SPLC's 'The Forgotten' Civil Rights Cold Case Roster Has Its Own Distinct Errors: Hilliard Brooks 1952 (Should Be 1950), Eli Brumfield 1961 (Should Be 1962), Ernest Jells 1964 (Should Be 1963), and 'Izell' Henry (Should Be Isaiah Henry) — Different From the LSU Errors But Overlapping on Brooks and Brumfield, Suggesting Both Rosters Inherit a Common Upstream Mistake on at Least Two Cases

USE CASE. Civil rights historians, journalism students, descendants, documentary filmmakers, and textbook authors routinely cite both the LSU Cold Case Project roster AND the SPLC 'The Forgotten' roster when framing the historical moment of these killings. Until this audit, no public cross-check of either roster against DOJ canonical Incident Date fields existed. The LSU audit (iter 103-110 in this same project) confirmed 19 date errors + 8 name errors in 127 entries. This iteration extends the methodology to SPLC's 63-entry 'Forgotten' roster and asks: does SPLC have its own errors? If so, do they overlap with LSU's? RESULT 1 (SPLC has at least 4 confirmed errors). (a) Hilliard Brooks: SPLC year 1952; correct 1950 (Brooks was a 22-year-old WWII veteran shot August 12, 1950 by Montgomery Police Officer M.E. Mills after refusing to re-board a city bus at the rear, dying August 13, 1950 — a documented prelude to the December 1955 Montgomery Bus Boycott). LSU also has 1952. Confirmed via DOJ /crt/case/hilliard-brooks. (b) Eli Brumfield: SPLC year 1961; correct 1962 (Brumfield was the African-American manager of a tire-recapping plant in Tylertown MS, shot by McComb police officer Britten Elmore on October 13, 1962 during a traffic stop — Elmore claimed self-defense after Brumfield allegedly jumped from his car holding a pocket-knife). LSU also has 1961. Confirmed via DOJ /crt/case/eli-brumfield and Civil Rights Digital Library. (c) Ernest Jells: SPLC year 1964; correct 1963 (Jells was a 21-year-old Black man shot by Clarksdale police on a rooftop on the night of September 20, 1963 after being accused of stealing a banana from a grocery store; family members disputed the police account because Jells was left-handed but officers testified he held the rifle on his right shoulder). LSU has 1963 correctly. Confirmed via DOJ /crt/case-document/ernest-jells-notice-close-file. (d) 'Izell Henry' / Isaiah Henry: SPLC lists 'Henry, Izell' in Greensburg LA 1954; the DOJ canonical name is Isaiah Henry, a 38-year-old African-American school-bus driver and farmer beaten July 28, 1954 in St. Helena Parish (he suffered permanent brain damage and died five years later). LSU has 'Isaiah Henry' correctly. Confirmed via DOJ /crt/case-document/isaiah-henry-notice-close-file. RESULT 2 (cross-database structure). Comparing SPLC against the five LSU multi-year errors: SPLC has the CORRECT year for Mahone (1956 vs LSU's 1967), Banks (1954 vs LSU's 1964), and Greene (1960 vs LSU's 1965). This proves that the LSU multi-year errors on Mahone, Banks, and Greene are LSU-isolated typos rather than propagated upstream errors — they were introduced inside LSU's own pipeline and went uncaught because LSU is rarely cross-referenced against any other source. Conversely, the LSU+SPLC shared errors on Brooks (1952 vs canonical 1950) and Eli Brumfield (1961 vs canonical 1962) suggest that BOTH rosters inherited those wrong years from a common upstream source — most likely an early NAACP report from the 1950s/60s era or the original Charles Evers list that catalogued cold case victims before DOJ began publishing Notice-to-Close-File memoranda in 2008. Once the DOJ memos started appearing with canonical Incident Date fields, neither LSU nor SPLC propagated the corrections back into their own rosters. RESULT 3 (third roster: Northeastern CRRJ Burnham-Nobles Digital Archive). I cross-checked the same victims against the Northeastern CRRJ Burnham-Nobles Digital Archive (https://crrjarchive.org/), which is the most rigorously curated public civil-rights cold case archive: it includes more than 20,000 pieces of primary-source evidence across 1,000 incidents from 1930-1954 (and more being added through 1970). The CRRJ Brooks page at /incidents/344 is titled 'Killing of Hilliard Brooks in Alabama in 1950' and gives the correct year. CRRJ also has Eli Brumfield at the correct October 13, 1962 date. So CRRJ matches DOJ on both shared LSU+SPLC errors. This means the upstream source of the LSU+SPLC shared 1952/1961 errors is