions of confinement are likewise unequal across the tiers, and the inequality runs the wrong way. Tier II’s out‑of‑cell floor is five hours per week of exercise, with no daily distribution specified — any lawful pattern across the week — which by arithmetic leaves 163 hours per week, or 23.3 hours per day, in cell as a policy floor; on the two days outside the entitlement, guaranteed out‑of‑cell time is zero ((GDC SOP 209.08 §IV.D.1.f; derived figure, GPS #19272)). The juvenile restrictive‑housing program (RHA‑JOAS) sets the identical five‑hours‑per‑week floor for people under 18 ((GDC SOP 209.11 §IV.E.1.g)). By contrast, Tier III guarantees four hours out of cell per day, Monday through Friday, of which at least one hour must be recreation, and Tier II STEP and Tier III STEP guarantee four hours out of cell seven days a week (though Tier II STEP expressly counts “shower, feeding” and similar routines toward the four hours) ((GDC SOP 209.09 §IV.E.1; GDC SOP 209.45 §IV.C.1; GDC SOP 209.55 §IV.C.1)). Tier II STEP therefore guarantees 5.6 times more out‑of‑cell time than Tier II itself, and Tier III four times more ((derived, GPS #19273)). Adults in Tier II are limited to non‑contact visits at every phase; juveniles in RHA‑JOAS receive contact visits ((GDC SOP 209.08 §IV.D.1.i; GDC SOP 209.11 §IV.E.1.j)). Tier II’s tablet is confiscated on placement until Phase III ((GDC SOP 204.10 §B.2)). The 2025 Tier III policy guarantees single‑occupancy cells in all five phases and single occupancy in Tier III STEP; Tier II STEP permits single or double occupancy; and SOP 209.08 is silent on cell occupancy altogether — while SOP 209.06, the generic administrative‑segregation policy, treats a single cell as an enumerated exception requiring assignment by medical, mental‑health or classification staff ((GDC SOP 209.09 §IV.E.1.a; GDC SOP 209.55 §IV.C.1.a; GDC SOP 209.45 §IV.C.1.a; GDC SOP 209.06 §IV.G)). On GDC’s own written minimums, and setting the derived floors against the Nelson Mandela Rules’ 22‑hour definitional threshold and 15‑day prolongation threshold, Tier I, Tier II, generic administrative segregation and the juvenile program all sit inside the Mandela definition of solitary confinement as a matter of policy floor, while the reformed unit sits outside it on weekdays and inside it on weekends ((UN Standard Minimum Rules for the Treatment of Prisoners, Rules 43–45; GPS #19274, #19329)). Two conditions of confinement are being run in the same program under the same rules — 23 hours a day alone in a cell, and 23 hours a day locked in a cell with another person — and GDC’s Tier II policy distinguishes neither, guarantees neither, and reports neither; whether Tier II meets Rule 44’s second element, absence of meaningful human contact, therefore turns on cell occupancy, which GDC does not specify for Tier II and does not publish for any facility ((GPS #19331)).

The mental‑health routing runs against the grain of need. Tier III excludes offenders at Mental Health Level III and Level IV, diverting them to specialized mental‑health treatment units; Tier II excludes only Level IV and admits Level III on mental‑health recommendation ((GDC SOP 209.09 §IV.B; GDC SOP 209.08 §IV.B). The written effect is that prisoners with the higher mental‑health classifications are routed away from the unit whose conditions were reformed under litigation and toward the unit with a five‑hour‑per‑week out‑of‑cell floor. GDC’s own mental‑health policy places Tier II in the highest‑need tier of its staffing hierarchy alongside death row, the SMU and acute care, and assigns it a 1:20 mental‑health counselor ratio ((GDC SOP 508.01 §§IV.C.1, IV.C.3.c.(5), eff. 07/27/2023)). Suicide‑risk questioning is monthly in Tier I and weekly in Tier II — the unit with the shorter maximum stay carries the less frequent screening requirement ((GDC SOP 219.01 §IV.A.3, eff. 11/15/2019)). Statewide officer training exists for the SMU/Tier III program and none for Tier I or Tier II, where training is delegated to each Warden ((GDC SOP 506.08 §G, eff. 05/27/2020)).