NEITHER the DOJ canonical memos NOR the CRRJ archive — both of which are correct. The shared errors must therefore have originated in some other publication (most likely an early NAACP report, a Charles Evers / Mississippi Black Paper-era catalog from the 1960s, or an even older Tuskegee Institute lynching record) that has been propagated into LSU and SPLC and never corrected. Identifying that exact upstream source remains an open research question for a future iteration. STRUCTURAL READING. Civil rights cold case rosters published by major journalism and advocacy organizations (LSU, SPLC) drift from the DOJ-canonical and CRRJ-canonical record over time because no organization systematically cross-checks them against the DOJ memos, and the DOJ memos and CRRJ archive have both been the authoritative sources since at least 2010 (DOJ) and 2022 (CRRJ public release). The audit methodology used here (parse public roster → diff against DOJ /crt/case/ pages → verify with second primary source) is straightforward and would catch most of these errors in any of the major rosters. The fact that SPLC and LSU have BOTH correct entries that the other got wrong AND shared errors that neither has corrected is itself a useful structural finding: there is no canonical 'corrected' civil rights cold case roster currently maintained by any public-facing organization other than DOJ itself. CAVEATS. (a) SPLC's 'The Forgotten' roster is year-only — I could not check day-level or month-level errors against it, so SPLC may have additional finer-grained errors that this audit cannot detect. (b) SPLC has only 63 entries vs LSU's 127, so SPLC's coverage is narrower. (c) The 'Izell' / Isaiah Henry discrepancy could in principle be an alternate name rather than an error, but DOJ uses Isaiah and so do all other primary sources I checked. (d) For the LSU and SPLC name errors that diverge (Carie/Carrie Brumfield, Ernest/Earnest Jells, Clarence/Charles Pickett, etc.), the DOJ canonical spelling is treated as authoritative.

Two major organizations publish public lists of civil-rights-era cold case victims: the LSU Cold Case Project (a journalism program at Louisiana State University) and the Southern Poverty Law Center's 'The Forgotten' list. Both are widely cited by historians, descendants, and textbook authors. In a previous loop iteration I audited the LSU list of 127 victims against the U.S. Department of Justice Civil Rights Division case pages and found 19 date errors plus 8 name errors. This iteration does the same audit on the SPLC list. The SPLC list is shorter (63 victims) and only gives years, not full dates. I found at least 4 errors in the SPLC list. Two of them are SHARED with the LSU list: Hilliard Brooks listed in 1952 when he was actually killed in 1950, and Eli Brumfield listed in 1961 when he was actually killed in 1962. Both organizations have these wrong, which means they probably both inherited the wrong years from a common upstream source (probably an old NAACP report from the 1950s or 60s) and neither has updated to match the DOJ since the DOJ started publishing canonical 'Notice to Close File' memos around 2010. Two of the SPLC errors are unique to SPLC: Ernest Jells listed in 1964 (correct: 1963), and 'Izell Henry' (correct: Isaiah Henry — different first name). LSU has Jells and Henry correct. Conversely, SPLC has the CORRECT year for three of LSU's biggest errors: Maybelle Mahone 1956 (LSU has 1967, off by 11 years), Isadore Banks 1954 (LSU has 1964, off by 10 years), and Mattie Greene 1960 (LSU has 1965, off by 5 years). This is actually a useful finding by itself: it proves that LSU's worst errors are LSU-internal typos, not bugs inherited from any upstream source. Why this matters: if you're a researcher trying to find primary records for any of these victims, and you start from the LSU list, you'll get the wrong year for Mahone, Banks, and Greene; if you start from the SPLC list, you'll get the wrong year for Jells, Brooks, and Brumfield. There is currently no canonical public roster of these victims that doesn't have errors — the only authoritative source is the DOJ Civil Rights Division case-page sidebar, which is well-organized but not packaged as a single browsable list. A genuinely correct roster would require systematically cross-checking either of these public lists against the DOJ memos one entry at a time, which is what this two-iteration audit has done.