The litigation that produced these reforms is narrower than it is often described. Gumm v. Ford (later Gumm v. Jacobs; companion Daughtry v. Emmons), No. 5:15‑cv‑00041‑MTT‑CHW (M.D. Ga., Chief Judge Marc T. Treadwell), was filed 2015‑02‑12 as a handwritten pro se complaint and terminated 2019‑05‑07; Sarah Geraghty of the Southern Center for Human Rights was appointed counsel in 2016 and Kilpatrick Townsend & Stockton joined as co‑class counsel ((GPS #19321)). On April 19, 2024, the court found that GDC had “no desire or intention” to comply with the December 2018 settlement and ordered an independent monitor, fines of $2,500 per day — $75,000 every 30 days for six months — and additional attorney’s fees, extending the settlement past its initial three‑year term; the order also documents that officials falsified therapy records, including recording that a man attended treatment sessions after he was already dead ((GPS #19322)). At the 2017 inspection the SMU held approximately 180 people, of whom 70 were designated mentally ill; Timothy Gumm himself was held continuously for seven and a half years despite fourteen transfer recommendations over four years ((GPS #19326)). The court adopted Dr. Craig Haney’s expert findings that residents were confined in cells “smaller than the average parking space,” with as little as five hours per week out of cell, no outside light, no congregate religious or educational programming, and frequently locked food‑port flaps ((GPS #19326)). Every operative settlement term — the general 24‑month limit, the bar on housing Mental Health Level III or above, the 60‑ and 90‑day evaluations, the quarterly panel, the Commissioner‑level approval — is written in SMU and Tier III vocabulary, the class arose from a single facility, and GDC codified it into SOP 209.09 alone. SOP 209.08, governing Tier II, has none of it and has not been revised since 2016‑04‑11 ((GPS #19324)). A correction is recorded rather than applied: existing GPS datapoint 6598 describes the Haney findings as applying to “Tier II/SMU residents,” when the inspection and certified class were the SMU at GDCP; the five‑hours‑per‑week figure in the Gumm record is an SMU finding from 2017 that happens to coincide with the figure SOP 209.08 still sets as the Tier II floor today, and the two must not be presented as one ((GPS #19325)).

Section 2 — Georgia’s Grievance System and the PLRA Exhaustion Gauntlet

Georgia’s internal grievance machinery — codified in GDC Standard Operating Procedure 227.02 (effective May 10, 2019) — functions as the compulsory gatekeeper for every lawsuit against state prison officials. The sheer financial residue of harm that survives this gate underscores the stakes: Georgia has paid $50,633,556 across at least 261 settled claims, a floor rather than a total, because the amount does not capture claims defaulted at the exhaustion stage or those that never survived the procedural gauntlet ((Collection: The Grievance Machine)). The PLRA’s chilling effect is starkly visible in filing rates. Schlanger’s analysis of Federal Judicial Center data shows that nationally, the prisoner civil‑rights filing rate fell from 24.6 per 1,000 incarcerated people in FY1995 to 10.5 per 1,000 by FY2012 (revised to 10.1 in the 2022 update). In Georgia, the drop was even sharper: from 1,496 filings (22.6 per 1,000) in FY1995 to 984 (9.7 per 1,000) in FY2019, a 57% decline that far outpaces the national trend ((Collection: The Grievance Machine, citing Schlanger/FJC data)).

The grievance process itself is riddled with procedural tripwires. An incarcerated person must file an original grievance within 10 calendar days of the incident; the warden has 40 calendar days (plus one 10‑day extension) to respond; a Central Office appeal must be lodged within 7 calendar days of the warden’s response; and the Commissioner has 120 calendar days to decide. A prisoner is limited to two active grievances at a time; a third is closed unless one is dropped within 5 days. Emergency grievances — ostensibly for imminent harm — require a Duty Officer response within 48 hours and a written answer within 5 days. Yet the policies contain self‑defeating exceptions: the very device through which grievances are filed, the GOAL Device/JPay Kiosk (SOP 204.10), is listed as a non‑grievable item under SOP 227.02 § IV.B.2, which states that “Use of the Tablet is a privilege, not a right.” Loss of tablet access therefore eliminates the primary mechanism for filing a grievance, and that loss cannot itself be grieved — a tablet paradox that can render the administrative remedy practically unavailable from the start ((Collection: The Grievance Machine)). Further, a grievance alleging staff use of force that is deemed “Non‑Compliance” and forwarded to the Office of Professional Standards yields a final, non‑appealable decision, blocking the external review that exhaustion supposedly ensures.

Segregation is not grievable

SOP 227.02 states the anti‑retaliation rule and then removes the principal retaliatory act from the grievance system. Section IV.A.3 provides that “[r]etaliation against an Offender for filing a grievance is strictly prohibited,” including disciplinary action taken for filing; Section IV.B.3 provides that an offender “may file a grievance alleging retaliation or harassment, regardless of the form of the alleged retaliation or harassment.” But Section IV.B.2.d lists as non‑grievable “[i]nvoluntary assignments to Administrative Segregation,” routing the appeal to SOP 209.06 instead, and Section IV.B.2.g lists as non‑grievable “[h]ousing assignments, program assignments, security classifications or work assignments, unless there is an alleged threat to the Offender’s health or safety” ((GDC SOP 227.02 §§IV.A.3, IV.B.2.d, IV.B.2.g, IV.B.3; GPS #19319)). The practical result is that a person moved into administrative segregation after filing a grievance may file a grievance alleging retaliation in the abstract, but may not grieve the segregation assignment itself unless able to frame a threat to health or safety. The remedy is routed into the SOP 209.06 classification appeal, which tests whether the placement criteria are met — not motive — and which terminates with the Warden: under SOP 209.06 §IV.E.2 the initial 24‑hour placement decision “will not be subject to appeal by the offender,” and under §IV.F.5 “[t]he Warden’s/Superintendent’s review and decision on the appeal is final” ((GDC SOP 209.06 §§IV.E.2–3, IV.F.5; GPS #19294)). There is no neutral or external decision‑maker anywhere in that process and no right to counsel or to a non‑staff advocate.

The Tier policies compound the asymmetry at the point of entry and exit. A facility Classification Committee can place a person in Tier II on the Warden’s approval, and the Warden may authorize immediate assignment “[i]n emergency situations”; release, by contrast, requires a Tier II Unit Manager recommendation, Warden and Regional Director review, a Criminal Investigations Unit STG clearance completed within ten business days, and finally the personal decision of the Director of Field Operations ((GDC SOP 209.08 §§IV.C.4, IV.C.7, IV.F.5.c)). Asymmetry is also written into the appeal structure: the appeal of initial Tier II assignment goes to the Director of Field Operations, but the appeal of the 90‑day review — the decision that continues confinement — goes only to the Warden, who is the same official who approved the placement in the first instance, with no route to Central Office and a seven‑business‑day review window ((GDC SOP 209.08 §§IV.C.6, IV.F.4; GPS #19258)). Every Tier II condition is modifiable by the Warden or the Warden’s designee on a finding of undefined “extenuating operational circumstances,” with no time limit, no documentation requirement, no notice to the prisoner and no reporting obligation to Central Office; the 2025 Tier III policy by contrast requires comparable exceptions to be recorded in the Electronic Logbook and justified in writing, and requires written notice to the North Region Director whenever out‑of‑cell time is cancelled or shortened for three or more consecutive days ((GDC SOP 209.08 §IV.D.1.l; GDC SOP 209.11 §IV.E.1.n; GDC SOP 209.09 §IV.E.1.l; GPS #19276, #19282)). GDC’s discipline policy designates Tier II as the route around its own 30‑day disciplinary‑isolation cap, in a section whose stated premise is a finding of guilt, while SOP 209.08’s published summary states that the Tier II program “is not a punishment measure” ((GDC SOP 209.01 §W.2; GDC SOP 209.08 §I; GPS #19292, #19291)).

The record that would show retaliation is excluded by policy

The single most consequential structural finding for this page is that GDC policy forbids the grievance from entering the file that determines Tier II release. SOP 227.02 §IV.E.2.b provides: “A Grievance must not be placed in the Offender’s file or referred to in SCRIBE case notes.” SOP 209.08 §IV.F.5.c.1 conditions release from Tier II on the Unit Manager’s review, in SCRIBE, of the offender’s disciplinary reports, profiles, alternative entrée meal profiles, case notes, Security Threat Group history, incident reports, progress notes, movement history, mental‑health status, personal history sheet, medical profile, court production orders, and county of conviction and residence. A keyword search of SOP 209.08 for “grievance” returns nothing: the Tier II policy does not mention grievances at all and contains no anti‑retaliation provision ((GDC SOP 227.02 §IV.E.2.b; GDC SOP 209.08 §IV.F.5.c.1; GPS #19320)). The one record that would show a grievance preceded a Tier placement is, by GDC policy, absent from both the institutional file and the SCRIBE case notes — and SCRIBE case notes are among the records that determine whether a person leaves Tier II. Any sequencing test — did the grievance precede the placement — must therefore be reconstructed from a record the policy says does not contain it.

Two further features of the rulebook bear on how a person in Tier II can contest anything. SOP 209.08 as published runs its major sections A, B, C, D, D, F, G — two consecutive D sections and no E — instructs the Classification Committee to apply “the eligibility criteria cited above in VI.B.1‑14” when the criteria are in Section IV.B and there are thirteen of them, and assigns the same former code (IIB09‑0001) to two different policies; the record‑retention list omits Attachments 4 and 8, including the Phase 3 privileges handout ((GDC SOP 209.08; GPS #19295)). A person checking the rules governing indefinite segregation against the copy in the inmate library is working from a document with duplicate section letters, a broken internal cross‑reference, and a miscount of the criteria that justify their confinement. GDC nonetheless characterises Tier III as incentive‑based and states affirmatively in SOP 209.45 and SOP 209.55 that the STEP programs “are not a restrictive housing unit” — a definitional carve‑out that will exclude the step‑down population from any record production framed on a single term ((GDC SOP 209.09 §IV.A.1; GDC SOP 209.45; GDC SOP 209.55; GPS #19333)).

What Georgia does not publish

Despite SOP 227.02 requiring semi‑annual grievance reports to the Commissioner and monthly, per‑institution data retention, GDC does not publish grievance statistics, rejection rates, or dispositions on its public research portal ((Collection: The Grievance Machine, Data Gap)). This secrecy prevents the public — and incarcerated people — from knowing how many grievances are filed, how many are dismissed on procedural grounds, or what proportion raise allegations of retaliation. The same opacity extends to discipline for retaliation: SOP 227.02 prohibits retaliation for grievance filing, but no SOP specifies the disciplinary consequences staff face for violating that prohibition, leaving the ban toothless ((Collection: The Grievance Machine, Data Gap)).

The opacity extends to segregated housing. A systematic review of the GDC‑published report corpus GPS holds — 110 non‑monthly GDC reports and all 34 distribution sections of the monthly Inmate Statistical Profile series — found no published count of the Tier or restrictive‑housing population. The Annual Fiscal Reports carry restrictive‑housing language in five years (FY2014, FY2017, FY2018, FY2019, FY2020), a reduction percentage and a delta in two, and never a population; from FY2021 onward the subject disappears as a program narrative altogether. The Daily Population by Facility reports consist of eleven facility‑type rows and three columns, with no housing type, no security level and no segregation status. None of the Inmate Statistical Profile’s 34 distributions reports segregation, restrictive housing or Tier status. The PREA Annual Reports, all eleven editions 2014–2024, report allegation and substantiation counts by facility and physical‑plant improvements; the SMU appears only as a facility row. The phrase “restrictive housing” is absent from all 22 GDC productions in GPS’s open‑records archive, and “Tier II” is absent from all 261 productions ((GPS #19306, #19303)). GDC reports how many people left extended restrictive housing and what percentage that was, and never states the population before or after: FY2018 reports a 41 percent reduction (671 people) since January 2017; FY2019 reports a 40 percent reduction (675 people) since the same January 2017 baseline — statements that are not reconcilable to a single baseline, and from which GPS’s arithmetic yields roughly 1,640 and 1,690 respectively, figures that must never be cited as GDC‑published population data ((GDC Annual Fiscal Reports 2018, 2019; GPS #19298, #19299)). The one Tier population GDC does publish is a side effect: the Special Management Unit at Jackson appears as a facility row in the monthly Inmate Statistical Profile, recording 154 people as of 2026‑09‑01, ranging between 149 and 195 across 75 monthly snapshots from July 2020 to September 2026, and 5 admissions in FY2025 (0.06 percent of state‑prison admissions) — a figure that counts people admitted to GDC custody directly at the SMU, not transfers into the Tier III program ((GDC Profile of All Inmates, Aug. 2026; GDC Profile of Inmate Admissions FY2025; GPS #19304, #19305)). One facility’s Tier units alone (178 cells, Valdosta State Prison) approach the size of the entire SMU ((GDC Annual Fiscal Report 2018; GPS #19301)).

The structural reason the Tier population cannot be recovered from what GDC publishes is documented in GDC’s own rules. Every person in Tier III, Tier II STEP and Tier III STEP is classified Close Security, and the Inmate Statistical Profile’s only housing‑adjacent field (“supervision”) reports Close, Medium and Minimum — so the Tier population is contained within the published Close Security count and cannot be isolated from it, because GDC publishes security classification and not housing status ((GDC SOP 209.09 §IV.A.2; GDC SOP 209.45; GDC SOP 209.55; GPS #19307)). And no policy instrument requires anyone to compile a Tier count into a report: a review of the 209 series found no SOP naming any Tier roster, periodic report or Central Office aggregate submission. The term “Central Office” does not appear in SOP 209.08 at all. Placements flow upward as individual approvals; the Statewide Classification Committee exists in SOP 220.09 only for transgender and intersex housing decisions and has no Tier function. The reason no Tier population figure is published is not that the data does not exist — GDC’s own annual reports say a SCRIBE module tracks it — but that no policy instrument requires anyone to compile it ((GPS #19308)). There is also no document in the SOP corpus listing which facilities are designated Tier I, Tier II, Tier II STEP, Tier III and Tier III STEP sites; SOP 209.08 refers to “Level 5 facilit[ies]” without defining the term or naming them, and SOP 508.01 refers to “designated SMHTU/TIER II facilities” without naming them — the single most load‑bearing missing document for the question of Tier’s system‑wide scale ((GPS #19264)).

The DOJ’s October 2024 findings confirm what the data gaps suggest: incarcerated people “do not always report incidents because they do not expect staff to take any action in response,” and GDC failed to protect those who cooperated with the investigation from ongoing retaliation ((DOJ Findings Report, October 2024)). In April 2024 the DOJ expanded its Civil Rights of Institutionalized Persons Act investigation of Georgia prisons to include restrictive housing, disciplinary practices and special education services. The findings report issued 1 October 2024 addressed only the violence and sexual‑abuse prongs; it found that “GDC fails to control violence even in its segregated housing units and exposes incarcerated persons to an unreasonable risk of harm due to its inappropriate use of segregated housing.” The restrictive‑housing findings have not been released. A CRIPA findings report on restrictive housing would ordinarily contain population, duration and per‑facility figures obtained by compulsory process, which is why it is the most likely external source of a Tier population count; the absence of those findings is a live, dated gap rather than evidence about Georgia’s Tier population either way ((GPS #19311)). In 2024, the Georgia Senate Study Committee on prison conditions adopted recommendations focused on staffing, mental health, and physical facilities but did not adopt any grievance‑transparency or independent‑oversight recommendation, leaving the structure of the grievance machine intact ((Collection: The Grievance Machine, Finding)).

GDC already compiles the data that would provide transparency — the semi‑annual reports, the monthly counts, and since March 2019 a SCRIBE module expressly built to track people in extended restrictive housing, with a Warden’s Dashboard field for days spent in Tier II since FY2020. A legislative mandate to publish annual, facility‑level grievance statistics with breakdowns by type and outcome, and to publish a Tier placement count, would require no new data collection, only public release. In the absence of such transparency, the $50.6 million settlement ledger and the four years of published restrictive‑housing reduction figures that then stopped remain the clearest public signals of the scale of harm that the grievance system either fails to redress or actively suppresses.

Section 3 — Reform Models and Recommendations

Other states have begun to address the grievance‑accountability gap through independent oversight structures. Virginia provides the most directly relevant model: its statute mandates that grievance data be a published data stream held by a body independent of the corrections agency, with the ombudsman’s office publishing annual reports ((Collection: The Grievance Machine, citing Virginia model)). Maryland created the Office of the Correctional Ombudsman in 2024 (SB 134, Chapter 836), an independent unit with authority to review complaints, inspect facilities, and make recommendations ((Collection: The Grievance Machine, Legal Fact)). Washington’s Office of the Corrections Ombuds (OCO) sits within the Governor’s office with statutory access to DOC facilities and records, and it investigates complaints and conducts monitoring ((Collection: The Grievance Machine, Policy)). New Jersey established a Corrections Ombudsperson under the 2020 Dignity Act, with an advisory board ((Collection: The Grievance Machine, Policy)). At the federal level, the Federal Prison Oversight Act of 2024 created an independent DOJ Inspector General inspection regime tied to facility risk scores and an Ombudsman for the Bureau of Prisons ((Collection: The Grievance Machine, Legal Fact)).

Federal PLRA reform remains a goal of advocacy organizations. The ACLU and MacArthur Justice Center have called for softening § 1997e through a good‑faith/substantial‑compliance exhaustion standard, elimination of the physical‑injury bar for constitutional violations, and repeal of the attorney’s‑fee caps that starve prison‑condition litigation of experienced counsel ((Collection: The Grievance Machine, Finding)). In the Georgia context, a first practical step would be a state bill requiring GDC to publish annual, facility‑level grievance data — numbers filed, rejected, and the reasons for rejection — modeled on the Virginia structure, housed within an independent entity such as a correctional ombudsman office. That reform would not eliminate retaliation, but it would end the data blackout that currently shields the grievance machine from public scrutiny and signals to staff that the processing — or non‑processing — of complaints will never be measured. Two Georgia‑specific additions follow from the record above. First, the same mandate should reach Tier placement: GDC’s own annual reports establish that the data exists in a SCRIBE module and a Warden’s Dashboard, and GDC has already demonstrated that it will run and produce per‑facility housing count queries on request — GPS open‑records request R024615‑101925 (filed 2025‑10‑19, completed) requested current counts of inmates by security designation housed at each GDC state prison, and GDC produced per‑facility count tables. That precedent materially undercuts any burden or new‑record objection to a Tier placement count request ((GPS #19310)). Second, because SOP 227.02 makes involuntary administrative segregation non‑grievable and routes the appeal to a process that ends with the Warden, any reform of the grievance system has to address the classification‑appeal channel as well; transparency about grievance processing without an external decision‑maker on the placement itself leaves the retaliation question where it began.

Records that would test the retaliation question

Because no court has found that Georgia uses Tier placement to retaliate for protected activity, and because GPS’s internal corpus cannot establish it, the primary route to an answer is the agency’s own records ((GPS #19312)). The internal limit should be stated plainly: GPS’s case‑file schema has no Tier value. Its retaliation method field offers “segregation” and nothing finer, so Tier I, Tier II, Tier III, disciplinary isolation and administrative segregation all collapse into a single code; of the distinct alleged courses in which segregation is coded as the method, exactly one names a Tier program, and that one concerns a death on Tier II rather than a retaliatory placement onto Tier ((GPS #19313)). Roughly two‑fifths of GPS’s negative retaliation determinations rest on a keyword screen rather than substantive review, so any count derived from the corpus is a floor, not a measurement, and counts derived from it that reach a public surface must be an aggregate count only — never narrative and never a combination of facility, date and method fine enough to identify a person in custody ((GPS #19315)). A keyword search of all published GPS posts for “Tier II” returns zero results: GPS has published extensively on retaliation and on conditions without ever publishing on the Tier system ((GPS #19316)).

The primary ask is a report or query extract from SCRIBE listing, as of a date certain, every person assigned a Tier II, Tier II STEP, Tier III, Tier III STEP, administrative segregation, disciplinary isolation or protective custody housing status or profile, with facility, housing unit or wing, phase, date of initial assignment and cumulative days in that status, produced in the electronic format in which it is maintained (O.C.G.A. § 50‑18‑71(f) supports native‑format production). The Tier II SCRIBE module (live March 2019) and the Warden’s Dashboard field for “number of days an offender has spent in restrictive housing Tier II” (FY2020) should be named explicitly, because GDC’s own annual reports establish both exist. Custodian: GDC Open Records, Office of Legal Services; the record‑holding units are the Facilities Division and the Classification Unit, Offender Administration ((GPS #19309)).

Fallback asks, all existing records named in GDC policy: the Accountability Log required by SOP 209.06 §IV.N and SOP 209.09 §IV.R.5, recording “cell number, bed number, offender name, offender number, race, in or out time, assignment of offender, and destination of offender,” printed daily after 1800 hours — a same‑day, cell‑by‑cell, bed‑by‑bed census of each Tier unit, with the same requirement imposed for Tier I and Tier II by SOP 209.07 §VI.K and SOP 209.08 §IV.P.6 ((GPS #19296)). The segregation admissions‑and‑releases register required by SOP 209.06, recording name and GDC number, housing location, date and time of action, reason for admission and tentative release date, and authorizing official or committee — which yields both population and length of stay. The Restrictive Housing Rounds Log, form M40‑01‑01, required weekly at every facility by SOP 508.20 Attachment 1 (eff. 2022‑08‑02), which at facilities without mental‑health services requires entry of the offender’s name and GDC identification number, restrictive‑housing unit identifier, officer on duty, and time of entry and exit (a partial roster at facilities with mental‑health services, and the request should say so). The Sick Call Encounter Log for Restrictive Housing, form P‑03‑0005‑11, one of seventeen mandatory tracking logs required by SOP 507.04.04 at all facilities “including private and county prisons.” Division Director approvals for isolation beyond 30 days, required by Board of Corrections Rule 125‑3‑2‑.10(f) — counting these for a year gives a floor on long‑stay isolation. The quarterly over‑24‑month Tier III panel reviews, SOP 209.09 Attachment 12, plus the Commissioner’s or Assistant Commissioner’s personal approvals, which together yield the number held in Tier III beyond 24 months. The SMU Warden’s written notices to the North Region Director under SOP 209.09 §IV.E.1.l. The Electronic Logbook (NoteActive) door sheets and unit logs. The SCRIBE “Daily lock down” report named in SOP 508.01 §IV.C.5.a. And the current GDC document designating facilities as Tier I, Tier II, Tier II STEP, Tier III and Tier III STEP sites, together with the current facility security‑level assignment list ((GPS #19309, #19289, #19264)).

Two framing points govern these requests. Ask for existing records rather than analysis, because GDC’s standard objection is that compiling a count creates a new record — an objection answered by GDC’s own FY2019 and FY2020 annual reports and by the Accountability Log being printed daily under its own policy. And ask across the whole vocabulary — Tier I, Tier II, Tier III, extended restrictive housing, restrictive housing, administrative segregation, disciplinary isolation and protective custody — because GDC’s terminology drifts by year and because SOP 209.45 and SOP 209.55 state affirmatively that the STEP programs “are not a restrictive housing unit,” a definitional carve‑out that will shrink any production framed on a single term ((GPS #19309, #19333)).

For any individual case where Tier is named, the SCRIBE Tier assignment record and the Classification Committee 90‑day review forms (SOP 209.08 Attachments 5 and 6) set against that person’s grievance‑filing dates is the sequencing question — the two documents whose order would establish or defeat the causal element. In aggregate: Tier II and Tier III placements by facility and month, cross‑tabulated against grievance filings by the same individuals, and the Director of Field Operations’ Tier release decisions with stated reasons ((GPS #19314)). Note the obstacle in GDC’s own policy: SOP 219.01 §§IV.E and IV.F assert that offender records are confidential state secrets under O.C.G.A. § 42‑5‑36(c) and direct that all requests for offender‑related records be forwarded to the Office of Legal Services for review. A request framed around individual records will meet that exemption; a request framed around aggregate counts, unit rosters stripped of names, and the designation and capacity documents will not be answered by it ((GPS #19296)). The highest‑value single records request arising from this research is for actual out‑of‑cell hours and cell‑occupancy status for Tier I and Tier II by facility — the two facts that determine whether the program as run meets the Mandela Rules’ definition of solitary confinement, and the two facts GDC’s Tier II policy neither guarantees nor reports ((GPS #19331)).

Finally, the closest thing in GPS’s corpora to Tier II‑specific litigation is not a retaliation case. The Southern Center for Human Rights’ Georgia State Prison conditions class action, filed September 2021, alleged approximately 300 people in Georgia State Prison’s “Tier II” program, over 70 percent of them with serious mental illness; Acute Care Unit cells in which suicidal residents were “confined for days without clothes, hygiene items, or toilet paper, in cells smeared in the previous occupants’ feces and blood”; at least 12 suicides at Georgia State Prison between September 2019 and May 2021, nearly 30 percent of all GDC suicides in that window; and a correctional officer vacancy rate above 70 percent at filing. These are allegations in a complaint, not findings. Georgia State Prison at Reidsville closed 2022‑02‑19. No ADA or Rehabilitation Act claim specific to Georgia Tier II and serious mental illness appears anywhere in the corpora ((GPS #19328)).